developed and fabricated biodiesel production...

I

APPENDIX I

DEVELOPED AND FABRICATED BIODIESEL PRODUCTION PLANT

CALCULATION OF MOLAR RATIO

Molecular weight of thumba vegetable oil = 870

Molecular weight of methanol = 32

Hence, 1 gm mole of thumba oil = 870 gm

and 1 gm mole of methanol = 32 gm

Catalyst (NaOH / KOH) = 1% by weight of oil

Amount of methanol for 1 kg of thumba vegetable oil

• For 1:6 molar ratio = (32 / 870) × 1 × 6 = 0.221 kg

• 1:4.5 molar ratio = (32 / 870) × 1 × 4.5 = 0.166 kg

II

APPENDIX II

TEST ENGINE SPECIFICATIONS

Description Specifications

Make Legion Brothers, Bangalore (India)

Type

Single Cylinder, Direct Injection, Four-Stroke,

Vertical, Water-Cooled, Naturally Aspirated

Variable Compression Ratio Multi-Fuel Diesel

Engine

Power 3 to 5 HP

Rated Speed 1450 to 1600 rev/min (Governed Speed)

Number of Cylinders One

Compression Ratio 5:1 to 20:1 (Variable Compression Ratio)

Bore 80 mm

Stroke 110 mm

Injector Pressure 203 bar

Injection Timing by Spill 23⁰ CA BTDC

Method of Loading Eddy Current Dynamometer

Method of Starting Manual Crank Start

Method of Cooling Water Cooled

Over all Dimensions 1400 x 1300 x 1100 mm

Weight 225 kg

Air Tank Size (mm) 400 x 400 x 400 mm

Orifice Size (mm) 20 mm

Valve Timings

Inlet Valve Opens 4.5⁰ CA BTDC

Inlet Valve Closes 35.5⁰ CA ABDC

Exhaust Valve Opens 35.5⁰ CA BTDC

Exhaust Valve Closes 4.5⁰ CA ATDC

III

APPENDIX III

TECHNICAL SPECIFICATIONS OF CRANK ANGLE ENCODER


Make Omron Electronics, USA

Protection Circuit

Output Reverse Polarity Protection, Power Source

Circuit Reverse Polarity protection, Surge

Suppressor, Short-Circuit Protection

Ambient Air

Temperature Operating -40° C to 70° C, Storage: -40° C to 85° C

Voltage Influence ±1% Max. of Sensing Distance in Rated Voltage

Range ±15%

Insulation Resistance 50 MΩ

Parts and Case

Dielectric Strength 1,000 VAC at 50/60 Hz for 1 min. Between Current

Carry Parts and Case

Vibration Resistance 10 to 55 Hz, 1.5 mm Double Amplitude for 2 hours

Each in X, Y and Z Directions

Shock Resistance 500 m/s2, 10 Times Each in X, Y and Z Directions

1,000 m/s2, 10 Times Each in X, Y and Z Directions

IV

APPENDIX IV

SPECIFICATIONS OF PRESSURE PICKUP AND CHARGE AMPLIFIER


Make / Type Kistler 6613CQ 09

Pressure Range 0 to100 bar

Type Piezoelectric

Cooling Air Cooled

Calibration at 200 °C 0 to 100 bar

Sensitivity (0.5 %) 25 mV/bar

Frequency range (-3 db) 0.016 to 20000 Hz

Linearity 1 % FSO

Shock 2000 g

Operating Temperature Range -50 to 300 °C

Sensitivity Shift 1%

Tolerance 5%

Thermo Shock

( 9.5 bar, 1500 1/min) 0.5 bar

Time Constant

a) For Cylinder Measurement

b) For Calibration

5 s

>2500 s

Signal Output (at 1mA load) max. 4.4 to 5 V

min. > 0 V

Signal Span 3.0 V

Zero Line 1.9 to 2.2 V

Supply Voltage 7 to 32 V DC

Supply Current 6 mA

Output Impedance 100

Mounting Torque of Sensor 15 Nm

V

APPENDIX V

SPECIFICATIONS OF FIVE GAS EXHAUST ANALYZER


Make/Modal No. Airrex HG-540, Hephzibah Co. Ltd., Korea

Application CO2, CO, O2, NOX and HC Gas Measurements

Measuring Method, Measuring Range and Resolution

CO2 NDRI, Range 0-20 %, Resolution .01%

CO NDRI, Range 0.000- 9.999 %, Resolution .001 %

O2 NDRI, Range 0- 25 % , Resolution .001%

NOX Electrochemical, Range 0-5000 ppm, Resolution 1 ppm

HC Electrochemical, Range, 0-10000 ppm, Resolution 1 ppm

Warm-up Time 5 Minutes

Response Time Less Than 10 Seconds

Sampling System Direct Sampling From Tail Pipe

Power Supply 90-260 V AC, 50/60 Hz

Weight 5 kg

Dimensions 270 x 340 x 165 mm

SPECIFICATIONS OF SMOKE METER


Make / Modal Neptune / OPAX 2000 II DX 200 P

Application Diesel Smoke Measurement

Light Source Halogen Lamp 6 V/10W/3000 °K

Smoke Equivalent Column 400 mm

Temperature of Smoke Chamber 90° C ± 10° C

Warm-up Time 10-15 minutes

Sensor Silicon Photodiode With Corrected Response

Sampling System Partial Flow Sampling System

Response Time Less Than 10 Seconds

Operating

Temperature/Humidity

2 to 45° C / < 95 %

Power Supply 220/240 V AC and 12 V DC

Weight 18 kg

Dimensions 420 x 210 x 400 mm

VI

APPENDIX VI

HIGH SPEED DATA ACQUISITION SYSTEM DETAILS

Make: Legion Brothers, Bangalore (India), Type: Class A

14 Analog Inputs (12-16 Bits Depending on Speed)

UE9-Pro Adds 24-bit Low-Speed ADC for 20-Bit Effective Resolution

±5 or 0-5 Volt Maximum Analog Input Range

2 Analog Outputs (12-Bit, 0-5 Volts)

23 Digital I/O

Up to 2 Counters (32-Bits Each)

Maximum Input Stream Rate of 50+ kHz (Depending on Resolution)

Capable of Command/Response Times as Low as 1.2 Milliseconds

USB 2.0 / 1.1 Full Speed Interface

Command/Response (Software Timed) Reads / Writes 1.2-4.0 Milliseconds

Ethernet 10 Base-T Interface

Dual-Processor Design with 168 MHz of Total Processing Power

Electrical Isolation Possible with Ethernet Interface

Enclosure Size Approximately 3" x 7" x 1" (75mm x 185mm x 30mm)

Rated for Industrial Temperature Range -40 to +85° C

Power can be Provided by the USB Cable or an External 5 volt Supply

No Power Supply Needed for USB Operation

Supports Software or Hardware Timed Acquisition

VII

APPENDIX VII

GLOBAL BIODIESEL STANDARDS AND SPECIFICATIONS

Standard /

Specification

Austria Czech

Repub. France Germany Italy Sweden USA India

ON

C1191

CSN

656507

Journal

Official

DIN E

51606

UNI

10635

SS

155436

ASTM

6751

BIS

Draft

Date Jul-97 Sep-98 Sep-97 Sep-97 Apr-96 Nov-96 Jul-03 May-05

Application --- FANE RME VOME FAME VOME VOME FAMAE FAMAE

Density 15° C G/CM3 .85-.89 .87-.89 .87-.90 .875-.90 .86-.90 .87-.90 -- .86-.90

Viscosity40° C cSt 3.5-5.0 3.5-5.0 3.5-5.0 3.5-5.0 3.5-5.0 3.5-5.0 1.9-6.0 2.5-6.0

Distillation ° C -- -- <360 -- <360 -- -- <360

Flashpoint ° C >100 >110 >100 >110 >100 >100 >130 >120

CFPP ° C 0/-15 -5 -- -10/-20 -- -5 -- --

Pour point ° C --- -- <-10 --- <-15 -- --- ---

Sulphur %Mass <0.02 <0.02 -- <0.01 <0.01 <0.001 <0.05 >0.005

CCR 100% %Mass <0.05 <0.05 -- <0.05 -- -- <0.05 >0.05

10% dist.

Resid. %Mass -- -- <0.3 -- <0.5 -- -- --

Sulphate ash %Mass<.02 <0.02 <0.02 -- <0.03 -- -- <0.02 >0.02

Ash %Mass -- -- -- -- <0.01 <0.01 -- --

Water Mg/kg -- <500 <200 <300 <700 <300 <0.05% <500

Total Contam. Mg/kg --- <24 -- <20 -- <20 -- <24

Cu-Corros.

3h/50° C

--- -- 1 -- 1 -- -- <No 3 <No 1

Cetane No. --- >49 >48 >49 >49 -- >48 47 min. 51 Min.

Neutral No. mgKOH/g <0.8 <0.5 <0.5 <0.5 <0.5 <0.6 <0.8 <0.5

Methanol %Mass <0.20 -- <0.1 <0.3 <0.2 <0.2 -- <0.2

Ester Content %Mass -- -- >96.5 -- >98 >98 -- <96.5

Monoglyceride %Mass -- -- <0.8 <0.8 <0.8 <0.8 -- <0.8

Diglyceride %Mass -- -- <0.2 <0.4 <0.2 <0.1 -- <0.2

Triglyceride %Mass -- -- <0.2 <0.4 <0.1 <0.1 -- <0.2

Free Glycerol %Mass <0.02 <0.02 <0.02 <0.02 <0.05 <0.02 <0.02 0.02

Total Glycerol %Mass <0.24 <0.24 <0.25 <0.25 -- -- <0.24 0.25

Iodine No. --- <120 -- <115 <115 -- <125 -- <115

Phosphorus Mg/kg <20 <20 <10 <10 <10 <10 <10 <10

Alkaline Met.

(Na,K) Mg/kg -- <10 <5 <5 -- <10 -- --

FAME = Fatty Acid Methyl Ester, VOME = Vegetable Oil Methyl Ester,

FAMAE = Fatty Acid Mono Alkyl Ester, RME = Rapeseed oil Methyl Ester

VIII

APPENDIX VIII

EXPECTED ACCURACY OF MEASUREMENTS

Although, the information about the type of measuring instruments used for various

measurements involved are discussed in respective chapters, the accuracies of the

measurements are also summarized in appendix as follows:

Measurements Accuracy

Load ± 0.2%

Speed ± 2 rev/min

Time ± 0.5%

Temperatures ±1⁰ C

CO ± 0.001%

CO2 ± 0.01%

NOX ± 2 ppm

HC ±1 ppm

O2 ±0.5%

Smoke Opacity ± 0.1%

Pressure ±0.2 bars

The experiments were repeated at optimum conditions several times and

repeatability was found to be satisfactory.

IX

APPENDIX IX

THE LPG GAS SUPPLY SYSTEM DEVELOPED FOR RESEARCH

WORK

LPG Gas Cylinder

Flame Trap

Weighing Machine

Test Engine Setup

Universal Journal of Environmental Research and Technology All Rights Reserved Euresian Publications © 2011 eISSN 2249 0256 Available Online at: www.environmentaljournal.org Volume1, Issue 2: 124-134

124

Mathur Y. B. et al.

Open Access Review Article

Economics, Formulation Techniques and Properties of Biodiesel: A Review

1Mathur Y. B.,

2Poonia M. P. and

3Jethoo A. S.

1Government Polytechnic College, Bikaner (India) 2 Department of Mechanical Engineering, Malaviya National Institute of Technology, Jaipur (India)

3Department of Civil Engineering, Malaviya National Institute of Technology, Jaipur (India)

Corresponding author: [email protected]

Abstract: Due to limited resources of fossil fuels and the environmental concern, there has been improved focus on

vegetable oils and biodiesel fuel as an alternative source of energy. Governments across the world are

injecting huge amount of money into the development of this sector in an attempt to reduce their

dependency on fossil fuels. However, the alternative diesel fuels must be technically and environmentally

acceptable, and economically competitive. From the viewpoint of these requirements, triglycerides and

their derivatives may be considered as viable alternatives for diesel fuels. One of the main problems of

vegetable oil use in diesel engine is their higher kinematic viscosity due to which problems occur in pumping

and atomization, ring-sticking, carbon deposits on the piston, cylinder head, ring grooves, etc. Hence,

straight vegetable oils have to be modified to bring their combustion related properties closer to diesel. This

fuel modification is mainly aimed at reducing the viscosity in order to get rid of flow/ atomization related

problems. In the present work, efforts have been made to understand and compile the outcome of

researches on economics of biodiesel fuel, issues associated with use of vegetable oil in diesel engine by

using some well known techniques available to overcome higher viscosity related problems for making them

compatible with the hydrocarbon-based diesel and biodiesel fuel properties.

Keywords: Biodiesel, Blending, Micro-Emulsion, Pyrolysis, Transesterification, Vegetable Oil

1.0 Introduction: The conventional fossil fuel resources are depleting continuously at an alarming rate resulting in increasing dependency on renewable sources of energy. Fossil fuel combustion has also degraded the environment and earth has become a subtle place for survival of mankind. Among the various renewable energy options, biodiesel from seed oil crops have a great potential for meeting the future increasing demands of petroleum and its products. Biodiesel is gaining worldwide acceptance as an environment-friendly solution to the energy and environment degradation crisis. It is an accepted option for achieving energy security, reduction in imports, rural employment, and for improving the agricultural economy (Parikh, 2005). Biodiesel results in substantial reduction of un-burnt hydrocarbons, carbon monoxide, and particulate matter (Srivastava and Prasad, 2000; Banapurmath et al., 2008; Baiju et

al., 2009; Beepanraj and Lawrence, 2011). Hence at this particular juncture, much emphasis has to be given for exploitation of sustainable, long lasting, renewable sources of energy.

1.1 Historical Background of Biodiesel: Dr. Rudolf Diesel, who invented the first Diesel Engine in 1895, used only biofuel in his engine. His visionary statement was “The use of vegetable oils

for engine fuel may seem insignificant today, but

such oils may become in course of time, as

important as petroleum and coal tar products of

the present time”. The above prediction is becoming true today as more and more biodiesel is being used all over the world (Hossain and Davies, 2010; Dileep et al., 2007). Despite the widespread use of fossil petroleum-derived diesel fuels, interest in vegetable oils as fuel for internal combustion engines was reported in several countries during the 1920s and 1930s. The use of biodiesel was recognized much later and became technically relevant only after the energy crisis in the year 1973 and afterwards. More recently, in 1977, Brazilian scientist Expedito Parente invented and submitted first industrial process for the production of biodiesel for patent (Gerhard, 2005).

1.2 Feedstock for Biodiesel Production: Biodiesel, derived from the oils and fats of plants like Soybean, Cotton, Sunflower, Jojoba, Rapeseed, Canola, Jatropha Curcas, Thumba and animal fat by transesterification process can be used as a complete substitute or an additive to diesel up to maximum possible extent (Sahoo and Das, 2009; Xuezheng et al., 2009). Owing to their availability, various oils have been in use in different countries as feedstock for biodiesel production. The

Universal Journal of Environmental Research and Technology

125

Mathur Y. B. et al.

vegetable oils used for biodiesel are mainly Rapeseed or Sunflower oil in Europe, the USA and Canada uses Soyabean, Rapeseed, other waste oils and fats; frying oil and animal fat is the chosen option in Ireland; Castor oil and Soyabean oil is used in Brazil; Coconut oil is preferred in Malaysia and Philippines; Palm oil in Thailand, Malaysia, Indonesia and the Philippines; Cotton Seed oil in Greece; Linseed and Olive oil in Spain; Jatropha and Karanja in used in India, Nicaragua and Africa to produce biodiesel (Frank et al., 2009; Altin et

al., 2001, Arjun et al., 2008). Several other non-edible plants such as Neem (Azadirachta Indica), Meswak (Salvadora Species), Mahua (Madhuca

indica), Rubber (Hevea Species), Castor (Ricinus

communis), Diploknema Butracea, Garcinia Species and Thumba (Citrullus colocynthis) can also be used for producing biofuels in India (Barnwal and Sharma, 2005; Mohibbe et al., 2005; Augustus et

al., 2003).

1.3 Biodiesel as Fuel for Diesel Engines: Biodiesel, the renewable liquid fuel produced from biological raw material is a good substitute for petroleum diesel. Biodiesel contains no petroleum, but it can be blended at any level with petroleum diesel to create a biodiesel blend. It can be used in compression ignition engines with little or no modifications. As an alternative fuel, biodiesel can provide almost same power output as of diesel (Gumus and Kasifoglu, 2010; Zafer and Mevlut, 2008; Christian et al., 2009, Pall et al., 2009). The conversion process of vegetable oil into biodiesel is quite efficient, in nearly one-to-one ratio. Among the many advantages of biodiesel fuel which includes its safe use in all conventional diesel engines, offers almost the same performance and engine durability as conventional diesel fuel, non-toxic and reduces tailpipe emissions, visible black smoke and noxious fumes and odours and higher cetane number which improve the combustion characteristics. However NOX emission of biodiesel increases because of rapid combustion and other fuel characteristics. Diesel fuel blends with biodiesel have superior lubricating properties with reduced wear and tear on the diesel engine components. The most common diesel fuel blends are B10 containing 10 percent biodiesel and B20 containing 20 percent biodiesel (Sirisomboon et al., 2007; Nag et al., 1995). Lot of research work is being done to analyse the engine combustion and performance using straight vegetable oils as well as biodiesel but fewer researchers have analysed and compiled the

economics of biodiesel, its formulation techniques and fuel related properties in recent past. In this paper, a detail survey of literature is therefore undertaken to comprehensively review the different recently published research papers on economical viability, conversion of straight vegetable oil into biodiesel and combustion related properties of biodiesel fuel for their utilization in diesel engines.

2.0 Economics of Biodiesel: Economical feasibility of biodiesel depends on the price of the crude petroleum and the cost of transporting diesel through long distances. It is certain that the cost of crude petroleum is bound to increase due to increase in its intensive demand, limited supply, strict regulations on the aromatics and sulphur contents in diesel fuels which will result in higher cost of production of diesel fuels as removal of aromatics from the distillate fractions requires capital-intensive processing equipments. The major economic factor to consider for input costs of biodiesel production is the feedstock (price of seed, seed collection and oil extraction, transport of seed and oil), which is about 75–80% of the total operating cost. Other important cost related factors are labour, methanol and catalyst for biodiesel conversion for straight vegetable oil, which must be added to the feedstock. Cost recovery will be through sale of oil cake and of glycerol (Mulugetta, 2009). The volatile oil prices due to increased demand have necessitated for continuous research and development into the biodiesel sector so as to increase the production of biodiesel of suitable quality and at reasonable price so that it can compete with diesel fuel. Between 2001 and 2006 alone, the global annual production of biodiesel grew up by 43% (Victor et al., 2010). India has rich and abundant forest resources with a wide range of plants and oilseeds. The production of these oilseeds can be stepped up many folds for producing diesel fuels. Economical feasibility of biodiesel depends on the price of the crude petroleum and the cost of transporting diesel to long distances to remote markets in India. Further, the strict regulations on the aromatics and sulphur contents in diesel fuels will result in higher cost of production of conventional diesel fuels. The reason for high biodiesel prices are the limited availability of biodiesel feedstocks. The biodiesel program in any country has a time lag between policy planning and actual implementation and hence the introduction could be gradual, gaining the maturity only after 4-5


years. This is especially applicable to India where biodiesel is proposed to be made from nonoils. Presently, the availability of these oils is very limited and the price of such oils is higher than petro-diesel. For successful launch of biodiesel availability of oil on large scale has to beenat reasonable price and abundance availability of biodiesel feedstocks. Palm oil and refined soyoil are the main options that areinternationally. The costs for biodieselfrom palm oil, soybean oil and Jestimated about Rs 45.0 /litre, Rs. 44.0/litre and Rs 39.0 /litre, respectively (Demirbas, 2009)sources in India, estimate current biodiesel finished production costs at anywhere betwee32.0 to 45.0 per litre (Joseph, 2006; The Ministry of Petroleum and Natural Gasannounced the biodiesel purchase pOctober 2005. The Policy provided for the purchase of biodiesel at 20 specified purchase centres in 12 different states in India /litre (inclusive of taxes/duties) from January 2006. This price was subject to review every six months. The last price revision of biodiesel was done in December 2006, when it was fixed at Rs 26.5(Altenburg et al., 2008). The Indian government also announced, on 23rd Decattractive incentives to encourage biofuels plantation in wastelands and to utilise indigenous bio-mass feed stocks for production of bioaddresses the issues across the entire value cfrom plantations, and processing to marketing of biofuels. India’s new policy on biofuelsblending at least 20 percent biofuels in diesel and petrol by 2017. This implies that 13.38 million tons of biodiesel will be required. Detail economic analysis of vegetable oil based biofuels iwas done by Dorado et al. (2006). They identified that the price of the feedstock was one of the most significant factors. Also, glycerol was found to be a valuable by-product that could reduce the final manufacturing costs of the process up to 6.5% depending on the raw feedstock used.other researchers (Prueksakorn et al.

al., 2006; Zhou and Thomson, 2009Govindasamy et al., 2009; Wu et al.,

et al., 2009; Comporn and ShabbirHuanguang et al., 2010; Foidl et al.,

al., 1999) stated the possibilities of development of energy crops and biodiesel in Europe, Mexico, Japan, Thailand, China, and India.


126

Mathur Y. B. et al.

years. This is especially applicable to India where biodiesel is proposed to be made from non-edible oils. Presently, the availability of these oils is very limited and the price of such oils is higher than

aunch of biodiesel availability of oil on large scale has to been assured at reasonable price and abundance availability of

Palm oil and refined soybean options that are traded

internationally. The costs for biodiesel production atropha oil are

Rs 45.0 /litre, Rs. 44.0/litre and Rs Demirbas, 2009). Industry

sources in India, estimate current biodiesel finished production costs at anywhere between Rs

; Singh, 2009).

The Ministry of Petroleum and Natural Gas announced the biodiesel purchase policy in October 2005. The Policy provided for the purchase of biodiesel at 20 specified purchase

in India at Rs 25.0 /litre (inclusive of taxes/duties) from January 2006. This price was subject to review every six months. The last price revision of biodiesel was done in December 2006, when it was fixed at Rs 26.50/litre

The Indian government December 2009,

attractive incentives to encourage biofuels plantation in wastelands and to utilise indigenous

ed stocks for production of biofuels. It addresses the issues across the entire value chain from plantations, and processing to marketing of

policy on biofuels targets els in diesel and

. This implies that 13.38 million tons . Detail economic

analysis of vegetable oil based biofuels in Spain . They identified

that the price of the feedstock was one of the significant factors. Also, glycerol was found

product that could reduce the final manufacturing costs of the process up to 6.5% depending on the raw feedstock used. Some

et al., 2010; Hill et

, 2009; Ben, 2009; et al., 2009; Naoko

and Shabbir, 2009; et al., 1996; Bona et

stated the possibilities of development nd biodiesel in Europe, Mexico,

3. Fuel Formulation TechniquesOne of the main problems of vegetable oil use in diesel engine is their higher kinematic viscosity because of heavier triglycerides and phospholipids, due to which problems occur in pumping and atomization, ring-sticking, carbon deposits on the piston, cylinder head, ring grooves, etc. Straight vegetable oils are not suitable as fuels for diesel engines; since they have to be modified to bring their combustion related properties closer to mineral diesel. This fuel modification is mainly aimed at reducing the viscosity in order to get rid of flow/ atomization related problems. Heating/ pyrolysis, dilution/ blending, microand transesterification are some well known techniques available to overcome higher viscosity related issues associated with use of vegetable oil in diesel engine and to make them compatible with the hydrocarbon-based diesel fuels2007; Bajpai and Tyagi, 2006 The petro-diesel molecules are saturated and nonbranched molecules with carbon atoms ranging between 12 and 19 whereas, usually triglycerides generally with a number of branched chains of different lengths andmixture of organic compounds ranging from simple straight chain compound to complex structure of proteins and fattypical molecular structure of vegetablemolecule is shown in Figure represent hydrocarbon chain of fatty acids.chemical composition of different vegetable oils is presented in Table 1.

Figure 1: Molecular Structure of a Typical

Vegetable Oil M

Fuel Formulation Techniques: One of the main problems of vegetable oil use in diesel engine is their higher kinematic viscosity because of heavier triglycerides and phospholipids,

to which problems occur in pumping and sticking, carbon deposits on the

piston, cylinder head, ring grooves, etc. Straight vegetable oils are not suitable as fuels for diesel engines; since they have to be modified to bring

on related properties closer to mineral diesel. This fuel modification is mainly aimed at reducing the viscosity in order to get rid of flow/ atomization related problems. Heating/ pyrolysis, dilution/ blending, micro-emulsification,

n are some well known techniques available to overcome higher viscosity related issues associated with use of vegetable oil in diesel engine and to make them compatible

based diesel fuels (Agarwal, 2007; Bajpai and Tyagi, 2006; Sharma, 2009).

diesel molecules are saturated and non-branched molecules with carbon atoms ranging between 12 and 19 whereas, Vegetable oils are usually triglycerides generally with a number of branched chains of different lengths and are the

of organic compounds ranging from straight chain compound to complex

oteins and fat-soluble vitamins. The structure of vegetable oil

igure 1, where R1, R2 and R3 represent hydrocarbon chain of fatty acids. The chemical composition of different vegetable oils is

Molecular Structure of a Typical

Molecule


127

Mathur Y. B. et al.

Table 1: Fatty acid Composition of Some Vegetable Oils

Vegetable

Oils

Fatty Acids Composition (%)

14:0 16:0 18:0 18:1 18:2 18:3 20:0 22:0 22:1 24:0

Jatropha - 12-17 6.70 37-63 19-41 - - - - -

Karanja - 3.7-7.9 2.4-8.9 44.5-71.3 10.8-18.3 - - - - 1.1-3.5

Rapeseed - 3 1 64 22 8 - - - -

Neem 0.2-0.26 13.6-16.2 14.4-24.1 49.1-61.9 2.3-15.8 - 0.8-3.4 - - -

Sunflower - 9 2 12 78 - - - -

Soyabean - 6-10 2-5 20-30 50-60 5-11 - - - -

Corn 1-2 8-12 2-5 19-49 34-62 Traces Traces - - -

Peanut - 11 2 48 32 1 1 2 - 1

Mahua 16-28.2 20-25.1 41-51 8.9-13.7 - 0-3.3 - - -

Rice Bran 0.4-0.6 11.7-16.7 1.7-2.5 39.2-43.7 26.5-35.1 - 0.4-0.6 - - 0.4-0.9

Palm 1.5 43 5 40 10 - 0.5 - - -

Castor - - 2-3 3-5 3-5 80-90 - - - -

Olive - 9-10 2-3 73-84 10-12 Traces - - - -

Tallow 3-6 24-32 24-31 37-43 2-3 - - - -

3.1 Heating/ Pyrolysis:

Heating/ Pyrolysis is the process in which high molecular weight compound breaks in to smaller compounds by means of heat with or without catalyst. Pyrolysis refers to a chemical change caused by the application of heat energy in the absence of air or oxygen. The liquid fractions of the thermally decomposed vegetable oils are likely to get converted into liquid oils. Many investigators have studied the Pyrolysis of triglycerides to obtain products suitable for diesel engines (Babu, 2008; Bridgwater, 2003). The Pyrolyzate oils have almost same viscosity, flash point, and pour point compared to diesel fuel with equivalent calorific value. The cetane number of the Pyrolyzate oil has been found to be lower. The Pyrolyzate oils from vegetable oils contain acceptable sulphur content, water and sediment and give acceptable copper corrosion values but unacceptable ash and carbon residue. Mechanisms for the thermal decomposition of triglycerides are likely to be complex because of many structures and multiplicity of possible reactions of mixed triglycerides (Manurung et al., 2009). 3.2 Dilution/ Blending:

High viscosity fuels like vegetable oils can be mixed with low viscosity fuel like diesel to overcome overall viscosity. These blends can then be used as diesel engine fuels. Dilution of vegetable oils can be accomplished with a solvent, methanol or ethanol. Vegetable oil can be directly mixed with diesel and may be used to run diesel engines. Blending of vegetable oil with diesel have been experimented successfully by many researchers.

The dilution of Sunflower oil with diesel fuels in the ratio of 1:3 by volume has been studied and engine tests were carried out by Ziejewski et al. (1983). They concluded that the blend could not be recommended for long term use in the direct injection diesel engines due to density difference. Pryor et al. (1983) had conducted the short term and long term performance tests with blends of vegetable oil with diesel. In short term performance test, crude-degummed Soybean oil and Soybean ethyl ester were found suitable substitutes for diesel fuel. A longer term evaluation of the engine when using 100% crude soybean oil was prematurely terminated severe injector chock led to decreases in power output and thermal efficiency. Parmanik (2003) studied the properties and use of Jatropha oil and diesel fuel blends in compression ignition engine. The heating value of the vegetable oil was comparable with ordinary diesel fuel and cetane number was slightly lower than diesel oil. He also studied the effect of blending Jatropha oil with diesel fuel in compression ignition engines. Significant improvement in engine performance was observed as compared to pure vegetable oil. The exhaust gas temperature was reduced due to reduced viscosity of the vegetable oil diesel blends. It was found that the fuel consumption was increase with a higher proportion of the Jatropha oil in the blends. Acceptable thermal efficiencies of the engine were obtained with blends containing up to 50% (by volume) of Jatropha oil.


The tests were conducted by Forson on a single-cylinder direct-injection eoperated on diesel fuel, Jatropha oil and blendes of diesel and Jatropha oil in proportions of 97.4%/ 2.6%; 80%/ 20%; and 50%/ 50% by volumetest results showed that Jatropha oil can be conveniently used as a diesel substitute in a diesel engine. The test results further showed increases in brake thermal efficiency, brake power and reduction of specific fuel consumption for Joil and it blends with diesel. 3.3 Micro-Emulsification:

A micro-emulsion is a system consists of a liquid dispersed, with or without an emulsifier, in an immiscible liquid, usually in droplets smaller then colloidal size. Micro-emulsions are isotropic, clear, or translucent thermodynamically stable dispersions of oil, water, surfactant, and often a small amphiphilic molecule, called coThe droplet diameters in micro-emulsions range from 100 to 1000 Å. A micro-emulsion can be made of vegetable oils with an edispersant (co-solvent), or of vegetable oils with an alcohol and a surfactant and a cetane improver, with or without diesel fuels. Microbecause of their alcohol content have lower volumetric heating values than diesel fuels, but the alcohols have high latent heat of vaporization and tend to cool the combustion chamber, which would reduce nozzle choking. A micromethanol with vegetable oils can perform nearly as well as diesel fuels. Ziejewski showed that the engine performance were same for a micro-emulsion of 100 % sunflower oil and the 25 % blend of sunflower oil in diesel. 3.4 Transesterification Process:

The major problems associated with the straight vegetable oil operation in diesel engines are due to their high fuel viscosity and poor volatility. Transesterification of vegetable oils provides a significant reduction in viscosity, thereby enhancing the physical and chemical properties of vegetable oil to improve the engine performance (Tapanes et al., 2008)ethyl esters of vegetable oils (called as biodiesel) revealed fuel properties similar to diesel. transesterification process involves reactingtriglycerides, present in vegetable oils with alcohols such as methanol or ethanol in the presence of a catalyst (usually sodium hydroxide) at about 70°C to give the ester and the by


128

Mathur Y. B. et al.

The tests were conducted by Forson et al.(2004) injection engine oil and blendes of

atropha oil in proportions of 97.4%/ 2.6%; 80%/ 20%; and 50%/ 50% by volume. The

atropha oil can be conveniently used as a diesel substitute in a diesel engine. The test results further showed increases in brake thermal efficiency, brake power and

specific fuel consumption for Jatropha

emulsion is a system consists of a liquid dispersed, with or without an emulsifier, in an immiscible liquid, usually in droplets smaller then

emulsions are isotropic, clear, ucent thermodynamically stable

dispersions of oil, water, surfactant, and often a small amphiphilic molecule, called co-surfactant.

emulsions range emulsion can be

made of vegetable oils with an ester and solvent), or of vegetable oils with an

alcohol and a surfactant and a cetane improver, with or without diesel fuels. Micro-emulsions, because of their alcohol content have lower volumetric heating values than diesel fuels, but the

lcohols have high latent heat of vaporization and tend to cool the combustion chamber, which would reduce nozzle choking. A micro-emulsion of methanol with vegetable oils can perform nearly as well as diesel fuels. Ziejewski et al. (1984)

engine performance were same emulsion of 100 % sunflower oil and

the 25 % blend of sunflower oil in diesel.

associated with the straight vegetable oil operation in diesel engines are due

heir high fuel viscosity and poor volatility. Transesterification of vegetable oils provides a significant reduction in viscosity, thereby enhancing the physical and chemical properties of vegetable oil to improve the engine

2008). Methyl and ethyl esters of vegetable oils (called as biodiesel) revealed fuel properties similar to diesel. The transesterification process involves reacting

vegetable oils with alcohols such as methanol or ethanol in the presence of a catalyst (usually sodium hydroxide) at about 70°C to give the ester and the by

product, glycerine and waterreported that the methyl and ethyl esters of vegetable oil can result in superior performance than neat vegetable oils. Biodiesel fuels naturally contain oxygen, which must be stabilized to avoid storage problems (Jon, 2005;2009; Vivek and Gupta, 2004; The transesterification reaction is represented by the general equation

The base catalysed production of biodiesel,

generally occur using the following steps:

The first step is the conversion of triglycerides to diglycerides, followed by the conversion of diglycerides to monoglycerides and of to glycerol, yielding one methyl ester molecule from each glyceride at each step.

glycerine and water. It has been reported that the methyl and ethyl esters of vegetable oil can result in superior performance

iodiesel fuels naturally contain oxygen, which must be stabilized to avoid

2005; Patil and Deng, 2009; Vivek and Gupta, 2004; Jha et al., 2007). The transesterification reaction is represented by

production of biodiesel,

generally occur using the following steps:

The first step is the conversion of triglycerides to diglycerides, followed by the conversion of diglycerides to monoglycerides and of diglycerides to glycerol, yielding one methyl ester molecule from each glyceride at each step.


Figure 2: Transesterification Process for Biodiesel Production

Meher et al. (2006) studied the effects of catalyst concentration, alcohol/oil molar ratio, temperature, and rate of mitransesterification of Karanja oil with methanol. They found that the optimum reaction for methanolysis of Karanja oil was 1% KOH as catalyst, MeOH/oil molar ratio 6:1, reaction temperature 65°C and rate of mixing was 360 rpm for a period of three hrs. The yield of methyl esters was found to be higher than 85% in minutes and reaction was almost complete in two hours with a yield of 97 %. With 12:1 molar ratio ofNaOH/oil or higher, the reaction was completed within one hr. The reaction was incomplete with a low rate of stirring (180 rpm). Further in optimization study, Meher et al. (2006)the yield of methyl ester from Karanja oil under the optimal condition was 97-98%. Rathore and Madran (2008) studied the kinetics of transesterification of Karanja oil into its alkyl esters in supercritical methanol and ethanol without using any catalyst. The effect of molar ratireaction temperature on alkyl esters formation was investigated. A micro-emulsion of methanol with vegetable oils can perform nearly as well as diesel fuels. Darnoko and Cheryan (2000)experimental data on Palm oil kinetics. It was observed that the rate of alkali-catalyzed (KOH) transesterification in a batch reactor increased with temperature up to 60°C. Further increase in temperatures did not reduce the time to reach the maximum conversion.


129

Mathur Y. B. et al.

: Transesterification Process for Biodiesel Production

studied the effects of catalyst concentration, alcohol/oil molar ratio, temperature, and rate of mixing on

aranja oil with methanol. They found that the optimum reaction conditions

aranja oil was 1% KOH as lyst, MeOH/oil molar ratio 6:1, reaction

temperature 65°C and rate of mixing was 360 . The yield of methyl

esters was found to be higher than 85% in fifteen minutes and reaction was almost complete in two

97 %. With 12:1 molar ratio of NaOH/oil or higher, the reaction was completed

. The reaction was incomplete with a low rate of stirring (180 rpm). Further in an

(2006) found that aranja oil under

98%.

studied the kinetics of transesterification of Karanja oil into its alkyl esters in supercritical methanol and ethanol without using any catalyst. The effect of molar ratio and reaction temperature on alkyl esters formation

emulsion of methanol with vegetable oils can perform nearly as well as

(2000) reported experimental data on Palm oil kinetics. It was

catalyzed (KOH) transesterification in a batch reactor increased with temperature up to 60°C. Further increase in temperatures did not reduce the time to reach the

The free fatty acid and moisture content in the starting material are the key parameters for determining the viability of the vegetable oil transesterification process. According to Freedman et al. (1984) the free fatty acid lower than 1% to carry out the alkali reaction. In their study, they have observed that ifthe acid value was greater than 1, more NaOH was required to neutralize the free fatty acids. Water also caused soap formation, which consumed the catalyst and reduced catalyst efficiency. The resulting soaps caused an increase in viscosity, formation of gels and resulted intoof glycerol quite difficult. Ma the effects of free fatty acids and water content in transesterification of beef tallow. The presence of water had more negative effect on transesterification than free fatty acids. They concluded that for best results, the water content and free fatty acids content be kept below 0.06 % w/w and 0.5 % w/w respectively. Zullaikah et al. (2005) had subiodiesel from Rice Bran oil with high free fatty acids content. A twomethanolysis process was employed for the efficient conversion of Rice Bran methyl esters. Hawash et al.

transesterification of Jsupercritical methanol in the absence of catalyst under different temperature conditions. Ramadhas et al. (2004) reported the use of acid catalyst followed by alkali catalyst in a single process using Rubber seed oil with high free fatty

The free fatty acid and moisture content in the starting material are the key parameters for determining the viability of the vegetable oil transesterification process. According to Freedman

the free fatty acid quantity should be % to carry out the alkali catalysed

reaction. In their study, they have observed that if the acid value was greater than 1, more NaOH was required to neutralize the free fatty acids. Water also caused soap formation, which consumed the catalyst and reduced catalyst efficiency. The resulting soaps caused an increase in viscosity,

resulted into the separation Ma et al. (1999) studied

of free fatty acids and water content in transesterification of beef tallow. The presence of water had more negative effect on transesterification than free fatty acids. They concluded that for best results, the water content and free fatty acids content in beef tallow should be kept below 0.06 % w/w and 0.5 % w/w

had successfully made ran oil with high free fatty

acids content. A two-step acid-catalyzed methanolysis process was employed for the

Rice Bran oil into fatty acid et al. (2009) studied the

transesterification of Jatropha oil using supercritical methanol in the absence of catalyst under different temperature conditions.

reported the use of acid catalyst followed by alkali catalyst in a single

er seed oil with high free fatty


130

Mathur Y. B. et al.

acids content. The objective of this study was to develop a process for producing biodiesel from a low-cost feedstock like crude Rubber seed oil. Iso et al. (2001) have studied the transesterification by immobilized lipase in non- aqueous conditions. Noureddini et al. (2005) have investigated the

biodiesel production by lipase catalyst. The time taken to get the 67 % yield of biodiesel was 72 hours at room temperature. However, the energy input was zero. The reaction time and the cost of lipase were hurdles to commercialize lipase processes.

Table 2: Fuel Related Properties of Biodiesel Produced From Different Vegetable Oils

Biodiesel Kinematic

viscosity

(cSt, 40°C)

Density

(kg/m3)

Heating

Value

(MJ/kg)

Cloud

point

(°C)

Pour

point

(°C)

Flash

point

(°C)

Cetane

Number

Carbon

Residue

(w/w)

Jatropha 5.65 879 38.5 13 - 175 50 - Karanja 6.87 897 37.9 - -1 187 49 0.05 Rapeseed 7.2 883 37.37 - -12 - 51 - Neem 15 882 38.5 - - 180 47 - Sunflower 4.6 868 40.58 1 - 183 45-52 - Soyabean 4.5 872 39.76 1 -7 178 37-45 1.7 Coconut 3.36 866 36.1 - -4 122 56 0.03 Peanut 4.9 883 - 5 - 176 54 - Palm 5.7 880 - 13 - 164 62 - Babassu 3.6 879 - 4 - 127 63 - Thumba 3.83-5.86 883 39.37 .5 - 142 53 - Tallow - - - 12 9 96 -- -

4. Properties of Biodiesel: Biodiesel is a better fuel than diesel fuel in terms of sulphur content, flash point, and aromatic content. The fuel characteristics of biodiesel are close to diesel fuels, and therefore biodiesel becomes a strong alternative to replace the diesel fuels (Demirbas, 2009). The conversion of triglycerides into methyl or ethyl esters through the transesterifcation process reduces the molecular weight to one-third that of the triglyceride reduces the viscosity by a factor of about eight and increases the volatility marginally. Biodiesel has viscosity close to diesel fuels. These esters contain 10 to 11% oxygen by weight, which may improve combustion as compared to hydrocarbon based diesel fuels in an engine (Ejaz and Younis, 2008). The cetane number of biodiesel is around 50 which are higher than diesel. The use of tertiary fatty amines and amides can be effective in enhancing the ignition quality of the diesel fuel without having any negative effect on its cold flow properties. Since the volatility increases marginally, the starting problem persists in cold conditions. Biodiesel has lower volumetric heating values (about 12%) than diesel fuels but has a high cetane number and flash point (Sylvain et al., 2009; Achten et al., 2008; Houfang et al., 2009) Some of the desirable fuel properties of biodiesel derived from different vegetable oils are presented in Table 2.5.

5.0 Conclusions: Many energy fuels are being recently investigated as potential substitutes for the current high-pollutant diesel fuel derived from diminishing commercial sources. In the past 15 years, biodiesel has progressed from the research stage to a full-production scale in many developed countries. Vegetable oil either from seasonal plant crops or from perennial forest trees origin, after being formulated, have been found suitable for utilization in diesel engines. Many traditional seed oils like Pongamia Glabra, Jatropha (Jatropha Curcas), Mallous Philippines, Garcinia Indica, Thumba, Karanja and Madhuca Indica etc. are available in the country, which can be exploited for biodiesel production purposes. Although biodiesel offers several advantages, including technical, environmental, and socio-economic, it has some disadvantages. Major disadvantages of biodiesel are higher viscosity, lower energy content, higher cloud point and pour point, marginal increase in nitrogen oxides (NOX) emissions, lower engine speed and power, engine oil degradation, injector coking, engine compatibility, high price of biodiesel and higher engine wear. Important operating disadvantages of biodiesel in comparison with petro-diesel are cold start problems, the lower energy content, higher copper strip corrosion and fuel pumping difficulty. Some of the salient conclusions are drawn on the basis of available literature for economics of bio-diesel,


131

Mathur Y. B. et al.

its formulation techniques and combustion related properties: 1. The costs for biodiesel production from palm oil,

soybean oil and Jatropha oil are estimated about Rs 45.0 /litre, Rs. 44.0/litre and Rs 39.0 /litre, respectively.

2. The major economic factor to consider for input costs of biodiesel production is the feedstock (price of seed, seed collection and oil extraction, transport of seed and oil), which is about 75-80% of the total operating cost.

3. Other important cost related factors are labor, methanol and catalyst for biodiesel conversion for straight vegetable oil, which must be added to the feedstock. Cost recovery will be through sale of oil, cake and of glycerol.

4. Mechanisms for the thermal decomposition (Pyrolysis) of for biodiesel production are likely to be complex because of many structures and multiplicity of possible reactions of mixed triglycerides.

5. Direct blending of vegetable oil with diesel could not be recommended for long term use in the direct injection diesel engines due to density difference.

6. A micro-emulsion of methanol with vegetable oils can perform nearly as well as diesel fuels.

7. Transesterification of vegetable oils provides a significant reduction in viscosity, thereby enhancing the physical and chemical properties of vegetable oil to improve the engine performance.

8. The fuel properties of biodiesel are close to diesel fuels, and therefore biodiesel becomes a strong alternative to replace the diesel fuel.

References: 1. Achten, W. M. J., Verchot, L., Franken, Y. J.,

Mathijs, E., Singh, V. P., Aerts, R. and Muys, B.(2008): Review: Jatropha Biodiesel Production and Use. J. Biomass and Bioenergy, 32: 1063-1084.

2. Agarwal, Avinash (2007): Biofuels (Alcohols and Biodiesel): Applications as Fuels for Internal Combustion Engines. Progress in Energy and Comb. Sci., 33: 233–271.

3. Altenburg, Tilman, Hildegard, Dietz, Matthias, Hahl, Nikos, Nikolidakis, Christina, Rosendahl and Kathrin, Seelige (2008): Biodiesel Policies for Rural Development in India: A Report. D.I.A, German Development Institute, Bonn, 27th May, 2008. 15-48.

4. Altin, R., Cetinkaya, S. and Yucesu, H. S. (2001): The Potential of Using Vegetable Oil Fuels as Fuel for Diesel Engines. J. Energy Con. Manag., 42: 529–538.

5. Arjun, B., Chhetri, Martin S., Tango, Suzanne M., Budge, K, Chris, Watts and Rafiqul, Islam M. (2008): Non-Edible Plant Oils as New Sources for Biodiesel Production. Int. J. Molecular Sci., 9:169-180.

6. Augustus, G. D. P. S., Jayabalan, M. and Seiler, G. J. (2003): Alternative Energy Sources from Plants of Western Ghats (Tamil Nadu, India). J. Biomass and Bioenergy, 24: 437-444.

7. Babu, B. V. (2008): Biomass Pyrolysis: a State-of-the-Art Review. J. Biofuels and Bioproducts, 2:393–414.

8. Baiju, B., Naik, M. K. and Das, L. M. (2009): A Comparative Evaluation of Compression Ignition Engine Characteristics Using Methyl and Ethyl Esters of Karanja Oil. J. Ren. Energy, 34; 1616-1621.

9. Bajpai, D. and Tyagi, V. K. (2006): Bio diesel: Source, Production, Composition, Properties and its Benefits. J. Oleo Sci., 55: 437-502.

10. Banapurmath, N. R., Tiwari, P. G. and Hosmath, R. S. (2008): Performance and Emission Characteristics of a DI Compression Ignition Engine Operated on Honge, Jatropha and Sesame Oil Methyl Esters. J. Ren. Energy, 33:1982-1988.

11. Barnwal, B. K. and Sharma, M. P. (2005): Prospects of Biodiesel Production from Vegetable Oils in India. J. Ren. and Sust. Energy Rev., 9: 363–378.

12. Beepanraj, B. and Lawrence, P. (2011): Performance and Emission Characteristics of a Diesel Engine Fuelled With Palm Oil Methyel Ester and Its Blends. J. Tech. Word, III:197-203

13. Ben, Phalan (2009): The Social and Environmental Impacts of Biofuels in Asia: An Overview. J. App. Energy, 86: S21–S29.

14. Bona, S., Mosca, G. and Vamerali, T. (1999): Oil Crops for Biodiesel Production in Italy. J. Ren. Energy, 16:1053-1056.

15. Bridgwater, A.V. (2003): Renewable Fuels and Chemicals by Thermal Processing of Biomass. J. Chemical Engg., 91: 87–102.

16. Christian, Rodriguez Coronado, Joao, Andrade de Carvalho Jr., Juliana, Tiyoko Yoshioka and Jose, Luz Silveira (2009): Review, Determination of Ecological Efficiency in Internal Combustion Engines: The Use of Biodiesel. J. App. Thermal Engg., 29: 1887–1892.

17. Comporn, Pleanjai and Shabbir, Gheewala H. (2009): Full Chain Energy Analysis of Biodiesel Production from Palm Oil in Thailand. J. App. Energy, 86: S209–S214.

18. Darnoko, D. and Cheryan, M. (2000): Kinetics of Palm Oil Transesterification in a Batch


132

Mathur Y. B. et al.

Reactor. J. American Oil and Chemist Soci., 77: 1263-1267.

19. Demirbas, Ayhan (2009): Progress and Recent Trends in Biodiesel Fuels. J. Energy Con. and Manag., 50, 2009:14–34.

20. Demirbas, Ayhan (2009): Political, Economic and Environmental Impacts of Bio fuels: A Review. J. App. Energy, 86: S108–S117.

21. Dileep, K., Birur, K., Thomas, Hertel W. and Wallace, Tyner E. (2007): Impact of Biofuel Production on World Agricultural Markets: A Computable General Equilibrium Analysis. Tenth Annual Conference on Global Economic Analysis, Purdue University, West Lafayette, USA, 7th -9th June, 2007.

22. Dorado, M. P., Cruz, F., Palomar, J. M. and Lopez, F. J. (2006): Data bank: An Approach to the Economics of Two Vegetable Oil-Based Biofuels in Spain. J. Ren. Energy, 31, 2006:1231–1237.

23. Ejaz, Shahid M. and Younis, Jamal (2008): A review of Biodiesel as Vehicular Fuel. J. Ren. and Sust. Energy Rev., 12: 2484–249.

24. Foidl, N., Foidl, G., Sanchez, M., Mittelbach, M. and Hackel, S. (1996): Jatropha Curcas L. as a Source for the Production of Biofuel in Nicaragua. J. Bioresource Tech., 58: 77-82.

25. Frank, Rosillo-Calle, Luc, Pelkmans and Arnaldo, Walter (2009): A Global Overview of Vegetable Oils, with Reference to Biodiesel. A Report for the IEA Bioenergy, U.S.A, 2009. 12-22.

26. Forson, F. K., Oduro, E. K. and Donkoh, Hammond E. (2004): Performance of Jatropha Oil Blends in a Diesel Engine. J. Ren. Energy, 29:1135-1145.

27. Freedman, B., Pryde, E. H. and Mounts, T. L. (1984): Variables Affecting the Yield of Fatty Esters from Transesterified Vegetable Oils. J. American Oil and Chemist Soci., 61:1638-1643.

28. Gerhard, Knothe, Jon, Gerpen Van and Jurgen, Krahl (2005): The Biodiesel Handbook. AOCS Press, Champaign, Illinois, United States of America. 1-218.

29. Govindasamy, Agoramoorthy, Minna, Hsu J., Chaudhary, Sunita and Po-Chuen Shieh (2009): Can Biofuel Crops Alleviate Tribal Poverty in India’s Dry lands?. J. App. Energy, 86:S118–S124.

30. Gumus, M. and Kasifoglu, S. (2010): Performance and Emission Evaluation of a Compression Ignition Engine Using a Biodiesel (Apricot Seed Kernel Oil Methyl Ester) and its Blends with Diesel Fuel. J. Biomass and Bioenergy, 34: 134–139.

31. Hawash, S., Kamal, N., Zaher, F., Kenawi, O. and Diwani G. El. (2009): Biodiesel Fuel from Jatropha Oil Via Non-Catalytic Supercritical Methanol Transesterification. Fuel, 88: 579-582.

32. Hill, J., Nelson, E., Tilman, D., Polasky, S. and Tiffany, D. (2006): Environmental, Economic and Energetic Costs and Benefits of Biodiesel and Ethanol Blends. Proce. Natural and Academic Sci., 103:11206-11210.

33. Hossain, A. K. and Davies, P. A. (2010): Plant Oils as Fuels for Compression Ignition Engines: A Technical Review and Life-Cycle Analysis. J. Ren. Energy, 35: 1–13.

34. Houfang, Lu, Yingying, Liu, Hui, Zhou, Ying, Yang, Mingyan Chen and Bin, Liang (2009): Production of Biodiesel from Jatropha Curcas Oil. J. Computers and Chemical Engineering, 33:1091–1096.

35. Huanguang, Qiu, Jikun, Huang, Jun, Yang, Scott, Rozelle, Yuhua, Zhang, Yahui, Zhang and Yanli, Zhang (2010): Bio ethanol Development in China and the Potential Impacts on its Agricultural Economy. J. App. Energy, Volume 87:76–83.

36. Iso, M., Chen, B., Eguchi, M., Kudo, T. and Shrestha (2001): Production of Biodiesel Fuel by Triglycerides and Alcohol Using Immobilized Lipase. J. Molecular Catalysis and Enzymatic, 53:53-58.

37. Jha, M. K., Gupta, A. K. and Kumar, Vipin (2007): Kinetics of Transesterification on Jatropha Curcas Oil to Biodiesel Fuel. Proceedings of the Word Congress on “Engineering and Computer Science”, Sen Francisco, U.S.A., 24th -26th October, 2007.

38. Jon, Van Gerpen (2005): Biodiesel Processing and Production. Fuel Processing Tech., 86: 1097– 1107.

39. Joseph, Gonsalves B. (2006): An Assessment of the Biofuels Industry in India. United Nations Conference on “Trade and Development” Geneva, 18 October 2006.

40. Ma, F. and Hanna, M. A. (1999): Biodiesd Prediction: A Review. J. Bioresourse Tech., 70: 1-15.

41. Ma, F., Clements, L. D. and Hanna, M. A. (1998): The Effect of Catalyst, Free Fatty Acid, and Water on Transesterification of Beef Tallow.Tran. ASAE, 41:1261-1264.

42. Manurung, R., Wever, D. A. Z., Wildschut, J., Venderbosch, R. H., Hidayat, H., Van Dam, E. G., Leijenhorst, E. J., Broekhuis, A. A. and Heeres, H. J. (2009): Vaporization of Jatropha Curcas L. Plant Parts: Nut Shell Conversion to Fast Pyrolysis Oil. J. Food and Bioproducts Processing, 87:187–196.


133

Mathur Y. B. et al.

43. Meher, L. C., Sagar, D. V. and Naik, S. N. (2006): Technical Aspect of Biodiesel Production by Transestrification: A Review. J. Ren. and Sust. Energy Rev., 10: 248-268.

44. Meher, L. C., Dharmagadda, V. S. S. and Naik, S. N. (2006): Optimization of Alkali-Catalyzed Transesterification of Pongamta Pinnata Oil for Production of Biodiesel. J. Bioresource Tech., 7: 1392-1397.

45. Mohibbe, Azam M., Waris, Amtul and Nahar, M. N. (2005): Prospects and Potential of Fatty Acid Methyl Esters of Some Non-Traditional Seed Oils for Use as Biodiesel in India. J. Biomass and Bioenergy, 29: 293-302.

46. Mulugetta, Yacob (2009): Evaluating the Economics of Biodiesel in Africa. J. Ren. and Sust. Energy Rev., 13:1592–1598.

47. Nag A.N. S., Bhattacharya, M. S. and De, K. B. (1995): Utilization of Vegetable Oils. J. American Oil and Chemist Soci., 72:1591-1593.

48. Naoko, Matsumoto, Daisuke, Sano and Mark, Elder (2009): Bio fuel Initiatives in Japan: Strategies, Policies and Future Potential. J. App. Energy, 86: S69–S76.

49. Noureddini, H., Gao, X. and Philkana, R. S. (2005): Immobilized Pseudomonas Cepacia Lipase for Biodiesel Fuel Production from Soybean Oil. J. Bioresource Tech., 96: 769-777.

50. Pall, Amit, Verma, Ashish, Kachhwaha, S. S. and Mali, S. (2010): Biodiesel Production Through Hydrodynamic Cavitation and Performance Testing. J. Ren. Energy, 35:619-624.

51. Parikh, J. (2005): Growing Our Own Oils. Biofuels India, 3:164-169.

52. Parmanik, K. (2003): Properties and Use of Jatropha Curcas Oil and Diesel Fuel Blends in Compression Ignition Engine. J. Ren. Energy, 28: 239-248.

53. Patil, Prafulla D. and Deng, Shuguang (2009) : Optimization of Biodiesel Production from Edible and Non-Edible Vegetable Oils. J. Fuel, 88:1302–1306.

54. Prueksakorn, Kritana, Shabbir, Gheewala H., Pomthong, Malakul and Sebastien, Bonnet (2010): Energy analysis of Jatropha plantation systems for biodiesel production in Thailand. J. Energy for Sust. Develop. 2010.

55. Pryor, R. W., Manna, M. A., Schnistock, J. L. and Bashford, L. (1983): Soybean Oil Fuel in a Small Diesel Engine. Tran. SAE, 1983: 333-342.

56. Ramadhas, A. S., Jayaraj, S. and Muraleedharan, C. (2004): Biodiesel Production from High FFA Rubber Seed Oil. Fuel, 2004.

57. Rathore, V. and Madran, G. (2008): Synthesis of Biodiesel from Edible and Non-Edible Oils in Supercritical Alcohols and Enzymatic Synthesis in Supercritical Carbon Dioxide. J. Fuel, 2008.

58. Sahoo, P.K. and Das, L. M. (2009): Combustion Analysis of Jatropha, Karanja and Polanga Based Biodiesel as Fuel in a Diesel Engine. Fuel, 88; 994-999.

59. Sharma, Y. C. and Singh, B. (2009): Development of Biodiesel: Current Scenario. Ren. and Sust. Energy Rev., 13:1646–1651.

60. Singh, Santosh (2009): Indian Bio fuels Annual Report 2009. Global Agriculture Information Network Report IN 9080, New Delhi, 15th June 2009.

61. Sirisomboon, P., Kitchaiya, P., Pholpho, T. and Mahuttanyavanitch, W. (2007): Physical and Mechanical Properties of Jatropha Curcas L. Fruits, Nuts and Kernels. J. Biosystems Engg, 97: 201– 207.

62. Srivastava, A. and Prasad, R. (2000): Triglycerides-Based Diesel Fuels. J. Ren. and Sust. Energy Rev., 4:111–144.

63. Sylvain, Leduc, Karthikeyan, Natarajan, Erik, Dotzauer, Ian, McCallum and Michael, Obersteiner (2009): Optimizing biodiesel production in India. J. App. Energy, 86: S125–S131.

64. Tapanes, N., Aranda, D., Carneiro, J. and Antunes, O. (2008): Transesterification of Jatropha Curcas Oil Glycerides: Theoretical and Experimental Studies of Biodiesel Reaction. Fuel, 87: 2286–2295.

65. Victor, Kraemer, Wermelinger Sancho Araujo, Silvio Hamacher and Luiz Felipe Scavarda (2010): “Economic Assessment of Biodiesel Production from Waste Frying Oils”, J. Bio-resource Tech., 2010.

66. Vivek and Gupta, A. K. (2004): Biodiesel Production from Karanja Oil. J. Scien. and Ind. Res., 63:39 – 47.

67. Wu, C. Z., Yin, X. L., Yuan, Z. H., Zhou, Z .Q. and Zhuang, X. S. (2009): The Development of Bio energy Technology in China. J. Energy, 4:1–6.

68. Xuezheng, Liang, Shan, Gao, Jianguo, Yang and Mingyuan, He (2009): Highly Efficient Procedure for the Transesterification of Vegetable Oil. J. Ren. Energy, 34: 2215–2217.

69. Zafer, Utlu and Mevlut, Sureyya Kocak (2008): The Effect of Biodiesel Fuel Obtained from Waste Frying Oil on Direct Injection Diesel Engine Performance and Exhaust Emissions. J. Ren. Energy, 33: 1936–1941.

70. Zhou, Adrian and Thomson, Elspeth (2009): The Development of Biofuels in Asia. J. App. Energy, 86: S11–S20.


134

Mathur Y. B. et al.

71. Ziejewski, M. Z., Kaufman, K. R. and Pratt, G. L. (1983): Vegetable Oil as Diesel Fuel. USDA Agri. Rev., 28:106 -111.

72. Ziejewski, M. Z., Kaufman, K. R., Schwab, A. W. and Pryde, E. H. (1984): Diesel Engine Evaluation of a Nonionic Sunflower Oil-Aqueous Ethanol Micro-emulsion. J. American Oil and Chemist Soci., 61: 1620-1626.

73. Zullaikah, S., Lai, C. C., Ramjan, S. and Ju, Y. H. (2005): A Two Step Acid Catalyzed Process from Rice bran Oil. J. Bioresource Tech., 96:1889-1896.

74. 4th R&D Report on Tree Born Oilseeds 2008 -2009, National Oilseed and Vegetable Oil Development Board, Ministry of Agriculture, Govt. of India, Gurgaon, Feburary 2009.

Technology TodayQuarterly journal

Thrust Force, Torque In Drilling The Natural Fiber Reinforced Polymer Composite Materials And Evaluate 3 – 18Delamination Factor For Bone Graft Substitutes-A Work Of Fiction ApproachMR.CHANDRAMOHAN.D1, DR. MARIMUTHU.K2, MR.RAJESH.S3, AND MR.RAVIKUMAR.M.M4

Effect Of Interpass Tempetature During The Welding Of Is2062-Is206 19-26 1.AJAY RAO U.G 2. ANOOP RAMESH 3. AVINASH T 4. KARTHIK N 5. DR. B.S AJAY KUMAR 6. M.V CHANDRASHEKAR

Modal Analysis Of Automobile Suspension System 27-30VIKAS R.KHALKAR1*, LALITKUMAR M.JUGULKAR 2, BHARTESH DANAWADE 3, SIDDHARTH WAGHMARE4

Productivity Improvement Potential Analysis Through Ergonomic Intervention In Arc Welding 31-38S.MALIKRAJ, DR.A.K.GANGULY, DR.T.SENTHIL KUMAR

Analysis Of Pressure Relief Valve Using Mat Lab Simulink 39-44DR. D.N.RAUT, MEHUL S PRAJAPATI

Development of aspect – impact matrix during the implementation of environmental management 45-54system for common effluent treatment plant at kagal, dist. Kolhapur, maharashtra. (india)SUNIL SHIVAJI MANE, PROF.DR.GIRISH KULKARNI

Experimental Analysis Of A Catalytic Converter Using Supported Copper Catalyst To Reduce 55-62Nox And Pm From Diesel EnginePL.S. MUTHAIAH, DR. M. SENTHILKUMAR, DR.N.MANOHARAN

Design Of High Bandwidth Charge Pump With On Chip Loop Filter For High Performance Phase Locked 63-68 Loop In 180 Nm Cmos Technology *PROF. R .H. TALWEKAR, PROF. (DR.) S. S LIMAYE

A New Oblivious Watermarking Method Using Maximum Wavelet Coefficient Modulation 69-74K.RAMANJANEYULU DR. K. RAJARAJESWARI

Design And Implementation Of Tam Controller And Wrapper For Embedded Cores 75 - 80K. CHARAPANI ,DR. P. NEELAMEGAM

A comparative analysis of Pixel level fusion of multi-focused images using Discrete and stationary wavelet transform 81-84

K. KANNAN1, S. ARUMUGA PERUMAL2, K. ARULMOZH

Detection of Unsolicited mail (Spam) in Application Layer 85-88

MRS.N.NAGADEEPA1,DR.R.S.D.WAHIDABANU2

Studies on friction stir welded aluminium alloy- Roughness and macro/microscopic perspective 89-94

G. RAGHU BABU1* DR.K.G.K.MURTI2 DR.G.RANGA JANARDHANA3

An Efficient Musical Instrument Identification on Solo Performance Music Using MFCC 95-100CHELLASWAMY . C 1 SUGANTHINI REKHA.R2

Novel Approach To Mobile Database Disseminationforbandwidth And Energy Conservation 101-104R.BOOPATHI RANI1, A.ADAIKALAM2

TECHNOLOGY TODAY QUARTERLY JOURNAL 3 DECEMBER 2010 VOLUME II ISSUE 4 ISSN: 2180-0987

Contents Page No

Performance Analysis Of Guaranteed Packet Transfer In Coverage Area And Less Coverage Area For MANET 105-110

¹ MRS. S.GOMATHI , 2 DR K.DURAISWAMY

Comparative Study Of Subgrade Soil Strength Estimation Models Developed Based On Cbr, Dcp And Fwd Test Results 111-122Dr. R. SRINIVASA KUMAR

An Experimental Study About The Lightweight Composite Bricks 123-128

D. JEGATHEESWARAN*, Dr R. MALATHY**

Security Architecture Using Key Distribution Algorithm In Vanet 129-136M. PRABAKARAN M.Tech. (Ph.D.), Dr. G. MAHADEVAN M.E. Ph.D.,

Design Of Quality Information System Based On Iso 9001:2000 Standard 137-142M. SAKTHIVEL# S.R. DEVADASAN$ M. SAKTHIVEL S.R.DEVADASAN

An optimal and impatient backoff routing algorithm for ad-hoc networks 143-152DR. JANGA REDDY

Design Of Intelligent Robotic Arm For Visually Challenged‘A boon for physical ailment’ 153-158

SEIMA SAKI S1, SABITA DEVI D2 MANOJ S

Clustering In High Performance Multifeature Query Processing 159-1621 SIVASUBRAMANIAN S. 2 Dr.UMAMAHESWARI M

Zernike Moments Based Texture Features For Recognition Task 163-166

M.SURESH1, V.SUBBIAH BHARATHI2,

Performance evaluation of c.i. engine using jatropha-diesel blends 167-172

Y. B. MATHUR+, ABDUL SAMAD*, RANJEET SINGH**, G.VYAS#

Performance of variable compression test rig using alcohol as fuel 173-178DR (MISS) S L SINHA, MR R K YADAV

Implementation of image processing algorithms on fpga 179-184

J.A.SADALGEA, M.D. PATILB, R.K.SHASTRIC

A novel agent based dynamic resource management and job scheduling in grid computing 185-1881S. JAGADEESWARI , 2 R. PUNIDHA

Performance evaluation of double pass solar air collector with and without porous material 189-194

BHARAT RAMANI1, GUPTA AKHILESH2, RAVI KUMAR3

Flexural behaviour of Glass fibre reinforced Self compacting concrete Slabs 195-200

DR SESHADRI SEKHAR. T1 DR. SRINIVASA RAO P2 CHANDRA MOULI3Network security using elliptic curve cryptography 201-208

MR.S.DHANDAPANI1, MRS.C.KAVITHA2


Contents Page No

SSTHRUST FORCE, TORQUE IN DRILLING THE NATURAL FIBER REINFORCED POLYMER COMPOSITE MATERIALS AND

EVALUATE DELAMINATION FACTOR FOR BONE GRAFT SUBSTITUTES-A WORK OF FICTION APPROACH

MR.CHANDRAMOHAN.D1, DR. MARIMUTHU.K2, MR.RAJESH.S3, AND MR.RAVIKUMAR.M.M4

1, 3, 4 Research scholar, Department of Mechanical Engineering,Anna University Coimbatore, Coimbatore, India

Email: [email protected],2 Asst. Prof. Department of Mechanical Engineering, Coimbatore Institute of Technology, Coimbatore,

India Email: [email protected]

ABSTRACTThis paper discuss about the Natural Fiber Reinforced Composite Materials Contribution as bone implants. The biomaterial science is an interdisciplinary field that represents one of the most sophisticated trends in the worldwide medical practice. In the last decades researchers have developed new materials for the improvement of human life quality. Due to the frequent occurrence of bone fractures, it is important to develop a plate material for fixation on the fractured bone. These plate materials have to be lightweight, have to allow stiffness and should be biocompatible with humans. Drilling is the most frequently employed operation of secondary machining for fiber-reinforced materials owing to the need for joining fractured bone by means of plate material in the field of orthopaedics. An effort to utilize the advantages offered by renewable resources for the development of biocomposite materials based on biopolymers and natural fibres. Fabrication of Natural fibre powdered material (Sisal (Agave sisalana), Banana (Musa sepientum) & Roselle (Hibiscus sabdariffa)) reinforced polymer composite plate material by using bio epoxy resin Grade 3554A and Gardner 3554B. Instead of orthopaedics alloys such as Titanium, Cobalt chrome, Stainless steel and Zirconium and this plate material can be used for internal fixation on human body for fractured bone. In this, present work focuses on the prediction of thrust force and torque of the natural fiber reinforced polymer composite materials and the values compared with the Regression model and the Scheme of Delamination factor / zone using machine vision system also discussed with the help of Scanning Electron Microscope [SEM].

Key words: Natural fibres, Bio epoxy resin, Thrust force, Torque, Regression Model, Delamination, SEM.

1. INTRODUCTIONA judicious combination of two or more materials that produces a synergistic effect. Composites: materials, usually man-made, that are a three-dimensional combination of at least two chemically distinct materials, with a distinct interface separating the components, created to obtain properties that cannot be achieved by any of the components acting alone. Composites are combinations of two materials in which one of the materials, called the reinforcing phase, are in the form of fibers, sheets, or particles, and are embedded in the other materials called the matrix phase.

The reinforcing material and the matrix material can be metal, ceramic, or polymer. Typically, reinforcing materials are strong with low densities while the matrix is usually a ductile, or tough, material. If the composite is designed and fabricated correctly, it combines the strength of the reinforcement with the toughness of the matrix to achieve a combination of desirable properties not available in any single conventional material. The downside is that such


mailto:[email protected]


composites are often more expensive than conventional materials.

Examples of some current application of composites include common replacement materials used for fixation are: Titanium, Cobalt chrome, Zirconia and Stainless steel.The selection of the material for implants depends on the type of stress and patience’s needs. Factors such as biocompatibility and high modulus of elasticity of the material make it difficult for the grafts to work properly, when they are implanted in humans. This characteristic features give birth to an increased intension of studies by the researchers in materials that can be used as the grafting material. Artificial grafts that meet all the requirements of the orthopaedic surgeons are not good enough right now, so this field can be explored by the researchers. This paper proposed suggestions of using Natural fiber reinforced composite as a plate material which uses pure natural fibers that are rich in medicinal properties like Sisal, Banana & Roselle (hybrid) fiber. In this project, rather than using pure biomaterial as reinforcement with polymer composite, coating over the plate material is also done by using highly biocompatible materials like calcium phosphate, calcium sulphate, hen eggshell powdered material and Hydroxy Apatite (hybrid) composite. This plate material can be used as an inside fixation over the fractured bone. The most important thing that the researchers have to take into account is that these step taken now, will help the mankind to develop and to have a more pleasant life.2. BACKGROUNDBiomaterials improve the quality of life for an ever increasing number of people each year. The range of applications is vast and includes such things as joint and limb replacements, artificial arteries and skin, contact lenses, and dentures. This increasing demand arises from an aging population with higher quality of life expectations. The biomaterials community is producing new and improved implant materials and techniques to meet this demand, but also to aid the treatment of younger patients where the necessary properties are even more demanding. A counter force to this technological push is the increasing level of regulation and the threat of litigation. To meet these conflicting needs it is

necessary to have reliable methods of characterization of the material and material/host tissue interactions.

The main property required for a biomaterial is that it does not illicit an adverse reaction when placed into service [2]. The various materials used in biomedical applications may be grouped into (a) metals (b) ceramics (c) polymers and (d) composites made from the above groupings. Metals and alloys that is successful as biomaterials include: gold, tantalum, stainless steel, NiTi (Shape memory alloy), Co-Cr, and Ti alloys. Machining of such Orthopaedic alloy implants, with high speed machining, will offers advantages, but also have their own disadvantages that include complexity, high machining cost. Titanium was used for bone replacements, during past decades, but those implants are simple geometric approximations of the bone shape. Mismatches can occur between real bone and implants often causing stress concentrations and premature implant failure. The machining stocks are uneven & more than the required levels. This leads to more weight at rough casting stage and contribution of more machining time and these castings are manufactured in green sand moulding process which leads to poor surface finish, more material in machining surfaces and less dimensional stability. The above said complexities give way for the next generation bone implants that are polymers and ceramics which have better biocompatibility and good tensile properties.

3. OBJECTIVES OF THE STUDY

An effort to utilize the advantages offered by renewable resources for the development of Biocomposite materials based on biopolymers and natural fibers.

Fabrication of Natural fiber (Sisal, Banana & Roselle (hybrid) composite fiber) reinforced polymer composite plate material by using bio epoxy resin. Instead of orthopaedic alloys (such as Titanium, Cobalt chrome, Stainless steel and Zirconium


Bone graft substitutes such as calcium phosphate, calcium sulfate, hen egg shell and hydroxyl apatite can be coated on NFRP composite (Biocomposite plate material) and this plate material can be used for inside fixation on human body for fractured bone.

4. NATURAL FIBERSNatural fibers present important advantages such as low density, appropriate stiffness and mechanical properties and high disposability and renewability. Moreover, they are recyclable and biodegradable. Over the last decade, composites of polymers reinforced by natural fibers have received increased attention. Natural fibers such as sisal, flax, jute and wood-fibers possess good reinforcing capability when properly compounded with polymers. One of the unique aspects of designing parts with fiber reinforced composite materials is that the mechanical properties of the material can be tailored to fit a certain application. By changing the orientation or placement of the fibers the material can be designed to exhibit properties that are isotropic or highly anisotropic depending on the desired end result. 4.1) Sisal Sisal is valued for cordage use because of its strength, durability, ability to stretch, affinity for certain dyestuffs, and resistance to deterioration in saltwater. Sisal is used by industry in three grades. The lower grade fiber is processed by the paper industry because of its high content of cellulose and hemicelluloses. The medium grade fiber is used in the cordage industry for making: ropes, baler and binders twine. Ropes and twines are widely employed for marine, agricultural, and general industrial use. The higher-grade fiber after treatment is converted into yarns and used by the carpet industry. Uses of Sisal: products made from sisal are being developed rapidly, such as furniture and wall tiles made of resonated sisal. a recent development expanded the range even to car parts for cabin interiors. Other products developed from sisal fiber include spa products, cat scratching posts, lumbar support belts, rugs, slippers, cloths and disc buffers. Sisal wall covering meets the abrasion and tearing resistance standards of the American society for

testing and materials and of the national fire protection association. Apart from ropes, twines and general cordage sisal is used in low-cost and specialty paper, dartboards, buffing cloth, filters, geotextiles, mattresses, carpets, handicrafts, wire rope cores and macramé. In recent years sisal has been utilized as a strengthening agent to replace asbestos and fiberglass as well as an environmentally friendly component in the automobile industry.

Products made from sisal fiber are purchased throughout the world and for use by the military, universities, churches and hospitals.4.2) RoselleThe Roselle (hibiscus sabdariffa) is a species of hibiscus native to the old world tropics. It is an annual or perennial herb or woody-based sub shrub, growing to 2–2.5 m tall. The leaves are deeply three- to five-lobed, 8–15 cm long, arranged alternately on the stems. The flowers are 8–10 cm in diameter, white to pale yellow with a dark red spot at the base of each petal, and have a stout fleshy calyx at the base, 1.5–2 cm wide, enlarging to 3–3.5 cm, fleshy and bright red as the fruit matures. It is an annual plant, and takes about six months to mature. Roselle is native from India to Malaysia, where it is commonly cultivated, and must have been carried at an early date to Africa. It has been widely distributed in the tropics and subtropics of both hemispheres, and in many areas of the West Indies and Central America has become naturalized. Uses of Roselle: The seeds are considered excellent feed for chickens. The residue after oil extraction is valued as cattle feed when available in quantity. Nutritionists have found Roselle calyces as sold in Central American markets to be high in calcium, niacin, riboflavin and iron.4.3) Banana Fiber Banana fiber, a ligno-cellulosic fiber, obtained from the pseudo-stem of banana plant (musa sepientum), is a bast fiber with relatively good mechanical properties. The ‘‘pseudo-stem’’ is a clustered, cylindrical aggregation of leaf stalk bases. Banana fiber at present is a waste product of banana cultivation and either not properly utilized or partially done so. The extraction of fiber from the pseudo stem is not a common practice and much of the stem is not used for production of fibers. The buyers for banana fibers are erratic and there is no systematic way to extract the fibers regularly. Useful


applications of such fibers would regularize the demand which would be reflected in a fall of the prices. Uses of Banana and Plantain: Culinary uses: banana leaves, pseudo stems, fruit stalks and peels can all be used for various culinary purposes. Bananas are primarily eaten as a fruit, either on its own or as a part of a salad. All parts of the banana have medicinal applications as well.

5. COMPOSITE PREPARATION

Materials Used:The specimen used in this study is a plate of 60x40 mm made of natural fiber reinforced

composite material. The composite is made of natural fibers. Commercially available natural fibers are taken.The materials used in this project are (as shown in fig 5.1 Specimens arranged in Left to Right 1, 2, 3,4,5,6 listed below)

1) Banana fibre reinforced composite2) Sisal fibre reinforced composite3) Roselle fibre reinforced composite4) Sisal & Roselle (hybrid) fibre reinforced composite5) Banana & Sisal (hybrid) fibre reinforced composite6) Banana & Roselle (hybrid) fibre reinforced composite

Fig 5.1 Specimens (60x40 mm)

5.1 Chemical Treatment

The fibers are powdered.Then the fibers are cleaned normally in

clean running water and dried. A glass beaker is taken and 6% NaOH

is added and 80% of distilled water is added and a solution is made.

After adequate drying of the fibers in normal shading for 2 to 3 hours the

fibers are taken and soaked in the prepared NaOH solution.

Soaking is carried out for different time intervals depending upon the strength of fiber required.

For our project the fibers are soaked in the solution for three hours.

After the soaking process is complete the fibers are taken out and washed in running water and dried for another 2 hours.

Now the fibers are taken for the next fabrication process namely the procasting process

5.2 Advantages of chemical treatment First and foremost chemical treatment

with NaOH will remove moisture


content from the fibers thereby increasing its strength.

Secondly the chemical treatment also enhances the flexural rigidity of the

fibers.

Thirdly this treatment clears all the impurities that are adjoining with the

fiber material and also stabilizes the molecular orientation.

5.3 Manufacturing process

A mould of 60 mm length and 40 mm diameter and is created using GI sheet mould.

An OHP Sheet is taken and Releasing agent is applied over it and fitted with the inner side of the mould and allowed to dry it.

A glass beaker and a glass rod or a stirrer is taken and cleaned well with running water and then with warm water.

Then calculated quantity of bio epoxy resin Grade 3554A and Hardner 3554B Resin is added and the mixture is stirred for nearly 15 minutes.

The reason behind this stirring is to create a homogeneous mixture of resin and accelerator molecules.

Then after the mixing is over, calculated quantity of fibers are added and stirring process is continued for the next 45 minutes.

Then the mixture is poured into the mould and rammed mildly for uniform settlement.

Then the mould is allowed to solidify for nearly 24 hours.

5.4 Fiber volume fraction: The volume fraction of fiber was calculated by a method which enables the rule of mixtures and analysis of measured composite properties. The

method involves measuring the density of the composite (ρc) of mass (MC ) at a given mass fraction of the resin (MR ) Volume fraction of resin (VR) was calculated using the formula

VR = MR * ρc / MC * ρR -------- (5.1)Where ρR = density of resin in g/m3

Then the fiber volume fraction is determined by the relation VF = 1 - VR ---------- (5.2)6. EXPERIMENTAL SETUPA number of drilling experiments are carried out on a CNC machining center (Maxmill) using HSS twist drills for the machining of natural fibre reinforced polyester composites. A two-component drill tool dynamometer is used to record the thrust force and torque. Conventional high-speed steel twist drills are used as much as cemented tungsten carbide drills. Tool geometry is a relevant aspect to be considered in drilling of fibre-reinforced plastics, particularly when the quality of the machined hole is critical.

6.1) Factorial Design:

A 33 full factorial design with a total of 27 experimental runs are carried out. The thrust force and torque were the response variables recorded for each run. The effect of the machining parameters is another important aspect to be considered. It can be seen that cutting speeds from 20 to 60 m/min are usually employed, whereas feed rate values lower than 0.3 mm/rev are frequent. Cutting speed is not a limiting factor when drilling polymeric composites, particularly with hard metals, therefore, the use of cutting speeds below 60 m/min may be explained by the maximum rotational speed of conventional machining tools, since drill diameters above 10mm are rarely reported. Another reason for keeping cutting speeds below 60 m/min may reside in the fat that higher cutting speed values lead to higher cutting


temperature, which in turn may cause the softening of the matrix. The use of feed rates below 0.3 mm/rev may be associated to the delamination damage caused

when this parameter is increased. Table shows the detail of variables used in the experiment.


Table 6.1 Assignment of the levels to the factors

Level Drill size,d (mm)

Revolution ,N (rpm)

Feed rate,f (mm/rev)

1 3 600 0.12 4 900 0.23 5 1200 0.3

Table 6.2 Comparison results of Sisal & Roselle Thrust force and Torque

Sl. No.

Drill dia Speed Feed Thrust Torque (RM) (RM)

(mm) (rpm) (mm/rev) (N) (N-m) Thrust Torque(N) (N-m)

1 3 300 0.1 3 0.93 3.105598 0.9368652 3 600 0.1 3.86 0.96 4.003891 0.9681113 3 900 0.1 4.48 0.98 4.645422 0.986874 3 300 0.2 4.86 1.72 5.036403 1.7253265 3 600 0.2 6.27 1.77 6.493179 1.7828696 3 900 0.2 7.27 1.81 7.533562 1.8174157 3 300 0.3 6.45 2.45 6.682641 2.4660388 3 600 0.3 8.31 2.54 8.61559 2.5482869 3 900 0.3 9.65 2.59 9.99604 2.59766310 4 300 0.1 5.04 1.2 5.226974 1.20404611 4 600 0.1 6.5 1.24 6.738873 1.24420412 4 900 0.1 7.54 1.26 7.818622 1.26831213 4 300 0.2 8.18 2.21 8.476675 2.21736614 4 600 0.2 10.55 2.28 10.92855 2.29131915 4 900 0.2 12.23 2.32 12.6796 2.33571816 4 300 0.3 10.85 3.15 11.24743 3.16931917 4 600 0.3 13.99 3.26 14.50074 3.27502218 4 900 0.3 16.23 3.32 16.82415 3.33848219 5 300 0.1 7.55 1.46 7.827643 1.46272920 5 600 0.1 9.74 1.5 10.09178 1.51151421 5 900 0.1 11.3 1.53 11.70876 1.54080222 5 300 0.2 12.25 2.68 12.69423 2.69375523 5 600 0.2 15.79 2.77 16.36602 2.78359724 5 900 0.2 18.32 2.82 18.9883 2.83753425 5 300 0.3 16.25 3.83 16.84356 3.8502326 5 600 0.3 20.95 3.96 21.71555 3.97864427 5 900 0.3 24.31 4.04 25.19496 4.055737

Where, RM-Regression Model

Table 6.3 Comparison results of Sisal & Banana Thrust force and Torque

Sl. No.

Drill dia Speed Feed Thrust Torque (RM) (RM)

(mm) (rpm) (mm/rev) (N) (N-m)Thrust Torque(N) (N-m)

1 3 300 0.1 6.93 0.85 6.95012 0.877632 3 600 0.1 9.27 0.82 9.29083 0.846933 3 900 0.1 10.98 0.8 11.0102 0.829464 3 300 0.2 17.99 1.56 18.0399 1.605625 3 600 0.2 24.05 1.5 24.1154 1.549446 3 900 0.2 28.5 1.47 28.5783 1.51757 3 300 0.3 31.43 2.22 31.5172 2.28618 3 600 0.3 42.02 2.14 42.1318 2.206129 3 900 0.3 49.79 2.09 49.9289 2.1606310 4 300 0.1 8.38 1.1 8.40218 1.1384411 4 600 0.1 11.2 1.06 11.2319 1.0986112 4 900 0.1 13.27 1.04 13.3105 1.0759613 4 300 0.2 21.75 2.02 21.8088 2.0827714 4 600 0.2 29.07 1.95 29.1538 2.009915 4 900 0.2 34.45 1.91 34.549 1.9684616 4 300 0.3 38 2.87 38.102 2.9654817 4 600 0.3 50.8 2.77 50.9342 2.8617318 4 900 0.3 60.2 2.72 60.3603 2.8027219 5 300 0.1 9.71 1.35 9.73433 1.3930320 5 600 0.1 12.98 1.3 13.0127 1.3442921 5 900 0.1 15.38 1.28 15.4209 1.3165822 5 300 0.2 25.2 2.47 25.2666 2.5485423 5 600 0.2 33.68 2.38 33.776 2.4593724 5 900 0.2 39.92 2.33 40.0267 2.4086625 5 300 0.3 44.02 3.52 44.143 3.6286426 5 600 0.3 58.85 3.39 59.0097 3.5016927 5 900 0.3 69.74 3.32 69.9303 3.42949

Table 6.4 Comparison results of Roselle & Banana Thrust force and Torque

Sl. No.Drill dia Speed Feed Thrust Torque (RM) (RM)

(mm) (rpm) (mm/rev) (N) (N-m)Thrust Torque(N) (N-m)

1 3 300 0.1 6.93 0.85 6.95012 0.87763

2 3 600 0.1 9.27 0.82 9.29083 0.846933 3 900 0.1 10.98 0.8 11.0102 0.829464 3 300 0.2 17.99 1.56 18.0399 1.605625 3 600 0.2 24.05 1.5 24.1154 1.549446 3 900 0.2 28.5 1.47 28.5783 1.51757 3 300 0.3 31.43 2.22 31.5172 2.28618 3 600 0.3 42.02 2.14 42.1318 2.206129 3 900 0.3 49.79 2.09 49.9289 2.1606310 4 300 0.1 8.38 1.1 8.40218 1.1384411 4 600 0.1 11.2 1.06 11.2319 1.0986112 4 900 0.1 13.27 1.04 13.3105 1.0759613 4 300 0.2 21.75 2.02 21.8088 2.0827714 4 600 0.2 29.07 1.95 29.1538 2.009915 4 900 0.2 34.45 1.91 34.549 1.9684616 4 300 0.3 38 2.87 38.102 2.9654817 4 600 0.3 50.8 2.77 50.9342 2.8617318 4 900 0.3 60.2 2.72 60.3603 2.8027219 5 300 0.1 9.71 1.35 9.73433 1.3930320 5 600 0.1 12.98 1.3 13.0127 1.3442921 5 900 0.1 15.38 1.28 15.4209 1.3165822 5 300 0.2 25.2 2.47 25.2666 2.5485423 5 600 0.2 33.68 2.38 33.776 2.4593724 5 900 0.2 39.92 2.33 40.0267 2.4086625 5 300 0.3 44.02 3.52 44.143 3.6286426 5 600 0.3 58.85 3.39 59.0097 3.5016927 5 900 0.3 69.74 3.32 69.9303 3.42949

7. PREDICTION TECHNIQUES7.1 Regression Model:The statistical tool, regression analysis helps to estimate the value of one variable from the given value of another. In regression analysis, there are two types of variables. The variable whose value is influenced or is to be predicted is called dependent variable and the variable, which influences the values or used for prediction is called independent variables. The tool, regression can be extended to three or more variables. If two variables are taken into account, then it is called simple regression. The tool ofregression when extended to three or more variables is called multiple regressions.

7.2 SPSS:SPSS (originally, Statistical Package for the Social Sciences) was released in its first version in 1968. SPSS is among the most widely used programs for statistical analysis in social science. It is used by market researchers, health researchers, survey companies, government, education researchers, marketing organizations and others. In addition to statistical analysis, data management (case selection, file reshaping, creating derived data) and data documentation (a metadata dictionary is stored with the data) are features of the base software. Statistics included in the base software:

• Descriptive statistics: Cross tabulation, Frequencies, Descriptives, Explore,

• Descriptive Ratio Statistics

• Bivariate statistics: Means, t-test, ANOVA, Correlation (bivariate, partial, distances), Nonparametric tests.

• Prediction for numerical outcomes: Linear regression

• Prediction for identifying groups: Factor analysis, cluster analysis (two-step, K means, hierarchical), Discriminant Statistical output is to a proprietary file format (*.spo file, supporting pivot tables) for which, in addition to the in-package viewer, a stand-alone reader is provided. The proprietary output can be exported to text or Microsoft Word. Alternatively, output

can be captured as data (using the OMS command), as text, tab-delimited text, HTML,

XML, SPSS dataset or a variety of graphic image formats (JPEG, PNG, BMP and EMF).

7.2.2 Regression equations:Thrust = k * d a * n b * f c ------------

(7.1)Torque = k * d a * n b * f c ------------

(7.2)Whered = Drill diameter in mmn = Speed in rpmf = Feed rate in mm/reva, b & c = Regression

constants

Table 7.1 Regression equations for thrust force

Material Thrust force R2

Sisal 1.479916792 X d ^ 1.695179761 X n ^ 0.218422665 X f ^ 0.927436761 0.93614

Banana3.072028878 X d ^ 1.347046253 X n ^ 0.233186110 X f ^ 0.886143709 0.89341

Roselle3.023946833 X d ^ 1.094546824 X n ^ 0 .324203073 X f ^ 0.951921753 0.88378

Sisal andRoselle(Hybrid)

3.0283346946 X d ^ 1.809727356 X n ^ 0.366531308 X f ^ 0.697522528 0.91185

Sisal and Banana (Hybrid)

7.346100813X d ^ 0.659519445 X n ^ 0.418769028X f ^ 1.376076865 0.95432

Banana andRoselle(Hybrid)

7.34610183X d ^ 0.266522184 X n ^ 0. 522769028X f ^ 1. 768086565 0.95745

Table 7.2 Regression equations for torqueMaterial Torque R2

Sisal 2.849305728 X d ^ 1.099468559 X n ^ -.066603083 X f ^ 0.962428131 0.87688

Banana 2.187147307 X d ^ 1.138389699 X n ^ -.037946866 X f ^ 1.061918604

0.88161

Roselle 1.534106581 X d ^ 1.348045235 X n ^ -.09453993 X f ^ 0.772847627

0.88963

Sisal + Roselle 2.085745348 X d ^ .872156974 X n ^ .047331957 X f ^ 0.880955972

0.88741

Sisal + Banana 3.239667434 X d ^ 0.90444033 X n ^ -0.051380071 X f ^ 0.871441681

0.867421

Banana andRoselle(Hybrid) 3.239667434 X d ^ 0.90444033 X n ^ -0.051380071 X f ^

0.871441681 0.81094

8. DELAMINATIONThe advantage of the composite materials over conventional materials is that they possess high specific strength, stiffness and fatigue characteristics, which enables structural design to be more versatile. Due to the inhomogeneous and anisotropy nature of composite materials, their machining behavior differs in many respects from metal machining. In recent years, customer requirements have put greater emphasis on the product development with new challenges to manufacturers, such as machining techniques. Machining of composite materials require the need for better understanding of cutting process regarding accuracy and efficiency. Though near net shape process have gained a lot of attention, more intricate products need secondary machining for the required accuracy. Drilling is the most frequently used secondary operation for fiber reinforced materials.Induced Delamination

occurs both at the entry and exit planes of the work piece. These delaminations could be correlated to the thrust force during the approach and exit of the drill. Delamination is one of the major concerns in drilling holes in composite materials. In order to understand the effects of process parameters on the Delamination, a large number of experiments have to be performed and analyzed mathematical models to be built on the same. Modeling of the formation of delaminations is highly complex and expensive. Hence statistical approaches are widely used over the conventional mathematical models.Drilling is the most frequently employed operation of secondary machining for fiber-reinforced materials owing to the need for joining fractured bone by means of plate material in the field of orthopaedics as shown in figure 8.1.

Fig 8.1 Internal Plate Fixation in Humerus Bone

8.1) Types of Delamination8.1.2) Peel up Delamination:Peel up occurs at the entrance plane of the work piece. This is shown schematically shown below. After the cutting edge of the drill makes contact with the laminate. The cutting force acting in the perpendicular direction is the driving force for delaminaton. It generates a peeling force in the axial direction through the slope of the drill flute that results in separating the laminas from each other forming a delaminaton zone at the top surface of the laminate. It mainly depends on speed, point angle.

Fig 8.1.2 Peel up delamination

8.1.3) Push down DelaminationPush out is the delamination mechanism occurring as the drill reaches the exit side of the material and is shown schematically shown below. As the drill approaches the end, the uncut chip thickness gets smaller and resistance to deformation decrease. At some point, the thrust force exceeds the inter laminar bond strength and delamination occurs.

This happens before the laminate is completely penetrated by the drill. It mainly depends on feed rate, drill diameter.

Fig 8.3 Push down delamination

8.1.4) Procedure to calculate the value of delamination factor:• Drilling is done in the CNC MAXMILL for three different drill diameters of 3mm, 4mm, and 5mm respectively.• Then the job is placed in the MACHINE VISION system to capture the digital image of the hole drilled. This is done by using various zoom factors (11x, 67x, 22x, 134x).• Circle is drawn using the draw tool available in the RAPID-I software for both maximum diameter and the nominal diameter.• From the values of the Dmax and Dnom

delamination factor is calculated using the following formula, (Fd ) = Dmax/Dnom

The first part of the equation represents the size of the crack contribution and the second part represents the damage area contribution.

Fda = α (Dmax / Dnom) + β (Amax / Anom)----- (8.2)

Where, Amax – Maximum area related to the maximum

diameter of the delamination zone.

Anom – Area of the nominal hole.In this work,

α = (1- β), ----- (8.3)

β = Amax / (Amax - Anom). ----- (8.4)Fda = (1- β)*Fd + ((Amax / (Amax - Anom)*(Fd

2- Fd)).----- (8.5)

8.2 Calculations:Delamination factors,

Delamination factor (Fd ) = Dmax/Dnom

----- (8.6)Where,Dmax – Maximum diameter

corresponding to the Delamination zone.

Dnom – Nominal diameter.

Adjusted Delamination factor (Fda) = Fd + (Ad/ (Amax-Ao)) ( Fd

2- Fd) ---(8.7)

Where,Fda – Adjusted Delamination factor. Fd – Delamination factor.Ad – Area of the Delamination zone.Ao – Nominal area.

8.6 Scheme of Delamination factor / zone using machine vision system

Fig 8.6.1 Delamination zone Fig 8.6.2 Delamination zone Fig 8.6.3 Delamination zone(for Roselle & Banana (for Roselle & Sisal (for Sisal &Banana

5mm dia) 5mm dia) 5mm dia)

Table 8.1 Delamination Factor (Roselle and Banana) for d=5mm

S.NO FEED SPEED DELAMINATON FACTOR(mm/rev) ( rpm ) Fd Fda

1 0.1 600 1.205 1.998512 0.1 900 1.2034 1.994333 0.1 1200 1.2022 1.99199

4 0.2 600 1.176 1.923425 0.2 900 1.17 1.908076 0.2 1200 10162 1.887717 0.3 600 1.098 1.728968 0.3 900 1.092 1.714459 0.3 1200 1.084 1.69521

Table 8.2 Delamination Factor (Roselle and Sisal) for d=5mm

S.NO FEED SPEED DELAMINATON FACTOR(mm/rev) ( rpm ) Fd Fda

1 0.1 600 1.21 2.011612 0.1 900 1.209 2.008993 0.1 1200 1.206 2.001134 0.2 600 1.196 1.975045 0.2 900 1.92 1.964666 0.2 1200 1.184 1.943987 0.3 600 1.148 1.852358 0.3 900 1.142 1.837319 0.3 1200 1.134 1.81735

Table 8.3 Delamination Factor (Sisal and Banana) for d=5mm

9. SCANNING ELECTRON MICROSCOPE

S.NO FEED SPEED DELAMINATON FACTOR

(mm/rev) ( rpm ) Fd Fda

1 0.1 600 1.193 1.96725

2 0.1 900 1.189 1.95689

3 0.1 1200 1.1804 1.93471

4 0.2 600 1.156 1.87251

5 0.2 900 1.15 1.85738

6 0.2 1200 1.142 1.83731

7 0.3 600 1.078 1.68085

8 0.3 900 1.073 1.66894

9 0.3 1200 1.065 1.64996

9.1 IntroductionThe Scanning Electron Microscope (SEM) is a type of Electron microscope that images the sample surface by scanning it with a high-energy beam of electrons in a raster scan pattern. The electrons interact with the atoms that make up the sample producing signals that contain information about the samples surface topography, composition and other properties such as electrical conductivity. SEM can produce very high-resolution images of a sample surface, revealing details about less than 2 to 5 nm in size. Due to the very narrow electron beam, SEM micrographs have a large depth of fielding a characteristic three-dimensional appearance useful for understanding the surface of a structure of a sample. For conventional imaging in the SEM, specimens must be electrically conductive, at least at the surface, and electrically grounded to prevent the accumulation of electrostatic charge at the surface. Metal objects require little special preparation for SEM except for cleaning and mounting on a specimen stub. Nonconductive specimens tend to charge when scanned by

thelectron beam, and especially in secondary electron imaging mode, this causes scanning faults and other image artifacts. They are therefore usually coated with an ultrathin coating of electrically conducting material, commonly gold, deposited on the sample either by low vacuum sputter coating or by high vacuum evaporation. Conductive materials in current use for specimen coating include gold, gold/palladium alloy, platinum, osmium, iridium, tungsten, chromium and graphite. Coating prevents the accumulation of static electric charge on the specimen during electron irradiation. The image may be captured by photography from a high resolution cathode ray tube, but in modern machines is digitally captured and displayed on a computer monitor and saved to a computers hard disc.All samples must also be of an appropriate size to fit in the specimen chamber and are generally mounted rigidly on a specimen holder called a specimen stub. Several models of SEM can examine any part of a 6-inch (25cm) semiconductor wafer, and some can tilt an object of that size to 45°.

Fig 9.1 SEM of sisal and Fig 9.2 SEM of Roselle Fig 9.3 SEM of Roselle banana (hybrid) and banana (hybrid) and sisal (hybrid)

CONCLUSIONBased on the experimental results obtained the following conclusions can be extractedEffect of Thrust Force:In general, the thrust and torque parameters will mainly depend on the manufacturing conditions employed, such as: feed, cutting speed, tool geometry, machine tool and cutting tool rigidity, etc. A larger thrust force occurred for larger diameter drills and higher feed rates.

In other words, feed rate and drill dia are recognized the most significant factors affecting the thrust force. The worn-out drill may be one

of the major reasons for the drastic increase of the thrust force and why the thrust forces appearing when using multifaceted drill were larger than those when using twist dril1 at high cutting speed. Although

tools are worn out quickly and the thrust force increases drastically as cutting speed increases, an acceptable hole entry and exit is maintained. This project concluded that the thrust force is drastically reduced when the hole is pre-drilled to 0.4mm or above. The thrust force increases with the increase of fibre volume fraction.

Although it is known that the thrust force increases with the increase of the feed, this work provides quantitative measurements of such relationships for the present composite materials. In general, increasing the cutting speed will decrease the thrust force. This work shows that the cutting speed has an insignificant effect on the thrust force when drilling at low feed values. At high feed values the thrust force decreases with an increased cutting speed.Effect of Torque:It can be seen that thrust force and torque increased with drill diameter and feed rate. Examining these results, it is found that the torque increased slightly as the cutting speed increased. However, the increase in torque was much smaller than that in thrust force as cutting speed increases. The average torque appearing when using a multifaceted drill was larger than that using twist dril1 at low drilling speed and the average torque when using a multifaceted drill was smaller than that when using twist dril1 at high drilling speed. It was noticed that the average torque decreased as drilled length increased for twist drill. It was noticed that the effect of feed, speed, and fiber volume fraction on the resulting torque in drilling the specimen. The results indicate that the torque increases as the feed increases. This increase is due to the increasing of the cross-sectional area of the undeformed chip .The results also indicate that the torque increases with the increase of the fiber volume fraction. Increasing fiber volume fraction increases the static strength, and, hence, the resistance of the composite to mechanical drilling increases. This leads to the increase in the required thrust force and torque. The result also indicates that the torque decreases when increasing the cutting speed.Finally this paper concluded that one of the best material is Sisal and Roselle (Hybrid) can be used for internal fixation compared to other materials. The reason could be the reduced fiber bridging effect resulting lower fiber pull in / pull out during drilling process which as shown in Delamination factors and delamination zone. The complete breaking of the fiber rather than pull out is observed through Scanning Electron Microscope SEM analysis.In future this plate material externally coated by calcium phosphate, hen eggshell powdered material and Hydroxy Apatite (hybrid) composite. This plate material can be used for inside fixation on human body for fractured bone. The most important thing that researchers have to have in mind is that these

step taken now, will help humans to develop and to have a more pleasant life.

ACKNOWLEDGEMENTI express my sincere thanks to my beloved parents for their invaluable love; moral support and constant encouragement in my life. I profoundly express my sincere thanks and heartfelt respect to my guide Dr.K.Marimuthu, Ph.D., Assistant Professor, Department of Mechanical Engineering, Coimbatore Institute of Technology, Coimbatore for his for giving precious suggestions and inspiring guidance thought of my research work and course study.I express my sincere thanks to my Doctoral committee members Dr.G.Sundararaj, Ph.D., Professor, Department of Production Engineering, P.S.G.College of Technology, Coimbatore and Dr.I.Rajendran, Ph.D., Professor and Head, Department of Mechanical Engineering, Dr.Mahalingam College of Engineering &Technology, Coimbatore for their valuable guidance and suggestion for this work.I own my immense gratitude to my principal Dr.V.Selladurai, Ph.D., Coimbatore Institute of Technology, Coimbatore for his moral support during the course of my Research work.This research was sponsored by the INSTITUTION OF ENGINEERS (INDIA), KOLKATA. I wish to acknowledge their support.I would also like to acknowledge Dr.L.Karunamoorthy, Ph.D., Professor and Head of Central Workshop at the ANNA UNIVERSITY, CHENNAI for his help in SEM and EDAX Analysis.The help of my friend Mr.D.Logendran, M.E., Lecturer, Department of Mechanical Engineering, Adhiyamaan College of Engineering, Hosur, with the experiments is greatly appreciated.I also thank the Reviewers for their inputs and comments.Special thanks to my Wife C.Saranya with whom I shared much pain and joy. Your presence has been a big part in this Research work.Finally I thank my friends who were directly and indirectly helped me during this Research work.

REFERENCES[1] A.M. Abrao, P.E. Faria, J.C. Campos Rubio, P. Reis and J. Paulo Davim (2006) ‘Drilling of fiber reinforced plastics: A review’ Elsevier -

Journal of Materials Processing Technology 186 (2007) 1–7

[2] M.M. Bain, Eggshell strength: a mechanical/ultrastrucutral evaluation. Ph. D. Thesis, University of Glasgow, Scotland (1992).

[3] Cornell CN, Lane JM, Chapman M, Merkow R, Seligson D, Henry S, et al. Multicenter trial of collagraft as bone graft Substitute. J Orthop Trauma 1991; 5:1–8.

[4] Craig M. Clemons and Daniel F. Caulfield (1994) ‘Natural fibers’ Sage Journals – Journal of Reinforced Plastics Composites - 1354-66

[5] U.A. Khashaba, M.A. Seif and M.A. Elhamid (2006) ‘Drilling analysis of chopped composites’ Elsevier - Composites: Part A 38 (2007) 61–70

[6] S.C. Lin and IK. Chen (2005) ‘Drilling carbon fiber-reinforced composite material at high speed’ Composite Structures 71 (2005) 407–413

[7] Marta Fernandes and Chris Cook (2006) ‘Drilling of carbon composites using a one shot drill bit - Part II: empirical modeling of maximum thrust force’ Elsevier - International Journal of Machine Tools & Manufacture 46 (2006) 76–79

[8] N.S. Mohan, A. Ramachandra and S.M. Kulkarni (2005) ‘Influence of process parameters on cutting force and torque during drilling of glass–fiber polyester reinforced composites’ Elsevier - Composite Structures 71 (2005) 407–413[9] Mohanty AK, Misra M, Hinrichsen G. Biofibres, biodegradable polymers and biocomposites: An overview. Macromol Mater Eng 2000; 276/277:1–24

[10] Murugan R, Ramakrishna S. Bioresorbable composite bone Paste using polysaccharide based nano hydroxyapatite. Biomaterials 2004; 25:3829–35.

[11] Murugan R, Ramakrishna S. Bioresorbable composite bone Paste using polysaccharide based nano hydroxyapatite. Biomaterials 2004; 25:3829–35.

[12] R. Murugan, S. Rama Krishna- Composite science and technology (Division of bioengineering), Faculty of bioengineering, National university of Singapore

[13] K. Palanikumar and J. Paulo Davim (2006) ‘Mathematical model to predict tool wear on the machining of glass fibre reinforced plastic composites’ Elsevier - Materials and Design 28 (2007) 2008–2014

[14] S. Panthapulakkal, A. Zereshkian and M. Sain (2006) ‘Preparation and characterization of wheat straw fibers for reinforcing application in injection molded thermoplastic composites’ Elsevier - Bioresource Technology Volume 97, Issue 2, January 2006, Pages 265-272

[15] Paul Wambua*, Jan Ivens, Ignaas Verpoest (2003) ‘Natural fibres: can they replace glass in fibre reinforced plastics?’ Elsevier - Composites Science and Technology 63 (2003) 1259–1264

[16] Ramakrishna S, Mayer J, winter mantel E, Leong KW. Biomedical Applications of polymer-composite materials: a review.Comp Sci Tech 2001; 61:1189–224.

[17] C.C. Tsao and H. Hocheng (2007) ‘Evaluation of Thrust Force and Surface Roughness in Drilling Composite Material Using Taguchi Analysis and Neural Network’ Elsevier - Journal of Materials Processing Technology

[18] M. Zampaloni, F. Pourboghrat, S.A. Yankovich (2007) ‘Kenaf natural fiber - A discussion on manufacturing problems and solutions’ Composites Part A: Volume 38, Issue 6 June 2007, Pages 1569-1580.

EFFECT OF INTERPASS TEMPETATURE DURING THE WELDING OF IS2062-IS2062

1.AJAY RAO U.G 2. ANOOP RAMESH 3. AVINASH T 4. KARTHIK N 5. DR. B.S AJAY KUMAR 6. M.V CHANDRASHEKAR

1-4: 8th Semester students, Mechanical Department, Bangalore Institute of Technology5: Professor, Dept of Mechanical Engineering, Bangalore Institute of Technology

6: Assistant Professor, Dept of Mechanical Engineering, Bangalore Institute of Technology

ABSTRACT:

Through literature survey, it has been found that the established procedure for carrying out welding

operations as provided by ASTM (American Society for Testing and Materials) is not being followed

properly at welding sectors due to the increase in downtime and decrease in production during the

process.An attempt has been made in this investigation to find effectiveness of Interpass temperature

on efficiency of welding and comparing it with the one obtained by welding continuously, without

maintaining Interpass temperature, through Tensile and Impact tests.

LITERATURE SURVEY:“Interpass temperature” refers to the temperature of the material in the weld area immediately before the second and each subsequent pass of a multiple pass weld. It is the lowest temperature of the deposited weld metal before the next run is started. In practice, the minimum specified Interpass temperature is often equal to the minimum specified preheat temperature. In fact [1] Position Statement on the Northridge Earthquake recommends that Interpass temperature not exceed 550 F (290°C) when notch toughness is a requirement.

HOW INTERPASS TEMPERATURE AFFECTS THE QUALITY OF WELDS:(1) To minimize the risk of hydrogen cracking for carbon, carbon-manganese, and ferritic alloy steels, in which the minimum Interpass temperature is specified to be the same as the minimum preheating temperature.(2) To prevent deterioration of mechanical properties for carbon, carbon-manganese, and ferritic alloy steels, in which the maximum Interpass temperature is specified.

(3) To minimize the risk of solidification or liquation cracking for austenitic stainless steel, nickel and nickel alloys, and aluminum and aluminum alloys, in which the maximum Interpass temperature is specified.(4) To maintain good wetting of the molten pool onto the base metal for copper and copper alloys, in which the minimum Interpass temperature is specified to be the same as the minimum preheating temperature. [2]

PROCESS PARAMETRES:• Type of Weld : MIG Welding Down

hand Welding • Current supplied : 280 A- 310 A• Voltage supplied : 26 V- 30 V• Wire composition : Copper coated steel

wire• Shielding Gas : Carbon di Oxide• Wire Diameter : 1.2mm• Root gap : 14mm• Work piece length : 300 mm• Work piece thickness : 25mm• Total length of weldments : 360 mm • Interpass Temperature : ≤ 150º• Tip to work distance : 19±3 mm• Travel speed : 5.5±1 mm/sec• Electrode feed rate : 190±10 mm/sec

MIG WELDING:

COMPOSITION OF THE PLATE IS2062:

Element Grade B

% C : 0.22 max% Mn : 1.5 max% S : 0.045 max% P : 0.045 max% Si : 0.40 maxCarbon Equivalent 0.41

It is cold rolled mild steel.Tensile strength: 410 N/mm² Yield strength: 250

N/mm²Elongation: 23% minImpact strength:27J

PLATE EDEGE PREPARATION AND

WELDING PROCEDURE:

1. Beveling of 30 degrees from one edge

of each plate and grind a 2mm land on

the bottom of each bevel.

2. Tack weld the two plates together so

that the bevels form a V. The distance is

accurately set at 14mm using vernier

calipers while keeping it on a base

plate.

3. Provide backing support (anti-distortion

plate) to prevent distortion during

welding. Provide coupon to prevent

under fill and over fill during welding

operation.

4. Position the weldment for a down hand

welding (1G position).

5. Beginning at the bottom and using

1.2mm copper coated steel wire

electrode (maker Hyundai) weld a root

pass up the length of the V.

6. Complete the weld ensuring a superior

surface finish.[3]

TABULAR COLUMN:

Trail 1:I. The following table shows the measured value of Interpass Temperature for IS

2062- IS 2062 CONTINUOUS WELDING when carried out by Welder 1

PassCurrentin amperes

Voltagein volts

Timein minutes

Travel Speedin mm/second

Interpass tempin degrees

1 283 28.6 1.27.13 4 292 283 28.6 1.21.14 4.3 2363 283 28.6 1.26.26 4 2734 283 28.6 1.12.02 4.9 3685 283 28.6 1.02.95 5.3 4056 283 28.6 1.00.00 5.4 4147 283 28.6 1.09.48 5.1 4578 283 28.6 1.01.57 5.3 3509 283 28.6 1.04.91 5.2 36410 283 28.6 1.28.90 4 499

I. The following table shows the measured value of Interpass Temperature for IS

2062- IS 2062 INTERPASS WELDING when carried out by Welder 1

PassCurrentin amperes

Voltagein volts

Timein minutes

Speedin mm/second

Interpass tempin degrees

Time for coolingin minutes

1 290 30 2.10.73 3.2 29 182 290 30 1.30.32 4 124 193 290 30 1.18.60 4.6 130 224 290 30 1.49.43 3.4 136 235 290 30 1.31.33 4 145 226 290 30 1.23.77 4.3 85 237 290 30 1.46.68 3.5 145 258 290 30 0.58.96 5.5 145 209 290 30 1.17.84 4.6 145 2210 290 30 1.40.18 3.6 145 20

Trail 2:

II. The following table shows the measured value of Interpass Temperature for IS

2062- IS 2062 CONTINUOUS WELDING when carried out by Welder 2

IV.The following table shows the measured value of Interpass Temperature for IS 2062- IS 2062 INTERPASS WELDING when carried out by Welder 2

PassCurrentIn amperes

VoltageIn volts

TimeIn minutes

SpeedIn

mm/second

Interpass

tempIn degrees

1 310 29.5 1.45.28 3.4 302 310 29.5 1.32.70 3.8 2503 310 29.5 1.25.09 4.2 2844 310 29.5 1.32.80 3.8 2655 310 29.5 1.23.34 4.2 3106 310 29.5 1.26.86 4.2 2827 310 29.5 1.41.11 3.5 2958 310 29.5 1.20.34 4.3 3209 310 29.5 1.39.15 4 35010 310 29.5 1.40.15 4 250

TESTS CONDUCTED TO DETERMINE THE MECHANICAL PROPERTIES OF THE WELDMENT:

• RADIOGRAPHY: To check for internal defects like porosity and lack of fusion.

• TENSILE TEST: To find Tensile and Yield strength of the work piece.

TESTSPECIMEN DIMENSIONS: Initial gauge length: 50mm Outer diameter: 12.5mm

Inner diameter: 0mm

• IMPACT TEST: To find theimpact energy of the specimen.TESTSPECIMENDIMENSIONS: Type of Notch: V NotTest temp: RoomSize (mm): 10×10×55

TEST RESULTS AND OBSERVATIONS:

PassCurrentIn amperes

VoltageIn volts

TimeIn minutes

SpeedIn mm/second

Interpass tempIn degrees

Time for coolingIn minutes

1 310 29.5 1.39.25 3.6 36 19

2 310 29.5 1.45.89 3.4 147 20

3 310 29.5 1.21.36 4.5 143 22

4 310 29.5 2.01.36 3.2 145 23

5 310 29.5 1.56.45 3.1 88 24

6 310 29.5 1.45.98 3.4 146 22

7 310 29.5 1.23.65 4.5 145 21

8 310 29.5 1.49.25 3.3 145 22

9 310 29.5 1.35.69 3.7 143 21

10 310 29.5 2.58.87 3 149 22

It was found that the tensile strength increased by 10.93% and impact strength decreased by 4.45%.NOTE: Here root gap of 14 mm is taken for which the dilution from parent metal is minimal.

Hence this is the minimum increase in tensile energy. Drastic improvement in tensile energy can be noticed while Interpass concept is adopted by maintaining lesser root gap.

TENSILE TEST GRAPHS:TENSILE STRENGTH:

YIELD STRENGTH:

Trail

Condition % Variations

Without Interpass With Interpass IncreaseDecreas

eIMPA

CT ENERGY in J

TENSILE

STRENGTH in N/mm²

YIELD STRES

S in N/mm²

IMPACT

ENERGY in J

TENSILE

STRENGTH in N/mm²

YIELD STRES

S in N/mm²

TENSILE

STRENGTH

YIELD STRES

S

IMPACT ENERGY

1 182 541.77 430.32 169 596.55 502.62 9.19 14.39 7.14

2 174 555.37 437.98 171 635.92 533.75 12.67 17.94 1.75

TENSILEN/mm²

IMPACT TEST GRAPHS:

YIELDN/mm²

IMPACTN/mm²

CONCLUSION:Through experimental investigation it has been

found that maintaining an Interpass temperature

while welding IS 2062- IS 2062 test specimen

results in an increase of Tensile Strength of the

resulting weldment. It has also been found that

maintaining an Interpass temperature reduces the

Impact Strength of the weldment.

BIBLIOGRAPHY:

1. Mr. R. Scott Funderburk: Paper on

the importance of Interpass

Temperature in 1999. Later it was

published Welding Innovation, Vol.

XV, No.1, 1999.

2. American Welding Society (AWS)

3. Industrial Welding Procedures by Frank

R. Schell/ Bill Matlock [Unit 2 Page 7]

MODAL ANALYSIS OF AUTOMOBILE SUSPENSION SYSTEMVIKAS R.KHALKAR1*, LALITKUMAR M.JUGULKAR 2, BHARTESH

DANAWADE 3, SIDDHARTH WAGHMARE4 1Gharda Institute of Technology. Lavel. Ratnagiri-415 708, India

2 Rajarambapu Institute of Technology, Sakharale, Sangli3. Gharda Institute of Technology. Lavel. Ratnagiri-415 708, India

4 Rajendra Mane College of Engg. and Tech. Devrukh. Ratnagiri-415 804, IndiaCorresponding author (Prof. L. M. Jugulkar ,e -mail: [email protected], Ph. No. 9970700939, Currently

working as Assistant Professor in Automobile Engineering Department, Rajarambapu Institute of Technology, Sakharale, Tal- walwa, Dist-Sangli, Maharashtra, India )

ABSTRACTIt is necessary to determine the vibration characteristics of any system like an automobile or a machine, Vibration characteristics are nothing but determination of natural frequencies and mode shapes of a system at the initial stage while it is being designed. In this paper modal simulation of an automobile suspension system is carried out using finite element software ANSYS 10. Analysis is done by using combination of different elements like BEAM 3, MASS 21 and COMBINATION 14 (Spring –Damper).This paper deals with the verification of the classical results of Automobile Suspension System with the software and experimental results for checking its engineering design. An automobile suspension system is modeled by using lines, key points commands. Solid boundary conditions, Master Degree of Freedom are applied to the model for reducing the editing work and for more approximate analysis. We can learn how to save the time as well as money for the same. From the results obtained through the theoretical, FEA and Experimental analysis it can be seen that it is possible to design and develop an automobile suspension system quickly.

1 INTRODUCTION

One will use modal analysis to determine the natural frequencies and mode shapes of a structure or any system which is probably subjected to the dynamic loading conditions. They are also required if you want to do a spectrum analysis or transient analysis .You can do modal analysis on a prestressed structure, such as a spinning turbine blade. Any nonlinearities, such as plasticity and contact (gap) elements, are ignored even if they are defined. The purpose of analysis software is to do different types of analysis i.e. Static, Modal, Thermal, Eigen Buckling, Transient, Harmonic, Coupled field, Spectrum, CFD etc. Hence varieties of analysis are possible in ANSYS. Any given problem can be formulated and solved by using suitable elements, real constant, material properties, appropriate boundary conditions and loading. Hence no need to create experimental model for the same for checking its engineering design. Thus it is obvious that such software can increases the productivity of Design and Development department.

2 CASE STUDY

An automobile suspension system is simplified for considering only two major motions of the given system: 1 up and down linear motion of the body, 2 pitching angular motion of the body. Assume the body is idealized as a lumped mass with weight W and radius of gyration r, determine the corresponding coupled natural frequencies f1 and f2.In case of vehicle, when the brakes are applied on a moving vehicle, two motions of a vehicle occur simultaneously [1,6]: one motion is rectilinear(x) and other motion is angular (θ). This type of combined motion in the system is due to the fact that C.G. and the centre of rotation do not coincide. Consider a body of mass ‘M’ and moment of inertia ‘I’ supported on two springs K1 and K2 as shown in figure. Let at any instant the body is displaced through a rectilinear distance (x) and the angular displacement (θ) as shown in fig. since the’ θ’ is small, therefore springs K1 and K2 are compressed through a distance ( x- l1 θ) and (x+ l2 θ) respectively. The free body diagram of the system as shown in above figure.The differential equations for the above system is as shown in figure are as follows

M x¨ + K1( x- l1 θ) + K2(x+ l2 θ) = 0 (1) I θ¨ + K1( x- l1 θ) l1 + K2(x+ l2 θ) l2 = 0 (2)

The equations above have both the terms x and θ, hence such equations can be called as coupled differential equations. From the above equation is cleared that an automobile system has angular and translatory motion.Special cases of the system, Case 1

WhenK1l1=K2l2 (3)Consider case 1: When K1 l1= K2 l2, the equations becomesM x¨ + (K1 + K2) x = 0 (5)

I θ¨ + (K1 l12

+ K2 l22) θ = 0

(6)

From above equations it is seen that, rectilinear and angular motions can exists independently. Therefore, such equations are called as uncoupled differential equations. The natural frequencies of such system are as follows

M x¨ + (K1 + K2) x = 0ω n1= [(K1 + K2) / M]1/2

(7)

I θ¨ + (K1 l12 + K2 l2

2) θ = 0 ω n2 = [(K1 l12 + K2 l2

2) /I]1/2

(8)

3 THEORETICAL ANALYSIS Calculations for frequencies:M, Mass of vehicle =3.950 kg, kf = 10200N/m, kr = 10200N/m, IT, Torsional moment of Inertia =

Mr2= 3.950 x0.12=0.0395 kg-m2

l1=0.112 m, l2=0.112 m

ω n1=[ (Kf + Kr) / M]1/2………… = [(10200 + 10200) / 3.950]1/2

= 75.36 rad/secf 1= ω n1/ 2π = 75.36 /2π = 11.43 HZ

ω n2=[ (Kf l12 + Kr l2

2) / I]1/2

= [(10200* 0.1122 + 10200 * 0.1122) / 0.03950]1/2

= 80.48 rad/ secf 2= 80.48/ 2π = 8.91/2π = 12.81 HZ

4.EXPERIMENTALANALYSISThe experimental set up of an automobile suspension system [7] is for different frequencies are as shown in fig.3 and 4. The natural frequencies are determined experimentally by using FFT Analyser, Accelerometer etc.

Fig.3, Experimental set up of AutomobileSuspension System for Bouncing

Fig.4, Experimental set up of Automobile

Suspension System for Pitching

Graph 1

Graph 2

5 CAE ANALYSIS

ELEMENTS DETAILS: 1. BEAM3: is a uniaxial element with tension,

compression, and bending capabilities. The element has three degrees of freedom at each node: translations in the nodal x and y directions and rotation about the nodal z-axis.

Special Features :Stress stiffening, Large Deflection,

Birth death

COMBIN14: Spring-Damper: COMBIN14 has longitudinal or torsional capability in 1-D, 2-D, or 3-D applications. The longitudinal spring-damper option is a uniaxial tension-compression element with up to three degrees of freedom at each node: translations in the nodal x, y, and z directions. No

bending or torsion is considered.

MASS21: Structural Mass: MASS21 is a point element having up to six degrees of freedom:

translations in the nodal x, y, and z directions and rotations about the nodal x, y, and z axes. A different mass and rotary inertia may be assigned to each coordinate direction.

CAE RESULTS

6 Result Comparison: Table 1Target Ansy

sExperimental

Bouncing Frequency, f 1

11.43 11.43 9.76

Pitching Frequency, f 2

12.81 12.81 12.89

7 CONCLUSION:Ansys and Experimental Results result has been compared with the Classical Results and through the comparison it is cleared that results obtained by all the methods are approximate to each other. Hence one can conclude that CAE and Experimental method of checking engineering design are correct.

REFERENCES:

1 W. T. Thomson, Vibration Theory and Applications, 2nd Printing, Prentice-Hall,

Inc., Englewood Cliffs, NJ, 1965, pg. 181, ex. 6.7-1

2 Ansys 10.0, Release Guide

3 Introduction toFiniteElementsinEngineering- Tirupati R.Chandrapatla & Ashok D. Belegunda

4 Introductory course on Theory and Practice of Mechanical Vibrations, 2nd Edition, pg. 203-206

5 S. Timoshenko vibration problems in Engg. S Van to strend Co.Inc. New Jersy. Third Edition, 1955

6 S.S. Rao, “Mechanical Vibration”, 4th

edition prentice, Hall 2003.

7 Automobile Engineering by Kirpal Singh. 8th edition, Technical Chassis Specifications

PRODUCTIVITY IMPROVEMENT POTENTIAL ANALYSIS THROUGH ERGONOMIC INTERVENTION IN ARC WELDING

S.MALIKRAJResearch Scholar/ Bharathidasan University

[email protected]

DR.A.K.GANGULYDGM/OHS BHEL/Trichy

[email protected]

DR.T.SENTHIL KUMAR Director-CUIC./ ANNA University/Trichy

[email protected]

ABSTRACT :Improvement in productivity can only be achieved by means of Systematic analyses and optimization of production processes in welding. Capable Production processes and ergonomic interventions may provide the basis for the Manufacture of welded products of consistently with high quality and productivity. This can be approached by determining the process parameters , proper outcomes measures , outcomes comparison with the goals, collecting new improvement suggestions, then make applications and provide progress by continuous assessment by questioning and other forms.

1. INTRODUCTION

Productivity Improvement and Ergonomic InterventionProductivity failures have various reasons in the aspects of Materials, process, Employee performance, Employee’s health, illness, motivation, management perspective, etc… As per this study, employee is important most considerable factor in the production point of view. However Productivity improvements is a series of actions taken to identify, analyze and improve existing system within an organization to meet new goals and objectives. These actions often follow a specific methodology or strategy to create successful results. Ergonomic interventions may be one of these actions to be used for this purpose. The benefits we can derive by

improving ergonomics in the workplace leads to a smoother functioning, more efficient work environment that can:

• Improve employee comfort,• Reduce fatigue• Reduce RSI (repetitive strain injuries)• Reduce accidents• Increase employee satisfaction

And also in detail ,Ergonomic interventions can reduce the incidence and impact of musculoskeletal injuries, illnesses, and disorders. Primary prevention occurs when the intervention is undertaken before members of the

• population at risk have acquired a condition of concern,

Ergonomic interventions is the application of the body of knowledge (about human abilities, human

limitations and human characteristics that are relevant to design) to the design of tools, machines, systems, tasks, jobs, and environments for safe, comfortable and effective human use. (Board of




Certification for Professional Ergonomists (BCPE), 1993)Covering a brief introduction to ergonomics; the ergonomics, engineers and designers; occupational health & safety problems arising from ergonomic deficiencies - namely postural or Work Related Upper Limb Disorders(WRLUD). A basic treatment of practical/applied human anatomy and physiology, concentrating on the musculoskeletal system, limb movements and their range, soft tissues and their structures, muscular fatigue, posture, etc

RISK FACTORSThe three main categories of work related physical risk factors (repetition, force and posture) which are known or suspected to be involved in these conditions are worked through, along with the implicated issues of the work environment, physical constraints, task design, synergy and psychosocial factors.

DATA GATHERING Building on the practical knowledge, the researcher planning to apply some tools or methodologies which can assist in examining for WRUL or related disorders. In particular, they will deal with the use of the Rapid Upper Limb Assessment tool or methodology and how to interpret its resultsThe primary work characteristics that are called ergonomic risk factors include task physical characteristics (interaction between the worker and the work setting; posture, force,velocity / acceleration, repetition, duration, recovery time, heavy dynamic exertion, and segmental vibration) and environmental characteristics (interaction between the worker and the work environment; heat stress, cold stress, whole body vibration, lighting, and noise).After evaluating and controlling the work risk factors, assessing the workplace for ergonomic risk conditions generally involves two steps of

identification of the existence of ergonomic risks and quantification of the degree of ergonomic risk. Engineering, administrative and work practice

controls are included in prevention and control of ergonomic risk conditions. Management and workers are educated to risk conditions (ergoweb, 2007).An improvement of working conditions can be a difficult objective. Notable advances are easy to achieve in the field of routine fabrication of small-to-medium-size items, but the situation is different with the fabrication of large one-off constructions (Gonnet, 1995).Klatte and others (1997) emphasized standardized, ergonomically designed work places and capable processes as the basis for improvement in production performance Govindaraju and others (2001) stated that ergonomic considerations can improve human performance, but compounding the problem is the current inability of most ergonomists to make ergonomic recommendations that do not run counter to the productivity and quality goals of system designers. That’s why, ergonomic studies are all the more efficient as they are preventive (Gonnet, 1995). Several ergonomic studies of particular importance have been carried out so far. Schaub et.al. (2000) introduced a screening tool “Design Check” jointly developed by several institutions which allows the detection of ergonomic deficits in workplace layouts, and tested it at the assembly lines of automotive industry. Mavrikios et al. (2006) discussed human motion modeling using experimental motion data analyzed by a statistical design of experiments approach. ANOVA was performed for the determination of the impact factor of the anthropometric parameters influencing the motion path. Gonzales et al. (2003) analyzed the quality results of a selected firm after varying the initial work method on de basis of the results of an ergonomic evaluation and evidenced Occupational injury and illness reduction Workers’ compensation costs containment ,Productivity improvement , Work quality improvement , Absenteeism reduction, Government regulation, Compliance Evaluation and control of risk factors, Identification & quantification of risk conditions, Recommendations of controls to reduce risk conditions, Education to risk conditions statistically an improvement thereon. While Eklund (1995, 2000) focused on quality and ergonomics in

general, various industrial applications were executed by other authors. Yeow and Sen (2003) conducted an ergonomic study to improve the workstations for electrical tests in a printed circuit assembly in an industrially developing country. The interventions implemented were simple and inexpensive but

resulted in many benefits. Cost effectiveness of ergonomics and quality improvement in electronic manufacturing was studied by Helander and Burri (1995). Quinn et al. (1991) redesigned ergonomically the packaging workstation for automotive component parts. Sen and Yeow (2001) attained ergonomic

improvements for the manufacture of printed circuit assemblies through design changes. Yeow and Sen (2000) discussed ergonomic improvements of workstations for visual inspection and electrical tests in a multimedia product factory. Yeow andn Sen (2002) investigated quality, productivity,occupational health and safety, and cost effectiveness of 5 ergonomic studies in electronic factories. Apart from these works, an original process improvement case study performed through ergonomic design in automotive industry.

2. METHODS2.1. Population study and data collectionA potential study was performed among all the managers and welding workers (excluding other workers) at 85 Engineering industries involved in non pressure parts manufacturing. A total of 110 managers and welders were asked to participate in the study. At baseline, the managers of the 85 industries were asked to fill out a questionnaire (objective type) about the present status of the company in view of Production rate , Skilled employees availability, welding, quality failures, production work place design, Employee turnover, fitness of EMPLOYEES, safety hazards associated with being welder. Time for which seated in welding job, Low back trouble, the reasons of taking leave, welder’s efficiency, Back pain ,Sleeplessness, Indigestion, Irritability, Aching Muscle, Neck pain, Shoulder pain, Arm Pain, body pain2.2METHODOLOGY2.2.1 Baseline Study A total of 64 managers answered the questionnaire, and 78 consented to distribute a questionnaire to the workers of their company.

An identical questionnaire was sent to these 85 managers later. A total of 7 companies were excluded in this follow-up survey because the company no longer had any employees or because the manager did not return the questionnaire. At baseline , Outcome as 86 % of respondents never achieve OPTIMAL PRODUCTION RATE; 81% Welder is not Performing well ; Welder efficiency is only in between 50- 75%; 72% Employees (welders) taking leave more than the eligible no. of days , in that 58% of employees taking Leave only due to ILLNESS , Specifically 100% employees

suffered by BODY PAIN & 45% companies responded employees suffered by EYE RRITATION;Employees fitness is only on MODERATE LEVEL ( that is100%);hazards associated with being welder is 100% radiation ;58% high Employee Turnover;78% of Layout are not designed as per requirement;Porosity is major failure always occur in weld due to employees carelessness..; 90% employees are only not qualified and trained in welding ,and only 10% employees are qualified & well Trained.

2.2.2 ANALYSIS.This base line questionnaire analyzed by using CHI-SQUARE TEST ANALYSIS ( Ψ2 –test) .It is non parametric test, measure of comparing experimental results with expected theoretical results based on hypothesis

NULL Hypothesis H0 : Employee performance ( productivity) independent of above five factors…

Computation:

Qn No.

Details of Qn. YES NO

1 Optimal Production

13.2 22.8

3 Employee Taking Leave

11 Employee turnover

12 Workplace design

14 Employees qualified &trained

Qn No.

Details of Qn.YES NO

1 Optimal Production 5.09 2.94

3 Employee Taking Leave

12.41 7.18

11 Employee turnover

4.6 2.66

12 Workplace design

2.04 1.18

14 Employees qualified &trained

3.92 2.27

Ψ 2 = 44.29

Degree of Freedom = ( 5-1)*( 5-1) = 16

DECISION :The tabulated Value of Ψ2 at α = 0.20 and 16 degree of freedom is Ψ2 = 20.265

The tabulated Value of Ψ2 at α = 0.001 and 16 degree of freedom is Ψ2 = 39.252

Since calculated Ψ2 = 44.29 > tabulated value Ψ2 = 20.265 => the NULL Hypothesis is rejected =>: Employee performance ( productivity) is not independent of above five factors…

Hence productivity is dependent of above five factors in which employees performance is very important aspect which should be considered and concentrated for reviewed in the research .

Ergonomic Interventions in the Workplace look at the assumption that prevention strategies in the workplace can reduce the incidence and impact of musculoskeletal injuries, illnesses, and disorders. Primary prevention occurs when the intervention is undertaken before members of the population at risk have acquired a condition of concern, for example, educational programs to reduce the number of new cases (incidence) of low back pain. Secondary prevention occurs when the intervention is undertaken after individuals have experienced the

Details YES NO

Optimal Production 05 31

Employee Taking Leave 26 10

Employee turnover 21 15

Workplace design 08 28

Employees qualified &trained 06 30

TOTAL 66 114

condition of concern, for example, introduction of job redesign for workers with symptoms of carpal tunnel syndrome. Tertiary prevention strategies are designed for individuals with chronically disabling

musculoskeletal disorders; the goal is to achieve maximal functional capacity within the limitations of the individual's impairments. In this we review the scientific details about interventions, evaluation and

control of work risk factors, identification and quantification of existing work site risk conditions,

Efficiency

From this chart , it is inferred most welders have efficiency only in the range of 50%- 75%. Then what are the reasons there made of is identified as Work Risk Factors which have Certain characteristics of the work setting have also been associated with WRULD and it include:

POSTUREPosture is the position of the body while performing work activities. Awkward posture is associated with an increased risk for injury. It is generally considered that the more a joint deviates from the neutral (natural) position, the greater the risk of injury. Posture issues can be created by work methods (bending and twisting to pick up a box; bending the wrist to assemble a part) or workplace dimensions (extended reach to obtain a part from a bin at a high location; kneeling in the storage bay of an airplane because of confined space while handling luggage).

Specific postures have been associated with injury. For example:

Wrist : Flexion/extension (bending up and down) , Ulnar

/radial deviation (side bending) Shoulder : Abduction/flexion (upper arm positioned out to the

side or above shoulder level) , Hands at or above

shoulder height Neck: flexion/extension or bending the neck forward and to

the back ,side bending as when holding a telephone

receiver on the shoulder Low back : Bending at the waist, twisting

FORCETask forces can be viewed as the effect of an exertion on internal body tissues (e.g., compression on a spinal disc from lifting, tension within a muscle/tendon unit from a pinch grasp), or the

physical characteristics associated with an object(s) external to the body (e.g., weight of a box, pressure required to activate a tool, pressure necessary to snap two pieces together). Generally, the greater the force, the greater the degree of risk. High force has been associated with risk of injury at the shoulder/neck (Berg et al., 1988), the low back (Herrin et al., 1986), and the forearm/wrist/hand (Silverstein et al., 1987). It is important to note that the relationship between force and degree of injury risk is modified by other work risk factors such as posture, acceleration/velocity, repetition, and duration.

Two examples of the interrelationship of force, posture, acceleration/velocity, repetition and duration are:

1. A 20-pound weight lifted in a smooth, slow manner one time directly in front of the body from a 28 inch shelf to a 32 inch shelf will be much less of a risk than a 20-pound weight lifted quickly 60 times in 10 minutes from the floor to a 60 inch shelf.

2. A 45-degree neck flexion position held for one minute will be much less of a risk that a 45-degree neck flexion position held for 30 minutes.

Better analysis tools (e.g., 1991 Revised NIOSH Lifting Equation) recognize the interrelationship of force with other risk factors relative to overall task risk.Five additional force-related injury risk conditions have been extensively studied by researchers and ergonomists. They are not "rudimentary" risk factors. Rather, they are a workplace condition that presents a combination of risk factors with force being a significant component. Their common appearance in the workplace and strong association with injury prompts their introduction here.

GLOVES

Depending on material, gloves may affect the grip force generated by a worker for a given level of muscular exertion. To achieve a certain grip force while wearing gloves, a worker may need to generate greater muscular exertion than when not wearing gloves. Greater force is associated with increased ris k of injury.

VELOCITY/ACCELERATION

Angular velocity/angular acceleration is the speed of body part motion and the rate of change of speed of body part motion, respectively.

REPETITION

Repetition is the time quantification of a similar exertion performed during a task. A warehouse worker may lift and place on the floor three boxes per minute; an assembly worker may produce 20 units per hour. Repetitive motion has been associated with injury (Hagberg, 1981; Armstrong et al., 1982) and worker discomfort (Ulin, 1990). Generally, the greater the number of repetitions, the greater the degree of risk. However, the relationship between repetition and degree of injury risk is modified by other risk factors such as force, posture, duration, and recovery time. No specific repetition threshold value (cycles/unit of time, movements/unit of time) is associated with injury.

DURATION

Duration is the time quantification of exposure to a risk factor. Duration can be viewed as the minutes or hours per day the worker is exposed to a risk. Duration also can be viewed as the years of exposure to a risk factor or a job characterized by a risk factor. In general, the greater the duration of exposure to a risk factor, the greater the degree of risk.

HEAT

stressHeat stress is the total heat load the body must accommodate. It is generated externally from environment temperature and internally from human metabolism.

Excessive heat can cause heat stroke, a condition that can be life threatening or result in irreversible damage. Less serious conditions associated with excessive heat include heat exhaustion, heat cramps, and heat-related disorders (e.g.,

3.0 ASSESSMENT METHOD AND TOOL

RULA - A RAPID UPPER LIMB ASSESSMENT TOOL

Rapid Upper Limb Assessment (RULA) is a survey method developed foruse in ergonomic investigations of workplaces where work related upper limb disorders are reported. RULA is a screening tool that assesses biomechanical and postural loading on the whole body with particular attention to the neck, trunk and upper limbs. Reliability studies have been conducted using RULA on groups of VDU users and sewing machine operators. A RULA assessment requires little time to complete and the scoring generates an action list which indicated the level of intervention required to reduce the risks of injury due to physical loading on the operator. RULA is intended to be used as part of a broader ergonomic study.

Step1: Observing and selecting the posture(s) to assess

A RULA assessment represent a moment in the work cycle and it is important to observe the postures being adopted whilst undertaking the tasks prior to selecting the posture(s) for assessment. Depending upon the type of study, selection may be made of the longest held posture or what appears to be the worst posture(s) adopted. In some instances, for example when the work cycle is long or the postures are varied it may be more appropriate to take an assessment at

regular intervals. It will be evident that if assessments are taken at set intervals over the

working period the proportion of time spent in the various postures can be evaluated.

Step2:Scoring and recording the posture

Decide whether the left, right or both upper arms are to be assessed. Score the posture of each body part using the software. Review the scoring and make any adjustments if required. Select calculation button.

Step3:ActionLevel

The grand score can be compared to the Action Level List however it must be remembered that since the human body is a complex and adaptive system, they provide a guide for further action. In most cases, to ensure this guide is used as an aid in efficient and effective control of any risks identified, the actions lead to a more detailed investigation.

The RULA action levels give you the urgency about the need to change how a person is working as a function of the degree of injury risk.

Action level 1 - RULA score 1-2 means that the person is working in the best posture with no risk of injury from their work posture

Action level 2 - RULA score 3-4 means that the person is working in a posture that could present some risk of injury from their work posture, and this score most likely is the result of one part of the body being in a deviated and awkward position, so this should be investigated and corrected.

Action level 3 - RULA score 5-6 means that the person is working in a poor posture with a risk of injury from their work posture, and the reasons for this need to be investigated and changed in the near future to prevent an injury

Action level 4 - RULA score 7-8 means that the person is working in the worst posture with an immediate risk of injury from their work posture, and the reasons for this need to be investigated and changed immediately to prevent an injury

Guidelines For A Good Working Posture While Welding

• Learn to recognize symptoms of work-related musculoskeletal disorders (WMSDs; also called repetitive strain injuries or RSIs). Repeated uncomfortable postures and tasks can cause injury.

• Avoid awkward body positions which cause fatigue, reduce concentration and lead to poor welds which may need to be repeated.

• Always use your hand to lower your helmet. Do not use a "jerking" motion of your neck and head.

• Position yourself in a stable, comfortable posture. • Position the welding item as flat as possible, on a

horizontal surface, between waist and elbow height.

• Position scaffolding at a comfortable height to allow working in a seated position.

• Avoid working in one position for long periods of time.

• Work with material slightly below elbow level when working in a sitting position.

• Work with material between waist and elbow heights for comfort and precision when working in a standing position.

• Always store materials and tools within normal reach.

• Use positioning aids to accommodate work posture

4.0 CONCLUSIONS:Productivity Analysis outcome support the value of a programmatic move towards the ergonomics. •A successful

ergonomics program requires the following key items: Management support , ergonomic experts–Assessment of needs -Ergonomic evaluations (facility, workstation),Ergonomic intervention ,corrective action, Training (management and employees),Tracking progress of program and individuals through surveys etc, Employee participation. 5.0 REFERENCES:

Jung, E., D. Kee, and M. Chung (1995). Upper body reach posture prediction for ergonomics evaluation models. International Journal of Industrial Ergonomics 16, 95–107.

Prentice, M. (1986). Orientation statistics without parametric assumptions. JRSS-B 48, 214–222.

Chong, I. (1996, March/April). The economics of ergonomics. WorkplaceErgonomics, pp. 26–29.

Mallett, R. (1995, July). Human factors: Why aren’t they considered?Professional Safety, pp. 30–32.

Eklund, J.A.E. (1995), “Relationship between ergonomics and quality in assembly work”,Applied Ergonomics, Vol. 26 No. 1, pp. 15-20.

Eklund, J.A.E. (2000), “Development work for quality and ergonomics”, AppliedErgonomics, Vol.31 No. 6, pp. 641-648.

www.ergoweb.com/resources/faq/concepts.cfm , viewed 15 January 2007.

ANALYSIS OF PRESSURE RELIEF VALVE USING

MAT LAB SIMULINK

DR. D.N.RAUT(Professor) [email protected] (9869699909) VJTI Mumbai

MEHUL S PRAJAPATI (P.G Student)[email protected] (9819292364) VJTI Mumbai

ABSTRACT:The goal of this paper is to investigate the principles and practical mechanics involved in the design of a hydraulic valve, using a computational software package. A pressure relief valve is chosen as the valve type to be modelled, and the MATLAB software package including the SIMULINK application is used to model the system. This report contains the assumptions, variables, mathematical modelling and analysis involved in modelling and simulating pressure relief valve.

I.INTRODUCTION

The world of hydraulics provides the technology to the industry which is ever improving and more complex. In order to reduce the human errors the engineering fraternity has taken a step forward towards hydraulics system. Fluid power systems are designed to operate within a preset pressure range. This range is a function of the forces the actuators in the system must generate to do the required work. Without controlling or limiting these forces, the fluid power components (and expensive equipment) could be damaged. Relief valves avoid this damage.

Fig 1

A simple relief valve is the ball sitting on the seat with

a



spring keeping it closed (Fig 1) until the area of the seat is enough to allow the valve to open and pass the fluid.The flow capacity is limited by the size of the seat and the pressure difference across the opening. To get more flow across the valve the ball has to move further back against the spring increasing the force and required pressure.

The pressure relief valve is used in almost every hydraulic system (Fig 2). The function of the relief valve is to limit the maximum pressure that can exist in a system. Under ideal condition, the relief valve should provide alternative flow path to tank for the system fluid while keepingsystem pressure constant.

There are two types of such valve that arecommercially available: direct and pilot type.Direct Acting relief valves are the simpler of the two and. The advantage of the direct acting relief valve is that its construction is simpler, and therefore the cost is not as much as the pilot operated relief valve.

The pilot operated relief valve has two “stages”. The first stage is the pilot stage. This stage contains the pressure-limiting device, basically a small direct acting relief valve, controlled by a spring-loaded poppet. The second stage is the balanced piston stage. In this stage, the user adjusts the pre-load set on the pilot spring, and this in turn regulates the force necessary to displace the poppet spring

II.OPERATION& CONSTRUCTIONA relief valve functions by diverting flow through a movable poppet. The poppet is the name given to the tip of the spool – the part that is acted on by the fluid pressure. The movement of the poppet is dependent upon the force imposed on the poppet by a spring and the force experienced over the poppet area from fluid pressure. The valve chosen was DRV08-3W-P-05 Direct-Acting relief valve from TucsonHydrocontrols. A diagram of the valve is displayed below in figure 3.

Fig 3 Cross Section of Valve

Specification:

Operating pressure: - 210 bar (3000psi). Maximum flow: -DRV08-3W-P-05 8gpm. Pressure range:5bar(75psi)-50bar(800psi). Knob screw: - M10 x 1.25. Density (hydraulic fluid):- 0.03303 lbm/in3

This particular valve has a knob adjustment for the spring displacement, which allows for 5 turns of adjustment. The knob turns on a M10 screw with 1.25 mm Pitch, which means each turn on the knob is 1.25mm (0.049 in) in spring displacement. The maximum spring displacement allowed is 0.246 inches. The cartridge material, internal poppet and balancing piston are made of stainless steel.

Lock nut

Knob screwPusher

Spring

Cartridge

Poppet

http://www.tucsonhydrocontrols.com/

http://www.tucsonhydrocontrols.com/pdf/010107.pdf

Asumptions:-

Mass of the spring was negligible, hydraulic fluid is 100% incompressible

III.MATHEMATICAL EQUATIONS

The valve’s cylindrical cavity consists of poppet, pusher, o-ring and spring. The poppet is forced by the spring against the inlet orifice for preventing fluid movement through the inlet. The spring force acting the poppet Fs is given by

Fs = K. Δx (1)

Where K is the spring constant in lbf/inch, and ∆x is the spring displacement. An initial preload is placed by the spring on the poppet against the cracking pressure fluid to displace the poppet. Opposite to the spring force is the fluid pressure ‘P’ acting on the poppet with force Fp, where

Fp = P . Apoppet (2)

Here, Apoppet is the area of the horizontal projection of the poppet tip area (D=4.5mm). The orifice area is in the form of a ring around the poppet which allows fluid to enter the relief chamber. If Fp < Fs, the fluid flow continues to bypass the relief valve and flow continues as normal through the rest of the system. If Fp > Fs,the poppet will displace, and fluid will enter the relief chamber and flow back to the reservoir. This allows some or even all of the fluid to continue flow, and reduce the pressure in the system. Flow rate through the orifice can then be found using the orifice equation

Q =Cd Aorifice .sqrt 2( Pa – Pb ) / ρ (3)Where,

Cd =0.7 – 1.0,the orifice equation constant ρ = The fluid density

Pa = Fluid Pressure Pb = Valve exit pressure Aorifice = Area of the orifice

A. Spring Force

The spring force is a function of the number of turns of the knob on the valve. As the knob is turned, the spring is compressed increasing the force that is exerted on the poppet, varying the cracking pressure. The model compares this spring force to the force exerted by the fluid to determine whether or not the spool moves.

For 1 turn the displacement is

(1turn)

= 0.05” (5)

To convert the displacement into the force of the spring, we used the equation:

Fs= Fi + k * Δx (6)

Fi is the initial cracking pressure before the knob is turned times the area of the poppet, k is the spring constant, and Δx is the displacement of the spring. To find the value of k we assumed that the maximum cracking pressure would occur when the spring reaches maximum compression.

Therefore, Fs = maximum cracking pressure times the area of the poppet

Fs = (50 bar) (0.159 mm2)

= 7.95 kgf (17.5 lbs)

Fi = (5 bar) (0.159 mm2)

= 0.795 kgf (1.749 lbs)

Δx = 0.25”

Using above equations k= 63

lbs/in (1.13kgf/mm).

IV.MATHEMATICAL MODELLING

The first step in determining the model is how far the spool moves. To determine this, a transfer function is derived from the free body diagram of the spool as shown below.

X denotes the displacement w.r.t reference

a) The relation between force and mass is

F α acceleration

F = M x a

F = M d²x/dt²

F = M d²/dt² (xi – x0)

b) Since spring undergoes elastic deformation

F α displacement

F = k x

F = k (xi – x0)

c) Since dampness is introduced in the system due to spring so

F α velocity

F = B dx/dt

F = B d/dt (xi – x0)

Where B=Damping constantThis gives the transfer function:

X = (Fs – Fp)/ (MS² + BS + K)

M is the mass of the spool. This was estimated as 0.01 lbs, based on the dimensions of the spool.

B = 2ζ sqrt (K x M)

= 2 x 1.1 sqrt (63 x .01)

(ζ=1.1 to represent a slightly over damped system)

B = 1.746 lbs/s.

Hence the input to the transfer function is the difference between the spring force and the force due to the pressurized fluid in the system. The output of the transfer function is the displacement of the spool in inches.

Due to the positive displacement of the poppet the fluid can pass through the open orifice. Hence a flow rate can be found out using the orifice equation. The inputs (inlet pressure, the outlet pressure, and the spool displacement) will be used to determine the flow through the orifice of the relief valve using the orifice equation:

(8)

Fig 5

The orifice area A(x) is dependent on spool displacement

F1Poppet B

M

K

Fp

X

Fig 4

(9)

The inlet radius (R) of the fluid acting on the poppet was measured and found to be 0.0078 inches. Xspool is the displacement of the spool. Theta was found and equal to 45 degrees. The discharge coefficient is equal to A2/Ao where Ao is the area of the orifice and A2 is the area after the fluid has passed through the orifice. Cd that used was 0.7.

IV. SIMULINK® MODEL

The test system was constructed using SIMULINK®. The number of turns, system pressure, and the Tank pressure of the relief valve were modelled as the inputs to the system. Using MATLAB® functions and transfer functions the model is operated.

A diagram of the SIMULINK® model is shown below in figure 5. The system is defined using flow chart with inputs and output parameters at each junction.

__

No of

TurnsSpring

ForceSpring

PressureSystem

Pressure

Force

difference

Cracking

Pressure

Spool

DisplacementFlow through

Valve (Q)

Tank

Pressure

+_

Displacement Transfer

Function

FUNCTIONS

OUTLET

INLET

V. ANALYSISThe valve model was ready for analysis and simulation. Certain parameters like system inlet pressure, tank pressure, spring stiffness were kept constant. Analysis was consisted with changing the number of turns on the relief valve and observing the flow rate through the relief valve and the pressure to the relief valve. Changing flow rate and pressure responses for 0,1,2,3,4, and 5 turns are displayed below.

VI. CONCLUSION

Summarizing the above facts it may be concluded that the important parameters like spring stiffness, Orifice Area, No of turns should be designed precisely. A keen understanding of how a valve operates is necessary in order to effectively model it. SIMULINK® was used to analyze this mathematical

Flow through

Valve (Q)

Tank

Pressure

system. Several were then run to confirm that our model matches results seen in the physical world.

It is further concluded that this study provides a good base upon which an inter-action study between the other components and such a relief valve of a complete hydraulic system can be initiated. The model can be used to predict trend under various loading conditions that are difficult to create experimentally. This consideration is important in future expansion of the study and loads.

VII. REFERENCES

1) Fluid Power Control Laboratory Adam Kalthoff . Andy Marass ,Shelley FabryCameron Muelling.

2) Dynamic analysis of proportional solenoid controlled piloted relief valve by bondgraph K. Dasgupta a, *, J. Watton b

3) H.E. Merrit, Hydraulic Control Systems, Wiley, New York, 1967

4) K. Dasgupta, R. Karmakar, Dynamic analysis of a pilot operated pressure relief valve, SimulModell. Practice Theory 10

5) Tucson Hydrocontrols product manual

6) Integrated hydraulics product manual

7) Kevin N. Otto, 2001, Product Design, Pearson Education,inc

8) Thermohydraulic effects of safety relief valvesA.A. Kendoush a,*, Z.A. Sarkisa, H.B.M. Al-Muhammedawi.

9) Numerical Method for Engineers using Matlab Schillimg & Harrris

10) Feed Back Control System Dr. S.D.Bhide,S.Satyanarayan,N.A.Jalgaokar.

11) Matlab An Introduction with Applications Amos Gilat.

DEVELOPMENT OF ASPECT – IMPACT MATRIX DURING THE IMPLEMENTATION OF ENVIRONMENTAL MANAGEMENT SYSTEM FOR COMMON EFFLUENT TREATMENT PLANT AT

KAGAL, DIST. KOLHAPUR, MAHARASHTRA. (INDIA)SUNIL SHIVAJI MANE

Lecturer, K.I.T.’s College of Engineering Kolhapur. Maharashtra(India) Shivaji University, Kolhapur.M.Tech. (Env.Sci.&Tech) E-mail- [email protected] Mobile No.-09423859404

PROF.DR.GIRISH KULKARNIDirector of Technology Department in Shivaji University, Kolhapur. Shivaji University, Kolhapur. PhD.(Energy)

E-mail- [email protected] Mobile No.-09823180139



INTRODUCTIONImplementation of the system like ISO14001 Environmental Management System (EMS) will improve the firm’s environmental performance. The case study is carried out for the implementation of the ISO14001 to the Common Effluent Treatment Plant at Five Star M.I.D.C., Kagal, Dist. Kolhapur. The bulk of industrial pollution in India is caused by the small scale industrial (SSI) sector. A small scale unit is defined as any industry whose plant and machinery are valued at less than 1 crore (Government is planning to increase this to 5 crores). Small Scale Industry account for over 40percent of the total industrial output in the country and generate over 44percent of hazardous wastes alone as compared to 13percent generated by the large scale industry. Common facilities and common effluent treatment plant (CETP) is widely believed as solution to the effluents problem for clusters of Small Scale Industries. World Bank under its "Pollution Prevention Programme" is promoting Common Effluent Treatment Plants as a viable solution to control industrial pollution. All is not well even with the Common Effluent Treatment Plants. There are very few Common Effluent Treatment Plants, which have been successful in tackling the water pollution problems from Small Scale Industries. A new approach to environmental protection is now available thanks to the development of new international standards on environmental management, in particular, ISO 14001. This approach relies less on command-and-control dictates from the Government and more on proactive efforts by all workers in the company.

The implementation of the environmental management system prescribed by ISO 14001 can lead to good compliance with environmental legislation and tangible, continual improvement in the environmental performance of enterprises thanks to the commitment and evolvement of top

management and all workers. Widespread implementation of these standards can go a long way toward improving the environmental performance of industry and promoting sustainable development in the countries of the world. ISO 14000 is a generic series of standards devised to harmonize environmental requirements for companies worldwide. The Swiss-based International Organization for standardization (ISO), under the guidance of ISO Technical Committee 207(TRC207) created the standard. ISO14001 defines the critical elements of Environmental Management Systems Standards certifiable under ISO14000. Those elements are; establishing an environmental policy, planning for policy’s implementation, implementing the plan, and finally checking to confirm the successful implementation of the plan.

OBJECTIVES OF STUDY:

1) Process planning for Environmental Management System and its impact on the environment.2) To rank (optimize) Common Effluent Treatment Plant wastewater environmental aspects and impacts.3) To develop an environmental program for significant aspects.

During the study the procedure and steps involved in getting ISO14001 Certification by common effluent plant at Five Star MIDC, Kagal, Kolhapur, were studied. The study was focused on development of the matrix for identification of significant aspect and impacts.

DETAILS OF COMMON EFFLUENT TREATMENT PLANT AT KAGAL,KOLHAPUR-

Kagal-Hathkanangale Industrial Area is one of the newly developing five star industrial areas in Kolhapur Region of state of Maharashtra,(India). The industrial area houses many large, medium and small scale industries and export oriented units. As a part of national environmental policy and to abate industrial pollution, it is proposed to set up common effluent

treatment plant (CETP) by Maharashtra Industrial Development Corporation (M.I.D.C). Design, construction, installation, commissioning of CETP is entrusted to SMS Infrastructure Ltd., Nagpur on turn-key basis along with the contractual and operation and maintenance for two years. The proposed CETP is designed for 10 MLD of waste water with the provision of 20% hydraulic overload for safety and provision is kept in layout so that in future it can be expanded to 15 MLD capacity

The various components of the CETP are as follows- 1. Screen Chamber

2. Collection sump

3. Oil and Grease Chamber

4. Equalization tank and submerged mixers

5. Primary treatment tank(Physico chemical Treatment tank)

6. Primary Clarifier

7. Aeration tank with blower

8. Secondary Clarifier

9. Acid and Alkali Tank(Chemical Oxidation Tank[COT])

10. Pressure Sand Filters(PSF) & Activated Carbon Filters(ACF)

11. Final Treatment Tank and Pumping to High Rate Transpiration System(HRTS)

12. Diesel Generators

13. Laboratory

14. Godown

15. Control room

16. Storage room of dried sludge

ENVIRONMENTALASPECTSIDENTIFICATION

All parts of the Planning phase build on identifying the significant Environmental aspects of your operations. An environmental aspect is any element of your business that can interact with the environment. The ISO 14001 EMS also refers to impacts. Impacts are the actual or potential changes to the environmental resulting from any of the environmental aspects. The relationship between aspects and impacts is one of cause and effect. While the ISO 14001 standard establishes the structure of the EMS, there is a lot of latitude for a company to develop the content that best fits its operations. For example, each company selects a method for identifying significant environmental aspects that vest suits its needs. For most businesses, it is critical to choose a method that is simple and easy to understand. The method you use also must be clearly documented. This documentation is needed for the ISO 14001 audit, and also, you will want to review your original list of environmental aspects to help set future objectives. Described below are two methods for identifying significant aspects.

a) Expertise-Based Method to Identify Significant Aspects

b) Criteria-Based Method to Identify Significant Aspects

As the CETP commenced from April 2009 it is not feasible to select the first method i.e. Expertise-Based Method to Identify Significant Aspects. The Criteria-Based Method is best suited method. The advantage of this systematic approach is that the selection process is clearly documented for future review (and for the ISO 14001 certification staff). Again, it is important to select criteria that are simple-the easier the system, the more likely it will be followed.

DEVELOPMENT OF RATING MATRIXA ‘traffic light’ system has been developed to prioritize the impacts. Red indicates that the aspect is significant; amber indicates that the impact could become significant and green indicates that the impact is not deemed significant. For the case of Common Effluent Treatment Plant at Kagal, we are

following the “Criteria-Based Method to Identify Significant Aspects” In this method the following Criteria are selected depending on need for the analysis,1) Quantity.2) Impact to natural resource.3) Toxicity.4) Cost.The Severity and rating is defined as Rating

1) Catastrophic -- 52) Severe -- 43) Critical -- 34) Marginal -- 25) Negligible -- 1The Probability of occurrence is defined as

1) Frequent -- 5

2) Probable -- 43) Occasional -- 34) Remote -- 25) Improbable -- 1The given activity is then analyzed in the matrix developed.In the analysis each activity is checked against selected criteria viz. Quantity, Impact to natural resource, Toxicity and Cost and its severity i.e. Catastrophic, Severe, Critical, Marginal, Negligible. The rating of these is taken as Total and based on the ‘traffic light system’, and probability of occurrence is taken into consideration for finding the total score. The various matrices are developed in the similar manner for all components of CETP and the rating is developed so that it is easy to define the significant aspect as well as the impacts. The actual developed matrix as well as the evaluation of all components of CETP is given below and one abstract table is prepared for significant aspects.

Actual Developed Matrix

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

R

esou

rce To

xici

ty

Cos

t

Tota

l

Max

. Sco

re

Probability1 2 3 4 5

Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Catastrophic 20 20 40 60 80 100Severe 16 16 32 48 64 80Critical 12 12 24 36 48 60Marginal 8 8 16 24 32 40Negligible 4 4 8 12 16 20

Actual Total=Actual Score= Total × Probability of Occurrence

INDICATOR OF SEVERITY

49-100 High Impact17-48 Medium Impact0-16 Low or No

Impact

SCREEN CHAMBER

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

R

esou

rce To

xici

ty

Cos

t

Tot

al

Max

. Sco

re


Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Screening

of Influe

nt

Disposal of Floati

ng matte

r

Land Pollution

Catastrophic 0 0 0 0 0 20 20 40 60 80 100

Severe 0 0 0 0 0 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60

Marginal 0 2 0 2 4 8 8 16 24 32 40Negligible 1 0 1 0 2 4 4 8 12 16 20

Actual Total= 6Actual Score= Total × Probability of Occurrence 6 × 3(Occasional) = 18

PUMPS FROM COLLECTION SUMP

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

R

esou

rce To

xici

ty

Cos

t

Tota

l

Max

. Sco

reProbability

1 2 3 4 5Im

prob

able

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Pumps

from Collection sump

Oil & Other spillages

from Pump

s

Land

Pollutio

n

Catastrophic 0 0 0 0 0 20 20 4

0 60 80 100

Severe 0 0 0 0 0 16 16 32 48 64 80

Critical 3 3 3 0 9 12 12 24 36 48 60

Marginal 0 0 0 0 0 8 8 16 24 32 40

Negligible 0 0 0 1 1 4 4 8 12 16 20Actual Total= 10

Actual Score= Total × Probability of Occurrence 10 × 5 = 50

OIL AND GREASE CHAMBER

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

R

esou

rce To

xici

ty

Cos

t

Tot

al

Max

. Sco

reProbability

1 2 3 4 5Im

prob

able

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Oil and

Grease

Separation

Disposal of Greasy matter

having high COD and BOD value to

the land filling(Hazardous

Waste)

Land Pollution

Catastrophic 0 0 5 0 5 20 20 40 60 80 100

Severe 0 0 0 0 0 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60Marginal 0 0 0 2 2 8 8 16 24 32 40Negligible

1 1 0 0 2 4 4 8 12 16 20

Actual Total= 9Actual Score= Total × Probability of Occurrence 9 × 2 = 18

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

Res

ourc

e

Toxi

city

Cos

t

Tota

l

Max

. Sco

re


Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Pumps from

Collection

sump

Sound of

Pump Assembly & motor

in operat

ion

Noise Pollution

Catastrophic 0 0 0 0 0 20 20 40 60 80 100

Severe 0 0 0 0 0 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60Marginal 0 0 0 0 0 8 8 16 24 32 40Negligible

1 1 1 1 4 4 4 8 12 16 20


EQUALIZATION TANK AND SUBMERGED MIXERS

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

Res

ourc

e

Tox

icity

Cos

t

Tot

al

Max

. Sco

re


Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Equalization and

uniform flow

maintenance

Mixers in continuous operation

Noise Pollution

Catastrophic 0 0 0 0 0 20 20 40 60 80 100Severe 0 0 0 0 0 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60Marginal 0 0 0 0 0 8 8 16 24 32 40Negligible

1 1 1 1 4 4 4 8 12 16 20


PUMPS FROM EQUALIZATION TANK

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

Res

ourc

e

Tox

icity

Cos

t

Tot

al

Max

. Sco

reProbability

1 2 3 4 5

Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Pumping of influent to Primary

treatment plant

Sound of

motor in

operation

Noise Pollution

Catastrophic 0 0 0 0 0 20 20 40 60 80 100Severe 0 0 0 0 0 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60Marginal 0 0 0 0 0 8 8 16 24 32 40Negligible 1 1 1 1 4 4 4 8 12 16 20

Actual Total= 4 Actual Score= Total × Probability of Occurrence 4 × 5= 20

Cos

t

Tot

al Probability1 2 3 4 5

Act

ivity

/O

pera

tion

Cau

se o

f im

pact

Effe

ct/P

oten

tial I

mpa

ct

Seve

rity

Qua

ntity

Impa

ct to

Nat

ural

Res

ourc

e

Tox

icity

Max

. Sco

re

Impr

obab

le

Rem

ote

Occ

asio

nal

Prob

able

Freq

uent

Pumping of influent to Primary

treatment plant

Spillage Of

Chemicals

Land Pollution

Catastrophic 5 0 0 0 5 20 20 40 60 80 100Severe 0 4 4 0 8 16 16 32 48 64 80Critical 0 0 0 0 0 12 12 24 36 48 60Marginal 0 0 0 2 2 8 8 16 24 32 40Negligible 0 0 0 0 0 4 4 8 12 16 20

Actual Total= 15Actual Score= Total × Probability of Occurrence 15 × 5= 75

IN THE SIMILAR MANNER THE REMAINING ACTIVITIES ARE ANALYZED AND THE ABSTRACT OF ALL ANALYSIS IS AS FOLLOWS-

Sr.No. Item Activity/Process(Aspect) Environmental Effect(Impact) Caused by

Impact Assessment

Score

1 Screen Chamber Screening of Influent Land Pollution

Disposal of Floating Matter and Soil/Sand from screens

18

2 Pumps from Collection sump

Pumping of influent to O&G Chamber

Land PollutionOil & Other

spillages from Pumps

50

Noise Pollution

Sound of Pump Assembly &

motor in operation

20

3 Oil and Grease Chamber Oil and Grease Separation Land Pollution

Disposal of Greasy matter

having high COD and BOD value to

the land filling(Hazrdous

Waste)

18

4

Equalization tank and

submerged mixers

Equalization and uniform flow maintenance Noise Pollution

Mixers in continuous operation

20

5 Pumps from Equalization

tank

Pumping of influent to Primary treatment plant Noise Pollution Sound of motor

in operation 20

Land Pollution Spillage of 75

Influent

6

Primary treatment

tank(Physico chemical

Treatment tank)

1)Lime Dosing 2)FeSO4 Dosing

3) Polyectrolyte Dosing

Land Pollution Spillage of Chemicals 30

Effect on Human

Irritation to Eyes and Effect on skin due to

spillage

60

7 Primary Clarifier Separation of solids & Thickening Land Pollution

Sludge removal and disposal by

land filling (Hazardous

Waste)

95

8 Aeration tank with blower

Carboneous B.O.D.Removal & Nitrification

Land PollutionBiosludge is

disposed off to the land filling

95

Noise Pollution Continuous operation 20

9 Secondary Clarifier

Clarification & Thickening of Activated Sludge from

Aeration tankLand Pollution

Sludge removal and disposal by

land filling95

10

Acid and Alkali Tank(Chemical

Oxidation Tank[COT])

1) Use of acid and alkali for pH correction

2) C.O.D.Removal by H O₂ ₂ 3) Oxidation with blowers

Land Pollution

Spillage of Acids/Alkalis like H SO ,₂ ₄ NaOH,H O ₂ ₂

18

Effect on Human

Irritation to Eyes and Effect on skin due to

spillage

45

Noise PollutionOperation of Blower for Oxidation

20

11

Pressure Sand Filters(PSF) &

Activated Carbon

Filters(ACF)

1)Filtration 2)C.O.D.Removal 3)Colour Removal 4)Suspended Solids

Removal 5) Trace element Removal

Land Pollution

Disposal of Damaged

Activated carbon (Once in three

months)

12

12

Final Treatment Tank and

Pumping to High Rate

Transpiration System(HRTS)

1) Liquid Disposal 2) Pumping to HRTS

Land Pollution Effluent disposal on land 80

Odour

If not treated properly

nuiscence of Odour

35

13 Diesel Generators Back-up Power Supply

Air Pollution Diesel used for power generation 65

Noise Pollution Sound because of Assembly of

Gene set20

14 Laboratory Checking inlet and outlet parameters Water Pollution

Disposal of chemicals in

water60

Effect on human health

Various chemicals

causing irritation to eyes, effect on

skin

27

15

Godown Storage of 1) Lime 2) Ferrous Sulphate 3) Polyelectrolyte

4) Alum 5) Acids

6) Alkalis 7) Hydrogen Peroxide

8) Urea 9) Diammonium Phosphate(DAP)

Land Pollution Spillage of Chemicals

60

Air Pollution Odour/Gases 70Effect on human

healthDuring handling

eye irritation,skin infection etc.

80

16 Control room Controlling Panels of all activities in the CETP

Air Pollution Hazard of short circuit

12

CONCLUSIONS• Detailed environmental management programs have been prepared for the various CETPs in India. The EMP is standardizing the environmental management activities of the CETP and improves the compliance status of the CETP and its member industries. Environmental awareness created through the EMS has laid a milestone for member industries of CETP, to implement pollution prevention measures and obtain positive results in reducing hydraulic and pollutant loads in their discharges. Various industries have formed waste minimization circles (WMC) and many are applying for ISO14001 certification. • Industrial Wastewater facilities, like CETP’s, have increasingly begun turning to Environmental Management Systems to improve their environmental performance and operational control and treatment efficiency which is the main objective of providing EMS. For study work CETP at Kagal was selected which was the ideal for implementing EMS.• This Work highlights the step by step implementation and formats of the Environmental Management System, making use of the same is very easy to follow and implement the routine procedures of CETP.• The report also highlights the procedures and formats to be followed in regular operations of day to day work and gives some suggestions for switching over to those procedures.

• Due to proper assessment of the environmental impacts by the system of criteria based method; it becomes very easy to take effective measures against major Environmental impacts identification. Thus Criteria Based Method found to be best Identification Method for CETP. • The reduction in the various recurring costs is possible because of the systematic approach implemented to all activities through EMS. • This project work ultimately lead towards achivement of Central Govt’s Goal to abate the pollution from SSI’s by improving the performances of CETP’s. Thus this project work is an effective and readymade tool.

• This project proves beneficial for Enviro - social awareness to the society which includes Industries as wells as civil people. • This project also concludes that, when people from education as well as industry go hand in hand it would be a To & Fro efforts & outputs for both.REFERENCES [1] Ashok Dhariwal, Anamika Sharma and Anil Vyas; (2009) “Case study for EMS (ISO 14001) for CETP Sangaria, Jodhpur”, Indian Journal of Environmental Protection, Vol.29, No.3, page no. 266-272.

[2] Cardiff And Vale Nhs Trust / Ymddiriedolaeth Gig Caerdydd A’r Fro Cardiff University / Prifysgol Caerdydd ;( 2002) “Environmental Management System Manual”[3] Cardiff And Vale Nhs Trust Ymddiriedolaeth Gig Aerdydd A’r Fro Cardiff University Prifysgol Caerdydd ;( 2002) “Environmental Management System Manual Ems Audit Procedure No.20”[4] Cardiff And Vale Nhs Trust / Ymddiriedolaeth Gig Caerdydd A’r Fro Cardiff University / Prifysgol Caerdydd; (2009) “Environmental Management System Manual, Training Strategy (Awareness And Competence)” Pageno.2to8[5] City of San Diego Internal Audit Findings Matrix Environmental Services Department WRAD-F-AUD-

06; Revision 2 Waste Reduction & Disposal Division Effective Date: July 1, 2007[6] City of San Diego EMS Internal Audit Report Environmental Service Department RDD-F-AUD-05; Revision 2 Waste Reduction & Disposal Division Effective Date: April 30, 2008[7] City of San Diego Internal Audit Notification Environmental Services Department, WRAD-F-AUD-02; Revision 4 Waste Reduction & Disposal Division Effective Date: March 24, 2009[8] Craig P. Diamond (1996); “NSF International Environmental Management System Demonstration Project”-Final Report.

[9] Department of Public Policy University of North Carolina PREPARED FOR: U.S. EPA, Office of Water and Office of Policy, Economics and Innovation(2003); “Environmental Management Systems: Do They Improve Performance?”- Project Final Report

[10] Donna J. Adams, Ellen R. Barrett, Richard E. Bickerstaff, Jr., Beth Eckert, James Naber ;( 2004) “Achieving Environmental Excellence: An Environmental Management System-Handbook for Wastewater Utilities”

[11] Maheswari, M. and Dubey, S. (2000), “Common Effluent Treatment Plant A solution or a problem in itself” Toxics Link (a nonprofit environmental organization)

[12] New York State Department of Environmental Conservation; (2006) “Understanding and Implementing an Environmental Management System-A step by step Guide for small and medium sized organizations.”

[13] Prasad B.M. (2000) “An overview of CETP’s in India”. CPCB, Delhi.

[14] R.Harapanahalli, B.Lokeshappa and N.T.Manjunath; (2005) “Wastewater management for

Textile industries in Common Effluent Treatment Plant- A case study”, Indian Journal of Environmental Protection.

[15] Sankar, U. (2000), “common effluent treatment plants: an institutional arrangement for pollution control in small-scale tanneries in India”. Madras school of Economics Allied Publishers, New Delhi.

[16] Sankar, U. (2001), “Economic Analysis of Environmental Problems in Tanneries and Textile Bleaching and Dyeing Units and Suggestions for Policy Actions”. Allied Publishers, New Delhi.

[17] S.Murthy and J.V.Kangane; (2005) “Performance Evaluation and operational enhancement of Common Effluent Treatment Plant: A case study”, Indian Journal of Environmental Protection

[18] S.Wayne Rosenbaum;(1997) “ISO 14000:Legal advantage of replacing command and control regulation”, Practice Periodical of Hazardous, Toxic and Radioactive Waste Management July-1997,page no.124-126.

[19] Technical EIA Guidance Manual for Common Effluent Treatment Plants Prepared for Ministry of Environment and Forests, Government of India by

IL&FS Ecosmart Limited Hyderabad September (2009)

EXPERIMENTAL ANALYSIS OF A CATALYTIC CONVERTER USING SUPPORTED COPPER CATALYST TO REDUCE NOX AND PM FROM

DIESEL ENGINE

PL.S. MUTHAIAHSathyabama University, Chennai

DR. M. SENTHILKUMARSKR Engineering College, Chennai

DR.N.MANOHARANSathyabama University, Chennai

ABSTRACTThe superior performance, higher output power and comparatively less-cost fuel make the diesel engines more popular in both heavy and light duty automobile applications. The main disadvantage in diesel engines is the emission of dangerous pollutants like Oxides of Nitrogen (NOX) and Particulate Matter (PM). The higher cost and high demand of presently used rare earth metal catalyst necessitate the need for finding out the alternatives.Cu-ZSM-5 zeolite has been proved to be one of the catalysts to detoxify gas emissions. Such catalysts have not been commercialized due to hectic ion exchange of ZSM-5 with Cu, which involves huge volume of water. Another major constraint is the transport of toxic gas emissions into the zeolite pore where the active sites are located is very much limited by the micro pores. In this context supported metal oxide catalysts appear to be advantageous. But in such supported metal oxide catalysts, the major problem like sintering has been encountered. As a result of sintering, the activity of catalyst decreases. Taking into account the above said problems, in the present study, alumino-phosphate supported copper phosphate was examined for the treatment of exhaust emissions. Instead of using wall flow ceramic substrates coated with platinum, steel wire meshes coated with newly prepared catalytic material was used to filter and oxidise the PM present in the exhaust gas. The experimental analysis was carried out on a 10 hp, four stroke, twin cylinder, water

cooled diesel engine with different load conditions, and the results obtained in reducing NOx and PM were studied and discussed in this paper.

1. INTRODUCTIONWith the introduction of the turbo charged high-speed diesel engines, the use of diesel engine vehicles in transport sector is increasing enormously. The main drawback in diesel engine is that it produces large amount of pollutants which include NOX, CO, unburned HC, smoke etc. Apart from these unwanted gases, air borne PM such as lead, soot, and other forms of black carbon are also produced in the diesel engine exhaust. All these pollutants are harmful to environment and human health. They are the main causes for green house effect, acid rain, global warming etc.SCR catalysts can be made of either vanadium based or iron- zeolite based. Vanadium is classified as toxic. Iron-zeolite catalysts are non-toxic and considered as a safer alternative for SCR system. Cu-lZSM-5 zeolite has been proved to be one of the catalysts to detoxify gas emissions. Such catalysts have not been commercialized due to hectic ion exchange of ZSM-5 with Cu, which involves huge volume of water.

Another major constraint is the transport of toxic gas emissions into the zeolite pore where the active sites are located is very much limited by the micro pores. In this context supported metal oxide catalysts appear to be advantageous, as they are free of such diffusion constrain. But in such supported metal oxide catalysts, the major problem like sintering has been encountered. As a result of sintering, the activity of catalyst decreases. Taking into account the above said problems, in the present study, alumino-phosphate supported copper phosphate was examined for the treatment of exhaust emissions. Use of copper phosphate is based on the fact that it can have enhanced sintering resistance and thermal stability. The active copper site chemisorbs NO and split into nitrogen and oxygen, and CO into CO2. Instead of using wall flow ceramic substrates coated with platinum, steel wire meshes coated with newly prepared catalytic material is used to filter and oxidise the PM present in the exhaust gas.

The experimental analysis is carried out on a 10 hp, four stroke, twin cylinder, water cooled diesel engine with different load conditions, and the results obtained in reducing NOx and PM are discussed in this paper.

2. PREPARATION OF CATALYSTSupported metal oxide catalysts are obtained by impregnating active ingredients like copper oxide on a thermally stable inactive support like alumina, magnesia etc. In this study, alumina phosphate is taken as supporting material and copper phosphate is selected as an active ingredient.The copper phosphate and aluminium phosphate powders are mixed with different proportions as shown in the Table 4, and are named as CAT 1, CAT 2, CAT 3, CAT 4, and CAT 5. A measured quantity (10-15% by volume) of orthophosphoric acid is added gradually and is thoroughly stirred for better blending. During this process, enormous heat is generated and after a few minutes, it starts hardening. Before it becomes fully hardened, the required shape can be formed by gently pressing it. The formed solid catalyst is dried at room temperature for a day. It is found that, during the chemical processing of the powders, they form small granules separated by small gaps, which brings the

porosity in the catalyst. The porosity in the catalytic material helps in better gas flow and better reaction between NOX and catalytic material, ensuring reduced NOX outflow.

Table 4: Catalyst Proportions

CATALYST TYPE

Aluminum Phosphate

% by Volume

Copper Phosphate

% by Volume

CAT 1 92 8

CAT 2 84 16CAT 3 76 24CAT 4 68 32CAT 5 60 40

Figure 1: Catalytic Beads

The flowing exhaust gas is free to move in all directions inside the canister. As the movement of exhaust gas is not abruptly obstructed anywhere in its path, the back pressure is limited to minimum level. The porous nature of catalyst beads also help the gas to flow over the larger surface area of the catalyst enabling better reaction to take place for reducing NOX as in the case of SCR system.

2.1 SCR CATALYST SYSTEMPresently, ammonia derived from urea is used to reduce NOX from diesel engines. This is achieved by allowing the ammonia plus exhaust gas to flow over the platinum coated ceramic substrates, and it has been proved as highly effective in reducing NOX in heavy duty applications [3]. Ammonia is produced on-board by rapid hydrolysis of non hazardous form of urea solution. The problem with this is, cost of on-board production of ammonia, cost of additional on-board air supply equipments, and high cost of rare metals like platinum etc. In this paper, rare earth metal catalyst is replaced by a specially prepared catalyst. This catalyst is prepared by combining Copper phosphate and Aluminium phosphate powder materials with suitable bonding agent. The catalyst selectively reacts with NO and NO2 species and effectively reduces them to form nitrogen and oxygen using SCR technology. The mechanism of decomposition of NOx, when the hot gases containing NOx meet with the catalyst is shown in figure 2.

Figure 2: Mechanism of decomposition

Simple NO molecules do not react by themselves to yield N2 and O2, because they are not activated. When NO molecules pass over the catalyst surface, they are absorbed on the active sites as shown above. The adsorbed molecules are activated which react laterally to yield O2 and N2.

2.2 DIESEL OXIDATION CONVERTERIn the present investigation, the diesel oxidation catalyst is made as a flow through device as shown in figure 3 which consists of newly-prepared solid catalyst and steel wire mesh material. These coated catalytic beads and steel wire meshes are kept inside

the canister in alternate layers. When the hot gases contact the catalyst and the coated wire mesh, most of the exhaust pollutants such as CO, gaseous hydrocarbons, unburnt fuel and lube oil, toxic aldehydes etc. are oxidized to CO2 and water, thus reducing harmful emissions. DOC does not collect or burn the soot particles in diesel exhaust. But it is accomplished by oxidizing the soluble organic fraction of diesel PM.

(SOF) + O2 CO2 + H2OSO2 + ½ O2 SO3 H2SO4

DOC can also produce sulphate particles by oxidizing the SO2 present in the exhaust gas and thus increases the PM emission. This may not be a problem, if the fuel contains <50ppm of sulphur [4].

2.3 VOLUME OF CATALYSTThe size of exhaust canister is based on the engine exhaust flow rates. For maximizing catalyst applied surface area, the volume of catalyst must be 1.5 to 2 times the engine displacement [4]. The engine selected for this study is a four stroke twin cylinder (80mm bore and 110mm stroke length) water cooled diesel engine. The engine displacement is calculated as 603 cm3/second, where as the total volume of catalyst used in the model is 1303 cm3. The total trap material (catalytic beads plus coated wire meshes) kept inside the canister occupies one third of its total

volume. This means, the remaining volume is used for the exhaust gas to flow out freely.

This helps for limiting the back pressure and ensuring effective DOC and SCR systems.

3.SELECTION OF FILTER MATERIALCeramic monolith, ceramic foam, steel wire meshes, ceramic silicon fiber, porous ceramic honey comb are the few types of filter materials reported in the literature. Out of these filter materials, steel wire mesh is selected as trap material because knitted steel wire mesh material is ranked first for its collection efficiency of PM [2]. The other reasons for its selection are thermal stability during regeneration, good mechanical properties, long durability, easy availability and less cost.

3.1 WIREMESH SPECIFICATIONSIn this analysis, the selected steel wire mesh grid sizes are 1.61 and 1.01 mm. The specifications of these wire meshes are shown in Table 1

Table 1: Wire Mesh Specifications

Wire Mesh size (gap) in mm

Wire Dia (d) in mm

Open Area %

Wt in Kg/m2

Mesh per inch

CPSI

1.61 0.51 57.7 1.56 12 1441.01 0.41 50.8 1.51 18 324

Since number of steel wire mesh pieces are stacked one over the other as a bunch, the gas flow cannot be straight, rather it is a zig-zag flow. This type of flow provides increased travel length and contact time so that more amount of DOC and SCR catalytic actions can take place and hence more PM reduction is achievable.

4. PHYSICAL MODELTo meet the less ground clearance available in the present heavy vehicles, the height of the canister is restricted to 110 mm. The cross section is made as rectangular of size 176 x 110 mm. The rectangular corners are suitably rounded off inside the canister for the smooth flow of exhaust gas and improved turbulence effect near the wall sides.

4.1 CONSTRUCTIONThe exhaust canister designed for this study comprises two compartments. The first one is meant

for filtration and DOC catalyst. The second compartment is meant for filtration and SCR system.

Figure 3: Sectional Plan Showing Two Compartments

Figure 4: Sectional Elevation at x-x (Enlarged view)

The upstream of the first compartment is filled with steel wire meshes for about one-fourth of its length. These wire meshes are placed vertically with their surface being parallel to the flow of gas. Sufficient gaps are provided amongst these bunch of wire meshes which make the gas to flow crosswise also for better filtration and reduce back pressure as shown in Figure 3. The enlarged view showing the flow directions of exhaust gas is shown as sectional elevation in Figure 4.The downstream of first compartment is filled with catalyst beads and steel wire meshes arranged in alternate layers. The beads are arranged horizontally in a single layer so that the gas can enter at one end of the bead and come out in the opposite end. During its travel, the gas follows the up and down movement and meets the entire inner surface area of the catalyst as shown in Figure 4. More contact surface area of the catalytic beads enables the better catalytic reaction to take place between the catalyst and the flowing exhaust gas. The beads are designed so that the top and bottom

portions are open and the gas can go up and down to the next layer of beads through the steel wire mesh layers which are also horizontally placed as shown in Figure 4.

Figure 5: Compartments filled with Catalyst& Filter Materials

Figure 6: Newly developed exhaust canisterThe catalytic converter assembled with catalytic beads and steel wire meshes placed inside the two compartments is shown in figure 5. The fully assembled catalytic converter before fitting it in the experimental setup is shown in figure 6.

4.2 OPERATING PRINCIPLE:A portion of the exhaust gas passes through the wire mesh layers which trap a portion of the soot. The remaining exhaust gas flows out straight to the neighboring bead placed in the same line – similar to a flow-through substrate. The soot trapped in the wire mesh material is combusted by the NO2 that is generated by the upstream catalyst and thus the filter is regenerated continuously. If a situation occurs where filter regeneration is stopped and a saturation

point occurred with the collected soot, the wire meshes placed over the catalytic beads will not plug as happened in wall flow filter.

In this condition, the exhaust gas can flow out through the catalytic beads which are similar to flow-through substrates. However, much larger part of exhaust gas flows through the catalytic beads in the direction as shown in Figure 4.As the path of the gas is not totally blocked, the back pressure developed inside the exhaust canister is very much limited and no further increase in back pressure can happen beyond certain limit, irrespective of the soot loading on the wire meshes over a period of time.

5. EXPERIMENTAL INVESTIGATION

Experiments have been conducted on the engine whose specification is given in table 5. and all the base line readings were taken without the catalytic converter. Then the catalytic converter system along with particulate trap was connected with proper sealing to the engine exhaust pipe. To avoid the exhaust heat loss, proper insulation was provided by asbestos rope. And the experiment was conducted with the catalytic converter fitted. The various parameters were set, and the readings were measured for different operating conditions like 14, 21, 42, 63, and 84% of load. The experimental setup is shown in figure 7.

Table 5: Test Engine Specification

Type Twin-cylinder, 4-stroke, direct injection, vertical, diesel engine

Nature of cooling Water cooling

Power/Hp 7.5 kW/10

Speed 1500 rpm

Piston diameter 80 mm

Stroke length 110 mm

Compression ratio 17.5:1

Brake power / hp 7.46 kW / 10

Figure 7: Experimental Setup

6. RESULTS AND DISCUSSIONS

The variation of NOx with the brake power of the engine with and without trap & converter is shown in Figure 8. It is seen that the Oxides of Nitrogen increases with increasing load. This is due to the increase in temperature inside the cylinder as the load increases. It is observed that the activity of the catalysts of NOX reduction depends on the density of the active sites of copper on the surface of the catalyst. More specifically, two copper sites are to be very close in order to adsorb and activate two NOX

molecules to yield N2 and O2. This condition becomes more and more probable with increase in the copper content of the catalysts. At low loads there is only a slight reduction in NOX level. Maximum NOX reduction is obtained at full load conditions. This is because of lower conversion efficiency of the trap at low loads due to low temperature.It is also seen that CAT 4 and CAT 5 are found to be more active than others as they give better NOX reduction. Comparing the performance of catalytic converter with CAT 4 and CAT 5, no much variation is noticed. This means the copper phosphate of 32% or 40% does not make much difference. From this, it can be concluded that 33.3% of copper phosphate and 66.6% of support material of aluminium phosphate may be the ideal combination for better performance of this catalytic converter.The variation of NOX

conversion efficiency with respect to brake power is shown in Figure 9.

NOX conversion efficiency with CAT 4 is found to be better than the other catalysts, and it varies from 21.92% at no load to 32.38% at full load.Figure 10 shows the variation of backpressure versus brake power. It is found that the back pressure increases gradually with increasing load. The increase in back pressure is due to the accumulation of particles on the surface of the trap. The backpressure of the engine with trap increases up to 5.45% than that of engine without trap at maximum load condition.In Figure 11, BSFC is shown against brake power. It is seen that BSFC increases at all loads, due to the presence of trap & converter than without trap & converter. Due to increased back pressure, it is found that specific fuel consumption is increased up to 6.26%. This may be due to more fuel consumption in the presence of trap & converter.The comparison in variation of PM with and without trap & converter is shown in Figure 12. It is seen that PM increases with increasing load. Up to 21% of load, the increase in PM was slow, later the increase was rapid. This increase of PM is due to incomplete combustion of the air-fuel mixture and at higher load the increase was rapid. In the presence of particulate trap, the decrease in particulate emission is observed for the increasing brake powers.

Figure 8: NOx Vs Brake Power

Figure 9: NOx Conversion Efficiency Vs Brake Power

Figure10: Back Pressure Vs Brake Power

Figure 11: Fuel consumption Vs Brake Power

Figure 12: Particulate Vs Brake Power

8. CONCLUSION

Based on the experiments conducted on DI diesel engine, the following conclusions can be drawn.NOX

reduction varies between 28.69% & 40.59% at no load and full load respectively. This can be further improved by effectively increasing the surface area of the catalytic beads.

In addition to this, EGR system is also to be added. With EGR attached system, the more NOx reduction can effectively be achieved to meet the Euro IV emission targets.With the help of DOC system and filtration, PM reduction is achieved up to 46.82% at full load condition.The maximum back pressure increase is found as 4.16% Due to increased back pressure, it is found that the specific fuel consumption is increased up to 5.62%.This after treatment technology for NOX and PM reduction is robust which needs no interaction with the engine management system and is totally independent.

ABBREVIATIONS

NOX – Oxides of Nitrogen

PM – Particulate Matter

DOC – Diesel Oxidation Catalyst

DPF – Diesel Particulate Filter

PFT – Partial Filter Technology

SCR – Selective Catalytic Reduction

CPSI – Cells Per Square Inch1. Eberhard Jacob, Rheinhard Lammermann,

Andreas Pappenherimer, Diether Rothe – “Exhaust Gas Aftertreatment System for Euro 4: Heavy Duty Engines” – MTZ 6/2005

2. Otto A. Ludecke et al (1983), “Diesel Exhaust Particulate Control System Development”, SAE Paper 830085.

3. K.V.R.Babu, Chris Dias, Shivraj Waje, Alfred Reck, Kim Wonsik – “PM Metalit – A Continuously Regenerating Partial Flow Particulate Filter – Concept and Experience with Korean Retrofit Programme”, SAE Technical Paper 2008-28-0008

4. Sougato Chatterjee, Andrew P. Walker, Philip G. Blakeman – “Emission Control Options to Achieve Euro IV and Euro V on Heavy Duty Diesel Engines”, SAE Technical Paper 2008-28-0021.

5. Allan C. Lloyd, “Diesel Engines Environmental Impact and Control”, DEER Conference 2002.

6. Jacobs, T., Chatterjee, S., Conway, R., Walker, A., Kramer, J. and Mueller-Haas, K., “Development of a Partial Filter Technology for HDD Retrofit”, SAE Technical Paper 2006-01-0213.

7. Fredholm, S., Anderson, S., Marsh, P., D’Aniello, M.J., Zammit, M.G.,Brear, F., “Development of a Diesel Oxidation Catalyst for Heavy Duty Diesel Engines”, SAE Technical Paper 932719.

8. Peter L.Herzog ET AL., (1992).” NOX Reduction Strategies for DI Diesel Engines”. SAE Technical paper 920470.

9. “Particulate TRAPS FOR Euro 4 commercial diesel vehicles” International Edition-Diesel Progress, May-June 2001.

DESIGN OF HIGH BANDWIDTH CHARGE PUMP WITH ON CHIP LOOP FILTER FOR HIGH PERFORMANCE PHASE LOCKED LOOP IN

180 NM CMOS TECHNOLOGY

*PROF. R .H. TALWEKAR, PROF. (DR.) S. S LIMAYE * Deptt. Of Electronics and Telecommunication, DIMAT, Raipur (CG), India.

Deptt. Of Electronics and Telecommunication, JIT, Nagpur (MS), India.E-mail: [email protected]*, [email protected]

ABSTRACT - A high band width Charge Pump with loop filter for Phase Locked loop is mostly used and preferred in wireless communication system One of the application in wireless system is the Orthogonal Frequency Division Multiplexing (OFDM) which is required the such type of high bandwidth Phase Locked Loop. Design of high performance phase locked loop require ,a design of high performance charge pump .We have designed a Charge Pump with loop filter for more than 2.4 GHz bandwidth in 180 nm CMOS technology by CADENCE VIRTUOSO tool. A high performance, low power consumption Charge Pump has designed to improve the jitter and bandwidth by connecting the feedback from loop filter to the Charge Pump. An on chip loop filter is designed for PLL and simulated in 180 nm CMOS Technology. A proposed architecture of Charge Pump with loop filter is designed and simulated by CADANCE spectre.


Key Words - Phase locked loop (PLL), Phase frequency detector (PFD), Voltage controlled oscillator (VCO), Charge pump (CP), Orthogonal frequency division multiplexing (OFDM)

I. INTRODUCTIONOrthogonal frequency division multiplexing (OFDM) is a special case of multicarrier transmission, where a single data stream is transmitted over a number of lower rate subcarriers [1]. Phase locked loop (PLL) is ca combination of a phase frequency detector (PFD), loop Filter (LPF), charge pump (CP) and voltage controlled oscillator (VCO). The operation of the Phase Locked Loop circuit is basically a feedback control system that controls the phase of a voltage controlled oscillator (VCO). A PLL based charge pump has been used to design because of its wide capture range and zero phase offset which is shown in Figure 1. The charge pump PLL frequency synthesizer has been used extensively to improve the performance of Orthogonal Frequency Division Multiplexing (OFDM). A noise rejection performance is also a big problem in the vicinity of the communication field and integrated circuit. Phase locked loop is mostly used in wireless communication and data recovery circuits. Particularly wherever a PLL has been using in wireless communication, an orthogonal frequency division multiplexing (OFDM) requires very compact and perfect synchronization of signal frequency [2]. A PLL is a part of this multi-carrier modulation system. Designing a 2.4 GHz phase locked loop for multi-carrier communication with high stability using 180 nm CMOS

technology is also a challenge in CMOS Technology. A second order low pass filter needed to supply the voltage to control the VCO in the PLL is designed with less die area and high performance capability and low power dissipation. The simulated PLL provides the loop parameters almost independent of divider factor and decrees the capacitance I/20th of conventional PLL. This paper describes the design. Of 2.4GHz CMOS PLL targeting OFDM application. The proposed PLL loop is designed with 1.8V supply and 0.18um CMOS Technology. The more details of the dynamic of the loop filter is described and simulated for better results with less non-ideality in the VCO input response. A 100μA charge pump is used to force the current pulses to low pass filter. The specific low pass filter designed and described here is very useful to avoid the disturbance to VCO. Here section-I described the introductory part of the Design of high performance phase locked loop. Section–II describes basics of Phase Locked Loop system. Section-III describes basics of charge pump. Section –IV describes the conventional Charge Pump. Section – V describes proposed high bandwidth Charge Pump for Phase Locked loop with loop filter. Section-VI describes loop filter dynamics and design criteria for better stability.

Section-VII gives result and waveforms and last section-VIII gives conclusion

II .BASICS OF PHASE LOCKED LOOP

Phase locked loop is mostly used in wireless communication and data recovery circuits. Now a day for above maintained application a low voltage low area and high performance integrated circuits are used which complicates the implementation of such type of integrated circuit. To design an on chip loop filter for better response is a challenge in CMOS Technology. In wireless communication wherever OFDM is being used requires very compact and perfect synchronization of signal frequency. Orthogonal frequency division multiplexing (OFDM) is a particular from of multi carrier transmission is suited for frequency selective channels and high data rates. This technique transform a frequency selective

wide band channel into a group of non selective narrow band channels which makes it robust against large delay spreads by preserving orthogonality in the frequency domain multi carrier modulation. A PLL is a part of this multi-carrier modulation system. Phase locked loop (PLL) basically consists of a phase frequency detector (PFD), loop Filter (LPF), charge pump (CP) and voltage controlled oscillator (VCO). A phase detector compares two input signals and produces an error signal which is proportional to their phase difference, the phase error output of PFD is fed to a charge pump and then to loop filter which integrates the signal to get a sharper and smooth signal so that the disturbances at the input of VCO get minimized [3] .The operation of the Phase Locked Loop circuit is basically a feedback control

system that controls the phase of a voltage controlled oscillator (VCO).

Figure 1. Basic Phase Locked Loop

III .BASICS OF CHARGE PUMPThe Charge Pump is an important part of the Phase Locked Loop. Basically CP consists of a current source, a current sink and switches. The CP is usually followed by a passive loop filter that changes the output current to a VCO control voltage; this voltage affects the charge pump output voltage. Figure 2 shows the circuit diagram of a conventional charge pump [4].

Figure 2. Basic Charge PumpA Charge Pump improves the performance of the Phase Locked Loop if it is designed with compensation circuit to reduce the errors due to non-ideal CP effects such as current mismatch and current noise which causes steady state phase error as well as reference spur .The channel-length modulation, and parasitic capacitors are the main reasons of the introduction of these types of spurs. [5].

IV. CONVENTIONAL CHARGE PUMP

To minimize the current mismatch a different equations and values have been applied and have found different Ron for different values of width of various transistors. In a conventional Charge Pump 15 transistors have been used to implement the Charge Pump along with the constant current source, to maintain the noise free voltage at voltage controlled oscillator. If either the zero or the pole moves toward the origin of the complex frequency plane, the phase margin will be degraded. The rule to size the capacitance can be easily carried out by merging expressions and to obtain both the desired settling time by means of impedance as supposed in and phase margin [6].

Figure 3. Conventional Charge Pump

Figure 4. Op Amp for Conventional Charge Pump

V. PROPOSED CHARGE PUMP

The charge pump circuits have non idealities, since the charge pump translates the digital timing information into an analog quantity for voltage-controlled oscillator (VCO) tuning, the non ideal behavior of the Charge Pump can contribute significantly to the PLL output-jitter [7].A new CP based on gain-boosting which reduced the channel length modulation effect. This output impedance is similar to triple-cascode output impedance but with less consumed voltage headroom. It is suitable for channel length modulation effect elimination but still have the problem of non-equal I Up and I Down [8]. In the conventional Charge Pump circuit, an Op Amp has been used was giving the more power dissipation because it was designed with large width of the transistor NM3 connected before output stage of Op Amp .The two output taken from the proposed Charge Pump have been used to supply the voltage to low pass filter and second output back to the transistor which was giving the poor bandwidth. If the discharging current mirror uses a cascade topology, the charging/discharging current curve in the effective output range will become flatter [9].Now by the concept of the feedback amplifier the midpoint of the loop filter has been connected to one of the input terminal of the conventional charge pump as an feedback which has improved the bandwidth from 500 MHz to 3.15 GHz .The conventional circuit was giving the high power dissipation which has down by providing the 100 uA current by constant current source.

Figure 5. Proposed Charge Pump.

Figure 6. Symbol of Proposed Charge pump. With loop filter

VI. LOOP FILTER DYNAMICS

In phase locked loops (PLLs) ,the loop filter sets the PLL’s closed loop bandwidth as well as its dynamic response. Most practical loop filters are second order & by the introduction of second order filter, a PLL becomes 3rd order system [2]. The loop transfer equation for the complete system is

2

2 2

2( )

2S

H sS S

η η

η η

ω ξωξω ω+

=+ + …………(1)

1 21 1 1 2

2

. 2VCO CPC CK R I C C N

Cω π += ………… (2)

i.e. 1 11

1 22 ( )VCO CPK R C IN C C

ωπ

=+

…………… (3)

1 11 2 1 21

1 21 2

tan tan (1/ )MC C C CRC

RC CRC Cφ − −+ += −

……………(4)

1 11 2 21

1 2 1 2

tan tanC C CRC C C C

ω − −+= −+ ……………(5)

To increase the stability of the phase locked loop a phase locked circuit with an additional pole is required and this is possible by adding a singlecapacitor in parallel with previous 1 1 R C network.

Figure 7. Loop Filter

From above equation it is clear that the stability of the third order charge pump phase locked loop depends on the ratio of two capacitors used in the loop filter .If the phase margin of two pole & three pole system is below60o , then it loses the stability. To improve the stability of the system and for fast settling, the phase margin must be greater than 60o , but for higher phase margin it slows down the frequency response. So the phase margin must be 60o .For maintaining 60o phase

margin, the capacitor ratio 1 2C C should be greater than 14 [10]. To achieve the same it has optimized for

different values of the capacitor from 1 pF to 20 pF and simulated the complete charge pump with loop filter circuit by CADENCE SPECTRE and obtained the various responses. Finally it is found that a very low capacitor does not responds to higher bandwidth which is shown in figure 8, but current responses for all the capacitors ranging from the 1 pF to 20 pF have been giving the near about same response .The current responses for all the values of the capacitors 1 pF ,5.75 pF ,10.5 pF, 15.25 pF, 20 pF n are shown in the figure 11 to figure 15 respectively . In the proposed loop filter the values have been reduced to nearly half of the conventional elements values, i.e. capacitor C1 20 pF , C2 1 pF at 10 K ohm resistor.

Figure 8 Optimize Waveforms of different Capacitors

Figure 9 constant current source

Figure 10. Optimized constant current source waveform

(1) VII. RESULT AND WAVEFORMS

The proposed high bandwidth Charge Pump with loop filter has designed using 180 nm CMOS

Technology and circuit has simulated by CADENCE Virtuoso tool. The frequency response of Charge Pump with loop filter and output waveform of Charge Pump with loop filter operated at 1.8 V has been shown in figure 7 and 8 respectively. A second capacitor which has been used for higher stability in conventional design, also used in the proposed design with one tenth of the main capacitor value. The smith chart for the impedance of the whole design has shown in the figure showing only the impendence curve

Fig 11 Current waveform of loop filter at 10.5 pF

Fig 12 Current waveform of loop filter at 1 Pf



Fig 15 Current waveform of loop filter at 20 pF

Figure 16. Frequency Response of Charge Pump with loop filter

Figure 17 Output waveform of Charge Pump

VIII. CONCLUSIONA high performance Charge Pump with loop filter for Phase locked loop has been designed in 180 nm CMOS Technology using CADENCE Virtuoso tool. The presented Charge Pump with loop filter operated up to the frequency of 3.15 GHz bandwidth .The given Charge Pump can be used UHF and High frequency band telecommunication application .

REFERENCES

[1] Jung-Yu Chang, Che-Wei Fan, and Shen-Iuan Liu “A frequency Synthesizer for Mode-1 MB- OFDM UWB Application”, National Taiwan University, Taipei Taiwan, 978-1-4244-2782-6/09/$25.00 ©2009 IEEE

[2] Zheng Yong Zheng,”A 3.96 GHz phase- Locked loop for mode-1 MB-OFDM UWB hopping carrier generation”Vol. 30, No. 7 Journal of Semiconductors July 2009

[3] B. Razavi, “Design of Analog CMOS Integrated Circuits.” McGraw- Hill, 2001.

[4] Tian Xia*, Stephen Wyatt, “High Output Resistance and Wide Swing Voltage Charge Pump Circuit”, 9th International Symposium on Quality Electronic Design, 0-7695-3117-2/08

[5] M.-S. Hwang, J. Kim and D.-K. Jeong, “Reduction of pump current mismatch in charge- pump PLL”, ELECTRONICS LETTERS 29th

January 2009 Vol. 45 No. 3

[6] A. Pugliese, G. Cappuccino and G. Cocorullo, “Nested Miller compensation capacitor sizing rules for fast-settling amplifier design”, ELECTRONICS LETTERS 12th May 2005 Vol. 41 No. 10

[7] Jianzheng Zhou 1, 2 Zhigong Wang 1* “A High- Performance CMOS .Charge-Pump for Phase-Locked Loops”, ICMMT2008 Proceedings, 978-1-4244-1880-0/08/$25.00 ©2008 IEEE.

[8] Rania H. Mekky 1 and Mohamed Dessouky 1,21 Mentor Graphic Egypt, Cairo, Egypt “Design of low mismatch gain boosting Charge Pump circuit for Phase Locked Loop” IEEE ICM - December 2007

[9] Tian Xia*, Stephen Wyatt, “High Output Resistance and Wide Swing Voltage Charge Pump Circuit”, 9th International Symposium on Quality Electronic Design, 0-7695-3117-2/08 $25.00 © 2008 IEEE

[10] Raghavendra R G Electrical Indian Institute of Science Bangalore,” Area Efficient Loop Filter Design for Charge Pump Phase Locked Loop” . GLSVLSI’07, March 11–13, 2007, Stresa-Lago Maggiore, Italy.

A NEW OBLIVIOUS WATERMARKING METHOD USING MAXIMUM WAVELET COEFFICIENT MODULATION

K.RAMANJANEYULU

Dept. of E.C.EBapatla Engineering. CollegeBapatla, [email protected]

DR. K. RAJARAJESWARIDept. of E.C.E, College of EngineeringAndhra University

Visakhapatnam, [email protected]

ABSTRACT In this work, a new robust oblivious image watermarking scheme based on local maximum wavelet coefficient modulation for copyright protection is presented. The original unmarked image is not required for watermark extraction. Wavelet coefficients are grouped into different blocks and one watermark bit is embedded in each block. LH sub band (third level) of a Discrete Wavelet Transform (DWT) of a host image is selected for embedding a watermark. This scheme demonstrates robustness to some common signal processing attacks such as average filtering, JPEG compression and additive Gaussian noise. Experimental results show that there is a significant improvement with the proposed algorithm over the existing schemes in terms of Normalized Correlation Coefficient (NCC) and Peak Signal to Noise Ratio (PSNR).

Key Words: Copyright Protection, Digital Watermarking, Discrete Wavelet Transform (DWT), NCC, PSNR

1. INTRODUCTION

Digital media has become the leading choice of format for storing audio and video information for multimedia applications on the internet. Media formats such as MP3, DVDs and JPEGs are widely available to millions of users’ everyday. Many researchers are aware of the issues of copyright protection, image authentication, proof of ownership etc. Hence there are many solutions that have been proposed. Watermarking Technique is one of these solutions. Watermarking Technique embeds information so that it is not easily perceptible. A watermarking technique is referred to as oblivious (blind or public) if the original image and the watermark are not needed during extraction. There are several important issues in the watermarking system. First, embedding process should not degrade

the quality of the host image and should be perceptually invisible to maintain its protective secrecy. Second, the watermark must be robust enough to resist common image processing attacks and not be easily removable and only the owner of the host to be able to extract the watermark. Third, the blindness is necessary in some application in which original image is not available at the time of extraction. Watermarking can be done either in spatial domain or in transform domain. Watermarking in spatial domain is simple but the watermarked image is hard to robust. Watermarking in transform domain is secure and robust to many attacks. Image watermarking algorithms using Discrete Cosine Transform (DCT), Singular Value Decomposition (SVD) and Discrete Wavelet Transform (DWT) are available.

Among these, the DWT approach remains one of the most effective and easy to implement techniques for image watermarking [1]. The most important issue in DWT-based image watermarking is how to choose the coefficients to be embedded. In [2][3][4] the watermark is embedded in the significant coefficients which are selected from the wavelet coefficients. In

[5] Wang et al. (1998) proposed a watermarking method, according to multi-threshold wavelet coding (MTWC) and the successive sub band quantization (SSQ) was adopted to search for the significant coefficients. The watermark is added by quantizing the significant coefficient in the significant band using different weights. In [6] two watermarking


methods were proposed. One is the watermark embedded by modifying the triplets of significant coefficients, according to a sequence of information bits. The other considers rectangular blocks of coefficients, and each block is used to embed one watermark bit. [2][7][8] Used the significant coefficients which are selected from global coefficients are showed that can effectively resist attacks. The problem is that the order of extracting the significant coefficients in the extraction process should be exactly the same as those in the embedding process. So they were unsuitable for the blind watermarking, especially for the attacked watermarked image. Wei-Hung et al. [9] used DWT for watermarking a 512 x 512 gray scale image. They quantized the maximum wavelet coefficient of a variable sized block of a selected sub bands. The watermark is a 32 x 16 binary image. In this work, embedding strategy is modified and increased the embedding capacity of the system and observed the improved performance for some attacks. In this paper, a blind and robust watermarking algorithm is presented to embed a binary watermark into a grayscale host image using the local maximum wavelet coefficient modulation. This is a block based approach. Third level LH sub band is selected for embedding the watermark. Selected sub band is divided into various blocks of equal size. Number of blocks must be equal to the number of watermark bits. Maximum coefficient of a block is modulated depending on whether the watermark bit is 1 or 0. Energy of the maximum coefficient is increased if the watermark bit is 1 and makes the energy of the maximum coefficient close to the second maximum if the watermark bit is 0. During the extraction process energy is subtracted from the maximum coefficient of a block. If the coefficient is still the maximum of the block then the decision is taken in favor of a 1, else in favor of a 0.

Experimental results show that the proposed method decreases the distortion of the host image and is effectively robust against JPEG compression, average filtering, Gaussian filtering and the PSNR value of the watermarked image is nearly equal to 40dB.The rest of the paper is organized as follows: In section 2, a brief background material about DWT is presented. In section 3, the proposed method is described. In section 4, we present some experimental results to demonstrate the much improved performance of the proposed method in comparison with the existing techniques. Robustness against the most common attacks is also presented. Finally, the conclusions are given in section 5.

2. DISCRETE WAVELET TRANSFORM

The DWT provides a number of powerful image algorithms including noise reduction, edge detection and compression. The fact that wavelet-based data structure has been adopted in the established image coding standard JPEG2000 encouraged extensive watermarking research in wavelet transform. Wavelet-based watermarking methods exploit the frequency information and spatial information of the transformed data in multiple resolutions to gain robustness.Wavelet transform decomposes an image into a set of band limited components which can be reassembled to reconstruct the original image without error. The DWT is computed by successive low pass and high pass filtering of the discrete time domain signal. Its significance is in the manner it connects the continuous time multi-resolution to discrete time filters. At each level, the high pass filter produces detailed information, while the low pass filter associated with scaling function produces coarse approximations. 2D version of the analysis and synthesis filter banks can be implemented by applying a 1D analysis filter bank to columns of the image and then to the rows. If the image has m rows and n columns, then after applying the 2D analysis filter bank, it produces four sub bands (LL, LH, HL and HH) each having m/2 rows and n/2 columns.

3. PROPOSED WATERMARKING SCHEME3.1. Watermark Embedding:In the proposed algorithm, a binary watermark image is embedded in a gray scale host image. The transform used is DWT. The embedding strategy is

based on the local maximum wavelet coefficient modulation. The steps of the proposed embedding algorithm are as follows.

1. Decompose the host image using 3 level DWT and obtain the sub bands (LL3, LH3, HL3, and HH3).

2. Represent the binary watermark as a vector. Let the number of watermark bits is N w .

3. Divide the LH3 sub band into N w number of blocks.

4. Compute mean value of the maximum wavelet coefficient (MWCmean) and adaptive embedding parameter (aj) as follows:

(a)MWC mean = 1

Nw 1

Nw

j =∑ M j

Where, M j = max j, if the watermark bit is ‘1’ = max j x t1, otherwise max j = maximum wavelet coefficient of the j th block t 1 = scaling factor

(b) a j = maximum |avg j | , | MWC mean x t q| for all j = 1 to N w

Where, avg j =average coefficient value of the j th block t q =scaling factor

5. Modulate maxj according to the Watermark bit for all j = 1 to N w max j new = max j +a j, if the watermark bit is‘1’ = sec j + t 2 x a j, otherwise Where, sec j denotes the second maximum coefficient value of the j th block t 2 = scaling factor (less than 1)

6. Get the modified LH 3 sub band by combining the modulated blocks.

7. Obtain the 3 level inverse DWT using modified LH 3 sub band to get the watermarked image.

3.2. Watermark Extraction:1. Decompose the possibly attacked watermarked image using 3 level DWT and obtain the sub bands (LL3, LH3, HL3, and HH3).

2. Divide the LH 3 sub band into N w number of blocks.

3. Compute the following

(a)MWCmean ׀ =

1Nw 1

Nw

j =∑ max j

׀

Where, max j | denotes the maximum coefficient of j th block.

(b)Mean block = 1

Nw 1

Nw

j =∑ |avg j | |

Where, avg j | is the average coefficient value

of the j th block excluding max j |

4. Detect the watermark bit using following Detection rule for all j = 1 to N w

Watermark bit = 1, if (max j | - a j | ) >= sec j |

= 0, otherwise Where, a j | =maximum

( )| |, | | + + avg Mean k k1 2blockj k1 = max j | / MWC mean

׀

k2 = avg j | / Mean block

sec j | = secondary maximum value of the j th block

4. EXPERIMENTAL RESULTSThe peak signal-to-noise ratio (PSNR) is used to evaluate the quality between an attacked image and the original image. For the sake of completeness, we list the PSNR formula as follows:

[ ]255 255

PSNR 10log10 M N1 2

f(x,y)-g(x,y)M N 1 y=1

dB

x

×=

∑ ∑× =

Where, M and N are the height and width of the image, respectively. f(x, y) and g(x, y) are the values located at coordinates (x, y) of the original image,

and the attacked image, respectively. After extracting the watermark, the normalized correlation coefficient (NCC) is computed using the original watermark and the extracted watermark to judge the existence of the watermark and to measure the correctness of an extracted watermark. It is defined as

1NCC ( , ) '( , )

m n 1 1

m nw i j w i j

i j∑ ∑= ×

× = =

Where, m and n are the height and width of the watermark, respectively. w(i, j) and w ,i)׀ j) are the watermark bits located at coordinates (i, j) of the original watermark and the extracted watermark. Three images are used for the experiments namely Lena, Barbara and Peppers (512 X 512 pixels, 8 bits/pixel). Scaling factors can be adjusted according to PSNR requirement. In these experiments 0.7, 0.1 and 1.0 respectively are used for tq, t1 and t2. The PSNRs of the watermarked Lena, Peppers, and Barbara images are 42.18, 39.61 and 39.17 dB, respectively. Fig (1) shows the original image and Fig (2) the binary watermark. Fig (3) shows the unattacked watermarked image and Fig (4) the extracted watermark. Here both geometric and non-geometric attacks are considered. Non-geometric attacks include JPEG compression, histogram equalization and sharpening. JPEG is one of the most frequently used formats on the Internet and in digital cameras. The JPEG quality factor is a number between 0 and 100 and associates a numerical value with a particular compression level.

Fig (1): Host Image Fig (3): Watermarked Image

Fig (2): Watermark Fig (4): Extracted watermark

When the quality factor is decreased from 100, the image compression is improved, but the quality of the resulting image is significantly reduced. Varied quality factors are applied in the experiments, and the results are shown in Table 1. The proposed method can detect the existence of a watermark through quality factors greater than 15. The results show that the value of NC is still greater than 0.62. For other non-geometric attacks for example median filters, Gaussian filter, average for example median filters, Gaussian filter, average filters, sharpening, and histogram equalization, the resulting images are blurred or sharpened at the edge and shown in Table 2. The proposed method can effectively resist attacks

such as median filter with a mask of size 3 x 3, or average filter with a mask of size 3 x 3. For geometric attacks, rotation, scaling, Gaussian noise and cropping are used. The results are shown in Table 3. We compare the proposed method to Wang and Lin’s [10], Li et al.’s [11], and Lien and Lin’s [12] methods using the Lena image; the results are shown in Table 4. Their watermarking approaches are blind and their methods are based on wavelet-tree quantization. Although Lien improves on Wang and Lin’s [10] approach to watermarked image quality, our method’s PSNR is higher than Lien’s. Moreover, our proposed method is better in terms of robustness.

TABLE 1Normalized correlation coefficients (NCC) after attacked by JPEG compression with quality Factor (QF) 10,15,20,25,30,35,40,50,60,70,80,90 and 100 on (a) Lena, (b) Peppers and (c) Barbara

QF Lena(PSNR=42.18)NCC

Peppers(PSNR=39.61)NCC

Barbara(PSNR=39.17)NCC

10 0.29 0.29 0.4915 0.60 0.47 0.6220 0.70 0.57 0.8025 0.89 0.73 0.8530 0.86 0.75 0.9035 0.85 0.77 0.8940 0.90 0.76 0.8750 0.94 0.87 0.9960 0.93 0.93 1.0070 1.00 0.94 1.0080 1.00 0.96 1.0090 1.00 1.00 1.00100 1.00 1.00 1.00

TABLE 2

Normalized correlation coefficients (NCC) after attacked by median filter (3 x 3), Average filter (3 x 3) Gaussian filter (3 x 3), sharpening and histogram equalization of the three test images (a) Lena, (b) Peppers and (c) Barbara

Non-geometric attacks

LenaNCC

PeppersNCC

BarbaraNCC

Median filter3 x 3

0.91 0.56 0.68

Average filter 3 x 3

0.72 0.46 0.60

Gaussian filter(3 x 3)

Variance =

0.4 0.5 0.6

1.00 0.99 1.00

0.93 0.87 0.99

0.77 0.67 0.91

Sharpening 0.67 0.71 0.60Histogram Equalization

0.76 0.58 0.85

TABLE 3

Normalized correlation coefficients (NCC) values after attack by scaling (512 x 512-256 x 256-512 x512), cropping ¼, rotation, and Gaussian noise for the three test images (a) Lena, (b) Peppers and (c) Barbara.

Geometric attacks

LenaNCC

PeppersNCC

BarbaraNCC

Scaling 50%

0.84 0.52 0.55

Cropping 25%

0.35 0.35 0.35

Rotation

Degree 0.25

0.80 0.70 0.85

Degree -0.25

0.80 0.69 0.80

Degree 0.30

0.79 0.70 0.86

Degree -0.30

0.80 0.71 0.88

Gaussian NoiseVariance = 0.001

0.61 0.65 0.79

TABLE 4

Comparing the proposed method with Wang and Lin’s [10] Li et al.’s [11] and Lien and Lin’s [12] method

Attacks Wang and Lin[10](PSNR =38.2 dB)NCC

Li et al [11](PSNR=40.6 dB)

NCC

Lienand Lin[12](PSNR=41.54dB)NCC

Proposed method(PSNR=42.18dB)NCC

Median Filters(3x3)

0.51 0.35 0.79 0.91

JPEG(QF=15)

NA 0.15 0.17 0.60

JPEG(QF=20)

NA 0.34 0.61 0.70

JPEG(QF=30)

0.15 0.52 0.79 0.86

JPEG(QF=50)

0.28 0.52 0.89 0.94

JPEG(QF=70)

0.57 0.63 0.97 1.00

JPEG(QF=90)

1.00 0.78 1.00 1.00

Sharpening

0.46 0.38 0.88 0.67

Gaussian Filter

0.64 0.70 0.84 1.00

Scaling (50%)

NA 0.35 0.79 0.84

5. CONCLUSIONSIn this paper, a novel oblivious watermarking method based on the DWT using maximum wavelet coefficient modulation is proposed. We use fixed block size and embed a watermark bit using 3 level LH sub-band. Perceptual quality of the watermarked image is good and the watermark can effectively resist common Image processing attacks, especially by JPEG compression (with a quality factor greater than 15) and Gaussian filter. Moreover, the proposed method is more robust in resisting common attacks such as median filter 3x3, average filter 3 x 3,and

Gaussian noise. In addition to copyright protection, the proposed scheme can also be applied to data hiding and image authentication.

6. REFERENCES

[1] Meerwald, P., & Uhl, A., “A survey of watermark-domain watermarking algorithms”, Proceedings of the SPIE, Electronic Imaging, Security and Watermarking of Multimedia Contents III, 2001, pp. 505–516.

[2] Dugad, R., Ratakonda, K., & Ahuja, N., ”A new wavelet-based scheme for watermarking images”, IEEE ICIP, Chicago, IL, 1998, pp. 419–423.

[3] Kim, J. R., Moon, Y. S., “A robust wavelet-based digital watermarking using level-adaptive thresholding”, Proceedings of the IEEE ICIP, Kobe, 1999, pp. 226–230.

[4]. Kwon, S. G., Ban, S. W., Ha, I.-S., et al., “Highly reliable digital watermarking using successive sub band quantization and human visual system”, Proceedings of IEEE ISIE, Pusan, 2001, pp. 205–209.

[5] Wang, H. J., & Kuo, C. C. J., ”High fidelity image compression with multithreshold wavelet coding (MTWC)”, SPIE’s Annual Meeting-Application of Digital Image Processing XX, San Diego, 1997, pp. 383–392.

[6] Davoine, F., “Comparison of two wavelet based image watermarking schemes”, International Conference on Image Processing, Vancouver, 2000, pp. 682–685.

[7] Hsieh, M., Tseng, D., & Huang, Y., “Hiding digital watermarks using multiresolution wavelet transform”, IEEE Transactions on Industrial Electronics, 48(5), 2001, 875–882.

[8] Temi, C., Choomchuay, S., & Lasakul, A., “A robust image watermarking using multiresolution analysis of wavelet”, Proceedings of the IEEE ISCIT, 2005, pp. 623–626.

[9] Wei-Hung Lin, Yuh-Rau Wang, Shi-Jinn Horng, Tzong-Wann Kao, Yi Pan, “A blind watermarking method using maximum wavelet coefficient quantization”, Expert Systems with Applications, Volume 36, Issue 9, November 2009, Pages 11509-11516.

[10] Wang, S.H., Lin, Y.P., “Wavelet tree quantization for copyright protection watermarking”, IEEE Trans. Image Processing 13(2), 2004, 154–165.

[11] Li, E., Liang, H., Niu, X., “An integer wavelet based multiple logo-watermarking scheme”, Proc. IEEE WCICA, 2006, pp. 10256–10260.

[12] Lien, B.K., Lin, W.H., “A watermarking method based on maximum distance wavelet tree quantization”, 19th Conf. Computer Vision, Graphics and Image Processing, 2006, pp. 269–276.

DESIGN AND IMPLEMENTATION OF TAM CONTROLLER AND WRAPPER FOR EMBEDDED CORES

K. CHARAPANI ,DR. P. NEELAMEGAM

School of Computing SASTRA University [email protected]

School of EEE SASTRA [email protected] :This paper presents the development of Wrapper and TAM for IEEE Std 1500 to test Embedded Cores. There are two compliance levels as IEEE 1500 unwrapped and IEEE 1500 wrapped. IEEE 1500 Wrapper allows the test interface of the core to be standardized and the test commands to be executed An FPGA kit is used to synthesis, simulate and for implementation. TAM and wrapper test throughput gains are examined on a core consisting of selected Benchmark circuits. The tolerance and area requirements of the core on various conditions with and without faults are studied.Keywords:Built-in self-testing (BIST), embedded cores-based system-on-a-chip (SoC), sequential circuits, Test Access Mechanism (TAM), test pattern generator (TPG), VHDL, wrapper.

1. INTRODUCTION

Large-scale integration has added enormous complexity to the process of testing modern digital circuits. Besides, during the past several years, integrated circuit technology evolved from chip-set philosophy to embedded cores based system-on-a-chip (SoC) concept [1] that simply refers to an IC, designed by stitching together multiple stand-alone VLSI designs to provide full functionality for an application. Though many aspects of these embedded cores-based systems and SoC are still evolving, they are revolutionizing the electronics industry. These innovations are already on their way to the next generation of cell phones, multimedia devices, and PC graphics chipsets. The cores-based design, justified by the necessity to decrease time-to-market, has created a host of challenges for the design and test community [4] [9]. Unlike traditional test design approaches, SoC makes it impossible to establish the demarcation lines between design and test. For mixed-signal devices and complex digital cores, engineers must use design tools that let them incorporate testability [7] early in the design process. Though the use of cores in a SoC serves a wide range of applications, they are far too complex to be tested by traditional methods. In practice, combinations of test methodologies such as functional test, full-scan test, logic BIST, RAM BIST, IDDQ, etc.

are normally used for SoCs. However, difficulties surround implementation of the methods for testing cores and SoCs as a whole. A wide range of circuits traditionally comes within the definition of SoCs including microprocessors and microcontrollers. The core test integration is a complex problem – the chip integrator can modify the test and add some design for test (DFT) and built-in self-test (BIST) features, if necessary. Specifically, in the context of embedded cores-based system testing, electrical isolation involving the input and output ports of the core from the chip or other cores is a necessity. The fundamental items of interest in core test is access, control, and isolation, and these are the issues which were addressed by the IEEE Technical Council on Test Technology Working Group 1500[2] entrusted with the responsibility of developing standard architecture for their solution. The embedded core test requires, in general, hardware components like wrapper around the core, a source and a sink for test patterns (on-chip or off chip) and an on-chip Test Access Mechanism (TAM) to connect the wrapper to the source or sink. The cores could be without boundary scan or with boundary scan. For design and test reuse, ASIC manufactures have suggested certain characteristics. In general, different DFT and BIST schemes like scan, partial scan, logic BIST and scan-based BIST are used to test various logic blocks within a SoC like microprocessor or microcontroller.


However, the main problem is still the resulting area overhead and performance penalties. Structural test methods like scan and BIST are desirable for test reuse, portability, and test integration into the SoC test set. The TAM includes on-chip test generation logic for cores with BIST. The DFT techniques involve adding optimized test logic within cores and at the chip level to enhance testability and DFT logic helps test pattern generation and application, and assist in the support test environment. In this paper, test methodologies are proposed for embedded cores-based SoC digital systems, which consist of wrapper and Test Access Mechanism (TAM). Some of ISCAS 85 combinational and ISCAS 89 sequential benchmark circuits are consider as a core. The fault model used is the conventional single stuck-fault model. In IEEE 1500-compliant [2], the wrapper separates each core under test condition. The Test Access Mechanism has an important role in transporting the test patterns to a selected core and this core responds to the output pin of the SoC. The faults are injected using a fault simulator, which in turn generates the test sets for the cores. The Test Access Mechanism (TAM) is common for all the cores in the SoC, acts as plain signal transport medium. Once the compilation of the cores is done, the fault simulation is carried out with the test patterns feeding the cores through the TAM. The program running in the background is responsible for the selection of the appropriate core. The automatic fault simulator generates tests for the core described at the gate and flip-flop level. There are very few constraints on the core being tested. The simulation process is completely automatic, and requires no intervention from the designer during the test generation process. The subject paper describes the architectures of the wrapper and Test Access Mechanism, applications of the fault simulator, together with models of the SoCs being used, based on application environment. Some partial simulation results using the designed cores are also provided to demonstrate the feasibility of the proposed implementations.

2. 1500-BASED SOC TEST INTEGRATION ARCHITECTUREThe IEEE 1500 wrapper [2] [3] has various modes of operation. There are modes for functional (nontest) operation, inward facing (IF) test operation, and outward facing (OF) test operation.

Different test modes also determine whether the serial test data mechanism (WSI–WSO) or the parallel test data mechanism (WPI–WPO), if present, is being utilized. Instructions loaded into the WIR, together with the IEEE 1500 wrapper signals, determine the mode of operation of the wrapper and possibly the core itself. There is a minimum set of instructions and corresponding operations that shall be supplied. Optional instructions and their corresponding behavior are also defined, together with the requirements for extension of the instruction set. All instructions that establish test modes that utilize the parallel port WPI and WPO are optional, as the presence of this port is optional. Furthermore, IEEE Std 1500 also allows for user-defined instructions.IEEE Std 1500[2] has a set of instructions that are defined to use only the serial interface (WSP) and a corresponding set of instructions that are defined for the parallel interface. IEEE Std 1500 must allow accessibility to test the core. There is one main core test instruction—Wx_INTEST (user-specified core-test instruction)—that is flexible enough to allow any core test to execute. IEEE Std 1500[2] does not define the test for the core, and this very flexible instruction was specified so that it could be defined differently for each core test. There are two other instructions that are mandatory: an instruction for functional mode (WS_BYPASS) and an instruction for external test mode (WS_EXTEST). WS_BYPASS puts the wrapper into the bypass configuration and allows access to all functional terminals of the core shown in Fig 1. WS_EXTEST is the serial EXTEST configuration of the wrapper. Even if there is a WP_EXTEST mode (for parallel access), there must still be a WS_EXTEST instruction capability.

Figure1: Standard IEEE 1500 Wrapper Components

3. TEST ACCESS MECHANISM (TAM) AND WRAPPERThe design of Test Access Mechanism (TAM) [4] and test wrapper is of critical importance in terms of system integration since they directly impact hardware overhead, test time, and tester data

Figure 2: TAM Architecture

volume. The main issues in this context are wrapper optimization, core assignment to TAM wires, sizing of the TAMs, and TAM wire routing. There are two important concepts related to TAM, viz. test pattern source and sink, and core wrapper. The test pattern source is responsible for generating the test vectors or test stimuli for the desired core under test. The test pattern sink compares the fault-free response to the faulty response of the core under test. The test pattern source and sink can be built on-chip or off-chip. In our case, we implemented this by using the fault simulator that generates the test vectors, and after getting the test response, compares it with the fault-free response. We will briefly discuss about fault simulation later. There are several ways to design and implement a TAM [5].

Some of the common TAM architectures are: 1) Daisy chained TAMs (that use serial

shifting of test data) 2) Bussed TAMs (based on use of complex

protocols) 3) Direct access TAMs (for cores with many

inputs and outputs)4) Multiplexer (MUX)-based direct access

TAMs. In this paper, we implemented MUX-based direct access TAM architecture, as shown in Fig. 2.

The TAM is used to drive the test vectors from the test source [5] [6], that is, from the fault simulator to the desired core under test and to transport the test response from the core back to the fault simulator.

The selection of the core in the SoC was implemented as part of the TAM architecture. We determined the width of the TAM by the core that has the maximum number of input/output (I/O) pins within the SoC. There are other issues for consideration at this phase, viz. the bandwidth of the TAM versus the cost of extra wires needed for its implementation, total test time depending on the TAM bandwidth, test vectors from the source, and ultimately test data for the individual core. The obvious mechanism to make embedded cores testable from the IC pins [10] is to make the core under test directly accessible from the IC inputs. Though this approach is mostly practiced for embedded memory cores, many block-based ASICs also use this test strategy.

4. TAM CONTROLLER DESIGN

As shown in Fig. 3, TAM controller is used to provide the dynamic control signals to wrapper [4], i.e., WIP. WIP is composed of six signals: WRCK, WRSTN, SelectWIR, ShiftWR, CaptureWR and UpdateWR. At the exception of WRSTN, all WIP signals are active high.

• WRCK (Wrapper Clock): The dedicated test clock.

• WRSTN (Wrapper Reset): The dedicated asynchronous reset signal, which resets Wrapper Instruction Register (WIR) and puts the wrapper in normal operation mode.

• SelectWIR: Selects whether WIR or one of wrapper data registers (WDR) is exclusively connected between WSI and WSO.

• ShiftWR, CaptureWR: When the corresponding signal is asserted, a shift or capture operation will occur on the rising edge of WRCK.

• UpdateWR: When it is asserted, an update operation will occur on the falling edge of WRCK.

Figure 3: TAM Controller State Diagram

TAM controller can be used, as long as its outputs fully conform to the requirement of WIP. TAM controller is a Finite State Machine (FSM) in nature. Since WIP signals mimic the output of TAP controller of IEEE 1149.1[9], we use the same state-diagram as TAP (Fig. 3) and just modified the output logic of FSM in order to make it suitable for WIP. The three inputs (tclk, tms, trst) of TAM controller have just the same meaning as in TAP, that is, TAM controller changes its state on the rising edge of tclk according to tms. In our design, tclk is also used as WRCK.

5. WRAPPER DESIGN

P1500 wrapper consists of wrapper instruction register (WIR), wrapper bypass register (WBY), wrapper boundary register (WBR), gating control logic, mandatory serial path, and optional parallel paths. The serial path is used both for wrapper control by loading instructions into WIR, as well as for low-bandwidth test data access to WBR. Parallel paths are used for internal scan-chains and high-bandwidth test data access to WBR. A minimal library of WBR cells is presented in and the design of WBR is not depicted in this paper. WBY used in our wrapper is just an ordinary flip-flop. We insert bypass registers both for the serial path and for parallel paths.WIR is composed of shift register; update register and decoding logic (Fig. 1). At the start of every test cycle, a new instruction is shifted into WIR through the serial path. Consequently, the operation of the wrapper is controlled by both WIP signals, as well as the presently active instruction in WIR. WIR outputs two static control signals for the wrapper. Sel identifies whether WBR or WBY is connected between WSI and WSO and M2_Mode determines whether the test data in WBR apply to system logic or not. WRSTN resets the content in the update

register and puts the wrapper in normal operation mode.

6. CORE TEST GENERATION AND TEST ENVIRONMENT

From ISCAS 85 combinational and ISCAS 89 sequential benchmark circuits [8] [9], C17, S27, and S298 are selected and designed in VHDL as a core. The problems related to testing sequential circuits are much more complicated than those for the combinational circuits are. The basic reason is the presence of storage elements in the realization of sequential circuits. The testing time is also generally higher in the case of sequential circuits. The test generation was focused at the gate level based on single stuck-fault model [11]. The first step in our SoC test generation procedure is to obtain specified sets of test vectors using high-level descriptions of circuit cores. We endeavored to improve upon the process by using automatic test generation for high fault coverage.

All the SoCs used in the paper were designed to operate in Altera Max Plus II development environment. Not only was the SoC cores, the Test Access Mechanism (TAM) also designed to work described in VHDL. The designed fault simulator carried out the process of fault injection for the different SoC cores. The fault simulator was written in C programming language, and injects faults in three distinct steps in a core. It first injects faults at the input lines, then at the output lines, which were followed by fault injection at the internal wires. The nature of faults is single stuck faults. Thus, each line can have only two types of stuck faults: stuck-at-1 and stuck-at-0. A line with a stuck-at-1 fault will always have a logical value 1 irrespective of the correct logical output of the gate driving it. We used here simulation based fault injection scheme to evaluate the dependability of the system based on the percentage of fault coverage. Each line is injected either with a stuck-at-1 or with a stuck-at-0 fault.data set is provided to the test program and the fault coverage is finally calculated. 7. IMPLEMENTATION RESULTTo demonstrate the feasibility of our developed test environment and test methodologies for embedded cores- based SoCs, independent simulations were conducted on the cores of the various SoCs that we constructed based on ISCAS 85 combinational and ISCAS 89 sequential benchmark circuits. We first provide here some simulation data on an SoC that

was constructed from the sequential circuits in ISCAS 89 benchmark circuits list. Next, simulation experience on a SoC comprised of the combinational circuits in ISCAS 85 benchmark circuits list is given. Finally, the selected benchmark circuits are being tabulated respectively and SoCs comprised of both sequential and combinational circuits are considered.

Table. 1 shows the number of injected faults, detected faults, and fault coverage for the first SoC (SoC_seq) using pseudorandom test vectors. Table. 2 shows the result for core area with wrapper for our experimental SoC.

Table.1: Test Data for the cores

Table.2: Core Area with Wrapper

8. CONCLUSIONS

Embedded cores-based design paradigm has evolved from the necessity to increase design productivity and decrease time-to-market, but the result creates numerous challenging problems for the test design community. Keeping in view the many formidable issues that arise in testing these cores-based SoCs, this paper furnishes, approaches for digital SoC, to develop test environments and methodologies. A 1500-compliant wrapper and TAM controller co-design scheme is presented in this paper. In our scheme, only one TAM controller is required at SoC level hence costing low area overhead.

The TAM plays an important role in transporting the test patterns for a desired cores in a digital SoC and the resulting responses to the output pins of the SoCs is implemented, as a plain signal transport medium, passes through the above mentioned systems. The faults are injected using fault simulator that generates tests for the tested core. Testing power consumption of the wrapper is also taken into account by using gated clocks to operate wrapper registers. The effectiveness of the proposed scheme has been confirmed by experiment results based on a sample SoC.

9. REFERENCES

1. Yervant orian, Erik Jan Marinissen, and Sujit Dey, Testing Embedded-Core-Based System Chip,IEEE Computer, 32(6):52–60, 1999.

2. IEEE P1500 WebSite.http://grouper. IEEE.org/groups/l500/index.html

3. Erik Jan Marinissen, Rohit Kapur, and Yervant orian, On Using IEEE P1500 SECT for Test Plug-n-Play, In Proceedings IEEE International Test Conference(ITC), pp.770 –777, Atlantic City, NJ, Oct 2000.

4. Vikram Iyengar, Krishnendu Chakrabarty, and Erik Jan Marinissen, Test Wrapper and Test Access Mechanism Co-Optimization for System-on-Chip, In Proceedings IEEE International Test Conference (ITC), pp. 1023–1032, Baltimore, MD, 2001.

5. Vikram Iyengar, Krishnendu Chakrabarty, and Erik Jan Marinissen, Test Access Mechanism Optimization, Test Scheduling, and Tester Data Volume Reduction for System-on-Chip, IEEE Transactions on Computers, 52(12): 1619-1632, 2003.

6. R. Rajsuman, System-on-a-Chip: Design and Test. Boston, MA: Artech House, 2000.

7. K. Chakrabarty, V. Iyengar, and A. Chandra, Test Resource Partitioning for System-on-a-Chip. Boston, MA: Kluwer, 2002.

8. K. Chakrabarty (Editor), SoC (System-on-a-Chip) Testing for Plug and Play Test Automation. Boston, MA: Kluwer, 2002.

9. S. Mourad and Y. Zorian, Principles of Testing Electronic Systems. New York: Wiley, 2000. Circuit

10. T. W. Williams and K. P. Parker, “Testing logic networks and design for testability”, Computer, vol. 21, pp. 9-21, Oct. 1979.

11. S. R. Das, C. V. Ramamoorthy, M. H. Assaf, E. M. Petriu, and W. –B. Jone, “Fault tolerance in systems design in VLSI using data compression under constraints of failure probabilities”, IEEE Trans. Instrum. Meas., vol. 50, pp. 1725-1747, Dec. 2001.

A COMPARATIVE ANALYSIS OFPIXEL LEVEL FUSION OF MULTI-FOCUSED IMAGES USING

DISCRETE AND STATIONARY WAVELET TRANSFORM

K. KANNAN1, S. ARUMUGA PERUMAL2, K. ARULMOZHI3 1 Asst. Professor, Department of Electronics and Communication Engineering, 3Principal ,

Kamaraj College of Engineering & Technology., Virudhunagar, Tamilnadu, [email protected],[email protected]

2Professor & Head, Department of Computer Science, St. Hindu College, Nagarcoil, Tamilnadu, [email protected]

ABSTRACT : The fast development of digital image processing leads to the growth of feature extraction of images

which leads to the development of Image fusion. Image fusion is defined as the process of combining two or more different images into a new single image retaining important features from each image with extended information content. There are two approaches to image fusion, namely Direct fusion and Multi resolution fusion. In Direct fusion, the pixel values from the source images are directly summed up and taken average to form the pixel of the composite image at that location. Multi resolution fusion uses transform for representing the source image at multi scale. The most common widely used transform for image fusion at multi scale is Wavelet Transform since it minimizes structural distortions. There are three levels in wavelet transform based multi resolution fusion scheme namely Pixel level fusion, Area level fusion and Region level fusion. Among these three levels, pixel level fusion provides best results. This paper compares the performance of pixel level fusion of multi focused images using discrete wavelet transform and stationary wavelet transform in terms of various performance measures.

key words: Image fusion, Discrete Wavelet Transform, Stationary Wavelet Transform, Mean Square Errors,

Signal to Noise Ratios, Correlation Measures and Structural Measures.

I. INTRODUCTIONImage fusion is defined as the process of combining two or more different images into a new single image retaining important features from each image with extended information content. For example, IR and visible images may be fused as an aid to pilots landing in poor weather or microwave and visible images may be fused to detect weapons or CT and MRI images may be fused as an aid to medical diagnosis. The fusion process must satisfy the following requirements such as it should preserve all relevant information in the fused image, should suppress noise and should minimize any artifacts in the fused image. There are two approaches to image fusion, namely Direct fusion and Multi resolution fusion. In Direct fusion, the pixel values from the source images are summed up and taken average to form the pixel of the composite image at that location [1]. Multi resolution fusion uses pyramid or wavelet transform for representing the source image at multi scale. There are three levels in multi resolution fusion scheme namely Pixel based fusion, Area based fusion and Region based fusion. Among these three levels, pixel based fusion provides best results and more

research taken place in this level. The performance measures which can be computed independently of the sub-sequent tasks express the successfulness of an image fusion technique by the extent that it creates a composite image that retains salient information from the source images while minimizing the number of artifacts or the amount of distortion that could interfere with interpretation. In this paper, it is proposed to compare the performance of pixel level fusion of multi focused images using discrete wavelet transform and stationary wavelet transform in terms of various performance measures.

II. DISCRETE WAVELET TRANSFORM

The Discrete Wavelet Transform (DWT) of image signals produces a non-redundant image representation, which provides better spatial and spectral localization of image formation, compared with other multi scale representations such as Gaussian and Laplacian pyramid. Recently, Discrete Wavelet Transform has attracted more and more

interest in image fusion. An image can be decomposed into a sequence of different spatial resolution images using DWT. In case of a 2D image, a single level decomposition can be performed resulting in four different frequency bands namely LL, LH, HL and HH sub bands. In general, N level decomposition results 3N+1 sub bands and it is shown in figure 1.

Figure.1: 2D-Discrete Wavelet Transform

III. STATIONARY WAVELET TRANSFORM

The Discrete Wavelet Transform is not a time- invariant transform. The way to restore the translation invariance is to average some slightly different DWT, called decimated DWT, to define the stationary wavelet transform (SWT). Let us recall that the DWT basic computational step is a convolution followed by decimation. The decimation retains even indexed elements. But the decimation could be carried out by choosing odd indexed elements instead of even indexed elements. This choice concerns every step of the decomposition process, so at every level we chose odd or even. If we perform all the different possible decompositions of the original signal, we have 2J different

decompositions, for a given maximum level J. Let us denote by j = 1 or 0 the choice of odd or even indexed elements at step j.

Every decomposition is labeled by a sequence of 0’s and 1’s: = 1, J. This transform is called the decimated DWT. It is possible to calculate all the decimated DWT for a given signal of length N, by computing the approximation and detail coefficients for every possible sequence. The SWT algorithm is very simple and is close to the DWT one. More precisely, for level 1, all the decimated DWT for a given signal can be obtained by convolving the signal with the appropriate filters as in the DWT case but without down sampling. Then the approximation and detail coefficients at level 1 are both of size N, which is the signal length. The general step j convolves the approximation coefficients at level j-1, with up sampled versions of the appropriate original filters, to produce the approximation and detail coefficients at level j. This can be visualized in the following figure 2.

Figure.2: 2D-Stationary Wavelet Transform

IV. WAVELET BASED IMAGE FUSION

Wavelet transform is first performed on each source images, and then a fusion decision map is generated based on a set of fusion rules. The fused wavelet coefficient map can be constructed from the wavelet coefficients of the source images according to the fusion decision map. Finally the fused image is obtained by performing the inverse wavelet transform. Let A (x, y) and B (x, y) are images to be fused, the decomposed low frequency sub images of A (x, y) and B (x, y) be respectively lAJ

(x, y) and lBJ

(x, y) ( J is the parameter of resolution) and the decomposed high frequency sub images of A(x,y)

and B(x,y) are hAjk (x, y) and hBj

k (x, y). ( j is the parameter of resolution and j=1,2,3….J for every j, k=1,2,3..).

Then, the fused high and low frequency sub-images Fj

k (x, y) are given as Fjk (x, y) = Aj

k (x, y) if G(Ajk (x,

y)) >= G(Bjk (x, y)), else Fj

k (x, y) = Bjk (x, y) and FJ

(x, y) = lAJ (x, y) if G(AJ (x, y)) >= G(BJ (x, y)), else

FJ (x, y) = lBJ (x, y) where G is the fusion rule and

Fjk (x, y) & FJ (x, y) are used to reconstruct the fused

image F′(x, y) using the inverse wavelet transform. The block diagram representing the wavelet based image fusion is shown in figure 3.

Figure.3: DWT/SWT Based Image Fusion

V. PIXEL LEVEL IMAGE FUSIONThis section describes five methods of pixel level image fusion based on multi scale representation of source images using wavelets[5]. To simplify the description of different pixel level image fusion methods, the source images are assumed as A & B and the fused image as F. All the methods described in this paper can be used in the case of more than two source images.Method 1: All the four sub bands of the fused image F is simply acquired by averaging the wavelet coefficients of source images A&B. i.e. Fj

k = (Ajk +Bj

k

)/2 & FJ = (lAJ+lBJ) /2. However when this method is

applied, there is chances of reduction of contrast of features uniquely presented in either of the images.Method 2: All the four sub bands of the fused image F is simply formed by taking the wavelet coefficients from source images which is having the maximum value. i.e. Fj

k = max (Ajk,Bj

k ) & FJ = max (lAJ, lBJ) .

Limitation is in the fusion of patterns that have roughly equal salience but opposite contrast, which results pathological [1].Method 3: Since larger absolute transform coefficients correspond to sharper brightness changes, the good integration rule is to select, at

every point in the transform domain, the coefficients whose absolute values are higher [2]. i.e. F j

k = max ( abs(Aj

k), (Bjk )) & FJ = max (abs (lAJ), abs(lBJ))

.Method 4: This method uses method1 for low frequency sub bands and uses method 3 for high frequency bands. i.e. Fj

k = max ( abs(Ajk), (Bj

k )) & FJ

= (lAJ+lBJ) /2 .

Method 5: This method is slight modification of method 4. It takes the average of approximation and diagonal details of source image to form the LL and HH sub band and uses absolute maximum criterion to form the LH and HL sub bands of fused image.

VI. EVALUATION CRITERIA

Over the past decade, researchers in a variety of fields have explored the potential benefits of applying wavelet transforms to fuse images. Objective image quality measures play important roles in various image fusion schemes. There are two classes of objective quality or distortion assessment approaches. The first are mathematically defined measures such as the widely used Mean Square Errors like Root Mean Square Error (RMSE), Laplacian Mean Square Error (LMSE) & Normalized Mean Square Error (NMSE), Signal to Noise Ratios like Signal to Noise Ratios (SNR) & Peak Signal to Noise Ratio (PSNR) and differences like Maximum Difference (MD), Average Difference (AD) & Normalized Absolute Difference (NAD). The second class of measurement methods considers correlation measures like Cross Correlation (NK) & Correlation Coefficient (CQ) and structural measures like Structural Content (SC), Luminance Distortion (LD), Contrast Distortion (CD), Quality Index (QI), Image Fidelity (IF), Histogram Similarity (HS) & Average Gradient (AG) [3,4]. This paper evaluates all the pixel level image fusion methods using above performance measures.

VII. EXPERIMENTAL WORK

The methods proposed for implementing pixel level image fusion using wavelet transform take the following form in general. The two source images to be fused are assumed to be registered spatially. The images are wavelet transformed using the HAAR wavelet, and transformed to the same number of levels. For taking the wavelet transform of the two images, readily available MATLAB routines are taken. In each sub-band, individual pixels of the two images are compared based on the fusion rule that serves as a measure of activity at that particular scale

and space. A fused wavelet transform is created by taking pixels from that wavelet transform that shows greater activity at the pixel locations. The inverse wavelet transform is the fused image with clear focus on the whole image.

VIII. CONCLUSIONThis paper presents the comparison of pixel based fusion of multi focused images using discrete and stationary wavelet transform in terms of various performance measures. The pixel level fusion using stationary wavelet transform provides computationally efficient and better qualitative and quantitative results than that of discrete wavelet transform. Hence using these fusion method one can enhance the image with high geometric resolution.

Reference1. P. J. Burt and R. J. Kolczynski, “Enhanced

image capture through image fusion”, proceedings of the 4th International Conference on Computer Vision, pp. 173-182, 1993.

2. H. Li, B.S. Manjunath, and S.K. Mitra, “Multi-sensor image fusion using the wavelet transform”, Proceedings of the conference on ‘Graphical Models and Image Processing’, pp. 235–245, 1995.

3. Zhou Wang and Alan C. Bovik, “A Universal Image Quality Index”, IEEE Signal Processing Letters, Vol. 9, No.3, pp. 81-84, March,2002.

4. Marta Mrak, Sonja grgic and Mislav Grgic, “Picture Quality Measures in Image Compression Systems”, EUROCON 2003, Ljubljana, Slovenia, PP. 233-237.

5. K. Kannan and S. Arumuga Perumal, “Optimal decomposition Level of Discrete Wavelet Transform for pixel based fusion of multi-focused Images”, Proceedings of the International Conference on “Computational Intelligence and Multimedia Applications”, ICCIMA’07, pp 314-318, Volume III, December,2007.

Table: 1. Results of Pixel Level Image Fusion using Wavelets

DETECTION OF UNSOLICITED MAIL (SPAM) IN APPLICATION LAYER

MRS.N.NAGADEEPA1

(Research Scholar)Lecturer in Computer Science,

Vivekanandha College of Arts and Sciences for Women, Tiruchengode (Namakkal

,DR.R.S.D.WAHIDABANU2

Dr.R.S.D.Wahidabanu2 HOD and Professor of ECE,Govt. College of Engineering, Salem-11.

ABSTRACT:The session initiation Protocol (SIP) is used for managing multimedia session in the Internet. As an emerging standard gaining more and more acceptance within the IT community, SIP will probably be the target of spammers. In this paper, a reputation based mechanism is proposed that builds trust between users within a SIP community and prevents spammers to carry out attacks. This technique uses a social network approach enhanced with some reputation ratings. A method for computing reputation is defined and an algorithm allowing the users to evaluate each other is also presented.

Key Words: Spam detection

1. INTRODUCTIONUnsolicited commercial email is called spam. The latter may hide action that varies from marketing advertisements to some computer virus spreading. Spammers send commercial advertisements for their products and services. For instance, some statistics [8] reveal that 52% of email users say spam has made the users less trusting and 25% say that the volume of spam has reduced the users usage of email. Spam appears as a real thread that the service providers try to combat using in particular some filtering mechanism. These filters are not always efficient as they have to make sure before discarding any message that the latter is a spam message with an extremely high confidence. In addition, these filters also need to be maintained and updated regularly.Reputation systems [15] [16] are mechanism that have beer, used in many different areas to build trust between members of a certain community. The reputation systems are in general lightweight and can be maintained easily.This paper describes a mechanism that uses the reputation concept for building trust between users in a SIP community. This mechanism allows the users to rate each other according to some predefined criteria. Based on this rating, a “SIP Social Network” is developed. This allows identifying the spammers and discarding the spammer from the network.

2. STATEMENT OF PROBLEM

Spam is an unwanted mail, sent by the unknown person for their benefits and services. Very often commercial advertisements are sent to the email user by the spammer. There are variety of solutions and techniques available to filter the spam messages. But there is no technique to filter the spam messages in application layer. So, technique must be devised to detect the spam in the application layer.

3. Definition of SIP

The SIP is a Protocol. It is presented in application layer for managing multimedia sessions in the Internet [1]. A session can be established between two end users, and the session is also used to make IP phone calls, conferencing and messaging. SIP is handled through messages such as INVITE and BYE. INVITE message is for initiating session and BYE message is to terminate the session.

4. Components of SIP

4.1. User AgentsThese are the end devices in a SIP network. They can originate SIP requests to establish a media session

and send receive media. A user agent can be either a SIP phone or SIP client software running on a PC.

4.2. Servers

These are intermediate SIP entities in a SIP network that assist the user agents in establishing media sessions and some other functions.

4.3. Location servers

Location server is a database where users information such as URLs, IP addresses, Scripts and other preferences are stored. It also contains routine information such as locations of proxies, gateways and other location servers.

SIP spam can take one of the following forms [5],5. Different forms of spam in SIP

5.1. Call spamThis is the case of unsolicited messages for establishing voice, video or IM session. The spammer proceeds to relay the message over the real time media. This form is the usual way used by telemarketers.

5.2. IM spamThis form is similar to email spam, unsolicited IMs whose content contains the message that the spammer is seeking to convey. The SIP MESSAGE request will be used here but also some other messages such as INVITE with text or html bodies.

5.3. Presence spamThis spam is similar to the previous one, i.e. unsolicited presence (subscribe) requests are sent to get on the buddy list of a user.

5.4. SIP social networksA network refers to a set of items called nodes. The nodes are connected with each other. A social network is a set of people with some pattern of contacts or interactions between them. These patterns can be friendship of individuals, business relationship between companies.

Figure 1: SIP Network

6. Methodology

A bunch of solutions to SIP spam is available. In general, these solutions are some mechanisms developed to deal with email spam and which could be adapted to SIP spam. Among these solutions, one can mention, content filtering, whitelists, content based communications and identity authentication [5]. Unfortunately, if these solutions can be applied to the SIP spam problem, they seem to be insufficient and need to be combined with each other or with some other techniques to provide robustness.SIP provider offers SIP based services such as VoIP calls and Instant Messaging. Let’s assume that any user of these services is managing a contact list to keep track of his friends that they are allowed to communicate with them. Usually, buddy lists are used for the Instant Messaging service and phonebooks are used for the telephony services. However, along this paper, contact list will be used as general terminology to describe lists of a user’s friends that are allowed to communicate with the user. Let’s also assume that any SIP user is able to assign a reputation value to all the persons figuring on his contact list. These reputation values simply refer to how trustful the assigned persons are.

7. Architecture

This architecture is based on the reputation. It is used to detect unsolicited calls or messages within the SIP user community. For detecting SIP spam, this architecture does not require anything on the user’s side except that when the users create their contact lists, they must score the person on this list according to evaluation criteria that the service provider has predefined.

Figure 2: Network Architecture

7.1. SIP repository (warehouse)SIP warehouse is a database. Contact lists of the different users are stored in the database. The first column of the record consists of users names or their URLs. The second column includes the reputation values that the owner of the list has assigned to each member of the user’s contact list. The contact list are created and updated by the user through a web interface.

7.2. Reputation Network ManagerFunction of the RNM is building the SIP network. Basedon the contact list the RNM construct the network as shown in figure 1 and also consequently accesses the reputationwarehouse to update the contact list and SIP network. The RNM receives the SIP server from the SIP server, and computes the reputation value related to this requestand compares it to a predefined threshold. If this reputation value is greater than the threshold, the RNM will inform the SIP proxy to process the request, otherwise, the SIP server will reject it.

8. ConclusionIn this paper, an approach combining social networks and reputation rating to deal with SIP spam is presented. After having described the problem space, a metric defined to be used in the rating operation as well as an algorithm designed for searching for eventual paths between the source and the destination and provide the corresponding reputation value. This mechanism can work in performance with some other filtering techniques to prevent spam problems within SIP communities.

References

[1] J. Rosenberg et al, “SIP: Session Initiation Protocol”, RFC 3261, June 2002

[2] J. Rosenberg et al, “Session Initiation Protocol

(SIP): Locating SIP Servers”, RFC 3263, June 2002

[3] Earthlink, http://www.earthlink.net

[4] Iptel.org, http://www.iptel.org

[5] J. Rosenberg et al, “The Session Initiation protocol (SIP) and Spam”, draft-rosenberg-sipping-spam-00, July 11, 2004

[6] J. Golbeck et al, “Trust Networks on the Semantic Web”,http://www2003.org/cdrom/papers/poster/p314/p314_golbeck.html

[7] J. Golbeck et al, “Reputation Network Analysis for Email Filtering”, http://www.ceas.cc/papers-2004/177.pdf

[8] P. Oscar et al, “Personal Email Networks: An Effective Anti-Spam Tool”, http://arxiv.org/abs/cond-mat/0402143

[9] M. E. J. Newman et al, “Email Networks and the Spread of Computer Viruses”, Physical Review E66, 0351001R (2002)

[10] M. E. J. Newman, “Properties of Highly Clustered Networks”, Physical Review E68, 026121 (2003)

[11] M. E. J. Newman et al, “Finding and Evaluating Community Structure in Networks”, Physical Review E69, 026113 (2003)

http://arxiv.org/abs/cond-mat/0402143

http://www.ceas.cc/papers-2004/177.pdf

http://www.ceas.cc/papers-2004/177.pdf

http://www.iptel.org/

http://www.earthlink.net/

[12] J. M. Pujol et al, “Reputation Measures based on Social Networks metrics for Multi Agent Systems”,

[13] D. Senn, “Reputation and Trust Management in Ad Hoc Networks with Misbehaving Nodes”, Diploma Thesis DA-2003.27, July 2004

[14] A. Alexa, “Reputation Management in P2P Networks: The EigenTrust Algorithm”, http://www.mpisb.mpg.de/units/ag5/teaching/ws03_04/p2p-data/01-20-paper2.pdf

[15] L. Xiong, L. Liu, “A Reputation-Based Trust Model for Peer-to-Peer eCommerce Communities”, Proc of the IEEE International Conference on E-commerce (CEC’03), 2003

[16] B. Yu et al, “A Social Mechanism of Reputation Management in Electronic Communities”, Proceedings of Fourth International Workshop on Cooperative Information Agents”, pp 154-165, 2000

[17] G. Zacharias et al, “Trust Management Through Reputation Mechanisms”, Applied Artificial Intelligence 14, pp 881-907, 2000

http://www.mpisb.mpg.de/units/ag5/teaching/ws03_04/p2p-data/01-20-paper2.pdf

http://www.mpisb.mpg.de/units/ag5/teaching/ws03_04/p2p-data/01-20-paper2.pdf

STUDIES ON FRICTION STIR WELDED ALUMINIUM ALLOY-ROUGHNESS AND MACRO/MICROSCOPIC PERSPECTIVE

G. RAGHU BABU1* DR.K.G.K.MURTI2 DR.G.RANGA JANARDHANA3

1. Associate Professor, Dept. of Mech. Engg., VNR Vignana Jyothi Institute of Engineering&Technology, Hyderabad.

2. Professor, Dept. of Mech. Engg., Gokaraju Rangaraju Institute of Engineering &Technology,Hyderabad.

3. Professor, Dept. of Mech. Engg., College of Engineering, Jawaharlal Nehru TechnologicalUniversity, Kakinada

* Corresponding author E-mail: [email protected]

ABSTRACT:Friction Stir welding is a promising solid state joining process and is widely being considered for aluminium alloys. FSW can eliminate casting defects and refine microstructures, enhancing formability and improving other properties. It can also produce fine-grained microstructures through thickness. Essentially, FSW is a local thermo-mechanical metal working process that changes the local properties without influencing the properties of the bulk material. Firstly, this paper introduces FSW fundamental parameters as advancing and rotating tool speed, approaching their influence on the processed surface roughness of Al6061.Single and overlapped friction stir welded joints were considered for the roughness analysis. Roughness mapping revealed existence of transitory and quasy stationary phases of FSW. A process monitoring method obtaining through roughness control is envisaged.

Keywords: Friction stir welding, aluminium alloys, material flow, roughness, microstructure

1. INTRODUCTION

Heat treatable wrought aluminium- magnesium-silicon alloys conforming to AA6061 are of moderate strength and possess excellent welding characteristics over the high strength aluminium alloys [1]. Hence, alloys of this class are extensively employed in marine frames, pipelines, storage tanks and aircraft applications.Compared to many of the fusion welding processes that are routinely used for joining structural alloys, friction stir welding (FSW) is an emerging solid state joining process in which the material that is being welded does not melt and recast [2]. Friction Stir welding (FSW) was invented at The welding Institute (TWI), UK in 1991. Friction stir welding is a continuous, hot shear, autogeneous process involving non-consumable rotating tool of harder material than the substrate material. Defect- free welds with good mechanical properties have been made in a variety of aluminium alloys, even those previously thought to be not weldable [3]. When alloys are friction stir welded, phase transformations that occur during the cooling cycle of the weld are of a solid state type. Due to the absence of parent metal melting, the new FSW process is observed to offer several advantages over fusion welding [4].The present investigation is aimed at the study of FSW fundamental parameters

as advancing and rotating tool speed, approaching their influence on the processed surface roughness of Al6061.Single and overlapped friction stir welded joints were considered for the roughness analysis.

2. EXPERIMENTAL PROCEDURE

The research was conducted on as-cast AA6061 aluminium alloy plates, simulating the flat-shaped ingots, of 5mm thickness. Their chemical composition is presented in Table1. The samples 200x300mm were longitudinally friction stir welded. As Fig.2 presents, FSW single and overlapped layers were made. A specially designed cylindrical tool (Fig.1) is rotated and plunged into the butt area of the plates. The tool has a small diameter threaded pin with a concentric with a larger diameter shoulder. FSW tool characteristics are presented in Table2. When the rotating pin contacts the surface it rapidly friction heats and softens a small column of metal. The tool shoulder and the length of the entry probe control the penetration depth. When the shoulder contacts the metal surface, its rotation creates additional frictional heat and plasticizes a larger cylindrical metal column around the inserted pin.

Fig.1 FSW of as-cast AA6061 aluminium alloy: a) tool design b) rotating tool prior to contact with the plate

Table 1 FSW tool characteristics

The shoulder provides a forging force that contains the upward metal flow caused by the tool pin. During FSW, the processed area and the tool are moved relative to each other such that the tool traverses, with overlapping passes, until the entire selected area is processed. The rotating tool provides a continuing hot working action, plasticizing the metal within a narrow zone, while transporting material from the leading face of the pin to its trailing edge. The process is carried out at low temperature, in solid state, typically below 0.3 of the fusion temperature (Tf) for aluminium alloys. The processed zone is deformed and under the thermal effect recovers or recrystallizes, forming a defect-free recrystallized, fine grain microstructure.

Table. 2 Processing Parameters

The process main operating parameters are friction pressure (tool up-setting force), tool rotation and travel speed are presented in Table2. The modified layer thickness is directly related to the friction pressure, whereas it is inversely affected by the rotation speed of friction tool and the traverse speed. The microstructure and properties of the modified layer depend on the amount of frictional heat generated by processing.

Table.3 Chemical composition of cast AA6061alloy

Samples of different layers cross-sections were prepared by using standard metallographic procedures. A modified Keller’s reagent was used for their etching. Observations of plastic deformation, material flow and microstructures were performed by using optical microscopy. Vickers micro hardness measurements were performed on processed layers cross-sections by using a micro hardness tester at a 100g load and a 15sec dwell time. Moreover, the roughness analysis was performed on processed surface.

Fig.2 Single and overlapped layers

3. RESULTS AND DISCUSSION3.1 Material Flow and MicrostructureFrom the macrographs shown in Fig.3, both the thermomechanically affected zone (TMAZ), which consists of the weld nugget and plastically deformed grains, and part of the heat affected zone (HAZ) are evident. The concentric rings seen within the nugget are also visible, as is the appendage to the nugget that stretches to the surface.The base metal (BM) microstructure of as-cast AA6061 (fig.4) consist of Al solid solution dendrites along with coarse silicon and intermetallic phases. Shrinkage porosity is also prevalent. According to Fig.4, FSW closed the shrinkage porosity and homogenized the as-cast microstructures by breaking up and evenly dispersing initial phases.

Fig.3 Macrostructure of FSW single and overlapped

layers

The base metal (BM) microstructure of as-cast AA6061 (fig.4) consist of Al solid solution dendrites along with coarse silicon and intermetallic phases. Shrinkage porosity is also prevalent. According to Fig.4, FSW closed the shrinkage porosity and homogenized the as-cast microstructures by breaking up and evenly dispersing initial phases.

Fig.4 shows the typical features of all different zones in a single processed layer cross-section of as-cast AA6061 under processing conditions of 1120 rpm for the rotational speed and 320mm/min for the welding speed. The positions 1-6 from Fig.4 are located in different microstructural zones. The micrographs show that the microstructure of the processed layer is complex and highly dependent on the position within the processed zone. This result arises because of the large local variations in the

plastic flow and from the thermal history resulted from the material interaction with the tool. The characteristic annular-banded structure is distinctly observed to be asymmetric and more obvious on the advancing side (A) of processed zone as shown in Fig.4, positions 1,2,4. a severe deformation has also occurred along the top surface of the processed layer where the shoulder of the tool is in contact with the material. The flowlines from Fig.4, positions 4,2,5 seams to represent plastic deformation increments that develop as the rotating tool moves through the processing line. Although it is well known that the material is transported from the advancing side to the retreating side (R), colligan [5] showed that with a threaded pin tool the material from the upper part of the processed zone is pressed down, whereas the material from the lower part processed zone is moved toward the top surface. The material may travel many cycles around the tool before being redeposited. A little flow of material was observed near the bottom of the processed zone.

Fig.4 Typical features of all different zones in a frictionstir welded single layer cross-section of as-castAA6061: BM – base metal pancake microstructure;1- flow patterns in the appendage zone; 2 – nuggetzone; 3 – the retreating side of TMAZ; 4 – theadvancing side of TMAZ; 5 – nugget bottom side;6– processed layer bottom sideThe microstructure in the stir zone is characterized by refined grains in a discrete series of bands and some precipitate mainly distributed at the grain boundaries (Fig.4). There is also still some debate concerning the origin of the annular rings observed within the nugget zone attributed to an abrupt variation in the grain size and precipitate density [6]. The nugget zone grains suggest effective strains together with a microstructural evolution that occurs by a combination of hot working and a dynamic recovery or recrystallization. The temperature reached in the nugget zone is known to be in the range of 450-500 0

C for the 6061-Al alloy [7]. Distinct precipitates and coarsened grains are observed at the deformed regions of TMAZ. HAZ grains are severely coarsened by FSW (Fig.4, positions 1, 4).

3.2 Hardness Distribution in the Processed Material

Fig.5 shows the typical microhardness distribution of the single layer processed cross-section of as-cast AA6061 alloy.

The base metal minimum hardness value was of 68 HV0.1.

Generally, there is an increase of hardness in the processed layer. A significant variation in hardness if found at 2.5 mm depth from the processed surface. A maximum value of 85 HV0.1 was registered at this depth, which corresponds to the interface nugget-thermomechanically affected zone. Although on the top surface of the processed layer maximum temperatures occurs, the hardness increase up to 78 HV0.1.

Fig.5 Typical microhardness distribution in a friction stir welded single layer cross-section of as-cast

AA6061

The hardness increase of as-cast AA6061 alloy by friction stir welding depends on the process parameters, tool rotating and advancing speed, influencing the degree of grains refinement and level of uniformity in distribution of precipitation hardening phase, Mg2Si.

3.3 Processed Material Roughness

Fig.6 shows typical roughness distribution of the single layer processed cross-section of as-cast AA6061 alloy. The analysis was made on A, B and C lines of 4mm length (Fig.6).The roughness profiles revealed the existence of transitory and quasy stationary phases of FSW. The transitory phases correspond to the starting and the ending moments of the process, due to the material increased heating. Maximum profile heights of 3.68 µm were registered on the surface of the process ending phase. In the quasy stationary phase, namely the process stable phase, maximum profile height of 2.2 µm was found out. A process monitoring method obtaining through roughness control can be developed.

Fig.6 Roughness distribution on a friction stir welded single layer of as-cast AA 6061

4. CONCLUSIONS

Several conclusions can be underlined about FSW of as-cast Aa6061:ª% The process makes an effective surface modification;

FSW closed the shrinkage porosity and homogenized the as-cast microstructures by breaking up and evenlydispersing initial phases.

ª% The material plastic deformation, flow and mechanical mixing exhibit distinctly asymmetric characteristics at advancing, retreating side zones.ª% The hardness increasing of as-cast AA6061 alloy by friction stir welding depends on the process parameters, tool rotating and advancing speed, influencing the degree of grains refinement and level of uniformity in distribution pf precipitation hardening phase, Mg2Si.ª% A global increased hardness was found out in the processed layer. A significant variation in hardness is found at 2.5 mm depth from the processed surface. A maximum value of 85 HV0.1 was registered at this depth. ª% The roughness profiles revealed the existence of transitory and quasy stationary phases of FSW. The transitory phases corresponding to the starting and the ending moments of the process, due to the material increased heating.ª% The process may address to the industry as a surface repairing

technique for different types of flaws and surface hardening.

5. REFERENCES

1. Thomas WM. Friction stir welding., 1991, International Patent application No. PCT/Gb92/02203 and GB Patent Application No.9125978.8 U.S. Patent No.5, 460,317.

2. Oosterkamp A, Djapic Oosterkamp L, Nordeide A. , 2004,”Kissing bond phenomena in solid state welds of aluminium alloys”, Weld J, 225s-231s.

3. Zeng WM, Wu HL, Zhang J.,2006, “ Effect of tool wear on microstructure, mechanical properties and acoustic emission of friction stir welded 6061 Al alloys”, Acta Metall Sinica, 19(1):9-19.

4. Olga Valerio Flores, 1998,”Microstructural issues in a friction stir welded aluminium alloy”, Scripts Mater, 38(5):703-8.

5. Shinoda. T., Kawai, M., takegami, H., 2004, “Novel process of surface modification of aluminium casts applied friction stir phenomenon”, IIW,Doc.III-1295.

6. Mishra R.S., Ma Z.Y., 2005, “Friction stir welding and Processing”, Materials Science and Engineering, Elsevier,1-78.

7. Vilaca, P., Santos, J.P., Gois, A., Quintino, L., 2005 “ Joining Aluminium Alloys Dissimilar in thickness by Friction Stir Welding anf fusion processes”, Welding in the World, Vol.49, No.3/4, 56-62.

6. NOMENCLATURE

AN EFFICIENT MUSICAL INSTRUMENT IDENTIFICATION ON SOLO PERFORMANCE MUSIC USING MFCC

CHELLASWAMY . C 1

Department of Electronics and Communication Engineering Srinivasa Institute of Engineering and Technology Chennai, Tamilnadu, India. email : [email protected] email

SUGANTHINI REKHA.R2

Department of Electronics and CommunicationEngineering, S.K.R Engineering College, Chennai,Tamilnadu,India. :[email protected]

ABSTRACT:

In this paper, a musical instrument analysis and identification method is proposed and it is useful for anyone who likes a particular type of music. The instrument identification is done by using Mel-Frequency Cepstral Coefficient (MFCC) algorithm. A familiar singer’s voice is easily recognizable for humans, even when hearing a song for the first time. But anyone who wants to listen the music of a particular musical instrument, the songs can be searched one by one from the database. New research questions are now emerging related to sophisticated human/computer interfaces, techniques and strategies for the control of information about music and creative musical processes. Existing systems for classification of automatic instrument which operates normally on a relatively large number of features, from which those related to the spectrum of the audio signal are particularly relevant. It is found that the cepstral features describing the audio signal spectral shape and high recognition accuracy can be achieved .Keywords: MFCC; musical instruments; spectrum; features.

I.INTRODUCTIONMultimedia information technologies provide comprehensive and intuitive information for a broad range of applications, which creates a great change in our modern life. Nowadays usage of digital multimedia (including audio, music, video and images) over the Internet and wireless communication are developed. Even more literatures are there to identify the musical parameters. The most applicable one is MFCC based musical instruments identification system. With digital music becoming more and more popular such as music CDs, MP3 music that can be downloadable from the internet, music databases both professional and personal are growing rapidly. It is mainly demand for the categorization and retrieval of music collections which provide the powerful functions such as browsing and searching musical content to the consumers. With this capability provided in a music system, the user can easily get to know about the particular music instrument songs contained in a distributed music databaseII. RELATED WORKThe musical information is essential in identifying browsing and retrieving musical collections. The major challenge for this study arise from the fact that musical signal tends to be changed from time to time arbitrarily and is inextricably intertwined with the signal of background accompaniment.

To determine the particular type of instrument in a musicals recording, methods are presented here there methods are used for indentifying musical instruments, separating vocal from non-vocal regions and differentiating from one instrument to other. [i] Further, this technology is used to grouping the songs of similar music instrument in terms of vocal segments. It also provides the information of a music instrument and its characteristics is manually shored into music files by professional records. It also provides the text information such as the type of musical instrument present in the music file. There information are often lack in music pieces downloaded from music clips grasped from digital music channels[2]. Using Gaussian Mixture model [GMM] the database can be trained efficiently with less time to converge[3]. The direct method to detect vocals is to note the energy within the frequency range of 200Hz to 2KHz, in which majority of the produces repetitions of the energy in the singing voice falls. This method produces repetitions of the samples. In order to avoid that, bootstrapping is done. It improves the performance of segmented samples. The noise performance can be evaluated using four different noise types by comparing MFCC and autocorrelation MFCC[4]. When noise is introduced, the MFCC’s accuracy gets degrades faster than AMFCC for all types of noise for this type improves the performance of segmented samples.

The noise performance can be evaluated using four different noise types by comparing and autocorrelation MFCC[4]. When noise is introduced, the MFCC’s accuracy gets degrades faster than AMFCC (autocorrelation MFCC) for all types of noise. For this type that have autocorrelation sequence, the AMFCC accuracy gets improved. A new MFCC algorithm is introduced and it will require half of the multiplication steps. The overlap function will affect the accuracy of recognition slightly which is performed after filter bank. The computation is reduced by half time than the original algorithm[5]. However, FFT based MFCC’s uses excitation to encode more information but LPC based MFCCs remove the excitation.. MFCCS are also used by Eronen and klapuri[6]. MFCC is utilized in audio and speech applications. In [7], Eronen analyzed and compared the effectiveness of several type of features to indentify different musical instruments. The best results are obtained with two sets of MFCCs. It is impossible to embed audio features such as timbre and spectral features of al songs. Therefore, it is impossible to group the songs which uses the similar musical instruments[8]. In general, the technique for musical instrument identification, vocal separation will generate important functions to a digital music management system.

We propose a scheme that finds the dominant frequencies which is present in a music file. We have found there is a strong influence between the energy and the temporal envelope. Because each music instrument will have different temporal envelope and the Cepstral coefficients while playing same note.

components. So that the high frequency components in pre-emphasis is artificially boosting up. After that the audio signal gets segmented to a length of 10 to 20 ms for analysis in the working phase ( L=160 samples at sampling rate of 8KHz ). The block diagram of proposed system is shown in fig 1.Each dividing segment is called as frame and each frame is multiplied by the window function. This multiplication has the effect of convolving the signal spectrum with the window spectrum. By using the same window function, we can further enhance the ability of FFT to extract spectral data from signals. It also reduces the leakage effect that occurs during an FFT of the data. For improving the sensitivity of the spectral measurement we have to perform window function before taking FFT routine. The window spectrum would have a narrow main lobe and large attenuation in the side lobes. The Hamming window has the lowest first side lobe level and after that it is decaying slowly. In the proposed system, the Hamming window with an overlap of 50% between adjacent frame, presented in equation (1) is used.

otherwise , 0

(1) 1-Nn0 , 1-Nn2cos 0.46 - 54.0

≤≤

=π

wn

Where N is the number of points in the frame. In the consecutive step used a 256 point FFT for obtaining a good frequency resolution. Because of the symmetry property of FFT, first 128 coefficients we can calculate. The filter bank consists

III. PROPOSED METHOD FOR MFCCPreprocessing is the act of processing the data before it is analyzed. The function of preprocessing is used to improve the quality of the input signal before it is processed for its application. The first step in preprocessing is pre-emphasis. It is the important step in the analysis of audio signals. The amount of

energy carried by high frequency components is small compared with low frequency components. So

of 33 triangular shaped band-pass filter, which are centered on equally spaced frequencies in the mel domain up to 4KHz. The Mel-scale equivalent value for frequency (f) expressed in Hz is used in equation (2) .

The term mel refers to a kind of estimation related to the perceived frequency. The mapping between the real frequency scale (Hz) and the perceived frequency scale (mel) is approximately linear below 1KHz and logarithmic for higher frequencies. The cepstrum of a time domain signal is the Inverse Fourier Transform of the log magnitude spectrum of the signal. So the cepstrum is normally computed through Discrete Cosine Transform. The MFCC parameters are obtained by calculating the discrete cosine transform over the logarithm of the energy in several frequency bands as used in equation (3) and (3a).

(3) )5.0(cosM

1k

−∑

==

MkmPcm

π

where ; L is the order and P is given by

( ) ( ) ( ) 3 a S k 1 , lo g 1

2

0

≤≤

= ∑−

=

jXjWP

k

jk

is the weight function associated with the k th mel band in mel scale. S is nband number in mel scale

IV. EXPERIMENT AND RESULTSThe above proposed method in section-3 can be applied to obtain the good results practically. The proposal can be applied with good examples and it is used to analyse the identification of musical instruments with these methods. The performance of musical instrument identification with MFCC for different carnatic musical instrument is analyzed. All data evaluation and implementation are done in MATLAB 7.1. Different experiments were conducted using the algorithm described in section-3. Here five different carnatic musical instruments namely Tabla, Flute, Guitar, Violin and Piano are considered and five different solo songs of each instrument and ten other songs (other than the list ) were studied. All the 35 song are added in the list box created using MATLAB. If the list is selected the corresponding song will be analyzed. The snapshot of the front end window is as shown in fig.4.

Fig.2. Response of Hamming Window

The input signal was down-sampled to a 32 kHz sampling rate, it was centered with respect to its temporal mean and its amplitude was normalized with respect to its maximum value. Figure 2 shows the output of Hamming windowing function. The analysis was performed over sliding overlapping windows. The frame length was 32 ms and the hop size is 16 ms. All spectra were computed with a FFT after a Hamming window had been applied. The need of Hamming Window is explained in section 3.

Fig.3. Mel-Frequency Filter Bank

It is seen from fig.3 that the Mel-Frequency filter bank output with the use of mel scale given in equation 2. In the figure variation in amplitude shows that the partial frequencies do not fall exactly on spectral bins. As far as cepstral features are concerned, the Mel coefficients were selected from the specified duration for each test signal (30 overlapping frames of 32-ms duration).

Fig.4.(a) Snapshot of front window -I

Fig 4-a and 4-b shows the main window of the Musical Instrument Identification system created using MATLAB tools. In figure 4-a the window listbox contains totally 35 songs of different instruments ( ie: five songs of five instrument what we choose and ten other songs ) If we select any song of the corresponding instrument from the list and click the search button, then the particular song is automatically analyzed and the corresponding number of songs is increased in the respective textbox provided. If we click the button author, the information about the author will be displayed as show in fig.4-b.

Fig.4.(b) Snapshot of front window –II

Fig.5. MFCC output for Tabla

Fig.6. MFCC output for Violin

Fig.7. MFCC output for Flute

Fig.8. MFCC output for Guitar

Fig.9. MFCC output for Piano

Figure 5 - 9 shows the MFCC output of Tabla,, Violin, Flute, Guitar and piano respectively. For example if we select Tabla_1 in the listbox then it will analyze by the algorithm and the coefficients of mel frequency is compared and the corresponding Tabla field in the status block present in the main window will increase. If the song is in the five desired songs the particular field automatically incremented by one and it is not in the above five then the alert display will display and the other field automatically incremented by one.

V. CONCLUSION

In this work, we have proposed a study on the efficiency of simple features for instrument identification systems in the context of solo musical performance of five carnatic musical instruments. The chosen approach gives higher recognition success and analyze the system performance in a more straightforward manner. It has been shown that the combination of cepstral coefficients with features describing the audio signal spectral shape results in high accuracy recognition for instruments belonging to the different families even over short-term decision lengths. It is planned to proceed this method of identification for multiple instrument used in a song. Moreover, an artificial neural network is used for more control over the system performance.

VI. RFERENCES

[1] L.Leue, and O.Izmirli,” Time-variable tempo tracking and beat Annotation for MIR,” ACM Journal, Vol.21, No.6 (june 2006) pp.306-307.

[2] K. S.-P. Heo, M. Suzuki, and A. Ito, S. Makino, “An Effective Music Information Retrival method Using Three-Dimensional Continuous DP,” IEEE Trans. Mltimedia, Vol.8, No.3 (2006), pp.663-669.

[3] H. Kameoka, T. Nishimoto, and S. Sagayama, “Separation of harmonics structures based on tied gaussian mixture model and information creterian for concurrent sounds,” International Conference on Acoustics , speech and signal processing (ICASSP04), 2004.

[4] Benjamin, J.Shannon and Kuldip K.Paliwal,”MFCC Computation from Magnitude Spectrum of Higher Lag Autocorrelation for Robust Speech Recognition,” Griffith University, Brisbane, QLD 4111, Australia.

[5] Wei HAN, Cheong-Fat CHAN, Chiu-Sing CHOY and Kong – Pang PUN, “An Efficient

MFCC Extraction Method in Speech Recognition,” IEEE Internatinal Conference on Acoustics, Speech and Signal Processing (ICASSP04), 2006.

[6] A. Eronen, and A. Klapuri, “Musical instrument recognition using cepstral coefficients and temporal features,” in proc. ICASSP 2000. Istanbul, June 5-9, 2000.

[7] A. Eronen, “Comparison of features for musical instrument recognition,” in proc. Of

WASPAA’01, New Platz, NY, USA, pp. 19-22, October 21 – 24.

[8] N. Kosugi, Y. Nishihara, T. Sakata, M. Yamamuro, and K.Kushima,” A practical query- by – humming system for a large music database,” ACM Multimedia 2000, pp. 333-342.

NOVEL APPROACH TO MOBILE DATABASE DISSEMINATIONFORBANDWIDTH AND ENERGY CONSERVATION

R.BOOPATHI RANI1, A.ADAIKALAM2

ECE Dept, Sudharsan Engineering College, Pudukkottai, India.1 - [email protected], 2 – [email protected]

ABSTRACT:-In the past few years there has been a rapid growth on wireless data applications in the commercial market. It shows that in the near future, a large number of mobile users carrying portable devices will be able to access a variety of information from anywhere and at any time. In this scenario, there are two fundamental information delivery methods for wireless data applications – point-to-point access (pull based way) and broadcast (push based way). This paper proposes two novel approaches for mobile database dissemination. One is broadcast i.e., data are sent simultaneously to all users residing in the broadcast area without request from the client. This will conserve bandwidth but to retrieve a data item, a mobile client has to continuously monitor the broadcast until the data item of interest arrives. This will consume a lot of battery power since the client has to remain active during its waiting time. Second approach provides a solution to solve this problem i.e., using indexing. The basic idea is that by including auxiliary information about the arrival time of data items on the broadcast channel, mobile clients are able to predict the arrivals of their desired data. Thus, they can stay in the doze mode and tune into the broadcast channel only when the data items of interest to them arrive. Several indexing techniques for wireless data broadcast have been discussed to conserve battery power. The work is being carried out in Network Simulator ns-2.

Keywords:-Broadcast Networks, Indexing, Mobile Database.I. INTRODUCTIONThe widespread adoption of wireless data services is disadvantaged by the relatively limited and costly wireless bandwidth. In regard to power capacity, it has been investigated that the life expectancy of a battery is anticipated to increase only 20% for every 10 years [3]. Consequently, the need to use power efficiently and effectively is crucial issues. To serve more nodes with less bandwidth, broadcasting techniques were proposed. These techniques prepare a data broadcast by appending individual information items together, and send it to a common channel for the mobile nodes to download. Each mobile node filters the downloaded broadcast for the information that they are interested in. This solution is efficient since it eliminates the need to send the same item multiple times when each item is used by many clients. It is often desirable for frequently requested data to be broadcast in a continuous cycle by mobile support stations, rather than transmitted to mobile hosts on demand. A typical application of broadcast data is stock-market price information. There are two reasons for using broadcast data [7]: 1.The mobile host does not have to invest on the energy cost for transmitting data requests Thus, the mobile client need to only receive data as and when those data are transmitted, rather than consuming energy by

transmitting a request.

The mobile client may also have the local nonvolatile storage for storing (cache) the broadcast data as and when received, for possible later use. This method conserves bandwidth and reduces a little energy consumption as well. To achieve a large amount of energy consumption, indexing techniques were proposed. By sending data preceded by an index, the client knows when the data of interest will be broadcasted. By knowing the time of broadcast, the client will be in doze mode most of the time and go to active state, after the data of interest is being broadcasted. Subsequent sections in this paper are organized as follows. Section II deals with network simulator ns-2. Section III describes the background of the approach, section IV deals about broadcasting, section V contains indexing techniques. It is then followed by analysis of indexing techniques at section VI and at section VII the simulation, Section VIII deals about results and findings. Finally, section IX concludes the paper.II .NETWORK SIMULATOR NS-2A network simulator is a piece of software or hardware that predicts the behavior of a network, without an actual network being present.

A. Examples of network simulators

1. ns2/ns3

2. OPNET3. GloMoSim / QualNet4. NetSim5. OMNeT++

Among the abovementioned simulators, the network simulator ns-2 [5] is a popular and widely used discrete-event based simulator for wireless and wired network simulation. NS-2 allows researchers to simulate a wide range of network configurations. It supports various protocols at the application layer, and mainly TCP and UDP at the transport layer. It also includes models for simulation of wired and wireless physical layers. Regarding its wireless physical layer implementation, ns-2 mainly offers an implementation for the IEEE 802.11 protocol [6].

III. BACKGROUND

With the increase number of mobile users in a cell, the required number of data items to be broadcast also increases accordingly. This situation may cause some mobile clients to wait for a substantial amount of time before receiving desired data item[1]. Consequently, the advantages of broadcast strategy will be eliminated. On the other hand, when the channel contains too few data items, a high percentage of mobile clients waste their time by listening to the channel until they find out that the desired data is not broadcast. This paper discusses data indexing in order to reduce the tuning time of the mobile client. Tuning time is the time clients listen to the

channel, which indicates their power consumption. The use of data indexing enables mobile clients to determine when the desired data arrives in the broadcast cycle. Thus, it allows them to switch into “doze mode” while waiting to the data and to safe power consumption [2]. Power consumption in doze mode is very much less than power consumption in idle mode.

IV. BROADCASTINGIndexing detail will be broadcasted first; it is then followed by data items. So, that each node receives the index information and calculates the location of data, i.e., the intended data location. From this approach, node will be in doze mode until the data of interest arrives. After receiving data, the node again go to doze ode.

V. INDEXING TECHNIQUES

The following diagram shows an indexing sheme. In this, all the clients receives index, calculates the needed data location, then receives data when needed data available on air.

Figure 2 Mobile database broadcast index technology

Signature and index-tree methods represent two classes of index methods for broadcast channels. The performance evaluation of these methods and broadcast without indexing has been made. Signature methods have been widely used for information retrieval. A signature Si of a data item is basically a bit vector generated by first hashing each data value into a bit string and then superimposing the bit strings together [4]. By examining the signature only, the client can estimate whether the data item contains the desired information. In index-tree methods, the distributed index is an index replication method in which only the relevant portion of the index tree is interleaved with the data in every fraction of the

broadcast cycle. It replicates some upper levels of the index tree to provide a global view of the divided fractions of the broadcast cycle and use lower level sub-trees for navigation within the fractions.Next, few indexing techniques such as bitmap index, hash index and optimum index have been compared.VI. THE ANALYSIS OF INDEX TECHNOLOGYA bitmap index is a special kind of index that uses bitmaps. It is a specialized variation of a B-tree index. It stores data in bit array and providing query service by calculating bitwise logical operation on the bitmaps. (e.g., AND, OR, XOR). Every key word in the bitmap is connected with a line, and is represented as 0 or 1. 0 represents that the key word does not exist in the data item, while 1 represents it exists in the data item. It requires less space. Hashing refers to an information storage and retrieval technique. In the ideal situation, mobile user’s machine can once acquire the needed hotpot data without any comparison. Thus, between the broadcast location of the data item and its key word, build a corresponding relation f(k), which is called Hash function. Hash function is in fact a mapping between the key word and its location, and different key word only gets one location. Optimum index combines bitmap index and Hash index in a same broadcast sequence, sufficiently utilizing the advantages of the two indexes, maximally reducing the tuning time of user’s machine.The following figure shows, node configuration in ns-2 and illustrates mobility of nodes.

Figure 3: Sample simulation Topology at different timings

Scenario 1: It consists of 8 mobile nodes, 4 source nodes and 4 destination nodes.

Figure 4: Throughput Vs Time for Point to Point Communication

In figure 4, the throughput of the system is decreasing as the number of communication increases.Scenario 2: It consists of 10 mobile nodes, 1 source nodes and 9 destination nodes.

Figure 5: Throughput Vs Time for Broadcast reception

In figure 5, the throughput of the system not depends on number of receivers in broadcast.

Figure 6: Energy Consumption Vs Time Broadcast Reception

VIII. RESULTS AND FINDINGS

Analysis: As figure 4 shows, Node 1 starts transmitting at time T =1.4 sec while Node 2 starts

transmitting at time T=10 sec. During the period of time [1.4 sec, 10 sec] Node 1 is the only transmitting node using the entire available bandwidth. This justifies the high performance of Node 1 during the specified interval of time. At time T=10 sec, Node 2 starts transmission hence sharing channel resources with Node 1. This explains the heavy reduction of bit rate. It concludes that if mobile database is broadcast upon request means, the throughput of the system will tend to decrease even if same data was requested by more clients and requires large bandwidth. The figure 6 and 5 shows that energy consumption of each receiver increases as throughput of each receiver increases. This is because node is receiving all data irrespective of its needed data alone. So, this disadvantage will be eliminated by using indexing. The figure 7 shows the comparison of tuning time between non-index, distributed index and signature method.

Figure 7: Tuning Time for different methodsFrom figure 7, the tuning time of signature method and distributed index is very much lower than non-index. The graph is drawn by taking tuning time of non-index as 100% and others are calculated from this. So, indexing, signature method will significantly reduce energy consumption.

The figure 8 shows tuning time of different indexing technologies such as single, bitmap and optimum index.

Figure 8 Average Tuning time for different indexing schemes

From figure 8, the average tuning time of optimum index is better than those of single index and (1:M)type of tree index the number of the data items data>5, but if the number of the data items is relatively small, it is unnecessary and of no importance to discuss the index technology.

IX. CONCLUSION

Data broadcast is a powerful tool for data dissemination in mobile and wireless environments. Broadcasting frequently required data item eliminates energy spent for transmitting request from client. It also provides efficient utilization of available bandwidth. The indexing schemes allow the client to listen to the channel after the data of interest broadcasted from the server. This will significantly reduce the tuning time. On comparing signature method, distributed index requires less tuning time. Among bitmap, hash and optimum index, optimum index requires extremely small tuning time. The better indexing technology enables the user’s machine to received needed data in shorter tuning time, saving the portable limited electric power.

REFERENCES

[1] Shou-Chih Lo and Arbee L.P. Chen. “Efficient index and data allocation for wireless broadcast services”[J]. Data & Knowledge Engineering, Volume 60, Issue 1, 2007.01, pp:235-255.

[2] Karthikeyan, Dr.V.Palanisamy, and Dr.K.Duraiswamy. “Performance Comparison of Broadcasting methods in Mobile Ad Hoc Network”[J]. International Journal of Future Generation Communication and Networking Vol. 2, No. 2, June, 2009.

[3] SHENG S., Chandrasekaran A. and Broderson R. W., “A Portable Multimedia Terminal for Personal Communication”[J]. IEEE Communications, pp.64-75, December, 1992.

[4] Lee W.C., and Lee D.L., “Using signature techniques for information filtering in wireless and mobile environments. Special Issue on Database and Mobile Computing”[J]. Journal on Distributed and Parallel Databases, 4(3):205–227, July 1996.

[5] The network simulator - ns-, http://www.isi.edu/nsnam/ns/

[6] The CMU Monarch Project’s Wireless and Mobility Extensions to NS.

[7] http://www.ignou.ac.in/virtualcampus/adit/course/cst302/block2/cst302-bl2-u1.htm

http://www.isi.edu/nsnam/ns/

PERFORMANCE ANALYSIS OF GUARANTEED PACKET TRANSFER IN COVERAGE AREA AND LESS COVERAGE AREA FOR MANET

¹ MRS. S.GOMATHI , 2 DR K.DURAISWAMY ¹ Professor / Department of MCA/ K.S.Rangasamy College of Technology / Tiruchengode -637215

[email protected] 2 Dean (Academic) /K.S.Rangasamy College of Technology / Tiruchengode -637215

ABSTRACT :The main challenge in mobile adhoc network is guaranteed transfer of packet within coverage area and less coverage area. The hierarchical clustering is used to increase coverage area to reduce the packet loss in MANET.The coverage area, packet loss is analyzed and reliability of coverage area and less coverage area is increased by hierarchical clustering.The traffic can be reduced and packet loss will also be minimized and it achieves guaranteed transfer of packets. Every node is updated with the neighbored information in a timely manner to achieve guaranteed coverage and to increase the packet delivery ratio and reduce end to end delay. key words:MANET ,QOS, Neighboring Nodes, Clustering, Graph, Adhoc, Guarantee, Network .

I. INTRODUCTION

MANET is formed by a set of mobile wireless devices with no fixed topology. The nodes can move freely, leave and enter the network at anytime. Mobile ad hoc networks (MANET) consist of hundreds to thousands of small, low-cost, low-powered sensor nodes, where the nodes are densely deployed across certain geographical areas. Since these sensor nodes have a severe limitation in their storage, radio communication, capabilities, bandwidth, and energy, the most important design consideration for MANET is to extend their operational lifetime by minimizing power consumption. This is the reason why protocols with low energy consumption have been an important research orientation in this field.

1.2 Manet Routing Protocols Dynamic Source Routing (DSR) ,Adhoc Ondemand DistanceVector(AODV), Zone Routing protocol(ZRP), Location Aided Routing(LAR), ProactiveRoutingPtocols(PRA) Temporally Ordered Routing (TOR).

2.QOS FACTORS 2.1 Throughput

Is the total number of packets received by the destination.

2.2 End to End Delay

is the average end to end delay of data packets from senders to receivers.

2.3 Media Access Delay

is the media transfer delay for multimedia and real time traffics’ data packets from senders to receivers.

2.4 Packet delivery ratio (PDR)is the ratio of the number of data packets received by the destination node to the number of data packets sent by the source node.2.5 Routing load specifies the load over communications links for traffic flow.

3. BROADCASTING METHODS

3.1 Probability Based Method - - Source node broadcasts the packet to all its neighbors. Some neighbors nodes rebroadcast the packet. The selection is based on predetermined probability. When the probability is 100%, this scheme is same as simple flooding. This method is well suited for dense networks. For sparse networks, all the nodes might not receive the packets. Based on the understanding that in a dense network, nodal and network resources can be save by having some nodes not rebroadcast the duplicate networks.

A more refined probabilistic scheme is a counter-based approach in which upon receiving a broadcasted packet, the current node applies a Random Delay Time (RDT) before it determines whether or not to rebroadcast packet[11].3.2 Area Based Method In the area based method, the packet is re-broadcasted only if the node covers a significant amount of area than the sender who sent the packet. Suppose a node receives a packet from a sender that is located very close to it. If the receiving node rebroadcasts, the additional area covered by the retransmission is low. If the receiving node is located at the boundary of the sender node’s transmission distance, then a rebroadcast would reach significant additional area. intermediate nodes will evaluate additional coverage area based on all received duplicate packet. We can image that in a dense network there may be multiple nodes which are located very close to each other. In such situations, the majority of the coverage areas of these nodes overlap each other. Based on estimated distance or location information, an intermediate node will determine whether or not to rebroadcast the received packet[11].

3.3 Distance-Based Method A node using the Distance-Based Scheme compares the distance between itself and each neighbor node that has previously rebroadcast a given packet. Upon reception of a previously unseen packet, a RAD (Random Assessment Delay) is initiated and redundant packets are cached. When the RAD expires, source node location is examined to see if any node is closer than a threshold distance value. If true, the node doesn’t rebroadcast. [12]

3.4 Broadcasting methods Broadcasting methods have been categorized into four families utilizing the IEEE 802.11 MAC specifications. Note that for the comparisons of these categories the reader is referred to[12]

3.5.Simple flooding Can be used as a simple protocol for broadcasting and multicasting in ad hoc networks with low node densities and/or high mobility .[12] 3.6 Neighborhood knowledge based methods A node will determine whether or not to rebroadcast based on its neighbor list. Upon receiving a broadcasted packet, a node will check the previous node’s neighbor list which is included in the packet header. If it turns out that it would not reach any additional nodes, it will decide not to rebroadcast the packet.[5]

4 .CLUSTERING Connecting two or more interconnected computers together in such a way that they behave like a single computer. Clustering is used for, and higher availability, higher scalability or both[4].

5.PROPOSED METHODThe basic principle behind hierarchical clustering is the following: If there are m input points (or data items), we start with n clusters where each cluster has a single point. From there on, the “closest” two clusters are identified. TheDistance between two clusters can be defined in many ways.

The two closest clusters are merged, resulting in a reduction in the number of clusters (by one). This process of merging is continued until the number of remaining clusters is q (where q is the target number of clusters and could be a part of the input). There are many ways in which the distance between clusters can be defined [12]. The metric that is commonly employed is the single link metric. Here, the distance between two clusters is defined to be the minimum weight of any edge connecting a point in one cluster with a point in the other. In this paper, a single link metric is employed as well.7.MOTIVATION AND RELATED WORKThere are several standard techniques proposed to identify clusters in a graph .These can be broadly classified into Identifier –Based and Connectivity –Based Cluster- head selection techniques. The popular Techniques are the lowest ID and the Maximum –Connectivity.In the Former case, if a Low-ID node ever gets highly mobile ,this could result in unacceptably large number of cluster head changes. In the Maximum connectivity as well as lowest distance value cluster, the clustering changes from one mobile node to another depending on the activity coverage area. The cluster based on mobile nodes observed over a long period of time.In the Proposed algorithm, a path is defined as the most reliable within the graph, through which delivery is guaranteed .In the ideal first identified the coverage area range .The transferring packet with in the coverage area and calculate the time taken for transferring a packet. Next compute less coverage area and transfer the packet .First step to identify the packet transferred time and packet loss within coverage area and less coverage area. The packet loss reduce the less coverage area.

Figure -1

Coverage area and less Coverage area

XII.SIMULATION RESULTS

Network Model Sample Data

Routing AODV,DSRProtocol

Coverage area 670 m x 670 mAlgorithm Random casting

Packet Size 512No. of Nodes 10,20,30

Figure3. Node Creation

Figure 4: Node cover for coverage area

Figure-5 Outside the coverage area packet delivery ratio and time taken to transfer the packets.

Figure 6 :Node cover for outside the coverage area

Figure 7: Node cover for coverage area and less coverage area

Figure 8 : Node cover for coverage area and less Coverage area packet loss

Figure 9 : Node spilt to coverage area and less coverage area

Figure 10 : Reduce the packet loss in less coverage area.

Figure11:Energy consumption for coverage area and less coverage area

Figure 12 : Packet delivery ratio for coverage area and less coverage area

ACKNOWLEDGMENTThe researcher would like to thank Lion. Dr.K.S.Rangasamy MJF ,Chairman of KSR Educational Institutions , Dr .K.Thiyagarajah Principal K.S.R College of Technology and Dr K.Duraiswamy Dean (Academic ) K.S.R College of Technology for their motivation and continuous encouragement .

[1] Banerjee and S.Khuller ,” A Clustering Scheme for hierarchical routing in wireless networks” , Technical Report CS-TR-4103,UMD, College patrk2001.

[2] C.Perkins ,E Beld ig-Royer and S.Das “ Ad hoc on Demand Distance Vector (AODV) Routing “ Request for Comments 3561 July 2003.

[3] Y.Ko and N.Vaidya “ Location Aided Routing (LAR) in mobile Ad hoc Networks . Proceeding of the ACM /EEE International conference on Mobile Computing and Networking(MOBIKON),66-75,1998.

[4] 1N.Karthikeyan, 2Dr.V.Palanisamy, and 3Dr.K.Duraiswamy “Performance Comparison of Broadcasting methods in Mobile Ad Hoc Network “International Journal of Future Generation Communication and Networking Vol. 2, No. 2, June, 2009 47.

[5] Heinzelman, W.R., Chandrakasan A, Balakrishanan H, “Energy-efficient communication protocol for wireless microsensor networks[J]” System Sciences,2000, In: Proceedings of the 33rd Annual Hawaii International Conference on 4-7 Jan 2000 Page:10 pp.vol.2.

[6] Heinzelman W, Chandrakasan A, Balakrishanan H, “An Application-specific protocol architecture for wireless microsensor networks[J]”, In: IEEE Transaction on Wireless Communications, 2002, (10):660-670.

[7] Jie Wu and Fei Dai “Mobility Management and Its Applications in Efficient Broadcasting in Mobile Ad Hoc Networks[J] “ 0-7803-8356-7/04 IEEE INFOCOM 2004

[8] S.V.M.G.Bavithiraja “Location Aided Broadcasting in Mobile Ad Hoc Networks[J]” International Journal of Recent Trends in Engineering, Vol. 1, No. 1, May 2009 pp 619.

[9] Mahboobeh Abdoos, Karim Faez, and Masoud Sabaei “Position Based Routing Protocol with MoreReliability in Mobile Ad Hoc Network[J]” World Academy of Science, Engineering and Technology 49 2009

[10] Hossein Ashtiani, Shahpour Alirezaee, seyed mohsen mir hosseini, HamidKhosravi”PNR: New Position based Routing Algorithm for Mobile Ad Hoc Networks[J]” Proceedings of the World Congress on Engineering 2009 Vol I WCE 2009, July 1 - 3, 2009, London, U.K.

[11] N. Karthikeyan, V. Palanisamy and Duraiswamy “Reducing Broadcast Overhead Using clustering Based Broadcast Mechanism in Mobile Ad Hoc Network[J]” Journal of Computer Science 5 (8): 548-556, 2009 ISSN 1549-3636 © 2009 Science Publications

[12] Md. Anisur Rahman, Md. Shohidul Islam, Alex Talevski “Performance Measurement of Various Routing Protocols in Ad-hoc Network[J]”Proceedings of the International MultiConference of Engineers and Computer Scientists 2009 Vol I IMECS 2009, March 18 - 20, 2009, Hong Kong ISBN: 97.

[13] Wei Lou, Member, Jie Wu, Senior Member”Toward Broadcast Reliability in Mobile Ad Hoc Networks with Double Coverage[J]” IEEE Transcations on Mobile Computing, vol. 6, NO. 2, Feb 2007.

AUTHORS

Gomathi. S received the B.Sc and MCA Degrees from Madras University and M.Phil (C.S) ,Mother Teresa Woman’s University Kodaikanal. She is working as a Professor in Department of MCA at K.S.R College of Technology,Tiruchengode. Her area of interest Computer Networks, Network Protocols, Mobile Ad hoc Networks, Unix and Network Programming. She is a member of ISTE.

Dr.K.Duraiswamy received the B.E,M.E and Ph.D (Communication Engineering ) in 1972,1974 and 1987 respectively. After working as a Lecturer in the Department of Electrical Engineering in Government College of Engineering ,Salem affiliated to Anna University and as an Asst.Professor in Government College of Technology ,Coimbatore and as a Professor and Principal at K.S.R College of Technology . He has been working as a Dean in the Department of Computer Science and Engineering at K.S.R College of Technology . His research interest includes Mobile Computing ,Soft Computing ,Computer Architecture and Data Mining .He is a Sr. member of ISTE,SIEEE and CSI.

COMPARATIVE STUDY OF SUBGRADE SOIL STRENGTH ESTIMATION MODELS DEVELOPED BASED ON CBR, DCP AND FWD

TEST RESULTSDr. R. SRINIVASA KUMAR

Civil Engineering Department, Osmania University, Hyderabad, India.ABSTRACT:Subgrade soil strength is an essential input used for design of new pavements and overlays. There are many models are available for determination of subgrade soil strength. Subgrade soil strength in terms of elastic modulus value is used in Mechanistic-Empirical methods and CBR value is used in empirical methods of design. Both the parameters can be used for structural evaluation of flexible pavement. Determination of elastic moduli values of pavement layers by back-calculation of deflections measured using Falling Weight Deflectometer (FWD) is a widely used scientific method. However, extensive use of FWD for large scale evaluation of in-service pavements will not be feasible, especially for low volume roads, unimportant roads and roads in developing and under developed countries. In the present investigation, an attempt has been made to develop relationships that can be used to estimate subgrade soil strength in terms of elastic modulus and CBR values using in-situ test results of Dynamic Cone Penetrometer (DCP). Because, DCP is a portable and low-cost equipment used for quick evaluation of unbound granular and subgrade soil. Comparative studies are also carried out with similar models available from the literature review.

Keywords:in-service pavements, falling weight Deflectometer (FWD), back-calculated subgrade modulus, California Bearing Ratio (CBR), dynamic cone Penetrometer (DCP) and regression models.

INTRODUCTION

Recent trend of flexible pavement design is being made using mechanistic and empirical (M-E) methods. The M-E methods use elastic moduli values and the empirical methods use CBR values of subgrade soil. Towards this objective, the task of selection of appropriate values of subgrade strength for the analysis of new as well as in-service pavements has been engaging the attention of pavement researchers for a number of years.Assessment of elastic moduli of granular and subgrade soil by using Falling Weight Deflectometer (FWD) or laboratory testing could be a costly procedure for evaluation of unimportant or low traffic volume roads, roads in developing and under developed countries. Due to this reason, Dynamic Cone Penetrometer (DCP) test has been extensively using by several countries around the globe for quick determination of an index value which represent the strength of compacted unbound granular and soil layers with reasonable accuracy at low cost. In the present investigation, an effort has been made to develop regression models by correlating the results obtained from the evaluation of the subgrade soil of pavements using test results of CBR, Dynamic Cone Penetrometer (DCP) and Falling Weight Deflectometer (FWD).

HISTORICAL DEVELOPMENT OF DCP

The first Dynamic Cone Penetrometer (DCP), which is similar to as now it is available, known as Scala Penetrometer, was developed by Scala (1956) in Australia. It was used for determination of in-situ CBR of cohesive subgrade soils of flexible pavement. The Scala Penetrometer consists of about 9 kg hammer which drops from 510 mm height through a vertical guide rod. The hammer impact energy is ultimately applied on to a cone (having apex angle 30o) fitted at bottom end of the guide rod. Later, a similar tool like DCP was developed by van Vuuren (1969). It consists of a 10 kg hammer which drops freely from 460 mm height. The impulse force is applied on to a cone having 30o. He has developed a relationship between the DCP test results and CBR values of subgrade soil. For rapid evaluation of flexible pavements a DCP was used by the Transvaal Roads Department of South Africa in 1973 (Kleyn 1975). The DCP consisted of 8 kg hammer which drops from height 574 mm. The cone’s apex angles considered were 30o and 60o and a comparative study was carried out between the DCP test results (Kleyn et al 1982). Another report (Kleyn and Savage, 1982) reveals that, several investigations were made on subgrade tested with DCP comprised of 8 kg hammer which drops from height 574 mm and the cone’s apex angle as 60o.

Since then, researchers in several countries like South Africa (Kleyn et al 1982; Kleyn and Van Heerden 1983), Australia (Smith and Pratt, 1983), New Zealand, Zimbabwe and many other countries (Livneh et al 2000) have widely used the DCP of Kleyn and Savage (1982) for evaluation of unbound granular and soil layers of pavement. In the recent past, the DCP has been adopted for design, construction quality control of flexible pavement granular base layer compaction (Siekmeier et al 1998), Subgrade acceptance criteria (Kremer and Dai 2004) and also using routinely for assessing in-situ strength of sand backfill trenches (Zhang et al 2004), narrow trenches (Ford and Eliason 1993), drain trenches beside of pavement (Siekmeier et al, 1998) and subgrade soil in many other countries viz. USA (Ayers et al, 1989; Webster et al, 1992, 1994; Grogan and Tingle, 1999; FHWA; AASHTO; ASTM D6951, 2004; KDOT; MnDOT; WDOT; MDOT; FLDOT; ODOT, INDOT and other), Israel (Livneh and Ishai, 1987; Livneh, 1991; Livneh et al, 1995, 2000), Ireland (McGrath et al, 1989), UK (TRL, Overseas Road Note 8, 1990), India (Sood et al, 1996; Kumar et al, 2001, 2003, 2006; IRC:SP:72-2007) and in numerous other countries.Instrumented DCP As another version, the DCP fitted with load-cell and accelerometer is termed as iDCP which can be mechanically operated and the hammer impulse and cone penetration data are electronically recorded (Chua and Lytton 1988; Nazarian et al 2000). Desired outputs can be obtained by analyzing the data using standard software packages supplied along with the iDCP manufacturer.WORKING PRINCIPLE OF DCPThe salient features of the widely using manually operated DCP are presented in Figure 1. The resistance offered by the compacted granular/soil layer due to penetration of a standard cone having 20 mm diameter and apex angle 60o 1 is driven by a 8 kg hammer dropped from a height of 575 mm, is recorded as average penetration per blow (IRC:SP:72 2007). The average penetration of the cone per blow is reported in as an index value and it may be represented in many forms viz. DCP index (DCPI), DCP, penetration rate (PR), penetration index (PI), DCPθ

o; or Number of blows required to penetrate a given thickness of layer as (DCPN), NDCP, Blow Rate (BR) and DCP Structural Number (DSN). In this paper, the cone’s average penetration per blow (in mm/blow) is denoted as DCPθ

o, where, θo denotes apex angle of the cone. During the cone penetration, the material shear strength is mainly accrued due to resistance offered by the shear displacement of soil particles. A smaller

value of DCP60o indicates harder material and vice

versa.with DCPIt is generally prescribed to conduct DCP tests along wheel path of pavement, during post monsoon season because, at that time, the pavement layers become soft or less stiff due to accumulation of rain or stagnated water. Based on initial positioning of the cone, the DCP testing methods can be broadly classified into different types such as testing in a pit excavated to depth up to (i) subgrade level (Figure 2a) or (ii) base or (ii) Sub-base level (Figure 2b).To begin with the testing procedure, initial reading on the DCP scale should be noted while holding the rod vertical and the cone is in contact of surface of compacted soil to be tested. While one person holding the DCP handle, another person should lift the hammer to the predetermined height (i.e. 575 mm) and allowed it to drop freely on to the anvil. Cumulative penetration value of the cone should be noted for each blow. Generally, the DCP testing should be stopped when the cone penetrating is not more than 1 mm per blow since the cone may be encountered with a rock or gravel larger than 20 mm size or hard strata. After the DCP test is over, the cone-rod should be extracted out and the tested soil sample may be collected for conducting other tests in laboratory.An extension rod of additional length of 1000 mm may be used in place of the standard penetrating rod where it is necessary to test the deep soil (Livneh and Livneh, 1994). A study has conducted by Livneh (1991) for determining of change in rate of penetration due to change in the cone apex angle from 30o to 60o and suggested the following relationship to convert DCP index value (in mm/blow) in terms of the other. = 0.006 + 1.092 (1)The DCP test can also be conducted in a laboratory on remolded material compacted in a steel mould which significantly eliminate the effect of confinement and such test can also be carried out as a prototype model (Ayers et al, 1989).REVIEW OF LITERATURE In different parts of the world, several researchers (De Villiers 1980; Livneh and Ishai 1987; Pen 1990; Chai and Roslie 1998; Chen and Latorella 1999; Livneh et al 2000; Roy 2007) have estimated strength properties of granular and subgrade soils by correlating DCP test results with standard test parameters. Such relationships can be used to convert the DCPθ

o value to different strength parameters such as CBR, elastic modulus and other standard test values. The following paragraphs briefly present these relationships.

Relationships between DCPθo and CBR

ValuesSeveral relationships are available to convert values of DCPθ

o to CBR (Table 1). Such variety of relationships do exists since large strain penetration takes place during testing of compacted soil by both the tests. A few salient details of different investigations made for development of the relationships (Table 1) are given below. • Development of Scala Penetrometer (1956) for estimation of in-situ CBR of cohesive soils has led to development of the present version of DCP.• Scala (1956) and Kleyn (1975) were initially identified straight line relationship between the DCPθ

o

and CBR values plotted on log-log chart. • Laboratory CBR values are used for development of the US Corps of Engineer’s relationship (Webster et al 1992). Many DCPθ

o to CBR relationships developed by different researchers around the globe were considered in this study and they were found to be in close agreement with the relationship developed by Webster et al (1992). Therefore, this relationship has been widely used by several researchers (Livneh 1995; Siekmeier et al 2000 and Chen et al 2001).• Ese et al (1994), extensively evaluated 23 granular base courses in Norway and correlated laboratory CBR and DCPθ

o values. They reported that, (i) difference in confining pressure in CBR mould and prevailing in-situ condition was accounted in development of the relationship and (ii) they suggested that, a critical stability value of 2.6 mm/blow may be taken for gravel base.• Nazzal (2003), conducted many laboratory CBR and field DCP tests on compacted granular material, clay and soils stabilized. During the laboratory CBR testing, the moisture content and density were maintained similar to in-situ condition. He reported that, by using the equation (No.15), the estimated CBR values were well matching with the values obtained by equation (No. 10) (Webster et al 1992), when the DCP60

o value is greater than 20 mm/blow.Relationships between DCP

θo and FWD Back-

calculated Subgrade Modulus Values

Elastic moduli of pavement layers are essential inputs required in the mechanistic design of new pavements and for the estimation of the remaining life of in-service pavements. Towards this aim, selection of appropriate values of base and subgrade layer moduli for the analysis of new as well as in-service pavements has been crucial issue concerning the pavement/soil researchers. Towards this objective,

several researchers in different countries have developed the following relationships between back-calculated subgrade moduli and DCP index values. Such relationships can be re-used where routine evaluation of pavement by FWD is not feasible for low traffic volume roads.§ Pen (1990) developed the following equations

based on field DCP values.

( ) ( )

( )system . P EAC Hby atedbackcalcul M P a)(in m odulus elastic subgrade theis E W here,

(18) )DC Plog(17.1653.3Elog

F W D. P HONIX using m easured sdeflection thefrom obtained M P a)(in m odulus elastic subgrade theis E where,

(17) DC Plog89.0250.3Elog

s

6 0s

s

6 0s

0

o

−=

−=

De Beer (1991) has prescribed the following equation.

Subgrade Elastic Modulus (MPa) = 1176 ×

()-1.082 (19)Where, DCP value is in mm/blow.

The following equations were prescribed by Chai and Roslie (1998):

Relationship between in-situ CBR and DCP: Es = 17.6(269/DCPV)0.64 (20)

Based on back-calculated results: Es = 2224 (DCPV)-0.996 (21)

Where, Es = Subgrade modulus value (in MPa) and DCPV = blows/300mm penetration of cone having apex angle 60o.

Chen et al (1999) conducted FWD, CBR and DCP tests on six sections of pavements in Kansas. Based on FWD back-calculated subgrade moduli (Es) and DCP test data, the following expression has prescribed.

Subgrade Resilient Modulus (Es in MPa) = 338 × ()-

0.39 (22)Where, value was in the range of 60 to 10 mm/blow.

Chen et al (1999) found that the subgrade moduli determined from their relationships between DCP to CBR and CBR to Es were widely varying along the length of the test sections.

Livneh (1991) developed the following relationships by correlating DCP values with back-calculated subgrade modulus obtained from Benkelman Beam deflection data.

George and Uddin (2000) developed the following equations (Cited in Roy 2007).

For fine grained soils:

For coarse grained soils:

Chen et al (2005) prescribed the following equation (After eliminating outlier data).

As an indirect approach, Siekmeier et al (2000) and Chen et al (2001, 2005), converted the DCP index value to CBR value by using the equation developed by Webster et al (1992) and the obtained CBR value is converted to the resilient modulus (Es) value using the following equation of TRL-UK (Lister and Powell 1987).

Chen et al (2001) conducted a series of FWD and DCP tests on base and subgrade layers of two pavement sections in Texas. The test results were found to be satisfactorily comparable. However, they found that the DCP index values can be affected by at least 10 percent due to variations in the test procedure.Several attempts were also made in India (Sood et al 1996, Srinivasa et al 2003, 2006), USA (Mi-DOT) and other parts of the world (Livneh et al 2000) for developing equations to estimate of modulus value of granular and subgrade soil using DCP index value and material parameters such as field density, density ratio, liquid limit, plasticity index, coefficient of uniformity, field moisture content, ratio between field to OMC and percent soil passing different sizes of sieves.

In addition to this, by using DCP test results, there are several relationships available for determining (i) AASHTO granular layer coefficients (cited in Roy 2007), (ii) unconfined compressive strength (UCS) of soil (De Villiers 1980; McElvaney and Djatnika 1991), using DCP test values of granular base and sub-base and (iii) shear strength of cohesion less materials using relationships available between the laboratory determined values of DCP60

o and triaxial deviator stress at failure (Ayers et al 1989).STRUCTURAL EVALUATION OF PAVEMENTSAs a part of a research project in India, Falling Weight Deflectometer (FWD) and Dynamic Cone Penetrometer (DCP) tests were conducted on different sections of National Highway (NH) 6, National Highway (NH) 60, Salua Road and IIT-perimeter Road. The above tests were conducted during post monsoon season because during this period the subgrade becomes soft or less stiff due to accumulation of rain and stagnated water.NH 6 and 60 are two-lane two-way highways. On NH 6, six different pavement stretches of about 300 m length each were selected and the type of subgrade soils ranged from clay to silt. The types of subgrade

soil encountered at NH 60 are of clay to silty clay. Salua road and IIT-perimeter road are lower volumes. Soil sample collected from one test pit excavated in Salua road indicated the subgrade soil is clay type. Along the IIT-perimeter road, the soil type varied from clay to silt.

Deflection Measurements Using Falling Weight DeflectometerPavement deflection measurements on selected test sections were taken using FWD (Srinivasa 2001). An impulse load of 40kN in magnitude was applied over a contact area of radius 150 mm. The FWD loading time ranges between 25 to 30 milli-seconds as usually. A load cell is used to measure the impulse load. Six geophones/deflection sensors are used to measure surface deflection basin and the sensors are placed radially at 300 mm intervals from the center of the loading plate.Analysis of FWD DataEffective layer moduli were backcalculated using a Genetic Algorithm based back-calculation computer program, BACKGA (Murthy 1999). A linear elastic layered system, ELAYER program (Reddy 1993), was used for forward calculation routine of the BACKGA program. All the FWD tested pavement sections were modeled as 3 layer system.DCP and CBR TestsDCP tests were conducted on subgrade soil of the pavement sections tested with FWD. Soil samples were collected from some of the sections and were tested in laboratory for determination of their CBR values and soil classification. The details of subgrade soil classification, DCP, FWD and CBR test results are given in Table 2-5.Development of Regression ModelsRegression analysis was carried out on the data collected in the present investigation and trials were made to develop different forms of relationships between the values of (a) CBR and DCP60

o and (b) back-calculated subgrade modulus (Es) and DCP60

o. From the trials, the following models were found to be the most suitable.Between CBR and DCP60

o:

Between Es and DCP60:

( )0 . 9 9 )R 3 2 ,( N

( 2 9 ) )D C Plo g (3 1 5 2.17 1 5 4 5.2c e n tp e rinC B Rlo g2

6 0s 0

==

−=

( )0 .7 1 )R 9 6 ,( N

( 3 0 ) )D C Plo g(6 4 3 8.05 5 3.2M P ainElo g2

6 0s 0

==

−=

COMPARATIVE STUDY

Comparative studies were conducted using the relationships developed in the present study with similar relationships available from the literature review. The Comparative study considered two types of relationships between (i) CBR and DCP60

o and (ii) Es and DCP60

o.(i) On relationships between CBR and DCP60

o

Relationships between DCP60o and CBR values are

presented in Figure 3. It is noted that, all the lines of the relationships are relatively located closer within a narrow bandwidth except the relationships prescribed by McElvaney et al (1985) and Nazzal (2003). Such disparities among relationships which were developed in different countries may be due to slight variations in the test procedure, differences in the soil moisture content, compacted density, surrounding confinement and type of soil. It is also noted that, all above equations are acceptable to the in-situ conditions prevailing and based on the adopted test conditions. Further comparison has been made on the relationships belongs to fine grained soils and other which were developed by Harison (1989), TRL-UK (1990), Webster et al (1992) and the present study. It can be observed from the Figure 4 that, (a) the relationships (by Eq. 6 and 7) developed by Harison are showing almost same slope (or trend) for the soil strength considered (b) The relationships of TRL-UK, Harrison and by the present study are closely matching with each other except with small differences (c) at the lower values of CBR, all the relationships are within a close bandwidth except the relationships (by Eq. 12 and 13) developed by Webster et al (1992) which were specifically prescribed for soil types of CL (where CBR <10 per cent) and CH. This indicates that, the DCP60

o values are sensitive at smaller CBR values and the subgrade consists of highly plastic clay.(ii) On relationships between Es and DCP:Relationships between back-calculated subgrade modulus (Es) and DCP60

o are presented in Figure 5. Larger variations are found among these relationships. A line drawn based on indirect method using relationships of Webster et al (1992) (by Eq 10) and TRL-UK (Lister and Powell 1987) (by Eq 28) is found to be located in the middle of the lines. Such scatter and disparities are not abnormal because, in different countries the subgrade soil properties may vary with many parameters such as variations in moisture content, intensity of compaction took place during construction and traffic loading, surrounding confinement and drainage conditions. It is also noted that, all above equations are acceptable which are derived based on the test conditions and prevailing material properties.

CONCLUDING REMARKSThe task of selection of appropriate values of subgrade strength for the analysis of new as well as in-service pavements has been engaging the attention of pavement researchers for a number of years. Towards this objective, the DCP as a low cost and alternative tool used for evaluating unbound granular and subgrade soil of low volume and roads in developing and under-developed countries where routine evaluation of FWD is not feasible. The relationships available are useful for estimation of soil strength properties used for design and structural evaluation of unbound granular/soil layers of pavement. The disparities among the different relationships developed for different countries may be due to variations in prevailing soil type, drainage condition, layers confinement, depth of testing and variations in test procedures adopted. However, the available relationships are acceptable which were derived based on test procedure adopted and likely reflects soil in their region.ACKNOWLEDGEMENTSThe Author is thankful to his teachers, Prof. B. B. Pandey, Rtd., Professor, Prof. M. Majumdar (Late) Rtd., Professor and Prof. K. Sudhakar Reddy, Professor of Civil Engineering, IIT, Kharagpur, India, for imparting basic training input.

REFERENCESAyers M. E., Thompson M. R. and Uzarski D. R., 1989, “Rapid Shear Strength Evaluation of In-Situ Granular Materials,” Transportation Research Record No. 1227, TRB, Washington, D.C., pp.134-146.

Chai G. and Roslie N., 1998, “The Structural Response and Behavior Prediction of Subgrade Soils Using Falling Weight Deflectometer in Pavement Construction,” Proc., of 3rd Int. Conf. on Road and Airfield Pavement Technology.

Chen D. H., Lin D. F., Liau P. H. and Bilyen J., 2005, “A Correlation Between Dynamic Cone Penetrometer Values and Pavement Layer Moduli,” Geotechnical Testing Journal, Vol. 28(1), pp.42-49.

Chen D. H., Wang J.N., and Bilyen J., 2001, “Application of Dynamic Cone Penetrometer in Evaluation of Base and Subgrade Layers,” Transportation Research Record 1764, TRB, Washington, D. C., pp.1-10.

Chen J., Hossain M. and Latorella T. M., 1999, “Use of Falling Weight Deflectometer and Dynamic Cone Penetrometer in Pavement Evaluation,”

Transportation Research Record No.1655, TRB, Washington, D. C., pp.145-151.

Chua K. M. and Lytton R. L., 1988, “Dynamic Analysis Using The Portable Pavement Dynamic Cone Penetrometer,” Transportation Research Record No. 1192, TRB, pp.27-38.

De Beer M., 1991, “Use of the Dynamic Cone Penetrometer (DCP) in the Design of Road Structures.” Geotechnics in the African Environment, Vol. 1, Blight et al. (Eds), Balkema, Rotterdam, pp.167-183.

De Villiers P. J., 1980, Dynamic Cone Penetrometer Correlation with Unconfined Compressive Strength, Univ. of Pretoria, Pretoria, 1980.

Ese D., Myre J., Noss P. and Vaernes E., 1994, “The Use of Dynamic Cone Penetrometer (DCP) for Road Strengthening Design in Norway,” Proc., the 4 th Int., Conf., on the Bearing Capacity of Roads and Airfields, pp.343-357.

Ese D., Myre J., Noss P. and Vaerness E., 1994, “The Use of the Dynamic Cone Penetrometer (DCP) for Road Strengthening Designs in Norway,” Proc., of 4th Int. Conf., on the Bearing Capacity of Roads and Airfields, Univ. of Minnesota, Minneapolis, Mn., pp.3-22.Ford G. R. and Eliason B. E., 1993, “Comparison of Compaction Methods on Narrow Surface Drainage Trenches,” Transportation Research Record No. 1425, Transportation Research Board, Washington, D.C., pp.18-27.

Gabr M. A., Hopkins K., Coonse J. and Hearne T., 2000, “DCP Criteria for Performance Evaluation of Pavement Layer,” Journal of Performance of Constructed Facilities, ASCE, Vol.14(4), pp.141-148.

Grogan W. P. and Tingle J. S., Evaluation of Un surfaced Airfield Criteria for Operation of C-17 Aircraft, Jl. of Transportation Engineering, ASCE, Vol. 125(1), 1999. Harison J. A., 1989, “In Situ CBR Determination by DCP Testing Using a Laboratory–based Correlation,” Australian Road Research, Vol. 19(4), pp.313-317.

IRC:SP:72-2007, 2007, “Guidelines for the Design of Flexible Pavements for Low Volume Rural Roads,” The Indian Roads Congress, New Delhi, India.

Jayawickrama P. W., Amarasiri A. L. and Regino P. E., 2000, “Use of Dynamic Cone Penetrometer to

Control Compaction of Granular Fill,” Transportation Research Record No. 1736, TRB, Washington, D.C.

Kleyn E. G. and Savage P. E., 1982, “The Application of the Pavement DCP to Determine the Bearing Properties and Performance of the Road Pavements,” Int. Symposium on Bearing Capacity of Roads and Airfields, Trodheim, Norway, pp.238-246.

Kleyn E. G. and van Heerden M. J. J., 1983, Using DCP Soundings to Optimise Pavement Rehabilitation, Paper submitted for Annual Transportation Convention, Johannesburg, Report LS/83, Materials Branch, Transvaal Roads Dept., Pretoria, South Africa, July.

Kleyn E. G., 1975, The Use of the Dynamic Cone Penetrometer (DCP), Report No. 2/74, Transvaal Roads Department, Pretoria, South Africa, July.

Kleyn E. G., Van Heerden M. J. J. and Rossouw A. J., 1982, “An Investigation to Determine the Structural Capacity and Rehabilitation Utilization of a Road Pavement Using the Pavement Dynamic Cone Penetrometer,” Int. Symposium on Bearing Capacity of Roads and Airfields, Trodheim, Norway, pp.1011-1026.

Kleyn, E. G., Maree, J. F. and Savage, P. F., 1982, “The Application of a Portable Pavement Dynamic Cone Penetrometer to Determine In-Situ Bearing Properties of Road Pavement Layers and Subgrades in South Africa,” The European Symposium on Penetration Testing, Amsterdam, Netherlands, pp.277-282, May.

Kremer, C., and Dai, S. 2004, “Improvement and Validation of MN/DOT Specification for Aggregate Base”, Minnesota Department of Transportation, Maplewood, MN.

Lister N. W. and Powell W. D, 1987, “Design Practice for Bituminous Pavements in the United Kingdom”. Proc. 6th International Conference on Structural Design of Asphalt Pavements, Vol. 1, pp.220-231.

Livneh M. and Ishai I., 1987, “Pavement and Material Evaluation by a Dynamic Cone Penetrometer,” Proc., 6th International Conference on Structural Design of Asphalt Pavements, Vol. 1, Ann Arbor, Michigan, 665-674.

Livneh M. and Livneh N. A., 1994, “Subgrade Strength Evaluation with the Extended Dynamic

Cone Penetrometer,” Proc., 7th Int., Congress Int., Association of Engineering Geology, Vol. 1, Lisbon.

Livneh M. Livneh N. A. and Ishai I., 2000, “The Israeli Experience with the Regular and Extended Dynamic Cone Penetrometer for Pavement and Subsoil-Strength Evaluation,” Nondestructive Testing of Pavements and Backcalculation of Moduli: Vol. 3, ASTM STP 1375, S. D. Tayabji and E. O. Lukanen, Eds., ASTM, West Conshohocken, PA, pp.189-213.

Livneh M., 1987, “Validation of Correlations between a Number of Penetration Tests and In-Situ California Bearing Ratio Tests”, Transportation Research Record No. 1219, TRB, Washington, D.C., pp.56-67, 1987.

Livneh M., 1991, “Verification of CBR and Elastic Modulus Values Derived from Local DCP Tests.” Proc., 9th Asian Regional Conference on Soil Mechanics & Foundation Engineering, Vol. 1, Bangkok, pp.45-50.

Livneh M., Ishai I. and Livneh N. A., 1995, “Effect of Vertical Confinement on Dynamic Cone Penetrometer Strength Values in Pavement and Subgrade Evaluations,” Transportation Research Record No. 1473, TRB, Washington, D.C., pp.1-8.

McElvaney J. and Djatnika I., 1991, “Strength Evaluation of Lime-Stabilized Pavement Foundations Using the Dynamic Cone Penetrometer,” Journal of ARRB, Vol. 21(1).

McElvaney J., Jayaputra A. and Harison J., 1985, “Correlation of CBR and dynamic cone strength measurements of soils,” Proceedings of 3rd Indonesian Conference on Geotechnics, Vol. 1, Jakarta.

Murthy S. M, 1999, “Application of Genetic Algorithms in Pavement Engineering”, M. Tech. Thesis submitted to the Indian Institute of Technology, Kharagpur, India.Nazarian S., Tandon V., Crain K. and Yuan D., 2000, “Use of Instrumented Dynamic Cone Penetrometer in Pavement Characterization,” Nondestructive Testing of Pavements and Backcalculation of Moduli: Vol. 3, ASTM STP 1375, S. D. Tayabji and E. O. Lukanen, Eds., ASTM, West Conshohocken, PA, pp.214-228. Nazzal M. D., 2003, “Field Evaluation of In-Situ Test Technology For Qc/Qa During Construction of Pavement Layers And Embankments,” A Thesis Submitted to the Louisiana State Univ. and Agricultural and Mechanical College.

Pen C. K., An assessment of the available methods of analysis for estimating the elastic moduli of road pavements. Proc., 3rd Int. Conf., on BCRA, Trondheim, 1990.Reddy K. S., 1993, “Analytical Evaluation of Bituminous Pavements”, Ph.D. thesis, Civil Engineering Department, IIT, Kharagpur, India.

Roy B. K., New Look at DCP Test with a Link to AASHTO SN Concept, Jl. of Transportation Engineering, ASCE, Vol. 133(4), pp.264-274, 2007.

Scala A. J., 1956, “Simple Methods of Flexible Pavement Design Using Cone Penetrometers,” New Zealand Engineering, Vol. 11(2), pp.34-44.Scala A. J., 1956, “Simple Methods of Flexible Pavement Design Using Cone Penetrometers,” Proc., 2nd Australian-New Zealand Conf Soil Mech. and Foundation Eng, pp.73.

Siekmeier J., Burnham T. and Beberg D., 1998, “Mn/DOT’s New Base Compaction Specification Based on the Dynamic Cone Penetrometer,” 46th

Geotechnical Engineering Conference, Univ, of Minnesota, Feb.

Siekmeier J., Young D. and Beberg D., 1999, “Use of DCP and portable FWD for subgrade characterization in Minnesota,” Proc., 3rd Symposium on Nondestructive Testing of Pavements and Backcalculation of Moduli, SPT 1375. 1999.

Siekmeier J., Young D. and Beberg D., 2000, “Comparison of the Dynamic Cone Penetrometer with other tests During Subgrade and Granular Base Characterization in Minnesota,” Nondestructive Testing of Pavements and Backcalculation of Moduli: 3rd Volume, ASTM SPT 1375., S. D. Tayabji and E. O. Lukanen, Eds.Smith R. B. and Pratt D. N., 1983, “A Field Study of In-situ California Bearing Ratio and Dynamic Cone Penetrometer Testing for Road Subgrade Investigations,” Australian Road Research, ARRB, Vol. 13(4), pp.285-294, December, 1983.

Sood, V. K., Vijayaraghavan, S. R., Joshi, P., Rao, Y. V. and Kannan S., 1996, “Evaluation of in-situ subgrade strength by dynamic cone Penetrometer”. Indian Roads Congress, Highway Research Board, No. 55, pp.39-49.

Srinivasa K. R., 2001, “Structural Evaluation of Pavements Using Falling Weight Deflectometer”, Ph.D. thesis, Civil Engineering Department, IIT, Kharagpur, India.

Srinivasa K. R., Ajay Kumar and Subba Rao, 2006, “Fabrication of DCP For Evaluation of Subgrade Soil”, National Conference on Civil Engineering Systems, Department of Civil Engineering, Osmania University, Hyderabad, 1-3 June.Srinivasa, Kumar, R., Reddy K. S., Mazumdar M. and Pandey B. B., 2003, “Models for Estimation of Subgrade Moduli from DCP Tests,” The International Journal of Pavement Engineering & Asphalt Technology, UK, Vol. 4 (2), December.Srinvasa, Kumar. R., 2009, A Text Book of Highway Engineering, under publication.

TRL, 1990, Transport Research Lab., A Users Manual for a Program to Analyse Dynamic Cone Penetrometer Data, Overseas Rd. Note 8, Dept., of Transport, UK.Van Vuuren, 1969, “Rapid Determination of CBR with the Portable Dynamic Cone Penetrometer,” Reprint from the “The Rhodesign Engineer”, September.Webster S. L., Brown R. W. and Porter J. R., 1994, “Force Projection Site Evaluation Using the Electrical Cone Penetrometer (ECP) and the

Dynamic Cone Penetrometer (DCP),” Proc., of 4th

Int. Conf., on the Bearing Capacity of Roads and Airfields, Univ. of Minnesota, Minneapolis, Mn.

Webster S. L., Grau R. H. and Williams T. P., 1992, “Description and Application of Dual Mass Dynamic Cone Penetrometer,” Instruction Report, GL-92-3, Army Engineers Waterways Experiment Station, Vicksburg, Miss.

Zhang Z. J., Abu-Farsakh Y. M. and Wang L., 2004, “Evaluation of Trench Backfill at Highway Cross-Drain Pipes”, Submitted to 83rd Transportation Research Board Annual Meeting, Washington, D. C. TABLE 1-Relationships between DCPè

o and CBR Values (from Srinivasa, 2009).

b and are the penetration values in mm/blow where the cone apex angles are 300 and 600 respectively; CBR is the California Bearing Ratio (%); CBRF

indicates field CBR; and CBRf indicates the field CBR value of cohesive residual soil.

c The prescribed range of value is 6.31 to 66.67 mm/blow.

TABLE 2-Test Results of IIT- Perimeter Road.

TABLE 3 -Test Results of Salua road.

TABLE 4 -Test Results of National Highway 60

TABLE 5-Test Results of National Highway 6.

.

LIST OF FIGURE CAPTIONS

FIG. 1-Details of DCP

FIG. 2-Different Methods of DCP Testing(Modified and elaborated based on Livneh et al 2000)

FIG. 3-Relationships for Converting DCP60 to CBR (Srinivasa 2009)

FIG. 4-Comparison Between a Few Selected Relationships of DCP60 - CBR.

FIG. 5-Relationships for Converting DCP60 to Elastic Modulus.

20 mm dia

θ

Handle

575

mm

Bottom Rod Meter Rule

Hammer

Guide Rod

Anvil Index

FIG. 1-Details of DCP.

Supporting Plate

Limit Plate

FIG. 2-Different Methods of DCP Testing.

(Modified and elaborated based on Livneh et al 2000)

DCP cone

Subgrade (a) On subgrade (Unconfined)

Subgrade

Bituminous Layer

Sub-base

Base

(b) Through granular layers (Confined)

Bituminous

Base

Sub-base

AN EXPERIMENTAL STUDY ABOUT THE LIGHTWEIGHT COMPOSITE BRICKS

D. JEGATHEESWARAN*

*Assistant Professor / Dept. of Civil EngineeringK.S. Rangasamy College of Technology, Tiruchengode - 637215, TamilNadu.

E-mail: [email protected] R. MALATHY**

Principal, Excel Engineering College, Komarapalayam -637 303, TamilNadu. E-mail: [email protected]

ABSTRACT: In this investigation, study the feasibility of using waste paper in brick. Three different proportions were used in this investigation. A total of 60 numbers of brick specimens were cast with and without admixture and were tested under compression and water absorption. These bricks are relatively light weight, good sound absorbent and more flexible, easily cut in to desirable shape. And also it does not expand or contract due to surround environment. but it has high percentage of water absorption than conventional bricks. And it controlled by using of water repellent admixtures. It can be the results show that the effect of high-level replacement of paper wastes does not exhibit a sudden brittle fracture, and it reduces the unit weight dramatically and introduces smoother surface compared to the current conventional bricks and concrete blocks in the market.

Key words: Waste Paper; Compression; Water absorption; Unit weight; Water Repellent Admixture

INTRODUCTION:Since the large demand has been placed on building material industry especially in the last decade owing to the increasing population which causes a chronic shortage of building materials, the civil engineers have been challenged to convert the industrial wastes to useful building and construction materials. Accumulating of unmanaged wastes especially in developing countries has resulted in an increasing environmental concern. Recycling of such wastes as building materials appears to be viable solution not only to such pollution problem but also to the problem of economic design of buildings. The increase in the popularity of using environmentally friendly, low-cost and light weight construction materials in building industry has brought about the need to investigate how this can be achieved by benefiting to the environment as well as maintaining the material requirements affirmed in the standards [2]. Many previous researchers [5-8] had undertaken obtained valuable results to use the industrial wastes in various forms of building material production.Some researchers carried out in the past used waste papers in a building material production [1,3]. In this paper, sand, cement, and waste materials like paper, fly-ash were mixed in a suitable proportion which gave more strength, less water absorbance and more eco friendly was found out.

And also, the character of that successive proportions were analysed in this paper.

Materials used:

Materials are the important criteria to produce a product. The materials should be easily available and eco-friendly. In this project waste materials were utilized to produce building bricks. The following materials were used in this investigation.

Fig.1 Materials used

o Paper

o Fly Ash

o Sand

o Ordinary Portland cement



o Water Repellent Admixture (Drfixit 105WP)

o Water Repellent Admixture (Drfixit 105WP)

Fig.2 Cellulose hydrogen bonds

Paper is principally wood cellulose. Cellulose is natural polymer. And Fig.2 shows the links of cellulose bonds. The cellulose chain bristles with polar-OH groups. These groups form hydrogen bonds with -OH group on adjacent chains, bundling, the chain together. The chains also pack regularly in places to form hard, stable crystalline region that give the bundled chains even more stability and strength. Fig.3 shows the network of cellulose fibres and smaller offshoots from the fibres called fibrils. In this, fibres and fibrils network forms a matrix, which becomes coated with Portland cement. When these networks of fibers and fibrils dry, they intertwine and cling together with the power of hydrogen bond. Coating this fiber with Portland cement creates a cement matrix, which encases the fibers for extra strength. Of course paper has more in it than cellulose. Raw cellulose has comparatively rough texture. Clay, rice husk ash is added to make the cellulose very smooth.

Fig.3 Microscopic view of cellulose

While adding more sand or glass to the mix results in a denser, stronger, more flame retardant material, but adds weight and reduces R- value. Heavy mixes with added sand, glass etc., increases strength and resistance to abrasion, but also reduces flexibility somewhat, adds weight and may reduce R-value. So the trick is finding the best mix for the application.

Fig.4 Mould

This mould was prepared with a non-water absorbing plywood material in the size of 230mm length, 110mm wide and 80mm deep.

The papers, which were collected, cannot be used directly. It should be made into paper pulp before mixing with other ingredients.

Fig.5 Tearing papers

Fig.6 Paper pulp

Mix proportions:

After all the ingredients were ready, the mixing was done. In this investigations, mixing was done manually. The following are the 3 proportions were used for manufacture the bricks with and without

water repellent powder Drfixit 105WP. That pidilite (Drfixit) company recommended the successive dosage of WP90 AS 20% (by the weight of Cement).

Fig.7 Moulding

After mixing, it should be placed in the mould within 30 minutes. So, six moulds were used at a time to make the process very fast. In this project, the bricks were moulded manually by hand. And these bricks were ground moulded bricks. And there was no special care was taken for curing, because paper used to hold water for long time. So only sun drying was allowed. The bricks were laid longitudinally in stacks of width equal to 2 bricks. The bricks were laid along and across the stack in alternate layer.

Fig.8 Curing

Result and Analysis :

Table:1 Details of Mix proportions

After casting the bricks, various tests were carried out to check the properties of the bricks. And the results of that test were analyzed with the existing and standard results. The following tests were carried out to check the brick.

Fig.9 Compression test

Compression test is the main and important test. This test only decides the strength of the brick. This test was carried out by a compression testing machine. This test was carried out on the 28th days from the date of casting of brick. Fig.9 shows the compression test was being done. While testing the brick great care must be taken, because the brick never failed catastrophically, it just compressed like squeezing rubber. So load was applied up to full compression. When the brick failed at the higher load, the structure was not fully collapsed. Only the outer faces cracked and peeled out. Water Absorption test is used to find out the amount of water absorbed in this brick. Because the brick, which are absorbing more water cannot be used in water logging area or exterior walls which is open to sky. The bricks from all the proportion were tested.

Fig.10 Fire test

A brick which is used for construction should not flammable in open flame, so fire test was carried out for the brick

Discussion :A brick that is used for construction should have compressive strength more than 3.5 N/mm2. In this above 3 bricks, only 2 type of bricks got compressive strength more than 3.5 N/mm2.But they are not suitable for load bearing walls. These bricks are having more elastic behaviour at failure.

It compresses like rubber. So at heavy load, the bricks may compress. Due to this compression the roof may be collapsed. So these bricks are suitable only for non-load bearing partition walls. As per standard, the bricks should not absorb water more than 20% of its weight, but these bricks exceeded water absorption ratio more than 20%. In this, the mix that contains

Fig.11 Compressive strength chart

Fig.12 Water absorption chart

only paper and cement absorbs water more than its weight.And the mixes that were mixed in 1:3:6 ratio absorbed more water. So these bricks are not suitable for water logging and exterior walls. These bricks can be used for interior partition walls only. Lightweight bricks are also the important objective of this project. So, all the bricks were tested whether they are having less weight or not. All the bricks were weighed in a well-conditioned electronic weighing machine.

The following are the weight of the bricks. The ordinary conventional bricks weight varies from 30 to 35 N. but these bricks weight varies from 10 to 20 N. The maximum weight is less than 20N only. So this bricks are light weight and it will also reduce total cost of construction due to the reduction in dead load.

Fig.13 Weight chart

From the fire test, it was observed that the fibrous concrete bricks did not burn with an open flame. They smoldered like charcoal. But these brick would be reduced to ashes after burning several hours. If the interior plaster and exterior stucco is provided on the fibrous concrete bricks, the bricks won’t burn. The

only weak point is inside the block, near electrical outlets, switches and other places where wires gives through walls, into boxes etc.,. Properly wired places never cause fire. If we apply the plaster without any hole or leakage on the bricks, it won’t burn or smolder inside. Because there will be lack of oxygen for burning.

In hardness test, a scratch was made on brick surfaces. While the scratch was made with the help of finger nail on the bricks, very light impression was left on the the brick surface. In this test two bricks from same proportion were taken and they were struck with each other. The bricks were not broken and a clear ringing sound was produced. So the bricks are good.The soluble salts, if presents in bricks will cause efflorescence on the surface of bricks. For finding out the presence of soluble salts in a brick, this test was carried out. In this test, these bricks were immersed in water for 24 hours. Then the bricks were taken out and allowed to dry in shade. And there was no any grey or white deposit on the bricks surface. It results that the bricks are free from soluble salts.

In site lot of bricks are wasted while cutting only. The labours could not able to cut the bricks exactly what they need. But, the bricks can be cut into exactly two pieces by using conventional saw blades. And the two fibrous concrete brick pieces can be hold together by putting a medium amount of glue on the bottom piece.

So, we can get any shape and size of the brick.

CONCLUSIONS

o These bricks can be used for non - load bearing partition walls. And these bricks are not suitable for water logging and external walls. It can be used in inner partition walls. Due to less weight of these bricks, the total dead load of the building will be reduced.

Since, these bricks are relatively light weight and more flexible.

These bricks are potentially an ideal material for earthquake prone areas.

o These brick does not expand or contract, so sheets of glass or glass block can be embedded in and trimmed with papercrete.

o These bricks are good sound absorbent, hence paper is used in these bricks. So, these bricks can be used in auditoriums.

o Since, the waste materials are used, it will reduce the landfills and pollution.

Acknowledgement:

Authors thanks AICTE, New Delhi, for sponsoring under career Award for young teachers scheme to Dr. R.Malathy to undergo this project.

REFERENCE:

1. B. Fuller, Fafitis and L. Santamaria (2006). “Structural Properties of a New Material of Waste Paper”, ASCE Civil Engineering, Vol. 76, No. 5, pp. 72 – 77.

2. Paki Turgut, Halili Murat Algin (2007) “Limestone dust and wood sawdust as brick material”, Journal of Building and Environment, Vol 42, pp 301 – 319.

3. Gordon Solberg (2002). “Building With Papercrete & Paper Adobe: A Revolutionary New Way to Build Your Own Home for Next to Nothing” Second Edition.

4. Topcu IB, Canbaz M (2004). “Properties of concrete containing waste glass”, Journal of Cement and Concrete Research, Vol 34, pp 267–274.

5. Shayan A (2004). “Value-added utilisation of waste glass in concrete”, Journal of Cement and Concrete Research, Vol 34, pp 81–89.

6. Shi C, Wu Y, Riefler C, Wang H (2005). “Characteristic and puzzolanic reactivity of glass powders”, Journal of Cement and Concrete Research, Vol 35, pp 987–993.

7. Corinaldesi V, Gnappi G, Moriconi G, Montenero A (2005). “Reuse of ground waste glass as aggregate for mortars”, Journal of Waste Management, Vol 25, pp 197–201.

8. Ronaldo S. Gallardo Mary Ann Q. Adajar (2006). “Structural performance of concrete with paper sludge as fine aggregate practical replacement enhanced with admixtures”-Symposium on Infrastructure Development and the Environment 2006 was held on 7th and 8th October 2006 at university of the Philippines.

9. Lex Terry (2006). “Papercrete Construction”-Building Environment and Thermal Envelope Council (BETEC) Symposium was held on 13th to 16th October 2006 at the Northen New mexico Community College in EIRito.

10. IS: 3495-1992 (Part 1 to 4), “Methods of Tests for Burnt Clay Bricks”, Bureau of Indian Standards, Third Revision, New Delhi.

SECURITY ARCHITECTURE USING KEY DISTRIBUTION ALGORITHM IN VANET

M. PRABAKARAN M.Tech. (Ph.D.),Asst. Professor , Dept. of CSE, Sambhram Institute of Technology, Bangalore 560 064

Dr. G. MAHADEVAN M.E. Ph.D.,Professor & HOD of CSE, AMC Engineering College, Bangalore 560 082.

ABSTRACT: The number of applications that involve wireless communications among mobile devices is rapidly growing. Many of these applications require wireless networks to be spontaneously formed by the participating mobile devices themselves i.e., mobile ad hoc networks (MANETs). The idea behind MANETs is to enable connectivity among any arbitrary group of mobile devices everywhere, at any time. Vehicular adhoc network (VANET) is one of the most relevant application domains of mobile ad hoc networks. The security issue of vehicular network is essential to make sure that life-critical information cannot be inserted or modified by an attacker. The system should be able to help establish the liability of drivers. At the same time, it should protect as far as possible the privacy of the drivers and passengers.Key Words: Vehicular adhoc network, security issue, parametric issues

INTRODUCTIONIn contrast to fixed networks, a centralized PKI or even a centralized certification authority is not feasible in vehicular ad hoc networks. Due to the absence of an on-line key distribution center in VANET’S, pre authentication in symmetric schemes requires either the proximity of the communicating devices, or some other kind of out-of-band channel.The thesis proposes a scheme of distributing the signing key and the functionality of a Certificate Authority over a number of different vehicular nodes by the means of secret sharing and threshold cryptography. The developed architecture addresses issues of authorization and access control, and a multi-level security model to adapt the complex capabilities of vehicular adhoc networks. The parametric issues considered in the vehicular network dynamics of the proposed scheme are semi-permanent mobility, high speeds, and very short connection times between neighbors.1. PROBLEM DEFINITION

The security issue of vehicular network is essential to make sure that life-critical information cannot be inserted or modified by an attacker. The system should be able to help establish the liability of drivers. At the same time, it should protect as far as possible the privacy of the drivers and passengers In existing fixed networks, a centralized PKI or even a centralized certification authority was used. Due to the absence of an on-line key distribution center in

VANET’S, pre authentication in symmetric schemes requires either the proximity of the communicating devices, or some other kind of out-of-band channel. The developed architecture addresses issues of authorization and access control, and a multi-level security model to adapt the complex capabilities of vehicular adhoc networks. The parametric issues considered in the vehicular network dynamics of the proposed scheme are semi-permanent mobility, high speeds, and very short connection times between neighbors.

2. METHODOLOGYKey Distribution is often a fundamental building block for security protocols. In vehicular networks, distribution poses several significant challenges. First, vehicles are manufactured by many different companies, so installing keys at the factory would require coordination and interoperability between manufacturers. If manufacturers are unable or unwilling to agree on standards for key distribution, then we could turn to government-based distribution. The government can impose standards, but doing so would require significant changes to the current infrastructure for vehicle registration, and thus is unlikely to occur in the near future. The use of wireless links makes MANETs susceptible to attack. Eavesdroppers can access secret information, violating network confidentiality. Hackers can directly attack the network to delete messages, inject erroneous

messages, or impersonate a node, which violates availability, integrity, authentication, and nonrepudiation. Compromised nodes also can launch attacks from within a network.

On-demand and link-state routing algorithms do not specify a scheme to protect data or sensitive routing information. Because any centralized entity could lead to significant vulnerability in MANETs, a security solution must be based on the principle of distributed trust. This is similar to the dilemma posed by the classic Byzantine generals problem, in which a general commands each division of the army, and some of the generals, who communicate via messenger, are traitors. All loyal generals must decide upon the same plan of action—that is, a small number of traitors cannot cause the loyal generals to adopt a bad plan. The same holds for MANETs: A number of compromised nodes cannot cause the network to fail. Although no single node in a MANET is trustworthy, threshold cryptography can distribute trust to an aggregation of nodes. This scheme lets n parties share the ability to perform a cryptographic operation such that any t parties can do it together, while up to t - 1 parties cannot perform the operation. However, dividing a private key into n shares and constructing t partial signatures is nontrivial given that traditional key distribution schemes either do not apply to the ad hoc scenario or are not efficient for resource-constrained devices.Combining identity-based techniques with threshold cryptography can achieve flexible and efficient key distribution. After distribution, a combiner can verify the t signatures and compute the final signature for the certificate. In this way, up to t - 1 compromised node cannot generate a valid certificate by them. If a large number of nodes are compromised, attributing fault to a specific malicious node is impossible.A proposed algorithm addresses this problem by limiting the possible fault location to the link between two adjacent nodes; as long as a fault-free path exists between two nodes, they can establish a secure communication link even if most nodes in the network are compromised. In addition, this algorithm can detect selfish nodes that refuse to cooperate with other nodes. If their behavior is the result of a denial-of-service attack rather than power-savings activity, the algorithm can isolate the selfish nodes. Ad-hoc network - and in particular wireless mobile ad-hoc network have unique characteristic and constraint that make traditional cryptographic mechanisms and assumptions inappropriate. In particular, it may not

be warranted to assume pre-existing shared secrets between member of the network and the presence of a common PKI. Thus, the issue of key distribution in ad-hoc networks represents an important problem. Unfortunately, this issue has been largely ignored; as an example, most protocols for secure ad-hoc routing assume that key distribution has already taken place. Traditional key distribution schemes either do not apply in an ad-hoc scenario or are not efficient enough for mall, resource-constrained devices. The proposed model combine efficient techniques from identity-based (ID-based) and threshold cryptography to provide a mechanism that enables flexible and efficient key distribution while respecting the constraints of ad-hoc network. The system discuss the available mechanisms and their suitability for the proposed task

3. SYSTEM MODEL

The system model classifies the applications into four groups covering the whole range of possible applications as depicted in Figure. These are

1) Car to Car Traffic Applications,2) Car to Infrastructure Applications,3) Car to Home Applications, and4) Routing based Applications.

The applications cover single hop and multi hop communication for the unicast case and therefore the relevant range of applications for the project. Car to car traffic applications include the transmission of traffic related information over multiple hops to a group of receivers, such as obstacles on the road, low friction, low visibility, etc.

Second, car to infrastructure applications represent the single hop broadcast case. An example application would be a low bridge warning application, where a bridge permanently broadcasts its height, or work zone warning, where cars are notified by the work-zone using radio beacons. Note that the name of these kind of applications may be misleading, as it is mostly the infrastructure that sends messages to cars3.

Third, car to home applications, which are seen among the prominent deployment applications for the system represent the single hop unicast case. For example a car to home media synchronization would be an application to be thought of.

FIG 1: A TYPICAL SYSTEM MODEL

Finally, as we assume that routing in VANETs will be an enabler for many applications, we include a rather general attack tree on routing based (multi hop unicast) applications. For the multi-hop applications, we assume both IP-based and position based routing mechanisms to be avail-able.4. VEHICULAR NETWORK CHALLENGESVehicular network challenges include technical problems like key distribution as well as more abstract difficulties, such as the need to appeal simultaneously to three very different markets. Authentication versus Privacy In a vehicular network, we would like to bind each driver to a single identity to prevent Sybil [7] or other spoofing attacks. For instance, in the congestion avoidance scheme, we would like to prevent one vehicle from claiming to be hundreds in order to create the illusion of a congested road. Strong authentication also provides valuable forensic evidence and allows us to use external mechanisms, such as traditional law enforcement, to deter or prevent attacks on vehicular networks. However, drivers value their privacy and are unlikely to adopt systems that require them to abandon their anonymity. For example, if we try to prevent spoofing in a manner that reveals each vehicle’s permanent identity, then we may violate drivers’ privacy expectations. Balancing privacy concerns with security needs will require codifying legal, societal and practical considerations. Most countries have widely divergent laws concerning their citizens’ right to privacy. Since most vehicle manufacturers operate in multinational markets, they will require security solutions that satisfy the most stringent privacy laws, or that can be customized to meet their legal obligations in each market. Authentication schemes must also weigh societal expectations of privacy

against practical considerations. Most drivers wouldresent a system that allows others to track their movements, but from a practical perspective, vehicles today are only partially anonymous. Each vehicle has a publicly displayed license plate that uniquely identifies it (and identifies the owner of the car, given access to the appropriate records). Thus, individual drivers have already surrendered a portion of their privacy while driving. Ideally, a secure vehicular network would build on these existing compromises instead of encroaching any further upon a driver’s right to privacy.Availability For many applications, vehicular networks will require real-time, or near real-time, responses as well as hard realtime guarantees. While some applications may tolerate some margin in their response times, they will all typically require faster responses than those expected in traditional sensor networks, or even ad hoc networks. However, attempts to meet real-time demands typically make applications vulnerable to Denial of Service (DoS) attacks. In the deceleration application, a delay of even seconds can render the message meaningless. The problem is further exacerbated by the unreliable communication layer, since one potential way to cope with unreliable transmission is to store partial messages in the hopes that a second transmission will complete the message. Current plans for vehicular networks rely on the emerging standard for dedicated short-range communications (DSRC) [2], based on an extension to the IEEE 802.11 technology [1]. Yin et al. provide a detailed, low-level evaluation of the performance of a simulated DSRC network and find that while the current DSRC standard provides an acceptable latency, the reliability is still lacking [22]. According to their simulations, on average, only 50-60% of a vehicle’s neighbors will receive a broadcast message. Since vehicles moving in opposite directions will remain within communications range for only a few seconds, opportunities to retry a

broadcast will be limited. On a positive note, DSRC features a high data rate.Low Tolerance for Errors Many applications use protocols that rely on probabilistic schemes to provide security. However, given the life-or-death nature of many proposed vehicular applications, even a small probability of error will be unacceptable. In fact, since the U.S. Bureau of Transportation Statistics estimates that there are over 200 million cars in the U.S. [21], even if only 5% of vehicles use an application that functions correctly 99.99999% of the time, the application is still more likely to fail on at least one vehicle than function correctly on all vehicles. Thus, to provide the level of guarantees necessary for these scenarios, applications will have to rely on deterministic schemes or probabilistic schemes with security parameters large enough to make the probability of failure infinitesimally small. Furthermore, for many applications, security must focus on prevention of attacks, rather than detection and recovery. In an ad hoc network, it may suffice to detect an attack and alert the user, leaving recovery and clean-up to the humans. However, in many safetyrelated vehicular network applications, detection will be insufficient, since by the time the driver can react, the warning may be too late. Instead, security must focus on preventing attacks in the first place, which will require extensive foresight into the types of attacks likely to occur (see Section 4).Mobility Traditional sensor networks frequently assume a relatively static network, and even ad hoc networks typically assume limited mobility, often focusing on handheld PDAs and laptops carried by users. For vehicular networks, mobility is the norm, and it will be measured in miles, not meters, per hour. Also, the mobility patterns of vehicles on the same road will exhibit strong correlations. Each vehicle will have a constantly shifting set of neighbors, many of whom it has never interacted with before and is unlikely to interact with again. The transitory nature of interactions in a vehicular network will restrict the utility of reputation-based schemes. For example, rating other vehicles based on the reliability of their congestion reports is unlikely to prove useful, a specific driver is unlikely to receive multiple reports from the same vehicle. Furthermore, since two vehicles may only be within communication range for a matter of seconds, we cannot rely on protocols that require significant interaction between the sender and receiver.Key Distribution Key distribution is often a fundamental building block for security protocols. In vehicular networks, distribution poses several significant challenges. First, vehicles are manufactured by many different companies, so installing keys at the factory would require

coordination and interoperability between manufacturers. If manufacturers are unable or unwilling to agree on standards for key distribution, then we could turn to government-based distribution. Unfortunately, in the U.S., most transportation regulation takes place at the state level, again complicating coordination. The federal government can impose standards, but doing so would require significant changes to the current infrastructure for vehicle registration, and thus is unlikely to occur in the near future. However, without a system for key distribution, applications like traffic congestion detection may be vulnerable to spoofing. A potential approach for secure key distribution would be to empower the Department of Motor Vehicles (DMV) to take the role of a Certificate Authority (CA) and to certify each vehicle’s public key. Unfortunately, this approach has many shortcomings. First, assuming the role of a CA is a challenging operation which is not in line with the DMV’s current functionality. Extensive anecdotal evidence [16] suggests that even specialized CAs offer questionable security against dedicated attackers trying to obtain a certificate for another institution/entity. Second, vehicles from different states or different countries may not be able to authenticate each other unless vehicles trust all CAs, which reduces security. Finally, certificatebased key establishment has the danger of violating driver privacy, as the vehicle’s identity is revealed during each key establishment. Incentives Successful deployment of vehicular networks will require incentives for vehicle manufacturers, consumers, and the government, and reconciling their often conflicting interests will prove challenging. For example, law-enforcement agencies would quickly embrace a system in which speed-limit signs broadcast the mandated speed and vehicles automatically reported any violations. Obviously, consumers would reject such intrusive monitoring, giving vehicle manufacturers little incentive to include such a feature. Conversely, consumers might appreciate an application that provides an early warning of a police speed trap. Manufacturers might be willing to meet this demand, but law-enforcement is likely to object.Bootstrap Initially, only a small percentage of vehicles will be equipped with DSRC radios and little infrastructure will exist to support them. Thus, in developing applications for vehicular networks, we can only assume that a few other vehicles are able to receive our communications, and the applications must provide benefits even under these limited conditions (with increasing benefits as the number of DSRC-equipped vehicles increases).

5. ADVERSARIES

The nature and the resources of the adversary will largely determine the scope of the defenses needed to secure a vehicular network. A realistic assessment of the vehicular environment suggests the following classes of adversaries (in increasing order of threat severity):Greedy Drivers While we might hope that most drivers in the system could be trusted to follow the protocols specified by the application, some drivers will attempt to maximize their gains, regardless of the cost to the system. In our congestion avoidance system, a greedy driver might try to convince the neighboring vehicles that there is considerable congestion ahead, so that they will choose alternate routes and allow the greedy driver a clear path to his/her destination.Snoops This category of adversary encompasses everyone from a nosy next-door neighbor to a government agency attempting to profile drivers. A burglar might try to use a vehicular network to detect which garages (and hence which houses) are empty. Companies may want to track consumers’ purchasing habits and use correlated data to alter prices and discounts. While data mining may be acceptable over aggregate data, it raises serious privacy concerns if one can extract identifying information for an individual.Pranksters Pranksters include bored teenagers probing for vulnerabilities and hackers seeking fame via their exploits. For example, a prankster targeting a collision-avoidance or platooning application might sit by the road and convince one vehicle to slow down while persuading the vehicle behind it to speed up. The need for real-time responses potentially leaves security mechanisms vulnerable to DoS attacks. A prankster could abuse this vulnerability to disable applications or prevent critical information from reaching another vehicle.Industrial Insiders Attacks by insiders are particularly insidious, and the extent to which vehicular networks are vulnerable will depend on other security design decisions. For example, if mechanics can update the software on a vehicle, they also have an opportunity to load malicious programs. If we allow vehicle manufacturers to distribute keys, then a single rogue employee at one manufacturer could create keys that would be accepted by all other vehicles.Malicious Attackers Malicious attackers deliberately attempt to cause harm via the applications available on the vehicular network. They may be individuals attempting to settle a score or

terrorists attacking our infrastructure. In many cases, these attackers will have specific targets, and they will have access to more resources than the attackers described above. Terrorists might manipulate the deceleration warning system to create gridlock before detonating a bomb. Criminals might spoof the congestion avoidance application to facilitate getaways. In general, while this class of attackers will hopefully be rarer than those outlined above, their combination of resources and directed malice makes them an important concern for any security system.

6. ATTACKS

While we obviously cannot anticipate every possible attack on vehicular networks, we can enumerate some of the more likely scenarios and ensure that applications are robust against this known set of potential attacks. While many of these attacks have appeared in other contexts, we list them here both for the sake of thoroughness and to examine the ramifications they may have in this new environment.Denial of Service (DoS) If the attacker can overwhelm a vehicle’s resources or jam the communication channel used by the vehicular network, then he can prevent critical information from arriving. Not only does this render the application useless, it could increase the danger to the driver if she has come to depend on the application’s information. For instance, if a malicious adversary wants to create a massive pileup on the highway, he could provoke an accident and then use a DoS attack to prevent the appropriate deceleration warnings from reaching other drivers. Message Suppression Attacks In a more subtle attack, the adversary may use one or more vehicles to launch a suppression attack by selectively dropping packets from the network. A prankster might suppress congestion alerts before selecting an alternate route, thus consigning subsequent vehicles to wait in traffic.Fabrication Attacks An adversary can initiate a fabrication attack by broadcasting false information into the network. For example, a greedy driver might pose as an emergency vehicle to speed up his own trip. An attacker may also choose to fabricate her own information, including her identity, location, or other applicationspecific parameters. Defending against fabrication attacks in a vehicular network is particularly challenging, since the traditional remedy of using strong identities along with cryptographic authentication may conflict with the need to preserve the privacy of participants in the network.Alteration Attacks A particularly insidious attack in a vehicular network is to alter existing data. This

includes deliberately delaying the transmission of information, replaying earlier transmissions or altering the individual entries within a transmission. For example, if the traffic congestion application requires a vehicle to collect “votes” from other vehicles at the site of the congestion, then an attacker might collect votes while traveling in normal traffic, but alter the locations and timestamps in the votes to make it appear that all of those vehicles were in the same place at the same time, deceitfully indicating a heavily congested highway. A malicious attacker might alter a message alerting vehicles to an obstacle ahead to persuade another vehicle that the lane is in fact clear. Clearly, applications on vehicular networks will need authentication of both the source of the data and the data itself.

7. PROPERTIES SUPPORTING SECURITY

While the previous discussion paints a grim picture of the security challenges facing vehicular networks, there are aspects of such networks that may aid the creation of secure applications. Regular Inspections In most U.S. states, all vehicles must pass inspection once a year. This yearly trip to the mechanic provides interesting possibilities for security maintenance in addition to the typical maintenance performed. For example, as part of the inspection, the mechanic might use SWATT [19] to verify the integrity of the software running on the vehicle’s processor. This would also be an opportune time to update/patch existing software, download new certificates, or receive the current list of revoked certificates. The inspection process affords the system an opportunity to return to a known, baseline state, and it serves as a firebreak against worm and virus infections. These inspections could also be included as part of the standard maintenance package anytime a driver brings the vehicle in for a tuneup or repair.Honest Majority Another advantage of vehicular networks is that the majority of drivers are likely to be honest. This will be reinforced by the fact that few people feel comfortable tinkering with their vehicles, so that most drivers will simply accept the default configurations. Assuming vehicles have adequate safeguards2 against worms and viruses, the trustworthiness of drivers will translate into vehicles that correctly follow established protocols. Applications can take advantage of this property via polling and aggregation, with the expectation that correct responses will outweigh incorrect or malicious ones. Additional Input Unlike traditional sensor networks, each node in a vehicular network

has a presumed intelligent operator available. Unlike ad hoc networks, the interaction with the operator must be minimal, since the application should not distract the driver. However, even minimal information from the driver may enable additional security techniques. For example, in the collision avoidance application, the vehicle can use the driver’s reaction to a warning as input to a reinforcement learning algorithm. If the driver ignores the warning and no crash results, the vehicle may decide that the warning was erroneous and adjust its parameters accordingly. If the driver does brake, it supports the vehicle’s decision to issue the warning. Information from the driver may also be supplemented with information from other sensors in the vehicle (e.g., proximity sensors).Central Registration Unlike ad hoc networks, all of the nodes in a vehicular network must register with a central authority, such as the state government (in the U.S.). These institutions have already developed extensive infrastructure for tracking and administering these registrations, and in the long term, we could potentially leverage this existing bureaucracy to strengthen the security of vehicular applications. However, given the vast nature of the existing infrastructure, any changes will take years of effort and require considerable funding. Thus, we cannot rely on this infrastructure to help bootstrap the system, but we can be prepared to take advantage of it when it comes online.Controlled Access Many portions of the transportation system already have access control mechanisms. For example, toll roads and many bridges have controlled entry and exit points, so it may be practical to equip these locations with the infrastructure necessary to distribute ephemeral identities and/or keys to allow vehicles to communicate while they remain on the controlled stretch of road. Since all vehicles must pass through the limited set of entrances, we can assume that any DSRC-equipped vehicle encountered in that segment of the highway will have received an appropriate key that will interoperate with everyone else’s key. We discuss other aspects of key distribution in Section 6.Existing Enforcement Mechanisms Finally, we have the presence of existing law enforcement to aid the security of vehicular networks. In many cases, an adversary will need to be within physical proximity of the victim to launch an attack. If the victim can unambiguously identify the attacker to local law enforcement, we can use existing legal remedies to provide disincentives for attacks. Also, since liability laws typically hold vehicle manufacturers to a far higher standard than software vendors, the manufacturers have a strong incentive to deploy secure systems that will resist attacks.

Figure 1: Secure Relative Localization Vehicle B can use broadcasts from vehicles C, D and E to determine A’s

location.

8. SECURITY PRIMITIVESTo secure vehicular networks, we propose the development of a set of security primitives, which can serve as the building blocks for secure applications. The ideal primitives should be specific to the vehicular setting, while remaining sufficiently general and composable to allow for reuse in a wide variety of protocols. Below, we describe some of the primitives that will be necessary for secure vehicular applications.Authenticated Localization of Message Origin Location will play a significant role in many vehicular applications, making it critical to determine that a message did indeed originate at a given location. For example, this primitive would prevent an attacker sitting on the side of the road from claiming to be a vehicle traveling on the highway. It would also prevent an adversary from using another communication medium (e.g., cellular) to replay a message heard in one location as though it had originated in a different location, thus preventing the types of attacks enabled by wormholes in sensor networks [9]. Such a primitive would assist in securing the congestion detection application, since the principal characteristic of congestion is the presence of multiple vehicles in a similar location. Anonymization Service An anonymization service would allow us to resolve some of the tension between authentication and privacy. It would rely on the observation that for almost all of the applications we envision, a vehicle does not need to authenticate the exact identity of the other vehicle sending the information, but only the connection between the information sent and a vehicle present on the road. Drivers could use their permanent identity to authenticate to an anonymization service. The service would then provide the driver with a temporary identification that cannot be traced back to the driver (although this could be modified to allow authorized entities to trace the connection between the temporary id and the original driver, either through sealed records or some form of escrow service). This primitive could help prevent spoofed identities, while still preserving drivers’ privacy expectations. As a sample implementation, consider a toll highway. Since all drivers must pass through the toll booth to enter the highway, they could also authenticate themselves to an anonymization service hosted in the booth. Thus, every time a vehicle passes a reanonymizer, it can acquire a new identity independent of its previous identity. The anonymization service would help secure the congestion detection application, since it would provide each vehicle with a single strong identity without compromising the vehicle’s anonymity. As an alternative to

a dynamic anonymization service, we could consider preloading each vehicle with a year’s worth of anonymous keys, as Raya and Hubaux propose [18]. The supply could be refreshed by the certifying authority. Unfortunately, this approach provides an attacker with a ready host of “legitimate” identities for use in a Sybil attack [7]. Tamper-resistant devices would make this attack more difficult but not impossible and would add to the expense of the vehicular network. Instead, we could consider a dynamic key distribution system, in which vehicles could create a new anonymous key pair every day.Additional Primitives Vehicular networks will also require primitives for performing key establishment and message authentication. In the context of sensor and ad hoc networks, key establishment continues to be an active research area. Unfortunately, these approaches do not readily translate to the vehicular setting, because the security requirements in vehicular networks are much more stringent than in either ad hoc or sensor networks and the trust assumptions are different. Message authentication is another important primitive for vehicular networks. Secure message authentication prevents an external attacker from injecting malicious messages into the network, and can prevent a relaying node from altering the message. In conjunction with message freshness and appropriate mechanisms, authentication prevents an attacker from replaying an old message.

9. RELATEDWORK Few researchers have examined the problem of security in vehicular networks. Zarki et al. [23] present the DAHNI (Driver Ad Hoc Networking Infrastructure) system for providing driver assistance. They show how they can use a vehicular network to track nearby vehicles and report potential hazards to the driver. In contrast to their work, we argue that privacy and key establishment are two vital issues that require additional work before vehicular networks can be securely deployed.

Hubaux et al. describe some of the attacks vehicular networks may face and propose a mechanism for providing secure positioning; they also suggest the congestion detection application discussed in this work [10]. In another work, Raya and Hubaux consider the issues involved with key management for vehicular networks, as well as the use of anonymous public keys. They also analyze the feasibility of using a PKI to support the security requirements of vehicular networks [18]

10. CONCLUSIONThe developed technique of key distribution for VANET security provides a useful foolproof system for the vehicle drivers driving in the highways. The top-down approach allows us to influence the system design at an early

development phase regarding security considerations, while on the other hand being able to generate a more detailed analysis as soon as the system’s specifications become more specific.Looking at the attacks, we found that two procedures would enhance overall security essentially, doing local plausibility checks in cars and regular system checks on the nodes. Plausibility checks could include comparison of received information to internal sensor data, evaluating messages from different information sources about a single event and scenario building, where single traffic events are related using statistics. Simulations have shown that this greatly increases the effort of an attacker, but it requires proper models for every application. Regular system checks would verify the proper function of a unit and therefore reduce the number of malfunctioning units. This could also include the option to update the software.

11. REFERENCES

[1] IEEE 802.11 WG, part 11: Wireless LAN Medium Access Control (MAC) and physical layer (PHY) specifications, August 1999.

[2] Standard specification for telecommunications and information exchange between roadside and vehicle systems - 5 GHz band dedicated short range communications (DSRC) Medium Access Control (MAC) and physical layer (PHY) specifications, September 2003.

[3] Car2Car Communication Consortium. http://www.car-2-car.org/, 2004.

[4] Jan Camenisch and Anna Lysyanskaya. Efficient non-transferable anonymous multi-show credential system with optional anonymity revocation. In Proc. of Advances in Cryptology - Eurocrypt, 2001.

[5] Sr ¯ dan ¡Capkun and J.P. Hubaux. Secure positioning of wireless devices with application to sensor networks. In Proc. of INFOCOM, March 2005.[6] John Chovan, Louis Tijerina, Graham Alexander, and Donald Hendricks. Examination of lane change crashes and potential IVHS countermeasures. http://www.itsdocs.fhwa.dot.gov/JPODOCS/ REPTS_TE/61B01!.PDF, March 1994.[7] John R. Douceur. The Sybil attack. In First International Workshop on Peer-to-Peer Systems (IPTPS), March 2002.[8] FCC. FCC allocates spectrum in 5.9 GHz range for intelligent transportation systems uses. http://www.fcc.gov/Bureaus/Engineering_Technology/ News_Releases/1999/nret9006.html, October 1999.

[9] Yih-Chun Hu, Adrian Perrig, and David B. Johnson. Packet leashes: A defense against wormhole attacks in wireless networks. In Proc. of IEEE INFOCOM, April 2003.[10] Jean-Pierre Hubaux, Sr dan ŽCapkun, and Jun Luo. The security and privacy of smart vehicles. IEEE Security & Privacy Magazine, 2(3):49–55, May–June 2004.[11] Markus Kuhn. Probabilistic counting of large digital signature collections. In Proc. of USENIX Security Symposium, 2000.[12] L. Lazos and R. Poovendran. SeRLoc: Secure range-independent localization for wireless sensor networks. In Proc. of ACM Wireless Security Workshop (WiSe), October 2004.[13] L. Lazos, R. Poovendran, and Sr ¯ dan ¡Capkun. ROPE: Robust position estimation in wireless sensor networks. In Proc. Of IPSN, April 2005.[14] Zang Li, Wade Trappe, Yanyong Zhang, and Badri Nath. Robust statistical methods for securing wireless localization in sensor networks. In Proc. of IPSN, April 2005.

[15] Donggang Liu, Peng Ning, and Wenliang Kevin Du. Attack-resistant location estimation in sensor networks. In Proc. of IPSN, April 2005.[16] PKI Forum. Verisign fraudulent certificates. http:// www.pkiforum.com/ resources/verisigncerts. html, Accessed on January 2005.[18] Maxim Raya and Jean-Pierre Hubaux. The security of vehicular ad hoc networks. In Proc. of ACM Workshop onSecurity of Ad Hoc and Sensor Networks (SASN), November 2005.[19] Arvind Seshadri, Adrian Perrig, Leendert van Doorn, and Pradeep Khosla. SWATT: Software-based attestation for embedded devices. In Proc. of IEEE Symposium on Security and Privacy, May 2004.[20] Trusted Computing Group. Trusted platform module main specification, Part 1:design principles, Part 2: TPM structures, Part 3: Commands. http://www.trustedcomputinggroup.org, October 2003. Version 1.2, Revision 62.[21] U.S. Department of Transportation, Bureau of Transportation Statistics. Transportation Statistics Annual Report, 2003.[22] Jijun Yin, Tamer ElBatt, Gavin Yeung, Bo Ryu, Stephen Habermas, Hariharan Krishnan, and Timothy Talty. Performance evaluation of safety applications over DSRC vehicular ad hoc networks. In Proc. of ACM workshop on Vehicular Ad Hoc Networks (VANET), 2004.[23] Magda El Zarki, Sharad Mehrotra, Gene Tsudik, and Nalini Venkatasubramanian. Security issues in a future vehicular network. In European Wireless, February 2002.

DESIGN OF QUALITY INFORMATION SYSTEM BASED ON ISO 9001:2000 STANDARD

M. SAKTHIVEL# S.R. DEVADASAN$ M. SAKTHIVELAssistant Professor Department of Mechanical Engineering

Anna University Coimbatore Jothipuram Campus Coimbatore TamilnaduPIN 641 047 Email : [email protected] Mobile :+91 9942912787

S.R.DEVADASANProfessor Department of Production Engineering PSG College of Technology

Email: [email protected] INDIA PIN 641 004Email: [email protected]

Mobile :+91 9842513209ABSTRACT:Right from the time, when the world began to experience the invasion of Japanese products and tough global competition, efforts were being exerted by various professionals, to improve the performance quality of organizations. Organizations are taking keen interest to quality due to global nature of competition Hence, companies have been implementing an array of quality programmes such as Total Quality Management (TQM) principles and quality system models to attain continuous quality improvement. With the advent of Total Quality Management (TQM) philosophy, both theoreticians and practitioners have been working on evolving and adopting plenty of techniques and tools. However, there has been little effort in designing and modeling quality information systems. In fact, even today, In spite of evolution of many smart models for implementing Enterprise resource Planning strategies. no such distinct model is available to incorporate Quality Information system (QIS) in companies. On realizing this theoretical and practical lacuna, a module of the research project reported in this paper has been initiated. The work has begun by designing and modeling a QIS. In this regard, it is observed that ISO 9000 standards have largely penetrated into companies situated in various parts of the World. Since the latest version ISO 9001 standard has been refined to evolve continuous quality improvement on focused manner, companies have been on track to implement this standard. Hence it was decided to model QIS by referring to ISO 9001:2000 standard

Keywords:

Quality Information System, Total Quality Management, ISO 9000.

INTRODUCTION One of the hottest topics in the world today is quality. Books, journals and seminars are repeatedly addressing it. ISO 8402 defines quality as “ the totality of features and characteristics of a product or service that bears on its ability to meet a stated or implied need”. Organizations are taking keen interest to quality due to global nature of competition. Hence, companies have been implementing an array of quality programmes such as Total Quality Management (TQM) principles and quality system models to attain continuous quality improvement and to face the global competition. Tobin defines TQM as the totality integrated effort for gaining competitive advantage by continuously improving every facet of organizational culture. TQM helps to meet customer requirements and expectations by the mutual cooperation of all the people in an organization.

The basic principle of TQM can be classified into Leadership, commitment, Total customer satisfaction, Continuous improvement, Total involvement,

Training and education, Ownership, Reward and recognition, Error prevention and Co-operation and teamwork. One of the imperatives of TQM is the integration of all resources and there by directing them towards effecting continuous quality improvement. One major way of integrating resources is by installing and managing of information system. Such information system when used for effecting continuous quality improvement is called as Quality information system (QIS). In spite of its importance in TQM projects, it is very rare to find either an engineer or a management professional contributing towards the design and development of QIS. In order to overcome this lacuna, a research project on designing QIS becomes the need of the hour. QIS can be developed from different perspectives in TQM projects. For example QIS can be developed for successfully implementing TQM


technique such as Quality Circle Programme, Quality Function Deployment, Failure Mode and Effect Analysis etc.

However, on considering the TQM scenario it should be noted that, if QIS is developed from the perspective of quality system, then it would offer highest value based continuous quality improvement oriented efforts. It should be noted that organizations throughout the world have been showing keen interest for implementing ISO 9000 series quality system models. These models have helped the organizations to systematize TQM efforts. Hence QIS developed from quality system point of view will offer supporting role in focusing an ISO 9000 based quality system elements towards achieving continuous quality improvement. On realizing this imperative, the design of QIS by referring to the recently released ISO 9001:2000 quality system standard models was carried out. The clauses of this standard were referred and major elements of QIS namely ‘Data on quality’, ‘Data processing for quality improvement’ and ‘ Information on quality’ were designed.

MIS and QIS

Management professionals have been discussing about managing information systems for more than six decades. All the management programmes like, M.B.A. have been incorporated with the course titled “Management Information Systems”. Through such courses various aspects of information systems are being addressed by the educators of management programmes. Besides, a large number of consultants have been involved in installing information systems in office environments of companies. Though quality improvement has been taking place throughout the world in parallel to the development of information systems, there has been no such concrete effort by management professionals towards integrating continuous quality improvement projects in information systems. In fact, no major discussions have taken place in managerial conferences and seminars about extending support to enhance the effectiveness of continuous quality improvement through the application of management information systems concepts. At this juncture, it should be noted that today’s companies have been investing heavily on TQM projects. Presumably, due to lack of proper guidance, not many companies have invested on developing information systems for enhancing the efficiency of TQM projects. Presumably, due to lack of proper guidance, not many companies have invested on developing information systems for enhancing the efficiency of TQM projects. Some experts and researchers in TQM field have advocated the need of developing information systems to support continuous quality improvement projects. The most noticeable contribution is by Juran who coined the term “Quality Information Systems” (QIS). Presumably, after his avocation over the use of QIS, some researchers worked in the direction of developing QIS during 1980s. After that the importance of QIS was not much felt by both theoreticians and practitioners. This is probably due to reason that from 1980s, the focus of companies was turned towards implementing ISO 9000 series standards based quality systems. Hence, today a situation has arisen which demands the bridging of both MIS and QIS principles to evolve high degree of improved results. In this regard, the contribution of ISO 9000 series quality systems standards based models in organizations should be recognized. The installation of those standards in organizations located in various parts of the world has created quality system consciousness among both managements and employees. Hence, it will be an effective work if QIS is developed by integrating with a suitable ISO 9000 series quality system based model. Of late, companies have been moving towards installing ISO 9001:2000 standard based quality systems. Hence, a QIS compatible in a company in

which ISO 9001:2000 standard has been implemented, will offer very powerful solutions towards achieving continuous quality improvement. The fundamental tenets of MIS suggest the processing of data to evolve useful information to target users. The research in QIS has addressed the development features of QIS in a different way. The difference in approaches will make it difficult or impossible to integrate MIS principles with TQM projects. Hence, QIS should be developed in accordance to the stipulations of MIS principles. Meanwhile, it is prudent to note that TQM field has grown to a very large extent in such way that a number of models techniques, tools and approaches have been developed. The volume of QIS will become large. If the scope of QIS includes all the above components of TQM, then managing it will become a different task. Hence QIS should be coupled with only the vital elements of TQM. However, such elements of QIS should not discord the other elements. Particularly such elements should facilitate in integrating the resources of the organisations for the purpose of attaining continuous quality improvement.

Quality Systems Throughout the world, organizations have been showing keen interest on installing quality systems. This phenomenon was not much prevalent before 1980s. The quality systems movement started only after ISO released the standards for quality systems. After that many organisations in developed countries implemented ISO 9000 series standards based quality systems and reported both tangible and intangible benefits.

Thereafter those companies began to insist that the companies situated in developing countries should also install ISO 9000 series standards based quality systems for receiving the supplies from them. Thus the quality system movement spread across the world during 1980s. However, during 1990s companies began to analyze the real benefits achieved after implementing ISO 9000 series standards based quality systems. At this juncture, ISO started the work of revising the ISO 9000 series standards based quality systems. Finally ISO 9000 released its revised version under the titles ISO 9000:2000, ISO 9001:2000 and ISO 9004:2000. Those standards facilitate in providing appropriate focus for achieving continuous quality improvement in companies. Moreover, only one implementation model called ISO 9001:2000 has to be referred. This has avoided the confusion over selecting the appropriate standard while implementing quality systems by different

categories of companies. Because of these merits, companies have been quickly attracted to implementing ISO 9001:2000 based quality systems. A comparative study on ISO 9001:1994 and ISO 9001:2000 has revealed that the latter is incorporated with more MIS elements than the other. For example, the clause 4.1 d of ISO 9001:2000 states the requirements of ensuring the availability of information on processes. The other clauses of ISO 9001:2000 which are incorporated with MIS elements namely data, communication and information are, 5.1a, 7.2.3, 7.3.2, 7.33b, 7.4.2, 7.5.1 and 8.4. This observation leads to an indication that ISO 9001:2000 standard itself is incorporated with the importance MIS elements, which are found to be insufficient. Hence, the inducement of ISO 9001:2000 based QIS would be relatively easier task with highly powerful output towards enhancing continuous quality improvement.

Design of QIS

It is imperative that ISO 9001:2000 based QIS should offer maximum gains to both management and employees in visible forms. For this purpose, it is prudent to refer to the traditional MIS principles. This is due to the reason that traditional MIS principles have been found to be powerful in managing information systems in organisations of all types. The fundamental tenet of MIS is that, the collected data shall be processed using appropriate models to provide information to the target users. This methodology is depicted in figure 1.

Data ®Processing ®Information Fig. 1 Traditional MIS methodology

In line with the traditional MIS methodologies, while designing ISO 9001:2000 based QIS, the data and processing models have to be identified. Further the target users and their required information should also be determined. These two major steps form the core principles behind in designing ISO 9001:2000 based QIS. So for, these principles have been used to design QIS pertaining to clause 4.1 of ISO 9001:2000.

The contents of clause 4.1 of ISO 9001:2000 are given below

The organization shall establish, document, implement and maintain a quality management system and continually improve its effectiveness in accordance with the requirements of this International standard.

The organization shall

a. Identify the process needed for the quality management system and their application throughout the organization

b. Determine the sequence and interaction of these processes

c. Determine the criteria and methods needed to ensure that both the operation and control of these processes are effective

d. Ensure the availability of resources and information necessary to support the operation and monitoring the of these processes

c. Monitor, measure and analyze these processes, and

f. Implement actions necessary to achieve planned results and continual improvement of these processes.

1. Data:

The data pertaining to this stipulation of clause 4.1 are given below: 1 . ISO 9001:2000 standard.

2. Quality management system manual of the company.

3. List of processes needed for the quality management system. 4. List of sequence and interaction required for these processes. 5. List of criteria and methods required for ensuring the effective operation and control of these processes. 6. List of planned results required for the quality management systems.

7. Data on continual improvement measures.

8. List of methods to measure, monitor and analyze these processes and required actions for implementation. ii. Database-

The database requirements pertaining to the “data” given in the previous clause are given below:

1. Posting of ISO 9001:2000 standard in an electronic environment with the provision for retrieving the data based upon the clause number,

product, processes and procedures etc. The hard copy of this standard may also be given to all employees or shall be deposited with section heads.

2. Preparation of quality manual with proper pagination and certification etc., (for the purpose of quick retrieval the required data). Besides, the quality manual shall be posted in an electronic environment with the provision to retrieve the data based upon the requirements of the departments, individual employees, management personnel etc. 3. Preparation of a handbook containing the processes needed for the quality management systems with proper chronological ordering and indexing procedures.

4. Preparation of a handbook containing the sequence and interaction needed for the quality management systems with proper chronological ordering and indexing procedures.

5. Preparation of a handbook containing the criteria and methods needed for the quality management systems with proper chronological ordering and indexing procedures.

6. Preparing a pamphlet on planned results and posting on it in strategic locations of the company.

7. Preparing a pamphlet on data on continual improvement measures and posting it strategic location of the company.

8. Preparation of a handbook containing the methods to measure, monitor and analyse these processes with proper chronological ordering and indexing procedures.

Electronic devices shall be chosen to accommodate the above database elements, which should enable easy retrieval of the required data within few seconds.

iii. Processing:

The methods and models recommended for processing the data are presented in this section:

1. An index of the keywords may be prepared with page number of 150 9001:2000 standard should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the

standard. Display boards may he designed to paint the required contents of ISO 9001 :2000 standard.

2. An index of the keywords may be prepared with page number of quality manual, which should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the quality manual.

3. An index of the keywords may be prepared with page numbers of hand book containing processes needed which should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the handbook.

4. An index of the keywords may he prepared with page numbers of handbook containing sequences and interaction, which should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the handbook.

5. An index of the keywords may be prepared with page numbers of handbook containing criteria and methods, which should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the handbook.

6. An impressive font size and color of the letters should be chosen for displaying planned results which should enable the user to obtain information.

7. An impressive font size and color of the letters should be chosen for displaying data on continual improvement measures which should enable the user to obtain right information.

8. An index of the keywords may be prepared with page numbers of handbook containing measure, monitor and analyze which should enable the user to obtain the right information. Besides, a list of FAQs may be prepared to guide the users to get appropriate information from the handbook.

IV. INFORMATION:

The information elements designed pertaining to clause 4. Id of ISO 9001:2000 standard are presented here:

1. Circulation of the stipulation ISO 9001:2000 standard to the target users using circulars and display of the same through electronic media.

2. Circulation of the stipulation of quality manual to the target users using circulars and display of the same through electronic media. 3. Circulation of the stipulation of processes needed for quality management systems to the target users using circulars and display of the same through electronic media. 4. Circulation of the stipulation of sequence and interaction to the target users using circulars and display of the same through electronic media. 5. Circulation of the stipulation of criteria and methods to the target users using circulars and display of the same through electronic media. 6. Pasting of the pamphlets, planned results, and continuous improvement measures in the locations where the target users are working.

7. Circulation of stipulation concerning methods to measure, monitor and analyze to the target users using circulars and display of the same through electronic media.

The same procedure followed to design QIS pertaining to the clause 4. Id will be followed to complete QIS. After this work, the design of ISO 9001:2000 based QIS will be evolved as a guidebook. This guidebook will be shown to academicians and practitioners who are associated with MIS and QIS activities. Based upon their reactions, suitable amendments and refinements will be made to the contents of the guidebook. After that, companies will be approached to develop QIS manual by referring the guidebook. If those companies agree to implement the ISO 9001:2000 based QIS, then the same has to be accomplished by following the developed QIS manual.

CONCLUSION

Due to the IT explosion, it has now become a reality to network all the facilities in a company, which may have its work centers at different parts of the world. This situation will enable the information systems to perform at a never before fast rate with accurate results. Meanwhile, there has been an ever increasing demand for higher and higher degrees of quality in the performance of organizations. In this situation, the exploitation of QIS will prove to be a promising solution to face the tough global competitive situation. As ISO 9000 series based quality systems have penetrated into large number of companies throughout the world, it is preferable to develop the foundation of QIS based on these quality system standards. In order to full fill these requirements, in this paper a preliminary work on designing, ISO 9001:2000 based QIS has been presented. While further work has to be come out to complete this design it is discernable to realize the importance of QIS in the current scenario has to be disseminated among the academicians, researchers and practitioners.

It is enlarge to note that companies have been supporting MIS projects through IT networking. A wider dissemination of importance of ISO 9001:2000 based QIS will motivate the companies to invest in this parts, such as change in organisations scenario will lead to the improved performance of ISO 9001:2000 based quality systems. In fact approaching the companies to support the projects reported in this paper will be a major further work in this direction

REFERENCES

1. Samuel K.M. Ho., “ Is the ISO 9000 Series for Total Quality Management”, International Journal of Quality and Reliability Management, Vol. 1 1 No. 9, 1994, pp. 74-89.

2. Davis G.B., Olson M.H., “Management Information System - Conceptual Foundation, Structure and Development”, MeGraw Hill Book Company,NY, 1985.

3. Juran J.M., Gryna F.M., “Quality Planning and Analysis”, Tata Megraw Hill, India, 1997.

4. Kathawala Y.A., “A Comparative analysis of selected approach to Quality”, International Journal of Quality and Reliability Management, Vol. 6 No. 5, 1989, pp. 7-17.

5. Tumnial V.M.R., Teng C.L., “Strategic Quality Management, Malom Baldriges and European quality

awards and ISO 9000 certification - core concepts and comparative analysis”, International Journal of Quality and Reliability Management, Vol. 13 No. 4, 1994, pp. 8-38.

6. PeterNeergaard, “Quality Management: a survey on accomplished results”, International Journal of Quality and Reliability Management, Vol. 16 No. 3, 1999, pp. 277-288.

AN OPTIMAL AND IMPATIENT BACKOFF ROUTING ALGORITHM FOR AD-HOC NETWORKS

DR. JANGA REDDYCMR Institute of Technology, Hyderabad

[email protected]

ABSTRACT:

Many distributed multiple access (MAC) protocols use an exponential backoff mechanism. In that mechanism, a node picks a random backoff time uniformly in an intervals that doubles in size after a collision. When used in an Ad-Hoc network, this backoff mechanism is unfair towards nodes in the middle of the network. Indeed, such nodes tend to experience more collisions than nodes with fewer neighbors; consequently, they often choose larger delays that those other nodes. We propose a different backoff mechanism that achieves a fairer allocation of the available bandwidth by decreasing the backoff delay upon collision or failure to send a packet. That is, a node becomes more aggressive after each failure. Accordingly, we call the mechanism the Impatient Backoff Algorithm (IBA). The nodes maintain the stability of the algorithm by resetting, in a distributed way, the average backoff delays when they become too small. We perform a Markov analysis of the system to prove stability and fairness in simple topologies. We also use simulations to study the performance of IBA in random Ad-Hoc networks and compare with an exponential backoff scheme. Results show that IBA achieves comparable mean throughput, while delivering significantly better fairness.

Keywords: Fairness, MAC, Ad-Hoc Networks. I. INTRODUCTION

It has been observed that the widely used exponential backoff mechanism (e.g., IEEE 802.11) is unfair towards nodes in the middle of an Ad-Hoc network with multiple interference domains (see [1] and [2]). This unfairness results from the higher degree of contention that these nodes face compared to nodes at the outer edges. We illustrate that unfairness and we propose a new backoff scheme to reduce it.

1.1.1 Unfairness of Exponential Backoff

To demonstrate the unfairness of the exponential backoff, we consider the network with three links as shown in Fig. 1.1. Transmissions A1-A2 and B1-B2 both interfere with X1-X2, while A1-A2 and B1-B2 do not interfere with each other. Link X therefore faces more contentions and is more likely to experience collisions than its neighbors.

Fig. 1.1 Simple topology demonstrating the unfairness of 802.11

We perform a simple experiment with three pairs of laptops located in three different rooms, as shown in Fig. 1.1, to implement this topology. The laptops use 802.11b with link rates of 11 Mbps. The table summarizes the achieved rates on each link. Note that MAC overhead limits the maximum possible capacity of a link to 6Mbps. We observe that links A and B can simultaneously achieve the maximum 6Mbps rate. When links A and X are on at once, they share the channel equally, each receiving approximately 3Mbps. On the other hand, if A, B and X are all on simultaneously, link X’s throughput drops significantly. This simple experiment indicates the unfairness of 802.11b towards nodes in the middle of the network. It is easy to see the cause of this unfairness. 802.11b follows a backoff mechanism whereby nodes try to capture the channel after waiting for a random backoff selected uniformly in an interval. Upon a collision, nodes are required to double the interval and try again (i.e. try less aggressively). Consider node X above. It faces more contention than nodes A and B, and thereby has a lesser chance of success. Consequently, it collides more often than nodes A and B, and backs off more. As a result, node X succeeds much less often – as demonstrated by the results in Fig. 1.1. More generally, this phenomenon biases the network against nodes in the middle of an Ad-Hoc network. This effect is particularly undesirable because multi-hop routes tend to use the middle of the network more often than the outer edges. The motivation for this approach is stability: by backing off exponentially fast, the rate of transmission attempts decreases quickly even if more nodes become active. Consequently, the likelihood that one node succeeds quickly approaches one. Of course, the backoff delay increases somewhat but remains generally small compared to the packet transmission times. When all the nodes share a collision domain, the fairness issue does not arise. Thus, such a backoff scheme is suitable for the shared collision domain of the Aloha network, the original Ethernet, and typical 802.11 configurations. As we just discussed, the situation is quite different in Ad-Hoc networks.1.1.2 Impatient Backoff Algorithm We propose a novel backoff mechanism, the Impatient Backoff Algorithm (IBA), that attempts to improve the fairness in a distributed MAC algorithm running across multiple interference domains. When using IBA, nodes decrease their average backoff delay upon collision – thereby becoming more aggressive in attempting to capture the next slot. Also, nodes increase their average backoff delay upon successful transmission. The danger of the scheme becoming unstable because of frequent

collisions is handled by resetting the average backoff delays when they get too small. We use a Markov chain model to show that IBA achieves fairness in simple topologies. We demonstrate the stability of the backoff scheme under reasonable assumptions by proving the positive recurrence of the Markov chain.We also evaluate the throughput tradeoff required in order to achieve fairness. Our analysis does not take into account collisions that happen due to propagation delays and imperfect knowledge of interference. We have built a simulation model that captures these effects to study the performance of IBA in an arbitrary topology and compare it against traditional 802.11-like exponential backoff mechanisms. Simulation results demonstrate that IBA is able to maintain a level of mean throughput comparable to exponential backoff in a random network – but achieves significantly better fairness. We use the simulation model to study the effect of a realistic reset mechanism which propagates imperfectly the reset control message hop by hop. The paper is organized as follows. We begin by describing related work in Section 1.2. In Section 1.3, we present the backoff model for IBA, which we evaluate analytically in Section 1.4. IBA design parameters and their effects are discussed in Section 1.5, while Section 1.6 presents simulation results before we end with sections on conclusions and future work.

1.2 Related Work

There is a wide body of literature dealing with 802.11 protocols (e.g. [3]), standards being available at [4]. Our focus is on the backoff mechanism utilized to handle congestion in these networks – and their resulting throughput and fairness. Bianchi [5] presents a two dimensional Markov model of the exponential backoff mechanism in 802.11. By assuming that the probability of collision of a node does not depend on its own state history, the author is able to derive expressions for the packet transmission probability and saturation throughput. Ergen and Varaiya extend that work in [6]. Their model incorporates carrier sense, non-saturated traffic and SNR, for both basic and RTS/CTS access mechanisms. Analysis of the model shows that the throughput first increases, and then decreases with the number of active stations. A valuable aspect of this work is the modeling of variable packet lengths and the un-slotted nature of the protocol (i.e. no synchronization between nodes). Varaiya et. al. have also looked at related problems in [7], [8] and [9]. Slotted media access protocols have been studied for several decades (see [10]for an overview).

In particular, a detailed analysis of Slotted Aloha MAC has evaluated its throughput and fairness. More recently, Yuan and Marbach [11] have proposed a rate control for random access networks. By controlling the rate at which the nodes attempt to transmit, by increasing it when idle and decreasing it after collision, the system is shown to be stable, i.e. positive recurrent. Our model for IBA differs from [5], [6] and also [11] in a crucial aspect – we consider Ad-Hoc networks spanning multiple interference domains. Consequently, the nodes do not all share the same medium, and so have different degrees of contention, and therefore different collision probabilities. It is this fact that biases such protocols against middle nodes in the network. Addressing this fairness issue is the primary goal of this paper, and leads us to propose the strategy of becoming more aggressive upon collision.1.3 Backoff Model

1.3.1 AssumptionsWe make simplifying assumptions on the MAC model. The main assumption is that packet transmissions occur in a slotted and synchronized fashion. Each packet time slot is divided into two phases: 1. Backoff Contention Phase. In this phase, each node that has a packet to send generates a random backoff value. It waits for these many backoff mini-slots. The backoff mini-slots are much much smaller than the packet transmission slot. If it has not heard a transmission from a neighboring node while waiting, the node sends out a short ‘Slot Capture Message’. All neighbors which hear this slot capture message (i.e., within transmission range), or carrier senses noise in the channel (i.e. within interference range) will keep quiet for this slot.2. Packet Transmission Phase. At the end of the Backoff Contention Phase, all nodes which successfully sent out the slot capture messages will transmit a constant sized packet. Thus, only the nodes that have generated a backoff delay smaller than or equal to those of their neighbors get to transmit a packet. A successful transmission is confirmed by an acknowledgement (ack).An example is shown in Fig. 1.2, where five nodes in a line contend for the channel. Each node is assumed to interfere with neighbors two hops away. The lower part of the picture shows one slot. During the Backoff Contention Phase, node A chooses a smaller backoff than B and C – consequently it is able to send a Slot Capture Message that is heard by B and C, which keep quiet. Nodes D and E are not affected by A’s slot capture, and E wins that contention. As a result, nodes A and E utilize the Packet Transmission phase in parallel.

Although C had a smaller backoff than E, C is quiet in this slot, allowing E to transmit. Note that nodes wait till the end of the entire Backoff Contention phase, before beginning the Packet Transmission phase. Collisions can occur in this scheme if two neighbors choose the same backoff. In this case, neither will hear the other’s slot capture message. Hence both nodes will try to send packets, and collide, resulting in a wasted packet transmission slot. A collision will also occur if a node within interference range is unable to discern the slot capture. Note that this scheme does not employ an RTS/CTS mechanism, hence hidden terminals [12] will not be accounted for. For instance, the transmissions of B and E do collide at C.

Finally, slotted transmission implicitly assumes synchronization between the nodes.We do not address the details in this paper, other than to indicate that the level of synchronization required is no more than any standard slotted MAC protocol.

1.3.2 Exponential Random Backoff Rate

Traditional backoff mechanisms choose a random backoff uniformly in [0;BL], where BL is the backoff limit. The mean backoff in this case is BL=2. Instead of using an uniform random variable, IBA chooses the backoff using an exponential random variable with mean BL=2. The exact number of backoff mini-slots is the determined by rounding the random variable and capping it at a maximum value, since the exponential random variable is unbounded. We utilize the exponential random variable since it offers some useful characteristics that we use in the analysis of the scheme in Section 1.4. In particular, when n nodes with mean backoffs b1; b2; : : : bn contend, node i wins the contention when backoff Bi < Bj j # i. The probability of this occurrence is calculated as follows below in 1.1 In this calculation, we ignore the probability that the smallest backoff delay falls in the same mini-slot as another one.The key principle of IBA is that nodes that face more contention become more aggressive so that they can get their fair share of the channel.

This is achieved by updating the backoff based on feedback received in the last slot. Assume that a node has a mean backoff delay b and has a packet to transmit. If the node fails to send in the current slot, either because of a collision or because it loses during the contention phase, it decreases its mean backoff delay by a multiplicative factor m > 1. On the other hand, if the node transmits successfully, it increases its mean backoff delay by the same factor. Note that decreasing the mean backoff delay makes a node more likely to win during the next contention phase.

1.3.4 Reset

The obvious problem of an aggressive backoff mechanism is one of collision meltdown. When there are a lot of contending nodes, all but one decrease their mean backoff delay. This increases the chance that several of them pick a small backoff delay in the same mini-slot and hence collide. In order to avoid this situation, we propose the IBA Reset mechanism. The idea is to increase the mean backoff delay of every node by a constant factor RF whenever any node’s mean backoff delay falls below a reset limit RL. A reset does not alter the result of the contention phase, since Eqn. 1.1 is unchanged by multiplying every term by RF . In Section 1.5, we discuss the choice of values for RL and RF . In reality, it is impossible to change the mean backoff delay of all nodes simultaneously; any reset message needs to propagate through the network. In Section 1.6.3, we simulate a reset propagation scheme, as also the loss of some reset messages, and see that their effect on overall throughput and fairness is minimal .

1.3.5 Fairness

Qualitatively, we understand fairness to imply that two nodes in similar situations ought to get the same share of the bandwidth. For simple networks, it is easy to see the “fair” allocation, e.g. in Fig. 1.1, we would like each of the threetransmitting nodes to get 1=3 of the channel bandwidth.

In more complicated and/or random topologies, evaluating fairness is a trickier proposition. There are several metrics of fairness in use e.g. Max-Min Fair [10], Proportional Fair [13], Jain’s Fairness Index [14] etc. In this paper, we utilize Jain’s index to measure scheduling fairness, and to compare against other MAC schemes. We prefer Jain’s index [14] since it measures the fairness in terms of a optimal desired throughput. Let the measured throughputbe Z1; : : : ;Zn and the desired throughput be D1; : : : ;Dn.

We first evaluate IBA in some simple topologies that are amenable to analysis. We especially consider the star topology shown in Fig. 1.3(a). We also analyze the triangle clique topology in which three nodes all interfere with each other.

1.4.1 Star Topology

The star topology is characterized by a middle node X that interferes with every other node, while the outer nodes do not interfere with each other. The figure shows a star with 4 arms, although we analyze a star with n outer nodes. Recall that we are modeling a non-RTS/CTS scheme. So we are concerned only with the interference between nodes as they transmit and need not consider the location of the receivers. The star topology is of special interest in terms of fairness, since node X contends with every other node in the network and traditional backoff mechanisms are unfair towards it. We want to ensure that our scheme is fair for this simple topology. When node X wins the backoff contention, it is the only one that transmits. However, when any other node wins the contention, node X has to keep quiet and so the remaining outer nodes believe they won the round and they all transmit. Thus, assuming that the nodes always have a packet to send, the mean backoff delays of all the outer nodes remain equal. Let the mean backoff of the middle node be b1 and that of each of the n outer nodes be b2. Since we use exponential random variables to generate the backoffs, the probability that the middle node wins the contention, as in equation (1.1), is given by 1.3 and 1.4

Note that for the purpose of analysis, we assume the backoff to be a real number, and not rounded off to integers, as described in Section 1.3.2. This would prevent collisions from ever occurring since two neighbors have 0 probability of choosing the exact same backoff. In Section 1.4.3, we analyze the star topology with collisions.

It is also worthwhile to consider the throughput-delay tradeoff in the star topology. The maximum throughput is achieved when the middle node is

quiet, and the outer nodes are always on, this leads to a total throughput of n. On the other hand, a fair share allows the middle node to be active 1=2 the time, while the other nodes are active during the remainder. Consequently, the throughput is only n+1/2

1.4.2 Fairness in Star Topology

Assume that all the outer nodes start with the same initial mean backoff delay. Then, the star topology can be evaluated by an appropriate Markov Chain whose states capture the ratio between the mean backoff delays (i.e. between the backoff rates). Let state Sk designate the state when ,

as shown in Figure 1.3(b). Note that the ratio of backoff rates is a Markov chain since the probability of success or failure, and therefore of moving to another state, is completely determined by the ratio of the rates r1 and r2 independently of the past. Following equations (1.3) and (1.4), the probability that the middle node X in state Sk succeeds in the next slot is given by n:m2k n:m2k+n = m2k m2k+1. If X wins, its backoff rate is updated to r1 := r1=m, while all the other nodes update their backoff rates to r2 := r2 £ m. As a result, the ratio between the rates becomes n:m2k¡2, i.e. the chain moves to state Sk¡1. Similarly, from state Sk, the probability that X loses in the next lot is n n:m2k+n = m2k+1, and this moves the state to Sk+1. The Markov chain has the structure of a birth-death chain, as shown in Figure 1.4 that is symmetric about state S0. This chain is seen to be positive recurrent for all m > 1 because the average drift of the absolute value of the state is negative whenever the state is not 0.We evaluate the steady state probabilities “i, I “ Z of the various states. We begin by calculating “1 in terms of “0

As shown above in eqn 1.5

We can evaluate “0 for various values of m.

Fig. 1.5. Modeling the star topology (with initial bias against middle node)

.4.4 Triangle Topology Model

The other topology that we analyze is the triangle topology, where each of three nodes interferes with the others. Clearly, the situation is symmetric, but we need to ensure that starting at any set of backoff rates, the system drifts back towards a state where all nodes have an equal probability of successful transmission. The evolution of the system is modeled by a Markov chain that represents the backoff rates of the three nodes. A state is denoted by the triple (1; b; c) that specifies the backoff rates of the three nodes in increasing order. We express each of the rates as a multiple of the smallest rate, so the first term is always 1. Thus the probability of the first node winning the contention is 1/1+b+c , while the other two nodes win with probability b/1+b+c and c/1+b+c , respectively.

Every transmission results in one backoff rate being divided by m, and the other two rates being multiplied by m. As a result, the ratio of rates between any two nodes changes by a factor of m2; consequently values b and c in the Markov chain are powers of m2. In Figure 1.7, we represent the Markov chain resulting from a triangle topology. In the figure, we use a value of m = 2 as an illustration. The Markov chain is clearly irreducible, but it is periodic with a period of 3.

1.5 Choosing Algorithm Parameters

IBA is characterized by three design parameters m, RL and RF whose choices affect the performance of the system. In this section, we present qualitative arguments for our choices.

The reset limit RL is the smallest value to which we allow the mean backoff to fall until it is reset. Having a very small value for RL allows nodes to maintain low mean backoff delays. Since the actual number of backoff slots is rounded to the nearest integer, multiple neighbors with low mean backoff delays are more likely to choose the same backoff, leading to frequent collisions. Choosing a large value for RL will alleviate this problem but leads to large backoff values. In our current model this has no ill-effect since all nodes always wait for the completion of the backoff contention phase before attempting to transmit. For simulation purposes, we choose RL = 16/5 = 3.2, which is 1/5th of the initial mean backoff value.

1.5.2 Reset Factor RF

Consider a dense sub graph in the Ad-Hoc network, e.g. a clique topology with q nodes. In such a situation, at each slot, at most one node succeeds while all the others fail. Consequently, q - 1 of the nodes divide their mean backoff delay by m. For a large enough q, at least one of the nodes repeatedly decreases its rate and cross RL, causing a reset. Quantitatively, resets occur approximately every logm RF slots, provided logm RF < q. Choosing a large RF indeed decreases the reset frequency, but this decrease happens only on a slower logarithmic scale. In our simulations we choose RF = 10.

1.5.3 Multiplicative Factor m

In Section 1.4.3 and 1.4.4, we showed that IBA is stable and fair in the star and triangle topologies for m > 1. In Section 1.6.2, simulations show this to be true even for random topologies. However, we need to select an optimal value for m. A small value of m allows nodes to maintain a lower mean backoff delay on average, since the backoff rate may be closer to RL without hitting it. This leads to more collisions, following the same argument as Section 1.5.1. On the other hand, a large value of m causes frequent resets. In a dense network, the frequency of resets is approximately logm RF . The compromise is to choose as large a value of m as possible (to minimize collisions), yet choose it small enough to keep resets under control. For a practical system, this is entirely a design decision based on the efficiency of the reset mechanism. We chose m = 1:2 for our simulations. In this case, the reset frequency is bounded above by the rate of one reset every log1:2 10 = 12:6 slots.

1.6 Simulation Results

we have built a simulation model using Matlab [16] that allows us to compare the performance of the Impatient Backoff Algorithm (Section 1.3) with a slotted exponential backoff algorithm (EBA). EBA chooses a backoff uniformly in a given range, starting with [0; 32]. The node with the smallest backoff in its neighborhood is able to transmit in that slot. A collision doubles the range, while a successful transmission brings it back to the initial range. No changes are made to the backoff when the node is quiet. We evaluate over a wide range of topologies, as we explain below.

1.6.1 Simple Topologies

We first evaluate our scheme by simulating simple topologies. The results are summarized in Table 1.6.1. For each topology, we list the mean throughput and the fairness, for both EBA and IBA. As explained earlier, we use Jain’s fairness Index (Section 1.3.5) to compare the two schemes, assuming an optimally fair allocation lets each node get an equal share of the throughput. In the star topologies, EBA is unfair, while IBA has a fairness of 1. However, in order to achieve fairness, IBA has to pay a throughput tradeoff. The rest of the topologies are symmetric, so both schemes are fair in such topologies. We observe that the mean throughput of IBA is comparable to that of EBA in all these cases.

1.6.2 Random Topologies

A more interesting comparison is visible in random topologies. We first consider a random field of size 4km * 4km and place 100 Ad-Hoc nodes at random on it. The nodes have a transmission range of 500m and an interference range of 1km. Every node is assumed to have packets to send at all times. Note that at any slot, multiple transmissions can take place at different parts of the field. Figures 1.9 and 1.10 express the simulation results for EBA and IBA, on the same topology. We denote a node by a circle with its center at the node’s location. The area of the circle is proportional to the throughput achieved by the node (i.e. number of successful transmissions) scaled by the total number of slots. By comparing the two graphs visually, we can qualitatively see the fairer nature of IBA. EBA in Figure 1.9 includes many nodes with very small throughput – all of whom are able to increase their throughput in IBA. Jain’s fairness index in the EBA simulation is only 0:66 and it increases to 0:75 for IBA. The lowest throughput achieved by a node in EBA is merely 0:0090, while the lowest throughput in IBA is 0:0490. The mean throughput for the two schemes is comparable, at 0:1066 and 0:1046.

Table 1.6.2 summarizes the results from simulations across several random topologies. In each case, we generated a different random topology and ran the simulations on both backoff schemes. We notice that the mean throughput is comparable, while IBA achieves a significantly higher fairness. Also, the minimum throughput received by a IBA node is 3 to 5 times higher than in EBA.

1.6.3 Reset Propagation and Lost Resets

As explained in Section 1.3.4, IBA requires all nodes to reset their mean backoff delays when any mean backoff goes below a reset limit. In a practical situation, this requires the propagation of the reset message through the network. We model this in our simulations by propagating the reset message hop by hop from the originating node. A node resets its backoff only when it receives a reset message. The reset messages have a time-to-live field to ensure expiration after a single reset. Also, a node does not reset more than once in a fixed number of slots, thus multiple reset messages starting from different parts of the network at around the same time cause the backoff delays to be reset only once. Furthermore, reset messages may get lost in the way. Our

simulations account for situations where up to 10% of reset messages get lost.

It turns out that reset propagation and loss has marginal effect on the throughput and fairness of IBA. The simulation results given above in Section 1.6.2 in fact take into account both these effects. Qualitatively, we can see why the reset propagation and loss does not affect the long term fairness of the system. A delayed or lost reset message implies that the intended recipient node does not reset its backoff in a timely manner and hence continues to have a low backoff. However, the result of this is that it wins a few contentions unfairly – which in turn causes it to increase its backoff. The unfairness only allows a few extra packet transmissions and does not persist. Thus the scheme is fair in the long run.

1.6.4 Variations in Simulation Scenarios

We also attempt a few variations in the simulation scenarios. In each of the cases, the throughput and fairness numbers are comparable to the previous cases, hence we do not repeat them explicitly.

Movement:

We want to investigate the effect of random walk movement on IBA. Every node moves randomly in discrete steps every 10 time slots.

Changing Number of Nodes:

We also consider the case when nodes do not always have packets to send. This is simulated by considering blocks of slots when a node is either awake or asleep. For simple topologies, we evaluate the fairness of IBA against the fair share of bandwidth based on the active nodes during each block of slots. The results show IBA to maintain its fairness even with a changing number of neighbors. An example is shown in Table 1.6.4 for a clique of five nodes. Different nodes have a different fair share since they are active for varying amounts of time, but the IBA scheme is able to provide them with that share. These experiments yield a fairness of

Biased Initial Backoff Rate:

We also extend the idea of an initial bias (Section 1.4.3) to a general topology. We start with an initial bias against a few selected nodes by making their initial mean backoff 64 times other mean backoff delays, and observe whether these nodes indeed get a fair share of the throughput. Under IBA, these backoff delays soon catch up and the nodes subsequently get an equal share of the bandwidth. While their initial backoff delays are high, these nodes tend to lose the contention, causing their mean backoff delays to be divided by m. Thus they catch up in approximately logm 64 slots.

1.7 Conclusions

Traditional distributed Ad-Hoc MAC protocols (e.g. 802.11b) use random backoff delays to avoid collisions and share bandwidth amongst contending nodes. The fundamental backoff rule is to become less aggressive upon collision. We observe that in a network spanning multiple interference domains, this approach leads to unfairness towards nodes in the middle of the network. We propose a novel backoff scheme that counters the conventional backoff wisdom.

Nodes in our Impatient Backoff Algorithmdecrease their backoff when they collide or are unable to send – thereby becoming more aggressive. This approach tends to help nodes with more neighbors and leads to a fairer allocation of bandwidth. The danger of the system becoming unstable due to frequent collisions is handled by resetting the mean backoff delays when they get too low. We use Markov chains to analyze IBA in simple topologies. We look at two extreme topologies – the unfair star topology, and the symmetric clique topology. By proving positive recurrence of the system, we show that IBA is indeed stable. We are also able to demonstrate fairness of the system, even as the number of nodes in the topology is changing dynamically. We compare the performance of IBA with an idealized slotted exponential backoff scheme. Results show that in simple topologies IBA is always able to achieve a fairness index of 1, albeit at the cost of some throughput tradeoff. In a random topology, IBA maintains the same mean throughput as EBA but has a significantly higher fairness index. Further extensions involving movements and nodes that switch between active and sleep phases also give similar results.

REFERENCES

1. J. Li, C. Blake, D. S. J. De Couto, H. I. Lee, and R. Morris, “Capacity of Ad Hoc Wireless Networks,” Proceedings ACM Mobicom 2001, Rome, Italy, July 2001.

2. R. Gupta, J. Musacchio, and J. Walrand, “Sufficient Rate Constraints for QoS Flows in Ad-Hoc Networks”, UCB/ERL Technical Memorandum M04/42, Fall 2004.

3. IEEE Standard for Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications, Nov. 1997. P802.11.

4. IEEE P802.11: The Working Group for Wireless LANs, http://grouper.ieee.org/groups/802/11/.

5. G. Bianchi, “Performance Analysis of the IEEE 802.11 Distributed Coordination Function,” IEEE Journal on Selected Areas in Communications, vol. 18, no. 3, March 2000.

6. M. Ergen and P. Varaiya, “Throughput Analysis and Admission Control for IEEE 802.11a,” to appear in ACM-Kluwer MONET Special Issue on WLAN Optimization at the MAC and Network Levels.

7. M. Ergen and P. Varaiya, “Throughput Formulation and WLAN Optimization in Mixed Data Rates for IEEE 802.11 DCF Mode,”GLOBECOM-CAMAD 2004.

8. M. Ergen, B. Dundar, and P. Varaiya, “Throughput Analysis of an Extended Service Set in 802.11,” GLOBECOM 2004.

9. S. Coleri, “PEDAMACS: Power Efficient and Delay Aware Medium Access Protocol for Sensor Networks,” Masters Thesis, Electrical Engineering and Computer Sciences Department, University of California, Berkeley, December 2002.

10. D. Bertsekas and R. Gallagher, “Data Networks (2nd Edition),” Prentice Hall, 1991.

11. C. Yuan and P. Marbach, “Rate Control in Random Access Networks,” preprint.

12. L. Kleinrock and F. Tobagi, “Packet switching in radio channels, Part II The hidden terminal problem in carrier sense multiple access and the busy tone solution, IEEE Transactions in Communications, vol. COM-23, no. 12, pp. 14171433, Dec. 1975.

13. F. P. Kelly, A. Maullo, and D. Tan, “Rate Control in Communication Networks: Shadow Prices, Proportional Fairness and Stability, “Journal of the Operational Research Society, vol. 49, pp. 237-252, March 1998.

14. D. Chiu and R. Jain, “Analysis of the Increase and Decrease Algorithms for Congestion Avoidance in Computer Networks,” Computer Networks and ISDN Systems, vol. 17, pp. 1-14, 1989.

15. A.G. Pakes, “Some Conditions of Ergodicity and Reurrence of Markov Chains,” Operations Research, vol. 17, pp. 1058-1061, 1969.

16. Matlab Simulation Environment, The Mathworks Inc. http://www.mathworks.com.

DESIGN OF INTELLIGENT ROBOTIC ARM FOR VISUALLY CHALLENGED‘A BOON FOR PHYSICAL AILMENT’

SEIMA SAKI S1, SABITA DEVI D2

ECE1, Mechatronics2 Sri Krishna College Of Engineering And Tech. Coimbatore, IndiaComputer Technology, PSG College of Technology, [email protected], [email protected]

MANOJ SCoimbatore, India [email protected]

ABSTRACT:

A physically disabled person nourishing himself was considered as a great deed in the 19th century. But this age has become an era where talents are considered as the matter of fact, in spite of their physical weakness. Here is a newfangled technology, through which a robot imitates human handwriting and acts as a proxy during strenuous circumstances. What the robot does is that, it acts according to the voice commands imposed on it, thereby fulfilling the physical ailment of the needy. Writing is brought by feeding a particular style in the form of images, which is stored as fonts in its memory. This not only assists disabled persons, but also supports people who need a proxy for their hand writing. When a particular font is fed, it writes what was dictated, thereby, acting as a human hand. This methodology that has been proposed here is much practical and the design shows improved efficiency.

Keywords:

Servo mechanism, manipulators, Bresenham’s algorithm, speech analysis.

INTRODUCTION Scientists fed their entire lifetime in developing technologies that would act as a substitute to hide physical impairment. Past achievements infer that there were devices which when coupled with a physically challenged person, would assist in erasing the traces of the abnormality. But still the originality of humans retains its color. No device could do the job with so much perfection as a human does. And every human, physically able or disabled, is unique. Why can’t a robot imitate human’s uniqueness, say, writing..! This paper presents the study of robotic arm that has been committed to paper writing. As a matter of fact, the system recognizes human language using speech analysis techniques and the signal obtained is converted into text. Texts are processed and converted into digital data. These digital signals control the servos present. Servos move in accordance with the information stored in the controller. This movement accomplishes writing.Fundamental Building Blocks Given below is the outline of technical blocks of the robotic humanoid hand. Our aim is to make the arm trace the letters in accordance with the speech. Arm movement can be brought up with manipulators and

servo motors. Since servos are fully self contained, the velocity and angle control loops are very easy to implement, while prices remain very affordable. For this, each alphabet, number or a special character is to be represented by a unique values, so called as ‘fonts’, and fed into a memory. And servo carries out the activity in relation to those values. And we need a microcontroller to take up memory and control activities. We require a speech analyzer to convert human language into text and text is processed further into digital data. Microcontrollers work with digital inputs. So the speech has to be converted into digital string.

To position and orient the robotic arm, manipulators of 6-DOF or 3-DOF can be employed. The DOF of the manipulator are distributed into sub assemblies of arm and wrist. The end effectors which aexternal to the manipulators are attached to it to hold thpen. The proper shape and size of the gripper andmethod of holding are determined by the objects tograsped and task to be performed. Inclination angle is to be specified to the motor, so that it writes accordingly.

Figure 1. Basic Building Blocks-Overview

A. Speech Analyses Speech recognition (also known as automatic speech recognition or computer speech recognition) converts spoken words to text. People with disabilities can benefit from speech recognition programs. Individuals with learning disabilities who have problems with thought-to-paper communication (essentially they think of an idea but it is processed incorrectly causing it to end up differently on paper) can benefit from this technique. A speech recognizer is a device that automatically transcribes speech into text. It can be thought of as a voice-actuated ‘typewriter’ in which a computer program carries out a transcription and the transcribed ‘text’ appears on a workstation display. The recognizer is usually based on some vocabulary that restricts the words that can be ‘printed’ out. Until we state otherwise, the designation word denotes a word form defined by its spelling. [2] The speech signal is analyzed and the feature is extracted for comparison with stored references in the pattern matching block. The reference patterns which characterize individual speech classes to be recognized are obtained by a training procedure using a large set of known examples. A decision scheme determines the word or class of the unknown input based on the matching scores or evaluated similarity measures. Methods to be discussed for dealing with adverse conditions may take place in various blocks in the diagram. [7].

Figure 1.Blocks Of Speech Analyses System

A. Servo MotorA servo mechanism or servo is an automatic device that uses error-sensing feedback to correct the performance of a mechanism. The term correctly applies only to systems where the feedback or error-correction signals help control mechanical position or other parameters. The servomotors uses optical encoders which are in the form of discs, the motor rotates through which the encoders tracks the count. In turn this servo is connected to MCU which executes in accordance to the track. AC servo motors are used because of their high efficiency and contrrol. They need not to be replaced often as DC servos due to absence of brushes. Servo Pulse width modulation (PWM) is a powerful technique for controlling servo with a controller’s digital outputs. The general concept is to simply send an ordinary logic square wave to the servo at a specific wave length, and the servo goes to a particular angle (or velocity if the servo is modified). The wavelength directly maps to servo angle. In our system, the motor is driven by the microcontroller which relates input digital string with its stored 2D representations of symbols. The standard time vs. angle is represented in this chart:

Figure 1. Angle variation of servo in relation to the width of clock

1)Servo Motor-Design:

A servo motor system, comprising a servo motor unit, a servo driver unit, power transmission lines for supplying electric power from servo driver unit to servo motor unit, and signal transmission lines for supplying electric signals to servo driver unit.

a) Servo motor unit comprises: It has a servo motor; a pole sensor for detecting a magnetic state, a rotary encoder for detecting a mechanical state of servo motor; a parallel-serial signal conversion circuit for converting an output from pole sensor and an output from rotary encoder

into a single serial signal; and a line driver for sending out serial signal through one of signal transmission line, a three-phase AC motor, and

outputs from pole sensor and rotary encoder include Pu-phase, Pv-phase and Pw-phase signals from pole sensor and a Z-phase signal from rotary encoder.

b) Driver logic circuit comprising:

Three AND circuits is connected to the input ends of parallel-serial signal conversion circuit, Pu-phase, Pv-phase and Pw-phase signals being each supplied to an input end of associated one of AND circuits whose other ends are commonly connected to receive sensor output signal while Z-phase signal is directly supplied to one of the inputs of parallel-serial signal conversion circuit; and a motor logic circuit comprising a NOR circuit is connected to the output ends of serial-parallel signal conversion circuit, detected Pu-phase, Pv-phase and Pw-phase signals separated from serial-parallel signal conversion circuit being supplied to different input ends of NOR circuit so as to produce an abnormal condition detection signal from an output end of NOR circuit in case an abnormal condition is detected by abnormal condition sensor while a detected Z-phase signal is directly obtained from one of output ends of serial-parallel signal conversion circuit.

c) Driver logic circuit:

It comprises of an AND circuit and three OR circuits is connected to the input ends of first parallel-serial signal conversion circuit, Pu-phase signal and Z-phase signal being supplied to different input ends of AND circuit while an output of AND circuit, and Pv-phase and Pw-phase signals are each supplied to an input end of associated one of OR circuits whose other ends are commonly connected to receive sensor output signal; and a motor logic circuit comprising a NOR circuit, an AND circuit and an OR circuit is connected to the output ends of said second serial-parallel signal conversion circuit, three outputs from second serial-parallel signal conversion circuit being connected to different inputs of NOR circuit as well as to different inputs of AND circuit while an output of NOR circuit and one of output ends of second serial-parallel signal conversion circuit being connected to different inputs of OR circuit so that a detected Z-phase signal, a detected Pu-phase signal and a detected sensor output signal may be obtained from output end of NOR circuit, and output end of OR circuit and an output end of said AND circuit, respectively, while detected Pv-phase and Pw-phase signals may be obtained directly from associated ones of output ends of second serial-parallel signal conversion circuit.

d)Servo driver unit comprises:

A line receiver for receiving serial signal transmitted from line driver through signal transmission line; a serial-parallel signal conversion circuit for separating output signals from rotary encoder and pole sensor from serial signal, and a controller for controlling supply of electric power to servo motor unit through power transmission lines according to separated parallel signals.

C. Manipulators

A robot is required to carry out specific by moving itsend effectors accurately and repeatedly. The dynamic control of manipulator motion and interaction forces requires know-ledge of forces and torques that must be exerted on a manipu-lated joints to move the links and the end effectors fromthe present location to the desired location, with or without the constrains of the particular planned end effectors trajectorand planned end effector force/torque.

1) Configurations of the arm of robot:Gantry: These robots have linear joints and are mounted overhead. They are also called Cartesian and rectilinear robots. Cylindrical - Named for the shape of its work envelope, cylindrical anatomy robots are fashioned from linear joints that connect to a rotary base joint.

• Polar - The base joint of a polar robot allows for twisting and the joints are a combination of rotary and linear types. The work space created by this configuration is spherical.

• Jointed-Arm - This is the most popular industrial robotic configuration. The arm connects with a twisting joint, and the links within it are connected with rotary joints. It is also called an articulated robot.

2) Degrees of freedom:To enhance robot hand dexterity, the robot should be designed to have a redundant number of degrees of freedom. In redundant robotic systems, inverse kinematics from task description space to joint space becomes ill-poised, making a difficult to determine joint motions. To avoid this ill-posed ness, most proposed methods introduce an additional input term calculated from an intentionally introduced artificial

index of performance. A 4 DOF redundant handwriting robot, whose endpoint is constrained on a two dimensional plane, such as a sheet of paper, is presented here. [9]

x,y and z are the Cartesian co-ordinates, q1,q2,q3 and q4 denotes vector of joint angles. L1,L2,L3 and L4 is the distance between two vector of joint angles. [1].

Figure 1. 4DOF Hand Writing Robot

The schematic diagram of a manipulator control system is shown. The dotted lines for the feedback indicate that the control system may or may not employ feedback of the actual joint locations and velocities. The parameters q,q’,q’’,t etc. are shown in the figure, where q gives the history of position and q’,q’’,t are its derivatives.

IV. ALGORITHM FOR TRACKING OF THE MANIPULATOR MOVEMENT

The Bresenham line algorithm is an algorithm which determines which points in an n-Dimensional raster should be plotted in order to form a close approximation to a straight line between two given points.function line(x0, x1, y0, y1)function line(x0, x1, y0, y1) int deltax := x1 - x0 int deltay := y1 - y0 real error := 0 real deltaerr := deltay / deltax int y := y0 for x from x0 to x1 plot(x,y) error := error + deltaerr if abs(error) e” 0.5 then y := y + 1 error := error - 1.0

B. FOR CIRCLE-MID-POINT CIRCLE ALGORITHMIn computer graphics, the midpoint circle algorithm is an algorithm used to determine the points needed for drawing a circle. The algorithm is a variant of Bresenham’s line algorithm, and is thus sometimes known as Bresenham’s circle algorithm, although not actually invented by Bresenham.

int error = -radius; int x = radius; int y = 0; while (x >= y) plot8points(cx, cy, x, y); error += y;++y; error += y; if (error >= 0) —x; error -= x error -= x; void plot8points(int cx, int cy, int x, int y) plot4points(cx, cy, x, y);

if (x != y) plot4points(cx, cy, y, x);void plot4points(int cx, int cy, int x, int y) setPixel(cx + x, cy + y); if (x != 0) setPixel(cx - x, cy + y); if (y != 0) setPixel(cx + x, cy - y); if (x != 0 && y != 0) setPixel(cx - x, cy - y);

Figure 1. A Sample character shown as a combintion of line and circle

V. RESULTS AND OBSERVATION

We used Digital Differential Analyzer Algorithm at the beginning but we did not end up with accurate results. So we ended up using Bresenham’s algorithm.

Positioning the write head after horizontal retrace and vertical retrace was not precise. Rectified it after minor adjustments with the coordinates.

65 rpm servo was found to miss match with the processing delays by the recognition system. Hence we ended up with 45 rpm motor.

VI.CONCLUSION

THE PROPOSED MODEL CAN BE EFFICIENTLY APPLIED FOR ALL CATEGORIES OF PEOPLE, MORE SPECIFICALLY PHYSICALLY CHALLENGED. THE WORLD ITSELF IS BECOMING AUTOMATED AND HENCE WE ARE FORCED TO GO FORWARD BY LEAPS AND BOUNDS ACROSS THE TECHNOLOGY TO MAKE UP A BETTER LIVING. THIS TECHNOLOGY WOULD SURELY ASSIST SUCH ESCALATING ADVANCEMENTS, NOT ONLY MADE FOR NORMAL PEOPLE BUT ALSO FOR THE DISABLED.

REFERENCES

[1] R K Mittal, I J Nagrath: “Robotics and control”, Tata McGraw-Hill Publications.

[2] Frederick: “Stastical methods for speech recognition” (chapter 1) -1997 MIT.

[3] IEEE transaction on “Audio, speech and

language processing”, sep 2006, vol. 14, issue 5.

[4] IEEE transaction on “Robottics and Automation”, march 2003, vol. 10, Issue 1.

[5] Third International conference on “Information technology and applications” 2005.Volume 2 issue 4 to 7 pages 21 to 24.

[6] IEEE transaction on “Speech And Audio

Processing”, July 2004 Volume 12, Issue 4.

[7] Ravi P. Ramachandran, Richard Mammone: “Modern methods of Speech processing”, Kluwer academic publications 1995. pg no. 236.

[8] Third international conference on “Information technology and applications”, july, 2005. Vol. 2, issue 4-7.

[9] IEEE journal of robotics and mechatronics: “Control of handwriting robot with DOF redundancy based on feedback in task coordinates”, Vol.16 No.4, 2004.

[10] www.tryengineering.org.

CLUSTERING IN HIGH PERFORMANCE MULTIFEATURE QUERY PROCESSING

1 SIVASUBRAMANIAN S.Dept of CSE,Bharath University,Chennai-73.

2 Dr.UMAMAHESWARI MProf & Head.,Dept of CSE,Bharath University,Chennai-73

ABSTRACT:The issue of clustering in high performance processing of multifeature queries are considered as one of the emerging process in database applications. The main aim of the paper is to develop the efficient processing of Multifeature queries. In this system the indices are built based on a static combination of feature weights. Clustering of the stored information which is used for accessing the information very effectively. Different queries may examine these features in different ways. This proposed system devise a representation that captures an f-feature point. This representation enables us to index the component based on weight ages. Natural language processing techniques will be used to identify the key entities in individual documents. This allows us to view the problem in a database/data mining context to identify related groups of items. To identify the related items based on traditional data mining techniques, item sets will be generated from the groups of items, followed by clusters formed with a hyper-graph partitioning scheme. Keyword: Multifecture queries, Clustering.

1.1 INTRODUCTIONThis paper deals with an efficient way of processing multifeature queries. In multifeature query processing the weight ages of features typically are different for different queries and different weight ages correspond to different results. In this process a user normally inputs a query with respect to more than one feature. The queries can be formulated in standard keyword form. From the given query the documents are searched and extracted by reading the full text in the query. Initially search process is carried out based on the single feature. The user can give the Query for searching the documents which are more relevant. The most relevant documents are fetched from the database. Weightages are applied to the documents by receiving each keyword and finding the number of occurrences of this keyword in each document. According to high preferential weightage, the documents are fetched accordingly. Thus we can efficiently process the multifeature query.

1.2 PROBLEM DEFINITIONA database is an organized collection of data. The term originated within the computer industry, but its meaning has been broadened by popular use. One possible definition is that a database is a collection of records stored in a computer in a systematic way, so that a computer program can consult it to answer questions. For better retrieval and sorting, each record is usually organized as a set of data elements (facts). The items retrieved in answer to queries

become information that can be used to make decisions. The computer program used to manage and query a database is known as a database management system (DBMS). The properties and design of database systems are included in the study of information science.The central concept of a database is that of a collection of records, or pieces of knowledge. Typically, for a given database, there is a structural description of the type of facts held in that database: this description is known as a schema. The schema describes the objects that are represented in the database, and the relationships among them. There are a number of different ways of organizing a schema, that is, of modeling the database structure: these are known as database models (or data models). The model in most common use today is the relational model, which represents all information in the form of multiple related tables each consisting of rows and columns (the true definition uses mathematical terminology). This model represents relationships by the use of values common to more than one table. Other models such as the hierarchical model and the network model use a more explicit representation of relationships. Strictly speaking, the term database refers to the collection of related records, and the software should be referred to as the database management system or DBMS. Database management systems are usually categorized according to the data model that they support: relational, object relational, network, and so on. The data model will tend to determine the query languages that are available to access the database. A

great deal of the internal engineering of a DBMS, however, is independent of the data model, and is concerned with managing factors such as performance, concurrency, integrity, and recovery from hardware failures. In these areas there are large differences between products. Data mining can be defined as “the nontrivial extraction of implicit, previously unknown, and potentially useful information from data” and “the science of extracting useful information from large data sets or databases”. Although it is usually used in relation to analysis of data, data mining, like artificial intelligence, is an umbrella term and is used with varied meaning in a wide range of contexts. It is usually associated with a business or other organization’s need to identify trends.Data mining involves the process of analyzing data to show patterns or relationships; sorting through large amounts of data; and picking out pieces of relative information or patterns that occur e.g., picking out statistical information from some data. A simple example of data mining is its use in a retail sales department. If a store tracks the purchases of a customer and notices that a customer buys a lot of silk shirts, the data mining system will make a correlation between that customer and silk shirts. In statistical analysis, in which there is no underlying theoretical model, data mining is often approximated via stepwise regression methods wherein the space of 2k possible relationships between a single outcome variable and k potential explanatory variables is smartly searched. With the advent of parallel computing, it became possible to examine all 2k models. This procedure is called all subsets or exhaustive regression. Some of the first applications of exhaustive regression involved the study of plant data. Generally, data mining (also called data or knowledge discovery) is the process of analyzing data from different perspectives and summarizing it into useful information - information that can be used to increase revenue, cuts costs, or both. Data mining software is one of a number of analytical tools for analyzing data. It allows users to analyze data from many different dimensions or angles, categorize it, and summarize the relationships identified. Technically, data mining is the process of finding correlations or patterns among dozens of fields in large relational databases. Web mining and web usage mining is the application of data mining techniques to discover usage patterns from the Web in order to better understand and serve the needs of users or Web-based applications. It is an activity that involves the automatic discovery of patterns from one or more Web servers. Organizations often generate and collect large volumes of data; most of this information is usually generated automatically by web servers and collected in server access logs. he

first web analysis tools simply provided mechanisms to report user activity as recorded in the servers. Using such tools, it was possible to determine such information as the number of accesses to the server, the times or time intervals of visits as well as the domain names and the URLs of users of the Web server. However, in general, these tools provide little or no analysis of data relationships among the accessed files and directories within the Web space. TNow more sophisticated techniques for discovery and analysis of patterns are now emerging. These tools fall into two main categories: Pattern DiscoveryTools and Pattern Analysis Tools 2.1 MULTI-DIMENSIONAL INDEXING

Multidimensional indexing and high-dimensional indexing have been extensively researched in the database literature. With the demand for even higher dimensional databases, consisting of hundreds of or more dimensions, earlier high-dimensional indexes face significant challenges. Indexing techniques have typically been designed for 30-50 dimensions and fail to improve the performance of sequential scan due to the known “dimensionality curse.” To tackle this phenomenon, recent proposals can be categorized into two approaches:· Dimensionality Reduction • One-Dimensional Transformation.

2.2DIMENSIONALITY REDUCTION

Dimensionality reduction methods map the high-dimensional space into a low-dimensional space which can be indexed efficiently using existing multi dimensional indexing techniques. The main idea is to condense the original space into a few dimensions along which the information is maximized. Such methods report approximate nearest neighbors, however, since dimensionality reduction incurs information loss. Representations of the original data points using smaller and approximate representations have also been proposed as a means of aiding high-dimensional indexing and searching by deploying the efficiency of sequential scan. The VA-file is one typical proposal. The VA-file (Vector Approximation file) represents the original data points by much smaller vectors. The VA-file is then sequentially scanned to obtain a very small set of candidates. However, the VA-file has several drawbacks. First, its performance is limited by the VA-file size. Second, it imposes a significant overhead. Third, the VA-file does not adapt gracefully to highly skewed data. One-dimensional transformations provide another direction for high dimensional indexing. Our work is more closely related to techniques based on

transforming multi dimensional data to single-dimensional space.2.3 MULTIFEATURE QUERY PROCESSINGAnother category of related work is on multifeature query processing. However, most of the existing techniques index each feature separately. The exception is the work by Ngu et al. that constructs a single M-tree index for all the features. However, this method may not be practical for real usage. First, it tedious and undesirable for very large data sets. Second, existing indexing structures, such as the M-tree ,are known to degrade in performance for dimensionality larger than 20. This may result in significant information lost that may affect retrieval effectiveness. More recently, the importance of each individual dimension has also been examined. In, a separate table for each dimension is created. Such a physical design enables the distance computation by adding dimensions one by one. Lower and upper bounds on the full dimensional distance can be exploited to discard the far away points based on the distance computed on a small number dimensions. This paper deals with an efficient way of processing multifeature queries. The issue of efficient processing of multifeature queries are considered as one of the emerging process in database applications. Different queries may examine these features in different ways. This proposed system devise a representation that captures an f-feature point. This representation enables us to index the component based on weightages. Natural language processing techniques will be used to identify the key entities in individual documents. This allows us to view the problem in a database/data mining context to identify related groups of items. To identify the related items based on traditional data mining techniques, item sets will be generated from the groups of items, followed by clusters formed with a hyper-graph partitioning scheme.

3.ADVANTAGESOF MULTIFEATURE QUERYIn this proposed “High Performance Multifeature Query Processing System”, the weightage of features typically are different for different queries and different weightage correspond to differ results. In this process a user normally inputs a query with respect to more than one feature. The queries can be formulated in standard keyword form. By reading the query, the keywords are extracted. Initially search process is carried out based on the single feature. This could be compared with searching the data, based on multifeature query. The most relevant documents are fetched from the database server. The

proposed system provides an efficient way of query processing that compactly capture features based on weightages. Documents will be clustered based on the keywords using hyper-graph partition algorithm. In normal Google search engine, we have to trace the number of links to get the exact result but in this proposed system we can get the exact link immediately. Fetched documents will be clustered which will be useful for retrieving the exact documents. The advantages of the proposed system: · Efficient filtering of data· Take less computational time· Data accuracy can be maintained· Multifeature can be maintained· Computational complexity reduced

Figure 1. Scheme of Proposed System

4.1 USER QUERY PROCESSING

In this module from the user submitted query, only the keywords are extracted. Thus the search process is based on keywords using modified porter stemmer algorithm and stored the relevant data in the database. The search process is carried out by traversing all the documents. The number of occurrences of each keywords are calculated for each document. In the porter-stemmer algorithm, it removes function words such as a, an, the, for, in etc and group morphological words such as police, policy that is it takes the stem of the word for searching. So it retrieves large number of documents which is difficult to retrieve the exact information. It increases computational complexity as well as time complexity. It leads to lack of data accuracy. Since it does not matches with exact documents, all features are merged as a single feature. Multifeature can not be maintained. In the modified porter-stemmer algorithm, it matches with exact keyword which then reduces the number of documents to be retrieved .We can able to get the exact link. It further reduces computational complexity as well as time complexity. Data accuracy can be maintained. Since it matches with exact documents, multifeature can be maintained.

4.2 FILTER PROCESSINGThe amount of Web information is growing rapidly, improving the efficiency and accuracy of Web

information retrieval is uphill battle. There are two fundamental issues regarding the effectiveness of Web information gathering: information mismatch and overload. To tackle these difficult issues, an integrated information filtering is introduced. In the first phase of the proposed scheme, an information filter that based on user search is used to quickly filter out irrelevant data. In the second data processing phase, data mining technique here clustering can be applied. This filter process improves effective access. In the filter processing module, the related documents are filtered based on the multifeature keyword by finding the number of occurrences in each document. 4.3 CLUSTERINGThe related documents will be grouped based on the keywords using hyper-graph partitioning algorithm. Hypergraph is an extension of a graph. Edges can connect multiple vertices. Hyper-graph can be applied to data mining, parallel computing, transportation problem. It can also be used for load balancing in networks.Here hyper-graph model is used to find the relations among the patterns. In clustering, the groups are not predefined. Here the database is partitioned into disjoint groups based on the keywords. Partition technique effectively use global as well as local information. Here the global measure is based on the number of occurrences of these patterns in the overall data. Since clustering is one type of classification problem, classes are formed based on the keywords. All the documents are read and if it matches with the existing pattern we can classify that document to the corresponding class. So the documents are effectively grouped based on keywords forming clusters. Finally a tree view of documents is generated. Algorithm:HyperGraph Partition:V : the set of vertices here the documents being clusteredE : the set of hyperedges ,which can connect more than 2 vertices here , a set of related documents.

A weight is assigned to each hyperedge here the weight is based on the number of occurrences of each keyword in all documents. Each document is viewed as an item. Each possible feature word as a transaction. The HyperGraph Partition Representation:

1.Generate classes based on the number of keywords.

2.Represent each document as a vertex item.

3.Calculate the number of occurrences of each keyword as a weightage

4.If the document matches with the keyword then classify that document in the corresponding class by adding hyperedge which forms clusters.

4.4 EFFICIENT INFORMATION RETRIEVEL

The requested query will be effectively fetched using modified KNN (K nearest neighbor) search algorithm.In the KNN search algorithm the highly selective feature is based on minimum distance,but my proposed system gives highest priority to the number of occurences of keyword in the fetched documents, it will effectively retrieve the documents.

The selected documents will be fetched from the database server and displayed to the user more effectively.CONCLUSION

From the proposed towards building a multifeature query processing, the future enhancement is far reaching implication in that it provides as with the new vantage point by exploring for adapt the search engine. Even though multifeature query processing offers high performance, still some drawbacks are there. We can more efficiently filter the documents, by applying various filtering technique. Performance may be improved by considering the factors memory space and retrieval time so that the system may be highly reliable in future. We have examined the issue of efficient processing of multifeature queries and we have developed a novel representation that efficiently captures an f-feature point. This system proposed modified porter stemmer algorithm to retrieve the exact documents and to reduce the number of documents. This system proposed hypergraph partition algorithm for efficient clustering. We have also proposed an efficient algorithm for processing KNN queries. Our extensive experiments on both real life and synthetic data sets showed that the proposed index structure offers significant performance advantages over existing methods. The User Queries are effectively processed and the documents are filtered. Finally the Documents are effectively grouped together and efficiently retrieved and displayed to the user.

REFERENCES

[1] Berchtold S.; Bo¨hm C.; Jagadish H.V., Kriegel H.-P. and Sander J., 2000. “Independent Quantization: An Index Compression Technique for High-Dimensional Data Spaces,” Proc. Int’l Conf. Data Eng., 577-588.[2] Beyer K.; Goldstein J.; Ramakrishnan R., and Shaft U., 1999. “When Is Nearest Neighbors Meaningful,” Proc. Int’l Conf. Data Transfer, 217-235. [3] deVries A.; Mamoulis N., Nes N., and Kersten M., 2002. “Efficient k-NN Search on Vertically Decomposed Data,” Proc. SIGMOD Conf., 322-333, [4] Guntzer U., Balke W.-T., and Kiessling W., 2000. “Optimizing Multifeature Queries for Image Databases,” Proc. Conf. Very Large Data Bases, 261-281. [5] Jin H., Ooi B., Shen H., Yu C., and Zhou A., 2003. “An Adaptive and Efficient Dimensionality Reduction Algorithm for High-Dimensional Indexing,” Proc. Int’l Conf. Data Eng., 87-98.

ZERNIKE MOMENTS BASED TEXTURE FEATURES FOR RECOGNITION TASK

M.SURESH1, V.SUBBIAH BHARATHI2,1 Research Scholar, Bharath University, Chennai 600 073, 2 Professor, DMI

College of Engineering, Chennai 602103

ABSTRACT :

Moment descriptors have been extensively employed as pattern features in character recognition, face

recognition, object identification and data compression. This paper examines the use of orthogonal moments

such as zernike moments for classifying real world texture images. The moments are computed from the

texture images after preprocessing them to normalise for first order statistical variations. Each unknown

texture is classified as one of the known class of textures in unsupervised manner. The experimental results

indicate the efficacy of the proposed features in texture classification.

KEYWORDS :Texture, Feature Vector, Orthogonal moments, Unsupervised Classification.

I. INTRODUCTION

In applications such as scene analysis, face recognition and remote sensing, it has long been recognised that texture analysis plays a fundamental role in establishing and classifying objects and regions. Texture could be defined as the surface property of objects. It is composed of large number of more or less identical patterns concatenated with some placement rules (Vangool et al., 1985). A checkerboard for instance is deterministic texture where the patterns are strictly ordered. In stochastic texture, the spatial distribution of the patterns is random. The real world images like Landsat images are known to contain different textured regions. Classification of these textured regions is vital for analysis and interpretation of the images. The major approaches for texture classification include statistical, structural, model-based and transform methods (Materka and Strzelecki,1998). A new

statistical approach to solve the problem of texture classification is proposed in this work, which is based on orthogonal moments. Much works on using moments for texture characterisation or representation have not been reported. The idea of using orthogonal moments for texture classification is motivated by the fact that the orthogonal moments have superior feature representation capability and low information redundancy. In this work, Zernike moments have been employed to classify the texture images.

II.COMPUTATIONOFZERNIKEMOMENTS

Regular moments have been used as shape features in a variety of applications like visual pattern recognition, object classification, template matching and data compression. Since the basis functions of regular moment are not orthogonal, it suffers from a certain degree of information redundancy. Teague (1980) introduced moments with orthogonal basis functions to overcome this problem.

A set of complex polynomials, which forms a complete orthogonal set over the interior of a unit circle, was introduced by Zernike (1934). These polynomials are expressed as

( ) ( ) ( )[ ] ( )1 /tanexp ,, 1 xyjmyxRyxV nmnm−=

where n : positive integer or zerom : positive or negative integer subject toconstraints that n-|m| is even and |m| £ n

Rnm (x, y) : radial polynomial defined as

( )( )( ) ( )

( )2 !

2 !

2 !

! 1,

2/

0

222/

−

−

−

+

−−=

∑ +−

=

−

smn

smn

s

snyxyxR

mn

s

sns

nm

The Zernike polynomials are orthogonal and satisfy,( )[ ] ( )

( ) ( )3 0

, 1/

, , *

122

==+π

=∫ ∫≤+

o th er w is emqpnifn

d yd xyxVyxV p qyx

n m

Zernike moments of order ‘n’ and repetition ‘m’ for a digital image function f (x, y) are defined as

( )[ ] ( ) ( )4 1x , ,, 1 22*

x≤+∑ ∑

+= yyxfyxV

nA

ynmnm π

Computation of Zernike moment requires an appropriate transformation of the image co-ordinate space to the unit disk over which the Zernike polynomials are defined.

III. EXPERIMENT AND RESULTS

The input images used in the classification

experiments are eight textures namely sand, calf

leather, pigskin, pic2, D77, fabric, t-ima2 and t-ima9

selected from Brodatz album (1968) and from other

texture databases. The samples of these textures are

shown in Fig.1. The texture images are 512 ´ 512

pixel size. All the input texture images are

preprocessed or requantised to have the same number

of gray levels i.e. 256. The moment features, which

could capture the textural properties, are identified by

a systematic feature selection procedure, during the

training session. Randomly selected set of training

samples per texture image is used for this purpose.

This step is necessary to reduce the dimension of the

moment feature vector without degrading the

classifier performance. Also the computation time is

saved during the testing phase.

Fig 1. Samples of Texture images used in the

classification Experiments

( ) ( ) ( ) ( )7 ... ,~ 2

1 1dVUJ ik

n

k

c

iik

mm ∑ ∑

= =

′= µ

( )

( )[ ] ( )8 ...

1

2 21

∑=

−=

=−m

jkj

ikik

vijx

VXdd

Unsupervised classification employed in this work aims to find natural groupings or clusters in multidimensional data based on measured or perceived similarities among the patterns. The data in this work are selected moment features and the clusters in the feature space represent a texture class. The test samples within a cluster are more similar to each other than are samples belonging to different clusters. A cluster consists of a relatively high density of points separated from other clusters by a low density of points. Fuzzy C-Means clustering method partitions the Universe F comprised of ‘n’ data samples (feature vectors) into ‘c’ clusters. The number of clusters ‘c’ could be 2 to ‘n’, since c= 1 denotes no cluster and c = n denotes each sample a cluster. The data universe F is

( )5 ... .............,.........,, 321 nffffF =

( )6 ... .............,.........,, 321 imiiii fffff = Unsupervised classification employed in this work aims to find natural groupings or clusters in multidimensional data based on measured or perceived similarities among the patterns. The data in this work are selected moment features and the clusters in the feature space represent a texture class. The test samples within a cluster are more similar to each other than are samples belonging to different clusters. A cluster consists of a relatively high density of points separated from other clusters by a low density of points. Fuzzy C-Means clustering method partitions the Universe F comprised of ‘n’ data samples (feature vectors) into ‘c’ clusters. Thnumber of clusters ‘c’ could be 2 to ‘n’, since c= 1 denotes no cluster and c = n denotes each sample a cluster. The data universe F is denoted as,

( )9 ... ,.... , , 321 vvvvV imiiii =where each entry fi is defined by ‘m’ number of moment features,Fuzzy clustering assigns a membership degree for each of the sample to every cluster. If the membership value is more for a cluster, then the sample belongs to that cluster. Therefore, a sample can have membership to more than one cluster. The clustering starts with an initial partition and proceeds as follows (Ross, 1997):

(i).A new partition is generated by assigning each sample to its closest cluster center.

(ii).New cluster centers are computed as the centroids

of the clusters.

(iii).The above two steps are repeated until an

objective function is minimized.

In this work, each texture image is partitioned into 4096 segments of 8 ´ 8 size segments. Out of which, 1500 segments have been randomly selected and used as training data. Moment featuresof sufficiently higher order are computed for the training data. Feature selection methods were applied on the moment features of training data leading to the identification of a compact set of 13 Zernike moment orders (A00, A11, A22, A33, A66, A77, A88, A99, A11,9, A12,10, A13,11, A14,12, A15,13 ). Composite images with heterogeneous texture regions are created for testing the discriminating power of Zernike moments with FCM clustering. The test images are partitioned into 8 ´ 8 pixel size windows. The moment features are computed for each window resulting in 256 numbers of feature vectors. In case of the selected Zernike moments orders, A00, A11, A22, A33, A66, A77, A88, A99, A11,9, A12,10, A13,11, A14,12, A15,13, computed for the

texture image, then the data universe F of the FCM clustering will have 256 data points. Each data point is then 13 dimensional, i.e. n = 256 and m = 13. After application of the FCM clustering, the final partition matrix is defuzzified. Each cluster is assigned with a pseudocolor. The output image as shown in Fig.2 is reconstructed by concatenating 8´8 pixel windows. The pseudocolor of the window is inferred from the final partition matrix. The classification accuracy achieved with FCM classifier using the moment features is depicted in Table 1. The class boundaries in the feature space are not fully separated and their overlapping is the reason for misclassification.

Table 1. Percentage of Correct Classification with FCM clustering method

Test image Output

Fig.2. Fuzzy C Means Clustering of test image

Pigskin

Sand

T_ima9

Calfskin

Pic2

Fabric

IV. CONCLUSIONAn eight-texture class classification problem is considered in this work. Experiments have been performed with selected feature vectors of Zernike moments, using FCM clustering. The experimental results establish that Zernike moments are efficacious texture descriptors.

V. REFERENCES[1] Vangool.L., Dewaele.P., Oosterlinck.A., “Texture analysis

Anno.1983.”, Computer Vision, Graphics and Image Processing ,29, 336-357, 1985.

[2] Materka.A., Strzelecki.M., “Texture analysis methods – Areview”, Technical University of Lodz, Institute of Electronics: COST B11 Report, Brussels, 1998.

[3] Teague.M.R., “Image analysis via the general theory of moments,” J.Opt.Soc.Am. ,70, 920-930, 1980.

[4] Zernike.F., Physica ,Vol.1., 689, 1934.

[5] Brodatz.P., “Textures: A photographic album for artists and designers”, New York, Dover, 1966.

[6] Ross.T.J., “ Fuzzy logic with Engineering applications”, Mc-Graw Hill, 1997.

PERFORMANCE EVALUATION OF C.I. ENGINE USINGJATROPHA-DIESEL BLENDS

Y. B. MATHUR+, ABDUL SAMAD*, RANJEET SINGH**, G.VYAS#

+ Lecturer, Department of Mechanical Engineering, Govt. Polytechnic College, Bikaner* Asst. Professor, Department of Mechanical Engg. Marudhar Engineering College, Bikaner

**Asst. Professor, Department of Mechanical Engineering, Engineering College, Bikaner# Asst. Professor, Department of Ceramic Engineering, C.E.T, Bikaner

ABSTRACT:

Petroleum fuels like diesel and gasoline are the primary sources of energy for sustainable development of any country in the world. Increasing industrialization, growing energy demands, limited reserves of fossil fuel, and increasing environmental pollution have joined necessitated exploring some alternative of conventional petroleum fuels. Among all renewable sources vegetable oils from both edible and non-edible plants such as jatropha, karanja, sunflower, rasp seed, neem, jojoba, mahua, coconut, caster, palm etc., exhibits a favorable alternative source for diesel engine operation.In the present investigation, jatropha oil and its blends with petro-diesel in various proportions were used in single cylinder 4-stroke C.I. engine to determine performance characteristics under various operating conditions. It has been practically observed that jatropha oil blended with diesel up to 20-25 % levels can be used successfully in diesel engines with no major detoriation in engine performance.

Key Words:

Diesel Engine, Vegetable Oil, Jatropha Oil, Substitute Fuel and Performance.

1. INTRODUCTIONIncreasing petroleum crises and the stringent

regulations on exhaust gas emissions has led to the

search for renewable, more efficient, cost effective

and environmental friendly alternative fuel for

compression ignition engines. Although there are

wide ranges of alternative fuels available, vegetable

oils like jatropha oil, karanja oil, castor oil, jojoba oil,

cotton seed oil, neem oil, mahua oil, palm oil,

soybean oil, and sunflower oil etc. are being

investigated as potential substitute for current high

polluting fuels obtained from conventional sources to

meet fuel crises. Vegetable oils have high-energy

content,comparable calorific values, non-toxic and

non-flammable in nature facilitates easier and safe

handling. Vegetable oils are less polluting renewable

fuels having properties quite compatible with petro-

diesel [1-3]. The major problem with direct use of

vegetable oil as a fuel in compression ignition engine

is there high viscosity. If the viscosity is reduced to

approximately that of petrodiesel, they can be used in

diesel engines without modification to the engine.

One of the possible methods to overcome the

problem of high viscosity is blending of vegetable oil

in proper proportions with diesel.

various blends with diesel were used to determine

engine performance in the present study [4-7].

Jatropha oil is considered to be one of the most

promising substitutes to diesel fuel [8-9]. The

properties of jatropha oil are quite close to diesel

[10]. Some of the important properties of jatropha oil

and petro-diesel are compared in table 1.

2. EXPERIMENTAL SETUP

Kirloskar AV1 naturally aspirated single cylinder

four-stroke water-cooled direct ignition diesel engine

developing 3.70 kW at 1500 revolution per minute

was used for the investigation. The engine

specifications are given in table 2.

The engine was coupled to the electrical

dynamometer for measuring the brake power of

engine. The fuel flow rate was measured on the

volumetric basis using burette and stopwatch. Air

consumption was obtained with the aid of U tube

menometer. Thermocouples in conjunction with a

digital temperature indicator were used to measure

temperatures. Digital tachometer is used to measure

the engine revolution. The experimental setup

employed for the present work is shown in figure 1.

3. EXPERIMENTAL PROCEDUREExperiments were initially carried out on the engine-using diesel as a fuel in order to provide base line data, subsequently experiments were conducted with blends of jetropha oil and diesel in different proportions and with the pure jetropha oil. The engine was stabilized before taking all measurements

and during the entire investigation the injection timing and injection pressure were set at 23 degree BTDC and 200 bars respectively. Different parameters of the engine such as brake power, brake specific fuel consumption, thermal efficiency, volumetric efficiency etc. have been determined with diesel oil, jetropha oil and various blends of diesel and jatropha.

.

4. RESULTS AND DISCUSSION The comparison of brake thermal efficiency of engine

with diesel oil, jatropha oil and various blends of

diesel and jatropha oil at varying load is illustrated in

figure 2. From the experimental results it was

observed that there was considerable increase in

efficiency with blends compared to pure

jatropha oil.

This is probably due to higher volatility, low density

and low viscosity of the blends leads to better

atomization and better combustion of fuel as

compared to pure jatropha oil. The blend with 20%

jatropha oil and 80% diesel showed the brake thermal

efficiency very closed to diesel mode operation.

Figure 2. Comparison of brake thermal efficiency for diesel oil, Jatropha oil and various blends

The variation of specific fuel consumption with

diesel, jatropha oil and various blends of jatropha oil

and diesel are presented in figure 3. It is reveled from

the figure that specific fuel consumption of jatropha

oil as well as various blends was decreased with

increase in load. The specific fuel consumption was

found to increase with higher proportion of jatropha

oil in the blend. The blend containing 20% jatropha

oil exhibited comparable specific fuel consumption

with diesel oil operation.

Though the blends as well as jatropha oil maintained

a similar trend to that of diesel, the value of specific

fuel consumption for jatropha oil was higher

compared to that of diesel in the entire load range.

This is attributed to the fact that the low calorific

value and high density of the blend containing a

higher percentage of jatropha oil might lead to more

discharge of fuel, there by increasing specific fuel

consumption.

Figure 3. Variation of BSFC with load for diesel, Jatropha Oil and various blends

Figure 4. Variations of volumetric efficiency with load

The figure 4 shows the variation of volumetric

efficiency with diesel, Jatropha oil and various

blends. Volumetric efficiency with pure jatropha oil

found to be higher than the diesel

and various blends.

5. CONCLUSIONS

The significant conclusions from the present

experimental investigation are summarized below:

1. Neat Jatropha oil results in slightly reduced

brake thermal efficiency as compared to

diesel fuel and the specific fuel consumption

of jatropha oil is greater then that of diesel

fuel operation under similar condition

2. The blends of jatropha oil and diesel fuel up

to 20-25% by volume competed favorably

with diesel fuel and offer a satisfactory

substitute for diesel fuel. Jatropha oil can be

adopted as alternative fuel for existing diesel

engines without any engine modifications.

Table 1: Properties of Jatropha oil & diesel

Table 2: Engine Specifications

6. REFERENCES

1. A.K. Hossain, P.A. Davies:” Plant Oils as Fuels for Compression Ignition Engines: A Technical Review and Life-cycle Analysis”, Renewable Energy, Volume 35, 2010, Pages 1–13.

2. Barnwal B K, Sharma M P: “Prospects of Biodiesel Production from Vegetable Oils in India”, Renewable and Sustainable Energy Reviews, Volume 9, 2005, Pages 363–378.3. Augustus G D P S, Jayabalan M, Seiler G J : “Alternative Energy Sources from Plants of Western Ghats (Tamil Nadu, India)”, Biomass and Bioenergy, Volume 24, Issue 6, June 2003, Pages 437-444.

4. Pramanik K: “Properties and Use of Jatropha Curcas Oil and Diesel Fuel Blends in Compression Ignition Engine”, Renewable Energy, Vol. 28, Issue 2, Feb 2003, Pages 239-248.

5. Forson F K, Oduro E K and Donkoh Hammond E: “Technical Note: Performance of Jatropha Oil Blends in a Diesel Engine”, Renewable Energy, Vol. 29, Issue 7, June 2004, Pages 1135-1145.

Jatropha oil (preheated and blends) in a Direct Injection Compression Ignition Engine.” Applied Thermal Engineering, Volume 27, Issue 13, September 2007, Pages 2314-232.

1. Deepak Agarwal, Avinash Kumar Agarwal : “Performance and Emissions Characteristics of 2. Avinash Kumar Agarwal : “Biofuels (alcohols and biodiesel) Applications as Fuels for Internal Combustion Engines”, Progress in Energy and Combustion Science , Volume 33, 2007, Pages 233–271.

3. Giibitz G M, Mittelbach M, Trabi M : “Exploitation of the Tropical Oil Seed Plant Jatropha Curcas L”, Bioresource Technology, Volume 67, Issue 1, January 1999, Pages 73-82.

4. Kandpal J B, Madan Mira : “Jatropha Curcus : A Renewable Source of Energy for Meeting Future Energy Needs”, Renewable Energy, Volume 6, Issue 2, 1995, Pages 159-160.

5. Openshaw Keith : “A Review of Jatropha Curcas : an Oil Plant of Unfulfilled Promise”, Biomass and Bioenergy, Volume 19, Issue1, July 2000, Pages 1-15.

PERFORMANCE OF VARIABLE COMPRESSION TEST RIG USING ALCOHOL AS FUEL

DR (MISS) S L SINHAAssociate Professor, Department of Mechanical Engineering, National Institute of Technology,

G E Road, Raipur, Chhattisgarh, India-492010, tel. 919826315780, Fax: 07712254200Email: [email protected]

& MR R K YADAV

Assistant Professor,Department of Mechanical Engineering, National Institute of Technology, G E Road, Raipur, Chhattisgarh, India-492010, tel. 919926555321, Fax: 07712254200

Email: [email protected]:Ever increasing consumption of fossil fuel and petroleum products has been a matter of concern for the country for the financial growth as well as environmental pollution abatement.Public are raising their concerns over the declining state of environment and health. The situation offers us a challenge as well as opportunity to look for substitutes of fossil fuels for both economic and environmental benefits to the country.Ethanol is one such substitute that can be produced from sugarcane and used in different blends with gasoline for automobiles. It has higher value of octane number which provides facilities of anti-knocking. But it has higher vapour lock tendency which create clogging problem in engine. It is a bio-fuel, therefore there will be no emission of NOx, CO, SOx etc. In the present investigation, different blends of ethanol in conventional petrol have been tested on the engine (Variable Compression Test Rig) and engine characteristics have been analysed. It has been observed that performance of engine is improved for 20% blend of ethanol in petrol.Keywords:

Blend, Petrol, Variable Compression test rig.

1. INTRODUCTIONThe world is presently facing the problems of depletion of fossil fuel and environmental degradation due to over use of underground based carbon resources. The search for an alternative fuel is the urgent need of present problems.The fuels of bio-origin such as alcohol, vegetable oil, bio-mass and biogas can be used directly in agricultural and transportation sector. Bio fuels are abundantly available alternative fuels, and have the potential to be produced from bio-mass sources. Alcohols (methanol and ethanol) along with their gasoline blends, have received considerable attention as alternative fuels. The main disadvantage of alcohol fuel, from an air quality stand point is the production of aldehydes during combustion. Under cold starting, alcohols decompose and produce aldehydes which pollutes the atmosphere. Alcohol (methanol and ethanol) along with their gasoline blends, have received considerable attention as alternative fuels ,the main disadvantage of alcohol fuel, from an air quality stand point is the production of aldehydes during combustion under cold start conditions. Alcohols decompose and produce aldehydes which

pollute the atmosphere. Ethanol is a pure substance. But gasoline is composed of C4-C12 hydrocarbon and has wider transitional properties. Due to the presence of an oxygen atom, it is considered as partially oxidized hydrocarbon.2. REVIEWThe use of alcohols and alcohol /petroleum blends in diesel has been shown to reduce the emissions of the potentially carcinogenic soot particles [1,2]. The studies of gasoline fueled conventional vehicles and M-85 fueled variable fuel vehicles showed the 37% reduction of organic emissions with M85 than with M-0 [3]. In this study, CO was observed to be 31% lower with M-85 than with M-0, where as NOx was found to be 23% higher. The emission of the hazardous gases were also observed to be lower.In comparison, methanol’s combustion product formaldehyde reacts much more rapidly to form ozone [4]. Thus the O3 forming tendency of methanol fueled emission is directly related to formaldehyde emissions which is directly proportional to methanol content [5]. Formaldehyde emitted from methanol fueled engine is highly soluble in aqueous aerosols and is available in the atmosphere as hydrated species called methylene glycol [6,7].

It react with SO2 emissions and can form hydroxy-methane. Sulfonate ions, while aqueous oxidation forms formic acid, both of which are highly toxic [8].It is found that the ethanol has had the greatest application out of all oxygenated fuels. It can also split in acetaldehyde during combustion. The use of gasohol (10% ethanol in gasoline) in motor vehicles have reduced the hydrocarbons, CO content in exhaust with an increase of NOx in it. The evaporative emissions of hydrocarbons also increased when ethanol was added to the fuel. A number of studies have been carried out to study the effect of emissions of engine when alcohols were used. In Brazil, blend of 20% alcohol in gasoline was used in motor vehicle with or without use of catalytic converter. It is found that HCHO and CH3CHO was emitted more in the atmosphere [9,10]. Similar results were observed in New Mexico also but level of aldehydes were found reduced when the 10% alcohol blend was used in vehicle with the use of catalytic converter [11,12]. The similar study were conducted in Dever, Colorado and it is concluded that oxygenated fuel usage has not had a major effect on atmospheric concentration of acetaldehyde [13,14].

3.PROPERTIES OF ALCOHOL & PETROL: A COMPARISON

The table 1 shows the comparison of petrol, methanol and ethanol. It is found that alcohols can be easily used as an alternative fuel as it behave much like petroleum and so, it can be stored and shifted in the same manner. It is more flexible fuel than hydrocarbon fuels permitting wider variation from ideal air fuel ratios.

4. RESULTS & DISCUSSIONThe performance test was conducted on Variable Compression Test Rig of Internal Combustion Engine of Mechanical Engineering Department at National Institute of Technology, Raipur, Chhattisgarh, India. The specifications of the engine are given below: Rated horse power 2.5HP Speed 2000 rpm Compression Ratio 10:1 to 2.5:1 Cooling Medium Air Dynamometer Hydraulic Type Hanger ( Load) 10.5cm Length

The performance parameter such as fuel consumption and exhaust temperature were measured by running the engine using petrol.

The same experimental procedure was repeated with alcohol and different blends of alcohol (5%, 10%, 15% & 20% ) with petrol. Figure 1 shows the variation of exhaust gas temperature for different blends of alcohol (0-20%) in petrol for different compression ratio (CR). It is observed that maximum exhaust gas temperature is found at about 15% of alcohol in petrol for compression ratio CR=5.5. As CR (4.5 – 5.0) of engine increases, maximum exhaust gas temperature also increases for the blends of alcohol (5%-10%). The increase in exhaust gas temperature clearly shows that due to presence of oxygen atom in alcohol, combustion of fuel is proper, it not only increases engine output but also reduces hydrocarbon emission.Figure 2-6 shows the variation of engine power with speed of engine for different compression ratio and for different blends of alcohol with petrol. The brake power at output shaft is observed to be very high when engine is fed with pure petrol without any kind of mixing. It is due to high calorific value of fuel. It is also observed that as % content of alcohol increases, engine brake power reduces. It is found that as compression ratio increases, brake power output also increases due to reduced value clearance volume. It increases proper mixing and combustion efficiency. As the engine speed increases, power output also increases up to 5% blend of alcohol. Beyond 5% alcohol content, brake power increases up to engine speed of 2400 rpm. After that it reduces up to 2600 rpm and then, it increases for CR=4.5 -5.0. Brake power increases, as the engine speed increases, but it reduces beyond 2500 rpm for CR=5.5. At 15 % blend of alcohol with petrol, brake power increases for CR=5 & 5.5, as the engine speed increases. Slight variation in brake power is observed when engine speed increases beyond 2500 rpm. For CR=4.5, brake power reduces, as the engine speed increases up to 2700 rpm, beyond this value, brake power increases.At 20% blend of alcohol for CR=4.5, there is no remarkable effect on brake power as the engine speed increases. For CR=5 & 5.5, brake power increases up to engine speed of 2500 rpm and after that slight variation in brake power is observed. Figure 7-11 shows the variation of brake specific fuel consumption (BSFC) with engine speed for different compression ratio and different blends of alcohol in petrol. It is observed that as CR increases, BSFC reduces as the higher value of CR ensures proper combustion in the combustion chamber. As the engine speed increases, BSFC also reduces.As the alcohol content increases in the fuel, BSFC reduces for the given speed.

5. CONCLUSIONAlternate fuels for spark ignition engines have become increasingly important due to decreasing petroleum reserves and the environmental consequences of exhaust gases from petroleum fuel engines. Alcohol can be adopted as an alternative fuel for the existing conventional spark ignition (SI) engine.The engine performance of a SI engine have been investigated by using ethanol-gasoline blends and pure gasoline. It is observed that when ethanol gasoline blend is used, the engine power and fuel consumption of the engine slightly increases. The engine performance improves as alcohol percentage increases in the fuel. It is observed that as the compression ratio of engine increases for different composition of alcohol and petrol, engine efficiency also increases. It is found that about 20% blend of alcohol in petrol gives the maximum efficiency. Alcohol fuel can be used along with the petrol in high compression ratio engines. It can also be mixed in petrol and can be used in engines without any modification. 6.REFERENCES[1] Gaffney, J.S., Sapienza, R., Butcher, T., Krishna, C., Marlow, W., O’Hare, T., 1980,” Soot reduction in diesel engines: a chemical approach. Combustion science and technology”, 24, 89-92.

[2] Wang, W.G., Clark, N.N.,Lyons, D.W., Yang, R.M., Gautam, M., Bata,M., Loth,J.L., 1997, “Emissions comparisons from alternate fuel buses and diesel buses with a chassis dynamometer testing facility”, Environmental science and technology 31, 3132-3137.

[3] Auto/Oil Air Quality Improvement Research Program, 1992, “Gasoline Reformulation and Vehicle Technology Effects on Exhaust Emissions”, In: Technical Bulletin Number 17. Coordinating Research Council, Atlanta, Georgia, pp. 1–18.

[4] Atkutsu, Y., Toyoda, F., Tomita, K.-I., Yoshizawa, F., Tamura, M., Yoshida, T., 1991, “Effect of exhaust from alcohol fuel on ozone formation in the atmosphere”, Atmospheric Environment 25A, 1383–1389.

[5] Gabele,P.A., 1990, “Characterization of emissions from a variable gasoline/methanol fuelled

car”, Journal of the Air and Waste Management Association 40, 296-304.

[6] Klippel, W., Warneck,P., 1980, “Formaldehyde content of the atmospheric aerosol” Atmospheric Environment 14, 809.

[7]Finlayson-Pitts,B.J.,Pitts Jr.,J.N., 1986, “ Atmospheric chemistry: fundamentals and experimental techniques.John” Wiley & Sons, New Yark.

[8] Boyce, S.D., Hoffmann, M.R., 1984, “Kinetics and mechanism of the formation of hydroxyl methane-sulfonic acid at low pH” Journal of Physical Chemistry 88, 4740–4746.

[9] Tanner,R.L., Miguel,A.H., deAndrade,J.B., Gaffney, J.S., Strit, G.E.,1988, “.Atmospheric chemistry of aldehydes: Enhanced peroxyacetylnitrate formation from ethanol-fueled vehicular emissions”, Environmental science and technology 22, 1026-1034.[10] Grosjean, D., Miguel, A.H., Taveres, T.M., 1990, “Urban air pollution in Brazil: acetaldehyde and other carbonyls”, Atmospheric Environment 24B, 101–106.

[11] Popp, C.J., Zhang, L., Gaffney, J.S., 1995,” Organic carbonyl compounds in Albuquerque, New Mexico air. A preliminary study of the effects of oxygenated fuel use”, In: Sterrett, F.S. (Ed.), Alternative Fuels and the Environment. Lewis Publishers, Boca Raton, FL, pp. 61–74.

[12] Gaffney, J.S., Marley, N.A., Martin, R.S., Dixon, R.W., Reyes, L.G., Popp, C.J., 1997, “Potential air quality effects of using ethanol-gasoline fuel blends: a field study”, in Albuquerque, New Mexico. Environmental Science and Technology 31, 3053–3061.

[13] Anderson, L.G., Wolfe, P., Barrell, R.A., Lanning, J.A., 1995, “The effects of oxygenated fuels on the atmospheric concentrations of carbon monoxide and aldehydes in Colorado”, In: Sterrett, F.S. (Ed.), Alternative Fuels and the Environment. Lewis Publishers, Boca Raton, FL, pp. 75–102.

[14] Anderson, L.G., Lanning, J.A., Wilkes, E., Wolfe, P., Jones, R.H., 1997, “Effects of usingoxygenated fuels on carbon monoxide, formaldehyde and acetaldehyde concentrations in Denver.”, Paper 97-RP139.05. In: The Air and Waste Management Association 90th Annual Meeting, Toronto, Canada, 11 pp.

Brief Profile of Authors 1. Dr S L Sinha has completed Bachelor, Master, and Ph D degree in Mechanical Engineering. She is a life member of Institution of Engineers (India) and Indian Society for Technical Education and has published 46 research papers in International /national journals/proceedings of conferences. 2. Mr R K Yadav has completed Bachelor and Master degree in Mechanical Engineering. He is a life member of International Journal of Multidisciplinary Research and Advances in Engineering and has published 5 research papers in International /national journals/proceedings of conferences.

Table 1 Properties of alcohols and petrol

Figure1: Variation of Exhaust Gas Temperature with

different blends of alcohol in petrol for

different compression ratio of engine cylinder

Figure2: Variation of Engine Power with speed for

petrol for different compression ratio of engine

cylinder

Figure3: Variation of Engine Power with speed for

5% blend of alcohol in petrol for different

compression ratio of engine cylinder

Figure4: Variation of Engine Power with speed for 10% blend of alcohol in petrol for different

compression ratio of engine cylinder

Figure5: Variation of Engine Power with speed for 15% blend of alcohol in petrol for different compression ratio of engine cylinder

Figure 6: Variation of Engine Power with speed for 20% blend of alcohol in petrol for different compression ratio of engine cylinder

Figure7: Variation of brake specific fuel consumption with speed for petrol for different compression ratio of engine cylinder

Figure 8: Variation of brake specific fuel

consumption with speed for 5% blend of alcohol in


cylinder




cylinder

Figure10: Variation of brake specific fuel

consumption with speed for 15% blend of alcohol

in petrol for different compression ratio of engine

cylinder




cylinder

IMPLEMENTATION OF IMAGE PROCESSING ALGORITHMS ON FPGA

J.A.SADALGEA, M.D. PATILB, R.K.SHASTRIC

a,b,c Faculty, Vidya Pratishthan’s College of Engineering, Baramati-413 133, Maharashtra, India

ABSTRACT:

This paper describes implementation of image processing algorithms on FPGA. Currently, implementation of image processing algorithms using software approach is slower due to limited speed of the processor. So a dedicated processor for implementation of image processing algorithms is required which is possible with VLSI technology. FPGAs have the advantages of speed and re-configurability which is required for image processing application.[1] This paper deals with image Edge detection and median filter algorithm. This paper presents two edge detection algorithms using Prewitt and Sobel opearators.The aim of this paper is to simulate and implement these algorithms using MatLab and VHDL. The device selected here for implementation is Spartan 3E500K from Xillinx. The proposed edge detection algorithm which executes a matrix area gradient operation to determine the level of variance through different pixels and result is displayed on a monitor. The image is read through MatLab which is input to digital hardware. The results are obtained with 25MHz clock frequency which is able to process 640×480 grayscale image. Comparision of results obtained from VHDL and Matlab is carried out by evaluating PSNR and processing speed of algorithms in MatLab and VHDL.

Keywords:

Edge Detection ,FPGA,VLSI

I. INTRODUCTIONRecently, Field Programmable Gate Array (FPGA) technology has become a viable target for the implementation of algorithms suited to image processing applications. The unique architecture of the FPGA has allowed the technology to be used in many such applications encompassing all aspects of video, image processing .[1] The aim of this paper is to develop FPGA realization for Edge Detection using different techniques such as Sobel, Prewitt and median filtering.Block Diagram:

Figure 1 System Block Diagram

As shown in block diagram of Figure 1, Source image can be fed to the algorithm through interface such as digital camera or any other source if image is pre-captured.

Here source image is read through MATLAB. Digital hardware consists of digital design for Edge detection using mask and threshold specified by user and digital design for median filter. Options like Sobel mask, Prewitt mask are there for selection. The processed image is given to VGA controller [2].VGA controller will generate Color signals (R, G, B), horizontal synchronization and vertical synchronization signals. VGA controllers are designed in FPGA architecture itself. The digital design operates at 25MHz clock frequency which is able to process 640×480 grayscale image. II FPGAs for Image ProcessingWith advances in the VLSI technology hardware implementation has become an attractive alternative. Assigning complex computation tasks to hardware and exploiting the parallelism and pipelining in algorithms yield significant speedup in running times. ASICs represent a technology in which engineers create a fixed hardware design using a variety of tools. Peter A. Ruetz and Robert W. Brodersen, proposed real time image processing Integrated circuits designed using CAD tool[4]. Once a design has been programmed onto an ASIC, it cannot be changed. Since these chips represent true, custom hardware, highly optimized, parallel algorithms are possible.

However, except in high-volume commercial applications, ASICs are often considered too costly for many designs. In addition, if an error exists in the hardware design and is not discovered before product shipment, it cannot be corrected without a very costly product recall. Recently, Field Programmable Gate Array (FPGA) technology has become a viable target for the implementation of algorithms suited to video and image processing applications. Technically, FPGAs are SRAM devices whose unique architecture has allowed the technology to be used in many such applications encompassing all aspects of video and image processing. FPGAs represent reconfigurable computing technology, which is in some ways ideally suited for image processing. Reconfigurable computers are processors which can be programmed with a design, and then reprogrammed (or reconfigured) with virtually limitless designs as the designer’s needs change [5]. FPGAs can be developed to implement parallel design methodology, which is not possible in dedicated DSP designs. ASIC design methods can be used for FPGA design, allowing the designer to implement designs at gate level. However, usually engineers use a hardware language such as VHDL or Verilog, which allows for a design methodology similar to software design. This software view of hardware design allows for a lower overall support cost and design abstraction.

III .Digital Design of Edge Detection

AlgorithmEdge detection is a common approach to detect meaningful discontinuities in grey level. In an image a point is classified as an edge point if magnitude (Gx + Gy) of first order derivative is greater than threshold specified by user. Gx is gradient in X-direction; Gy is gradient in Y direction. This gradient vector is formed by differentiation of image information in two spatial directions [3]. The Figure 2 shows the block diagram of Edge detection Unit. The system is built with major unit such as:

1) Input Memory.

2) Edge Detection Unit.

3) Output Memory.Input MemoryThe input memory is used to store the image data obtained by processing image using Matlab. This memory is used to store the grey level values of maximum of 640×480 sized image.The pixel intensity can vary from 0 to 255.

Figure 2 Edge Detection Unit

Edge detection unit

Edge detection unit consists of 1) X-edge Detector 2) Y-edge Detector 3) Adder4) Thresholding Unit

X-Edge Detector and Y-Edge DetectorThis unit calculates gradient Gx (change in x direction).If Gx value is found to be negative then two’s complement is taken to get the magnitude same method is applied for finding gradient in Y direction. There are different masks by which we can detect edges. The facility of mask selection is provided to this design either Sobel or Prewitt mask. AdderThis is the unit which is used to add magnitude of Gx and Gy. Thresholding unitThis unit does the important operation of thresholding. This also works on the basis of two control signals. Those are clk and threshold. Threshold is user controllable. With lower threshold we get more number of edge pixels as we increase threshold we get less number of edge pixels.Output MemoryThe output memory is used to store edge detected image. This memory contains only ‘0’or ‘255’.These values are then sent to the VGA to display edge detected output on monitor in synchronization with Hsync and Vsync signals. VGA signal generationThe simple VGA controller is designed in FPGA itself. This controller generates three color signals red, green, blue and two control signals Hsync and Vsync .Analog levels between 0V(completely dark) and 0.7V(maximum bright) on the color signals tells the monitor what intensities of these primary colors to combine. Each analog color input can be set to one of four levels by using two digital outputs. So a pixel can have one of 64 different color and only 4 grey shades.

Figure 3 Digital to analog VGA monitor interface

VGA clock frequency and resolution

The VGA controller operates at clock frequency of 25MHz. The actual pixels are sent to monitor within 25.17µs window. With 25MHz clock we get the horizontal resolution of 640 pixels and vertical resolution of 480 pixels.[2]

IV Median Filtering

Median filter is nonlinear filter which depends on ordering the pixels contained in the filter area. This paper employs 3×3 pixel window. In median filter median of 9 values within the window is found and median replaces the central pixel of that window. This window is applied over every pixel in the image. Bubble sort technique is used for arranging the pixels in ascending order.

V. Results

Figure 4 Test Bench waveform of Prewitt edge detection with threshold of 255.

Figure 5 Test Bench waveform of Sobel edge detection with threshold of 255

Figure 6 Result of VHDL simulation using Prewitt Mask.

Figure 7 Result of VHDL simulation using Sobel Mask.

Figure 8 Result of MatLab simulation using Prewitt Mask.

Figure 10 Result of VHDL simulation of Median Filter.

Figure 9 Result of MatLab simulation using Sobel Mask.

Figure 11 Result of MatLab simulation of Median Filter.

Table 1

VI. Conclusion and Future work

This paper demonstrates simulation of edge detection algorithm and median filter.Table1 shows the PSNR values for three algorithms. For MatLab simulation of edge detection algorithm requires 0.203 sec while simulation in VHDL requires 0.296 ms.The hardware implementation of Sobel edge detection algorithm is also carried out. Spartan 3E 500k board from Digilent technology is used for hardware implementation. Hardware implementation is carried out for 50×50 pixel image.Future work would be to interface external memory to FPGA to have larger sized images and extend capabilities of this system to video processing.

REFERENCES[1] Anthony E. Nelson “Implementation of image processing algorithms on FPGA hardware”, Graduate school of Vanderbilt University, May 2000.

[2] VGA signal generation with Xs Board, Application note by D. Vanden Bout, XESS Corporation.

[3] Rafael C.Gonzales,Richard E. Woods ‘Digital image Processing” Pearson Publication.

[4] Peter A Ruetz and Robert W Brodersen,” Architectures and design technique for real time image processing IC’s” ,fellow IEEE, Electronics research library, University of California. [5] Using Reconfigurable Computing Systems Danny Newport and Don Bouldin Electric and Computer Engineering University of Tennessee.

[6] Altera Edge Detection Using SOPC Builder and DSP Bulider Tool flow,Ver 1.0,Application Note377,May 2005. [7] Jemika Malik and Anand Ojha “Design of a VLSI FPGA integrated circuit” Department of Engineering, University of Denver 2005 IEEE Region 5 and IEEE Denver Section Technical, Professional and Student Development Workshop.

A NOVEL AGENT BASED DYNAMIC RESOURCE MANAGEMENT AND JOB SCHEDULING IN GRID COMPUTING

1S. JAGADEESWARI , 2 R. PUNIDHA1Asst. Prof, Dept of CSE Jayaram college of Engineering and Technology2Asst. Prof, Dept of CSE, Jayaram college of Engineering and Technology

1 Mail-Id: [email protected] 1 Mobile: 9047214420ABSTRACT:Grid computing has been a hot spot in recent year. In grid computing applications, resource management and job scheduling are the most crucial problems. In existing system widely used agent based technique is A4 (Agile Architecture Autonomous Agent). The drawback in using this kind of technique in a large-scale multi agent system is that the efficiency of the agents is not predetermined to work together. The Computational Grids provide Resource sharing, such as Personal Computers, Clusters, Data Base and Online instruments. This project explains about a novel agent-based dynamic grid resource management model considering both the resource management and job scheduling as coalition integrity, so as to gain good robust, scalability, efficiency, and high performance. This project use a carefully designed agent to obtain dynamic real time available computational ability of various nodes in the grid environment, so that new job could be assigned to the node that has the largest available computational power, and thus to improve system load ability and performance with good load balance. By constructing all grid computational resources into two-layered Heap Sort Tree (HST), which is based on Heap Sort algorithm; this makes the system more scalable, robust, fault-tolerant and high performance. By taking the advantages of agents in constructing and reconstructing the two-layered HST, this model is well fitted with the unpredictable changing grid environment.

Keywords:Heap sort Tree (HST), Grid Computing, Node State Monitoring Agent (NSMA), Autonomy Presentation Agent (ARA).

1. INTRODUCTION:In today’s pervasive world of needing information anytime and anywhere, the explosive Grid Computing environments have now proven to be significant that they are often referred to as being the world’s single and most powerful computer solutions. It has been realized that with many benefits of Grid Computing, this research have consequently introduced both a complicated and complex global environment, which leverages a multitude of open standards and technologies in a wide variety of implementation schemes. As a mater of fact the complexity and dynamic nature of industrial problems in today’s world are much more intensive to satisfy by more traditional, single computational platform approach. The Grid Computing discipline involves the actual networking services and connections of a potentially unlimited number of ubiquitous computing devices within a “Grid”. The Grid concept is defined as the controlled and coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organizations. This new innovative approach to computing can be most simply thought of as a massively large power “utility” grid, such as what provides power to our homes and businesses each and every day. This delivery of utility-based power has become second nature to many of us, worldwide. We know

that by simply walking into a room and turning on the lights, the power will be directed to the proper devices of our choice for that moment at that time. In the same utility fashion, Grid Computing openly seeks and is capable of adding an infinite number of computing capability and problem resolution tasks within the operational grid environment. As a whole the Grid Computing environment is a single, incredibly powerful, and effective computing solution. The Grid Computing systems provide a solution to complex Job Schedulers and Resource Management. The Grid Computing components such as Schedulers, Resource Broker, Load Balancing and Grid Portals we can concentrate on the critical areas such as Resource Management, Security, Data Management and Information Services. The potential to overcome the major challenges of grid resource management is discussed in [2]. An efficient P2P resource management approach to manage dynamic grid computational resources is proposed in [3]. In [2,4,5] pointed out the importance and feasible ability of agent in solving some key problems in grid computing environment. In [6,7,8] several agent-based resource management allocation model or mechanism and application is discussed. Job scheduling algorithms in different occasions are discussed in [9,10].

The rest of the paper is organized as follows. Section 2 presents the proposed model of agent-based dynamic grid resource management system. Section 3 gives out the algorithm of job scheduling adapt to the model of agent based dynamic resource management system. Section 4 presents an experimental prototype system and gives the analysis of the model and algorithm. Finally, in section 5, we summarize our work and points out future work we are preparing to do.

2. Model of Novel Agent-Based Dynamic Grid Resource Management System:2.1 General Architecture of the Grid Computing System:

General architecture of the grid computing system we considered in this paper is shown in figure 1. As is shown in figure 1, the whole grid computing system is logically divided into three layers – the application layer, the grid resource management layer, and the grid resource layer. In the application layer, grid users send their jobs to the system through the portal of the grid system. The portal then sends the job to the resource management system.

In succession, the grid resource management system selects the proper computational resources from the grids and submits the job to it to be fulfilled. Finally, the selected grid resources carry out the scheduled job and return the result to the requestor. The resource management layer is the crucial part of the whole grid computing system, and it is logically composed of Grid Resource Management Agents (GRMA) in our model. In the following paragraph, we’ll give out the detailed model of the grid resource management system based on agents.

2.2 Model of Novel Agent-Based Dynamic Grid Resource Management System:

The model of novel agent-based dynamic grid resource management system is shown in figure 2 and figure 3. As is shown in figure 2 and figure 3, the proposed model is logically constructed as a hierarchical two-layered Heap Sort Tree (HST), so as to make balanced and effective job scheduling for the grid computing system. In our model, every nodes of the grid computing system has two kinds of agents deployed on it.

The two kinds of agents are: Autonomy Representation Agent (ARA) and Node State Monitoring Agent (NSMA), which are both generally called GRMA in figure 1. The agent of ARA is only activated on the very node of an autonomy whose current available computational ability is the largest among all nodes of the autonomy. All agents of ARA should collaboratively construct themselves into one top layer HST, namely Top-HST (see figure 2), and the root node of the constructed Top-HST is the very ARA whose available computational ability is the largest among the whole grid computing system.

The other kind of agent of NSMA is deployed and activated on all nodes of the grid computing system to dynamically obtain available computational ability of the node on every certain period of time. The way that NSMA obtain MFLOPS of the node is that NSMA periodically runs a set of little benchmarks carefully designed for this case, so as to examine the current available processing ability of the node. All the physically hybrid computing resources in one autonomy area should be cooperatively constructed into one Sub-HST (see figure 3).

The root of the Sub-HST, i.e. ARA node, is the very node that has the largest available computing power among the whole autonomy area, and is selected out to represent the whole autonomy in the Top-HST. Hence, agent of ARA and NSMA are both activated in ARA node, while only the agent of NSMA is activated in NSMA node. So, in our agent-based grid resource management model, ARA is used to represent available computational ability is the largest among all nodes of the autonomy, and ARA is also used to represent the whole autonomy in the top layer of the model.On the other hand, NSMA is used to represent various kinds of hybrid computational resources of the grids, and it should always be activated to monitor the node’s dynamic available commuting power so as to cooperatively construct the Sub-HST for the autonomy. In one word, the agents of SMA and ARA deployed on every node of the grids are respectively activated, so as to dynamically reconstruct the Sub-HST and Top-HST cooperatively, i.e. Sub-HST and Top-HST would always reflect the real time state of the grid computing system.

3. Algorithm of Job Scheduling for the Model oF Novel Agent-based Dynamic GRID Resource Management System:3.1 Data Structure used by ARA and NSMA:In order to obtain high performance and good load balance, we construct the computational grid node into Heap Sort Tree (HST) in both top and sub layer of the resource management system. Both Sub-HST and Top- HST is a binary tree generated during the procedure of ARA and NSMA doing their heap sort respectively. After the Sub-HST being generated, the node with the largest available computational ability among the autonomy is selected to be the root node of the Sub-HST, and it is selected to represent the autonomy to join the Top-HST. In the same way, After the Top-HST being generated, the node with the largest available computational ability among the whole grid system is selected to be the root node of the Top-HST, and it is ready for the scheduler to submit a job to it to be fulfilled effectively.

Figure 4. Data structure used by ARA*MFLOPS: Million Floating Point/Second,

used to indicate the real time available computational ability of the node;

*URL: The universal location information of the resource.

In figure 4, Left Child and Right Child are used to build links between nodes so as to construct Top-HST and Sub-HST for the system, Rsrc Infos is the plain text information of the resource (such as nodes’ name, CPU type and frequency, memory size, network bandwidth, etc.), MFLOPS is used to weigh the available process ability of the node, Flag is a bit symbol to identify whether it is an ARA agent or a NAMA agent, and URL is the universal location information of the resource so that grid system’s job scheduler can submit job to it conveniently.

3.2 Algorithm of Job scheduling in Agent-based Model of Grid Resource Management System:In this sub section, we’ll give out the algorithm of job scheduling based on the agent-based resource management model proposed by us in section 2. The algorithm consists of four major steps:

1) One end user sends a job to the grid system through the portal of the system; 2) The grid computing system begins

to construct the Sub-HST and Top-HST by activating all agent of NSMA and related agent of ARA. We design a sub algorithm named hst_Construct_Alg to construct the HSTs; 3) The job be submitted to the root node of Top- HST, who is ready and has the largest computation power at current time, to be fulfilled. 4) Periodically call step (2) to reconstruct the Sub-HST and Top-HST so that they are always ready for use. The main algorithm described in detail is as following:(1) An end user sends a job to the grid system through the grid portal of the grid system;(2) The grid portal sends the job to the grid system;(3) If (Sub-HST and Top-HST are not ready) then Call the sub algorithm of hst_Construct_Alg to construct Sub-HST and Top-HST for the system;(4) Submit the job to the root node of Top-HST to be executed, and result be returned to the caller;

(5) If (job submit success) Set zero to the value of available computational power of the node selected just now; Else Report error to the caller and end this procedure;(6) If (the prearranged time of reconstructing (i.e. calling hst_Construct_Alg) has not arrived yet) Call hst_Construct_Alg on time to reconstruct Sub-HST and Top-HST;(7) Tell the portal that job has been submitted and ready to receive new request.8) End;The sub algorithm of hst_Construct_Alg is called at five occasions, when: 1) the system are initialed, and Sub- HST and Top-HST are not ready; (8) time of reconstructing Sub-HST and Top-HST arrives; 3) one new grid node joins, 4) one node has been assigned a new job, and 5) fault occurs (such as network, nodes error, and so on), that Sub-HST and Top-HST should be reconstructed.(Normally, the procedure implementing this algorithm periodically runs in the NSMAs of the whole grid system so as to obtain dynamic and real time information of the nodes of the grids. The detailed description of sub algorithm st_Construct_Alg is as following:If ((Sub-HST and Top-HST not ready) or (the prearranged periodic time of reconstructing Sub-HST and Top-HST arrives) or (one

node newly join to the system) or (one node has been submitted a job) or (one node disabled or other error occurs)) (a) Activate all agent of NSMA deplored on all grid resources;(b) NSMA tries to get the static information of the node such as CPU, memory, network bandwidth, etc.;(c) NSMA tries to get the dynamic real time available computational ability (i.e. MFLOPS) of the node by running little benchmarks designed for this case;(d) All geographical neighboring grids resources join themselves into one autonomy area, so as to construct the Sub-HST of this autonomy;(e) If ((two or more Sub-HST has finished constructing) and (not all Sub-HST has finished constructing)) Activate ARAs in root node of all newly finished Sub-HSTs, and make those root nodes be ARA nodes, and construct all the ARA nodes into one Top-HST;(f) Set Sub-HST and Top-HST ready; // End of the outermost If.4. Experiment Evaluation and CommentsWe use Globus Toolkit 4 to construct the grid computing environment. The resource management model proposed by us is based on agent technology. By well utilizing the agent’s advantages, our resource management model is dynamic, scalable, robust and fault tolerant with high performance. The algorithm we design for job scheduling is well suitable for the complex grids environment for it is based on agents. With the help of agents, our algorithm can respond rapidly to the unpredictable fast changing environment of the grids. The use of HST, which is constructed by using well-known heap sort algorithm, makes the algorithm very suitable for the occasion of grids environment, for it has been considered that heap sort algorithm is very effective in large scale sorting applications (such as environment).5. CONCLUSIONSIn this paper, we try to give out a novel agent-based dynamic resource management model and job scheduling algorithm in grid computing environment. We firstly talk about related work done in the area. Then we give out the novel dynamic agent-based resource management model and job scheduling algorithm according the proposed model in grid environment. In succession, we briefly introduce the experimental prototype system based on the model and the general result of experiments.Experiments show that the model and algorithm are feasible and rational. Our work can provide a good paradigm for integrating agent technology with grid application. In the future, we will continue doing more work on transplanting the prototype system to real grid environment and doing more effort to improve the system’s performance.

REFERENCES[1] Ian Foster and Carl Kesselman, “The Grid: Blueprint for a New Computing Infrastructure,” Elsevier Inc., Singapore, Second Edition, 2004.

[2] Huaglory Tianfield, “Towards Agent Based Grid Resource Management”, IEEE International Symposium on Cluster Computing and the Grid 2005 (CCGrid 2005), 9-12 May 2005, Page(s): 590 – 597.[3] Shudong Chen, Xuefeng Du, Fanyuan Ma, Jianhua Shen, “A Grid Resource Management Approach Based on P2P Technology”, Proceedings of the Eighth International Conference on High-Performance Computing in Asia- Pacific Region (HPCASIA’05), 2005, Page(s): 362–369.[4] Carl Kesselman, “Applications of Intelligent Agent Technology to The Grid”, Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence (WI’04), 20-24 Sept.2004 Page(s): 4 – 4.[5] Junwei Cao, Daniel P. Spooner, James D. Turner, Stephen A. Jarvis, Darren J. Kerbyson, Subhash Saini, and Graham R. Nudd, “Agent-based Resource Management for Grid Computing”, Proceedings of the 2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID’02), 21-24 May 2002 Page(s): 350 – 350.[6] Hairong Kuang, Lubomir F. Bic, and Michael B. Dillencourt, “Iterative Grid-Based Computing Using Mobile Agents”, Proceedings of the International Conference on Parallel Processing 2002 (ICPP’02), 18-21 Aug. 2002 Page(s): 109-117.[7] Yanxiang He, Weidong Wen, Hui Jin, Haowen Liu, “Agentbased Mobile Service Discovery in Grid Computing”, Proceedings of the 2005 The Fifth International Conference on Computer and Information Technology (CIT’05), 21-23 Sept. 2005, Page(s): 351 - 355[8] S. S. Manvi, and M. N. Birje, “An Agent-based Resource Allocation Model for Grid Computing”, Proceedings of the 2005 IEEE International Conference on Services Computing (SCC’05), 11-15 July 2005, Page(s): 311 – 314.[9] SHUN-LI DING, JING-BO YUAN, and JI U-BIN JU, “An algorithm for agent-based task scheduling in grid environments”, Proceedings of 2004 International Conference on Machine Learning and Cybernetics, 26-29 Aug 2004, Page(s): 2809 - 2814.[10] Anne Benoit, Murray Cole, Stephen Gilmore, and Jane Hillston, Enhancing the effective utilization of Grid clusters by exploiting on-line performability analysis, 2005 IEEE International Symposium on Cluster Computing and the Grid (CCGrid2005), 9-12 May 2005, Page(s):317 – 324.

[11] The Globus Alliance’s Globus Toolkit Web Page: http://www.globus.org/toolkit/

[12]IBM Aglets web page: http://www.research.ibm.com/trl/aglets/index.html.

http://www.globus.org/toolkit/

PERFORMANCE EVALUATION OF DOUBLE PASS SOLAR AIR COLLECTOR WITH AND WITHOUT POROUS MATERIAL

BHARAT RAMANI1, GUPTA AKHILESH2, RAVI KUMAR3

Department of Mechanical Engineering1, Department of Mechanical & Industrial Engineering2,3

Sardar Patel University1, Indian Institute of Technology, Roorkee2,3

Assistant Professor1(A.D.Patel Institute of Technology), Professor2,3(IIT Roorkee)GUJARAT (INDIA)1, UTTRAKHAND (INDIA)2,3

[email protected] 1 ,[email protected] 2 ,[email protected]

ABSTRACT:

This paper presents an experimental investigation on heat transfer and pressure drop characteristics of double pass solar air collector with and without porous absorber material. Effect of various operational as well as geometrical parameters on thermal performance has been studied. Mathematical model has been developed considering a separate energy balance for solid matrix material and flowing fluid to predict the thermal performance of double pass counter flow solar air collector with and without porous material. Correlations have been developed for Nusselt number and fricational factor. It is revealed that the thermal performance of double pass solar air heating collector with porous material is upto 35% higher than that of without porous material.Key-Words: Solar collector, Double pass, Porous material, Thermal performance.

1. INTRODUCTION

Solar air collectors are widely used for low to moderate temperature applications like space heating, crop drying, timber seasoning and other industrial applications. Conventional solar air collectors have poor thermal efficiency primarily due to high heat losses and low convective heat transfer coefficient between the absorber plate and flowing air stream, leading to higher absorber plate temperature and greater thermal losses. Attempts have been made to improve the thermal performance of conventional solar air collectors by employing various design and flow arrangements [1, 2 3, 4]. The double pass counter flow arrangement with porous material in the second air passage is one of the important and attractive design improvement that has been proposed to achieve the objective of improved thermal performance. The major reason of interest in porous packing includes its high effective heat transfer area per unit volume of packed duct resulting in high heat transfer capability. Also solar radiations are progressively absorbed by the layers of wire mesh and the remaining radiations are absorber by the absorber plate. Thus not only the absorber plate but the porous material also works as the absorber of solar radiation. Therefore, the heat energy due to

absorption of solar radiation is now distributed throughout the porous material and the absorber plate

and dissipated now more effectively to the flowing fluid due to very large surface area being in contact with the flowing fluid in a packed collector as compared to a plane collector where in only the absorber plate surface is in contact with the flowing fluid. The packing material also provides an increased number of sharp corners in the flow passage which enhance the level of turbulence, causing proper mixing and prevent the build-up of slow moving layer of air. Therefore, the heat transfer rate from the porous material to the flowing air is much more resulting in its improved thermal performance. This paper peresents an experimental study involving thermal performance analysis of double pass solar air collector with and without porous material. Effect of porous material on heat transfer and pressure drop characteristics have disscussed.

2 Experimental set upThe schematic diagram of the experimental set-up used is shown in Figure 1. It mainly consists of a test section having two identical ducts, one smooth duct and the other one similar in size but packed with

porous absorbing material (black painted wire mesh screens).Both the ducts have an identical length of 2.1 m, width of 0.54 m and depth of 0.021 m (with L/De=53.8),oriented facing south and tilted with an angle of 30° with respect to the horizontal surface to maximize the solar radiation incident on the transparent glass covers all year round (Roorkee, India). The smooth duct has an absorber plate of 1.0 mm aluminium sheet, blackened with black board paint on the side facing solar radiation. The reason for using two ducts in the present investigation is to compare the performance at the same operating conditions like mass flow rate, insolation and inlet fluid temperature. Flow straighteners have been provided at the inlet side of the test ducts to enable the flow to be stabilized. The length of straighteners is 0.8 m in the accordance with design considerations of Beckman et al. [5, 6].A wooden exit section has been provided at the outlet of the test duct which is followed by a mixing device. Calibrated orifice plate fitted in 78 mm M.S. pipe is used to measure the air flow rate. A standard set of inlet and outlet pressure taps have been provided on

each side of the orifice plate. The temperature distribution in the absorber plate and temperature of the air at inlet, intermediate and at outlet have been measured by means of pre-calibrated copper-constantan thermocouples. A digital temperature indicator is used to display the output of all thermocouples. Air inlet, intermediate and outlet temperatures are also verified by inserting standard thermometer at respective measuring points. The intensity of solar radiation has been measured by means of a Pyranometer (PSP Model). The wind velocity is measured with help of an anemometer.

3 MATHEMATICAL MODELING

The basic equations used to describe the heat transfer characteristics and thermal performance of solar air collectors have been developed from the conservation of energy under quasi-steady state conditions [7]. Energy balance equations for double pass solar air collector with porous material for each component of the system considered is given in the Appendix.

Figure 1 Schematic diagram of the experimental set-up.

Set of these non-linear equations have been solved with help of computer program to compute top glass cover temperature, bottom glass cover temeperature, absorber plate temperature and air flow temeperature in both the passes can be determined.

4 .PERFORMANCE PARAMETERS

The experimental data have been used to determine

the desired parameters as given below. All the

properties of air, i.e. viscosity, density, specific heat,

Prandtl number used in the calculation, are evaluated

at the arithmetic mean of the inlet and the outlet

temperature of air.

The key parameters that affect the performance of solar air collector are useful heat gain and thermal efficiency. For the present system overall useful heat gain can be given as [8]

)( fifopu TTmcQ −=

(1)

Thermal efficiency of solar air collector can be

calculated as under

IATTmc

IAQ

c

fifop

c

u )( −==η

(2)

The porosity of collector duct packed with porous

material has been determined from following relation

[9, 10]( )

tVsVtV

tVoV

p−

==(3)

Using experimenal data correlations for volumetric heat transfer coefficient (hv) and friction factor (fp) have been developed in terms of Reynolds number (Rep), Prandtl number(Pr), duct porosity (p), number of wire mesh layers and pitch to diameter ratio (pt/dw) as given under:

3/121.056.46 2.0 (P r))()())(Re(4 5.0 −−=w

tp

ep

vv d

ppDk ah

(4)

52.091.267.0 )()()(Re25.73 −−−=w

tpp d

ppf

(5)5 RESULTS AND DISCUSSION

5.1 Effect of porous material

The thermal performance of solar air heating collector is low primarely due to higher thermal losses and poor heat tranfer coefficient between aborber plate and flowing air. The porous material can be effectively used to enhance heat tranfer rate in solar air heating collector. The porous material causes proper mixing and prevents the build-up of slow moving layer of fluid next to the wall, and, therefore enhances the rate of heat transfer within the fluid. In the present study an experimental analysis has been carried out for double pass solar air collectors; one packed with porous material and other without porous material. Data have been collected for various operating conditions like mass flow rate, porosity, solar radiation intensity etc. The comparison of results for a specific case has been presented in figure 2. It shows that the thermal

performance of double pass solar air collector with porous material is comparable higher than that of without porous material. It has been also observed that the thermal performance of double pass solar air collector with porous matrix is greatly influenced by the porosity; performance increases as the number of wire mesh layers increases or with decreases in porosity. This may be due to the fact that decrease in porosity increases the effective heat transfer area per unit volume of packed duct. Therefore, thermal performance of double pass solar air collector with porous material is significantly higher than that of without porous material.5.2 Effect of porosity on volumetric heat transfer coefficientBased on the experimental data, the variation of volumetric heat transfer coefficient hv, with mass velocity Go, for various porosity has been presented in figure 3 for a specific case. It has been observed that for all the cases, the volumetric heat transfer coefficient increases with increase in mass velocity due to higher level of turbulence. It has been also observed that the volumetric heat transfer coefficient increases with decrease in the porosity. The decrease in porosity i.e. more number of wire mesh layers increases the amount of material packed in the flow passage, which results in decrease in the volume of voids for flow of air in the duct. Consequently, contact area of flowing air and packing material increases. Also, decrease in porosity reduces flow channeling i.e. increase in tortuousness of the air flowing through the packed duct. Therefore the volumetric heat transfer coefficient between the matrix and air stream increase with decrease in porosity and with increase in mass velocity. 5.3 Effect of porosity on pressure drop

An important factor that has to be considered when employing porous material for the purpose of enhancing heat transfer rate is the penalty arising from the increased pressure drop. The design of good exchanger demands a high heat transfer rate and a low mechanical pumping power needed to overcome the fluid friction (pressure drop) and move the fluid through the porous material. Thus, a very important aspect, the attention should be focused on, when employing porous material for the purpose of heat transfer enhancement is the increased pressure drop. In the present investigation pressure drop has been measured with help of inclined U-tube manometer for all the experimental runs. It is shown from figure 4 that pressure drop increases as mass flow rate increases and with decreased porosity. This is because for lower porosity passage becomes more tortous which results in higher frictional factor.

Figure 2 Comparison of thermal performance

Figure 3 Variation of volumetric heat transfer coefficient with porosity

Figure 4 Variation of pressure drop with porosity

6 .CONCLUSION

The double pass solar air collector with porous material is an important design improvement that has been suggested to achieve enhanced performance. From the present analysis it is concluded that performance of double pass solar air collector with porous material is comparable higher than that of without porous material. The pressure drop arises due to presence of porous material is not very high for lower flow rates and higher porosity.

Nomenclature

Ac Collector area, m2

a Surface area density, m2/m3

Cp Specific heat of air, J/Kg-K

D Depth of solar collector duct, m

Dh,De Equivalent diameter of duct, m

d Wire diameter, m

f Friction factor

h heat transfer coefficient, W/m2-K

k Thermal conductivity, W/m-K

L Length of solar collector duct, m

m Mass flow rate of air, Kg/s

m¢ Mass flow rate, Kg/s-m

n Number of layers

Nu Nusselt number, ( hDh/ k )

p Porosity, (Void vol / total vol.)

Pr Prandtl number, (rCp/m)

Qu Useful heat gain, W

Re Reynolds number, ( GoDh/ µ)

T Temperature, K

V Volume, m3

Greek symbols

a Absoptivity

r Density, Kg/m3

t Transmittivity

m Viscocity, (Kg/m-s)

h Thermal efficiency, (Qu/AcI)

I Intensity of solar radiation, W/m2

Subscript

1 first

2 second

a ambient, air

b bed

f fluid, flow

g glass

i inlet

m matrix

o out, void

p plate, packed

r radiation

s solid, smooth

t total

v volumetric

w wire, wind

REFERENCES:

[1]. A. Ahmad, J.S. Saini, & H.K. Varma, Thermo-hydraulic performance of packed bed solar air heater, Energy Conversion and Management, 37 (2), 205-214, 1996.

[2]. H.P. Garg, G. Dutta & A.K. Bhargava, Performance studies on a finned air heater, Energy, 14, (2), 87-92, 1989.

[3]. M. N. Metwally, H. Z. Abou-Ziyan and A. M. El-Leathy, Performance of advanced corrugated-duct solar air collector compared with five conventional designs, Renewable Energy, 10, (4), 519-537, 1997.

[4]. A.A. Mohamad, High efficiency solar air heater, Solar energy, 60, (2), 71-76, 1997.

[5]. W.A. Beckman, S.A. Klein & J.A. Duffie, Solar heating design by the f-chart method, Wiley, New York, 1977.

[6]. ASHRAE Standard, 93-77, Methods of testing to determine the thermal performance of solar collectors, 1977.

[7]. J.A. Duffie & W.A. Backman, Solar engineering of thermal processes, Wiley, New York, 1980.[8]. S.P. Sukhatme, Solar Energy : principles of thermal collection and storage, Tata McGraw-Hill publishing company Ltd., New Delhi, 9th reprint 2003.

[9]. S. Kakac, R.K. Shah & A.E. Bergles, Low Reynolds number flow heat exchangers, Hemisphere publishing Corporation, New York, 1983.

[10]. S. Whitaker, Forced convection heat transfer correlations for flow in pipes, past flat plates, single cylinder, single sphere and for flow in packed beds and tube bundles, A.I.Ch.E.Journal, 18, 361-371, 1972.

Appendix

FLEXURAL BEHAVIOUR OF GLASS FIBRE REINFORCEDSELF COMPACTING CONCRETE SLABS

DR SESHADRI SEKHAR. T1 DR. SRINIVASA RAO P2 CHANDRA MOULI3

1. Principal , Dr Sammuel George College of Engineering & Technology , Markapur , A.P E-Mail: [email protected]

2. Professor , Dept of Civil Engineering J.N.T.U College of Engineering , Kukatpally , Hyderabad . Mail: [email protected]

3. Principal . Priyadashini College of engineering & Technology for Women , Guntur

ABSTRACT:Self compacting concrete (SCC) is the new category of high performance concrete characterized by its ability to spread and self consolidate in the formwork exhibiting any significant separation of constituents. The elimination of vibration for compacting concrete during placing through the use of Self Compacting Concrete leads to substantial advantages related to better homogeneity, enhancement of working environment and improvement in the productivity by increasing the speed of construction . Understanding of this concrete flow property is of interest to many researchers. Flow properties of concrete at green stage are significantly governed by pastecontent, aggregate volume and admixture dosage. The flow properties of concrete is characterized in the fresh state by methods used for Self compacting concrete, such as slump-flow, V-funnel and L- box tests respectively. Several attempts have been made in the recent years on studies on the behavior of plain SCC. However, only a few attempts have been made on the behaviour of structural elements made of SCC. Also, several studies in the past have revealed the usefulness of the steel fibres to improve the structural properties of concrete like ductility, post crack resistance, energy absorption capacity etc. An attempt has been made in the investigation reported in this paper to study the effect of glass fibres on glass fibre reinforced self compacting concrete using Alkali-Resistant glass fibres on the strength and behaviour of fibre reinforced SCC structural elements subjected to flexure for various grades of concrete mixes of M 30, M 40.Keywords:Ultimate strength, flexural elements, Self Compacting Concrete, Segregation Resistance, Filling ability, Passing Ability, Water-Powder Ratio, Glass fibres.

INTRODUCTION

Fibre reinforced concrete is relatively a new construction material developed through extensive research and development work during the last two decades. It has already found a wide range of practical applications and proved to be reliable construction material having superior performance characteristics compared to conventional concrete. Incorporation of fibre in concrete has found to improve several properties like tensile strength, cracking resistance, impact and wear resistance, ductility and fatigue resistance. Many fibres like asbestos, steel, nylon, coir, etc have been used in the past. Out of these asbestos fibres concrete is successful although its exposure is detrimental to the health of human beings. A steel fibre improves ductility, flexural strength and toughness. Corrosion damage and increased density are the drawback of the steel fibres.Further development in the field of fibre reinforced concrete was due to introduction of

high strength fibres like glass and carbon fibres. The initial studies showed deterioration of glass fibres due to corrosive alkali environment of the cement paste.

The alkali resistant glass fibre, which is developed, recently has overcome this defect and can be effectively used in concrete. The production of fibre reinforced concrete should always be considered in two well defined phases i.e the fresh phase and the hardened phase. Each phase must be considered carefully at mix design stage and each presents its own particular characteristics and related constructional or structural problems. It is necessary to understand the interaction between the fibres and the surrounding matrix in both phases to know how the fibre properties affect the properties of concrete.LITERATURE REVIEWAli R. Khaloo1 et.al investigated on the influence of length and volumetric percentage of steel fibres on energy absorption of concrete slabs with various concrete strengths by testing 28 small steel fibre


reinforced concrete (SFRC) slabs under flexure . Test results indicate that generally longer fibres and higher fibre content provide higher energy absorption. The results are compared with a theoretical prediction based on random distribution of fibres. The theoretical method resulted in higher energy absorption than that obtained in experiment. A design method according to allowable deflection is proposed for SFRC slabs within the range of fibre volumetric percentages used in the study. The method predicts resisting moment–deflection curve satisfactorily. Jeffery R. Roesler2 et.al conducted monotonic load tests on plain and fiber-reinforced concrete slabs on ground to monitor the effect of fiber type and dosage on the strength properties of concrete slabs. The results revealed that simple material tests do not always successfully predict the contribution of fibers in cases where structural geometry and boundary considerations control redistribution of load. The tensile cracking loads of plain and fibrous slabs were found to be similar, which had previously been reported for small- scale fiber specimens, but, there was a significant increase in the flexural strength of fiber-reinforced concrete slabs, relative to plain concrete slabs. Companion beam flexural strength tests also significantly underestimated the concrete slab flexural strength for both the plain and fibrous concrete slabs. The addition of fibers increased the collapse load of slabs, with the key factors affecting the magnitude of the collapse load being fiber type and quantity. Strain and deflection profile measurements showed that fibers assisted in crack propagation resistance, crack bridging, and load redistribution. The shape of the load deflection curves indicated that the synthetic and the steel-fiber-reinforced concrete slabs behaved similarly at different stages of cracking. Uday Kumar3 et.al. investigated p to study the flexural behaviour of RC slabs reinforced with glass fibre reinforced plastic (GFRP) rebars. The RC slabs had a width of 500 mm, depth of 100 mm and had an overall length of 2200 mm. In all, eight numbers of RC slabs, including control specimens reinforced with high strength deformed (HSD) rebars were investigated in the present study. The parameters investigated were two types of GFRP rebars and percentage of reinforcement. The cracking load and load-deflection behaviour up to cracking load were almost the same for the RC slabs reinforced with HSD and GFRP rebars. The load carried by the RC slab reinforced with GFRP rebar was only 40% - 50% of ultimate load of the RC slabs reinforced with HSD rebars at a deflection corresponding to ultimate load of RC slab reinforced with HSD rebar. However, at 50 mm deflection, the load carried by RC slab with plain GFRP rebars was nearly the same as the ultimate load

of the RC slab with HSD rebar. For the RC slab with ribbed GFRP rebars, it was 10% more than that of the RC slab with HSD rebar.

RESEARCH SIGNIFICANCE:For a newly development material like glass fibre self compacting concrete studies on slabs under flexure is of paramount important for instilling confidence amongst the engineers and builders. The literature indicate that while some studies on flexural behavior of steel fibre reinforced concrete slabs are available on normal concrete , a comprehension study which involve flexural behavior of glass fibre reinforced self compacting concrete are not available. Hence, considering the gap in the existing literature, an attempt has been made to study flexural behavior of glass fibre reinforced self compacting concrete.

EXPERIMENTAL PROGRAMME To study the Flexural behavior of glass

fibre reinforced self compacting concrete slabs .

MATERIALS

CementOrdinary Portland cement of 53 grades available from local source was used in the investigation. The cement was tested for various proportions as per IS 4031-1988 and found to confirming to the various specifications of are 12269-1987.The specific gravity was 2.96 and fineness was 3200cm2/gm.Coarse AggregateCrushed angular granite metal of 10 mm size from a local source was used as coarse aggregate. The specific gravity of 2.65 and fineness modulus 6.05 was used. Fine Aggregate River sand was used as fine aggregate. The specific gravity of 2.55 and fineness modulus 2.77 was used in the investigation. The aggregates have been tested for various proportions as per IS 383-1970.Viscosity Modifying Agent A Viscosity modified admixture for Rheodynamic Concrete which is colourless free flowing liquid and having Specific of gravity 1.01+0.01 @ 250C and pH value as 8+1 and chloride content nil was used as viscosity Modifying Agent.AdmixtureThe Modified Polycarboxylated Ether based Super Plasticizer (5)which is brown colour and free flowing liquid and having Relative density 1.08+0.01 and pH value as 7+ 1 and Chloride Content nil was used was Super Plasticizer .

Glass Fibres The glass fibres are of Cem-FIL Anti – Crack HD (6)

with Modulus of Elasticity 72 GPA, Filament diameter 14 microns, Specific Gravity 2.68, length 12mm and having the aspect ratio of 857.1. The number of fibres per 1 kg is 212 million. Fly Ash

Type-II fly ash from Vijayawada Thermal Power

Station was used as cement replacement material. MIX PROPORTIONS OF SELF COMPACTING MIXESIn a rational mix design method for self compacting concrete, the coarse and fine aggregates contents are fixed so that self compact ability can be achieved easily by adjusting the water powder ratio, super plasticizer dosage and viscosity modifying agent dosage only. The mortar or paste in self compacting concrete requires high viscosity as well as high deformability. This can be achieved by using the super plasticizer, which results in a low water powder ratio for high deformability and viscosity modifying agent for high viscosity. Therefore once the mix proportion is decided, self compactability has to be tested by slump flow test, L Box test, V Funnel test. The mix proportioning for self compacting concrete was arrived is illustrated in Table No 1.0. Flexural Behaviour of Self Compacting Concrete and Glass Fibre Self Compacting Concrete slabs: To study the suitability of the self compacting concrete and glass fibre self compacting concrete slabs investigations were carried out ultimate load and load deflection characteristics.Glass fibre reinforced self compacting concrete by varying the percentage of glass fibres from 0 % to 0.1 % of sizes 1400 X 1200 X 100 mm were cast with reinforcement of 8mm diameter HYSD bars with a spacing of 200mm c/c on either side of the slabs by varying the percentage of glass fibres from 0 % to 0.1 %. These specimens were cured in water for 56 days and tested for ultimate load, deflections and failure characteristics under one third point loading. The test setup is shown in plate 1.0. The results are tabulated in Table 2.0 and 3.0.

DISCUSSIONS:Fresh State Properties of Self Compacting Concrete and Glass Fibre Self Compacting Concrete MixesAs it is evident, the basic requirements of high flowability and segregation resistance as specified by guidelines on self compacted concrete by EFNARC are satisfied. The workability values are maintained by adding suitable quantities of super plasticizers.

Effect of Glass fibre on Bleeding On the basis of the experimental study it was concluded that addition of glass fibres in concrete gives a reduction in bleeding. A reduction in bleeding improves the surface integrity of concrete, improves its homogeneity, and reduces the probability of cracks.Slab designations of reinforced self compacting concrete and glass fibre reinforced self compacting concrete mixesTables 2.0 and 3.0 gives the central deflections of M 30 and M40 grade of reinforced self compacting concrete and glass fibre reinforced self compacting concrete slabs over the full depth. The slab designation represents that the slabs S1 to S4 of reinforced self compacting concrete slabs of M 30 grade with 0, 0.03, 0.06, 0.1 % of glass fibres. The slabs S5 to S8 represent self compacting reinforced concrete slabs of M40 grade with 0, 0.03, 0.06, and 0.1 % of glass fibres.Flexural strength of reinforced self compacting concrete slabs The slab 1(M 30 + 0% fiber) failed at a load of 60 KN and the first crack observed to be at 40 KN. The slab 5(M 40+ 0% fiber) failed at a load of 60 KN and the first crack observed to be at 30KN. Flexural strength of glass fibre reinforced self compacting concrete slabs of M 30 grade The slab 2(M 30+ 0.03% fiber) failed at a load of 70 KN and the first crack to be observed at 40KN. The slab 3(M 30+ 0.06% fiber) failed at a load of 60 KN and the first crack observed to be at 40 KN. The slab 4(M30+ 0.1% fiber) failed at a load of 70 KN and the first crack observed to be at 40 KN.

The ultimate flexural load of the slab 2(M 30+ 0.03% fiber) is more than that of slab 1(M 30+ 0% fiber). From the above we can observe that the ultimate flexural strength is observed in a slab having 0.03 % glass fibres . The plate 19 and 20 gives the failure pattern of M 30 grade of glass fibre reinforced self compacting slabs with 0.0% and 0.03% fibres. Flexural strength of glass fibre reinforced self compacting concrete slabs of M 40 grade The slab 6(M 40 + 0.03% fiber) failed at a load of 70 KN and the first crack to be observed at 40KN. The slab 7(M 40 + 0.06% fiber) failed at a load of 70 KN and the first crack observed to be at 30 KN. The slab 8(M30+ 0.1% fiber) failed at a load of 70 KN and the first crack observed to be at 30 KN.

The ultimate flexural load of the slab 6(M 40 + 0.03% fiber) is more than that of slab 5(M 40 + 0% fiber). From the above we can observe that the ultimate flexural strength is observed in a slab having 0.03 % glass fibres .

The plate 3 and 4 gives the failure pattern of M 40 grade of glass fibre reinforced self compacting slabs with 0.0% and 0.1 % fibres.Load deflection characteristics of reinforced self compacting concrete and glass fibre reinforced self compacting concrete slabs

Tables 2.0 and 3.0 gives the experimental investigation for studying the load deflection behaviour of 1400 x 1200 x 100 mm reinforced self compacting concrete and glass fibre reinforced self compacting concrete slabs under 1/3 rd point loading.

Table 3.0 gives ultimate load and deflections at maximum load for both reinforced self compacting concrete and glass fibre reinforced self compacting concrete slabs. From the data and cracks it can be seen that load carrying capacity of glass fibre reinforced concrete slabs having 0.03 % glass fibres are on higher side when compared with other slabs having 0%, 0.06 % and 0.1 % glass fibres.

Variation of load deflection characteristics of glass fibre reinforced self compacting concrete slabs over reinforced self compacting concrete slabs : It can be seen that the presence of glass fibres in reinforced self compacting concrete slabs have not improved in flexure because of the reason that the glass fibres are small in diameter and smooth in nature. The load carrying capacity of reinforced self compacting concrete slabs with fibres at 0.03 % is greater than the slabs without glass fibers . However development of multiple cracks and micro cracks is prevented with the use of glass fibres .

CONCLUSIONS: 1. The Ultimate flexural Strength of glass fibre reinforced self compacting concrete beams at 0.03 % are on higher side when compared with other beams having glass fibres 0%, 0.06% and 0.1 %.2. The presence of glass fibres in glass fibre reinforced self compacting concrete slabs have not improved any flexural strength.3. Development of multiple cracks and micro cracks is prevented with the use of glass fibres.

REFERENCES:

1. Ali R. Khaloo and Majid Afshari Flexural behaviour of small steel fibre reinforced concrete slabs Cement and Concrete Composites Volume 27, Issue 1, January 2005, Pages 141-149

2. Jeffery R. Roesler, M.ASCE David A. Lange, M.ASCE Salah A. Altoubat, M.ASCEKlaus-Alexander Rieder, M.ASCE and Gregory R. Ulreich, M.ASCE5 Fracture of Plain and Fiber-Reinforced Concrete Slabs under Monotonic Loading JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / SEPTEMBER/OCTOBER 2004 pp 452-460.

3. V Udhayakumar, Dr B H Bharatkumar, K Balasubramanian, T S Krishnamoorthy, Dr N Lakshmanan Experimental Investigations on Flexural Behaviour of RCSlabs Reinforced with GFRP Rebars Journal of Institution of Engineers Vol 88, November 2007 pp 23-28.

4. www.efnarc.org The Europeon Guidelines for “Self Compacting Concrete” ,specifications , production and use – May 2005. 5. www.mbt-india.com.

6.www.saint- gobainvetrotex.com.

Table 1.0 Quantities of Materials Required Per 1 Cum Of Self Compacting Concrete Mixes and Glass Fibre Self Compacting Concretes

Table 2.0 Load Deflection Characteristics of Glass Fibre Reinforced Self Compacting Slabs of M 30 Grade.

Table 3.0 Load Deflection Characteristics of Glass Fibre Reinforced Self Compacting Slabs of M 40 Grade

http://www.mbt-india.com/

Plate2.0 Slab 1 M 30 grade SCC with 0% glass fibres

Table: 4.0 Ultimate load and Deflection at Ultimate load of Glass Fibre Self Compacting Concrete Slabs

Plate No 1.0 Test Set UP for measuring Flexural Strength of Slabs.

Plate 3.0 Slab 2. M30 grade SCC with 0.03% glass fibres

Plate 4.0 Slab3. M 40 grade SCC with 0% glass fibres

Plate 5.0 Slab4. M 40 GRADE SCC with 0.1% glass fibres

Biography of the Author:

Dr P. Srinivasa RaoB.Tech., M.Tech., Ph.D , M.I.E., M.I.S.T.E., Professor Specialized in structural engineering. Research interests are Concrete Technology, Structural Design, High Performance Concrete, Prefabricating Structures, Special Concretes and use of Micro Silica, Fly Ash in Building Materials. He has been associated with a number of Design projects, for number of organizations and involved as a key person in Quality control and Mix Designs. Has 24years of academic, research and industrial experience published over 100 research papers. He guided one Ph.D and 75 M.Tech projects. Guiding 5 Ph.D students , delivered invited lecturers in other organizations and institutions. Member of ISTE, Member of ICI and Member of Institute of Engineers.

Dr . Seshadri Sekhar.T

B.Tech., M.Tech. M.S., Ph.D , F.I.E ., F.I.E.T.E., M.I.S.T.E

Graduated from KSRM college of Engineering affiliated to S.V. University, Tirupati in Civil Engineering, and Completed M.Tech in specialization of Structural Engineering from JNTU college of Engineering Hyderabad. Completed M.S.(Software Systems ) from BITS pilani . He completed his Masters thesis in the field of Special Concretes. He had completed his doctoral thesis on Glass Fibre Reinfoced Self Compacted Concrete . He has published 80 research publications.

NETWORK SECURITY USING ELLIPTIC CURVE CRYPTOGRAPHY

MR.S.DHANDAPANI1, MRS.C.KAVITHA2

1 Assistant Professor, ECE, Sree Sastha Institute of Engineering and Technology, Chennai, India2Lecturer, ECE, Vemana Institute of Technology, Bangalore, India

E-mail: kavithanamakkal @ yahoo.co.inE-mail: dhandapani_me @ rediffmail.com

ABSTRACT:

Nowadays the communication system needs the security because the unauthorized eavesdropper is dropping the entire message while transmitting and receiving the data. With the need for information security in today’s digital systems both acute and growing, cryptography has become one of their critical components. Cryptographic services are required across a variety of platforms in a wide range of applications such as secure access to private networks, electronic commerce, and health care. Security depends on the difficulty of mathematical problem, actually the discrete logarithm problem, over the points on an elliptic curve. Elliptic Curve Cryptography (ECC) can be used to provide both a digital signature scheme and an encryption scheme. Put another way, we can use a smaller key to get the same amount of security, which speeds up the computations we want to speed up. It is important to develop a basis for utilizing efficient encryption schemes in wireless communications and in devices with low power and resources. ECC fits well for an efficient and secure encryption scheme. It is more efficient because ECC utilizes smaller key sizes. So memory requirement is less.

Keywords: Elliptic Curve, Security, Cryptography.

1. INTRODUCTIONThe field of cryptographic algorithms can be divided into two basic types, symmetric and asymmetric, which have distinctly different properties. Symmetric or secret-key algorithms require two parties to share some secret piece of information (i.e., the key) that is then used to encrypt/decrypt messages between them. The existence of a shared piece of secret information enables secret-key algorithms to be very computationally efficient. Hence, symmetric algorithms are used to encrypt the bulk of the messages being passed. It is suitable for data communication. Asymmetric or public-key algorithms on the other hand rely on the presumed existence of hard number-theoretic problems that enable two sets of keys to be created: public (encryption) and private (decryption). Public keys are stored in the open so that anyone can encrypt a message. However, because of the number-theoretic properties of the algorithms used, only the intended recipient who generated the public–private key pair can decode the message correctly. Hence, no secret needs to be shared by the communicating parties. Thus, public-key algorithms are used primarily for establishing secret keys throughout the network in a secure manner.

It is suitable for private key exchange and authentication. ECC is the best candidate for the public key cryptography algorithm for its encryption efficiencies [1].

2. COMPUTATIONAL DIFFICULTYIt is important for cryptographic algorithms to be reasonably efficient for the good guys to compute. The good guys are the ones with knowledge of the keys.Cryptographic algorithms are not impossible to break without the key. A bad guy can simply try all possible keys until one works. The security of a cryptographic scheme depends on how much work it is for the bad guy to break it. If the best possible scheme will take 10 million years to break using all of the computers in the world, then it can be considered reasonably secure. Often a scheme can be made more secure by making the key longer. With cryptography, computers can be used to exhaustively try keys. Computers are a lot faster than people, and they don’t get tired, so thousands or millions of keys can be tried per second. Also, lots of keys can be tried in parallel if you have multiple computers, so time can be saved by spending money on more computers.

Sometimes a cryptographic algorithm has a variable –length key. It can be made more secure by increasing the length of the key. Increasing the length of the key by one bit makes the good guy’s job just a little bit harder, but makes the bad guy’s job up to twice as hard (because the number of possible keys doubles). Some cryptographic algorithms have a fixed –length key, but a similar algorithm with a longer key can be devised if necessary. If computers get 1000 times faster, so that the bad guy’s job becomes reasonably practical, making the key 10 bits longer will make the bad guy’s job as hard as it was before the advance in computer speed, However, it will be much easier for the good guys (because their computer speed increase far outweighs the increment in key length). So the faster computers get, the better life gets for the good guys.

3. PUBLIC KEY CRYPTOGRAPHY

A message is to be transferred from one party to another across some sort of network. The two parties, who are the principals in this transaction, must cooperate for the exchange to take place. Security aspects come into play when it is necessary or desirable to protect the information transmission from an opponent who may present a threat to confidentiality, authenticity, and so on. All the techniques for providing security have two components.

1. A security related transformation on the information to be sent. Examples include encryption of the message, which scrambles the message so that it is unreadable by the opponent, and the addition of a code based on the contents of the message, which can be used to verify the identity of the sender.

2. Some secret information shared by the two principals and, it is hoped, unknown to the opponent. An example is an encryption key used in conjunction with the transformation to scramble the message before transmission and unscramble it on reception.

A trusted third party may be needed to achieve secure transmission. For example, a third party may be responsible for distributing the secret information to the two principles while keeping it from any opponent. Or a third party may be needed to arbitrate disputes between the two principals concerning the authenticity of a message transmission.

Fig.1.Model for network security

Model for network security is shown in Fig.1.This general model shows that there are four basic tasks in designing a particular security service.

1. Design an algorithm for performing the security related transformation. The algorithm should be such that an Opponent cannot defeat its purpose.2. Generate the secret information to be used with the algorithm.3. Develop methods for the distribution and sharing of the secret information. 4. Specify a protocol to be used by the two principals that makes uses of the security algorithm and the secret information to achieve a particular security service.It is based on mathematical functions. Public key cryptography is asymmetric, involving the use of two separate keys. The use of two keys has profound consequences in the areas of confidentiality, key distribution, and authentication. The security of encryption scheme depends on the length of the key and the computational work involved in breaking a cipher. Public key algorithms rely on one key for encryption and a different but related key for decryption. The algorithm is computationally infeasible to determine the decryption key given only knowledge of the cryptographic algorithm and the encryption key.

The general block diagram for cryptography is shown in Fig.2. The elements are described below.

Sender

It is the device, which transmits the original data. The plain text may be a stream of bits, a text file, a stream of digitized video, and a digital video image that is fed into the algorithm as input.

Encryption

A plain text is converted to cipher text. The process of hiding the information by its substances. It performs various substitutions and transformations on the plaintext.

EK(M) =C.Decryption

This is essentially the encryption algorithm run in reverse. It takes the cipher text and the secret key and produces the original plaintext. The process of turning the cipher texts back into plain text. DK (C) =M.

Fig.2.General Block Diagram

ReceiverIt is the device, which receives the data.

EavesdropperIt is extracting the original data without others knowledge.

Cipher text

This is scrambled message produced as output. It depends on the plaintext and the secret key. For a given message, two different keys will produce two different cipher texts. The cipher text is an apparently random stream of data and, as it stands, is unintelligible.Public and private key

This is a pair of keys that have been selected so that if one is used for encryption, the other is used for decryption. The exact transformations performed by the encryption algorithm depend on the public or private key that is provided as input.

4. KEY RANGE AND KEY SIZE

The encryption/decryption algorithm is usually not a secret, everybody knows about it. Also, one can access an encrypted message by various means (such

as by listening to the flow of information over a network). Thus, only the actual value of the key remains a challenge for the attacker. If the key is found, the attacker can resolve the mystery by working backwards to the original plain text message, as shown in Fig.3. Brute force attack works on the principle of trying every possible key in the range, until eavesdropper get the right key. It usually takes a very small amount of time to try a key. The attacker can write a computer program that tries many such keys in one second. In the best case, the attacker finds the right key in the first attempt itself, and in the worst case, it is the 100 billion attempts. However, the usual observation is that the key is found somewhere in between the possible range. Mathematics tells us that on an average, the key can be found after about half of the possible values in the key range are checked. Of course, this is just a guideline, and may or may not work in real practice for a given situation [2].

Cipher text

Try with key 0

Cipher text Plain text

Decrypt Plain text

Cipher text Decrypt Plain text

Try with key =1

Up to

Try with key =100 billion

Decrypt

As the Fig.3.Shows, the attacker has access to the cipher text block and encryption/decryption algorithms. Attacker also knows the key range. Attacker now starts trying every possible key, starting from 0. After every decryption; he looks at the generating plain text. If he notices that the decryption has yielded unintelligent plain text, he continues the process with the next key in the sequence. Finally he found the right key, which yields the plain text. How does the attacker determine if the plain texts, and therefore key range, are the right ones? This can be determined depending on the value of the plain text. If the plain text seems reasonable, it is highly probable that the plain text is indeed what corresponds to the cipher text. This

means that our key and plain text message are now cracked. How can I prevent an attacker from succeeding in such attempts? The solution is expanding or increasing the key range to a size, which requires the attacker to work for more years to crack the key.In computer terms, the concept of key range leads us to the principle of key size. Key size is measured in bits, and represent is using the binary number system. The key size is 40 bits, 56 bits, 128bits and so on. In order to protect against brute force attack, the key size should be such that the attacker cannot crack it within a specified amount of time. At the simplest level, the key size can be just 1 bit. This means that the key can be 0 or 1. If the key size is 2, the possible key values are 00, 01, 10, and 11. Obviously, these examples are merely to understand the theory. Thus every incremental bit, the attacker has to perform double the number of operations as compared to the previous key size. The essential steps are the following,1. Each user generates a pair of keys to be used for the encryption and decryption of messages.2. Each user places one of the two keys in a public register or other accessible file. This is the public key. The companion key is kept private. It suggests, each user maintains a collection of public keys obtained from others.3. If sender wishes to send a confidential message to receiver, sender encrypts the message using receiver’s public key.4. When receiver receives the message, receiver decrypts it using its private key. No other recipient can decrypt the message because only receiver knows receiver’s private key.With this approach, all participants have access to public keys, and private keys are generated locally by each participant and therefore need never be distributed. As long as a system controls its private key, its incoming communication is secure. At any time, a system can change its private key and publish the companion key to replace its old public key. The two keys used for public key encryption are referred to as the public key and the private key. Invariably the private key is kept secret [3].Let us take a closer look at the essential elements of a public key encryption scheme as shown in Fig.4. There is a source that produces a message in plaintext X=[X1, X2…XM]. The M elements of X are letters in some finite alphabet. The message is intended for destination. It generates a related pair of keys: a public key, PB, and a private key nB. An nB is known only to destination, whereas PB is publicly available and therefore accessible by source.With the message X and the encryption key PB as input, source forms the

Cipher text Y= [Y1, Y2 … YN]. (1)

Y=E (PB(X)) (2)The intended receiver, in position of the matching private key, is able to invert the transformation:

X=D (nB(Y)) (3)An opponent, observing Y and having access to PB, but not having access to nB or X, must attempt to recover X and/or nB. It is assumed that the opponent does have knowledge of the encryption (E) and decryption (D) algorithms. If the opponent is interested only in this particular messages as well, in which case an attempt is made to recover nB by generating an estimate nB. Either of the two related keys can be used for encryption, with the other being used for decryption. This enables a different cryptographic scheme to be implemented. It provides confidentiality. Public key cryptography makes it easy with no prior arrangements. Two types of systems are considered secure and efficient: RSA, ECC [4].The RSA scheme is a block cipher in which the plain text and cipher text are integers between 0 and n-1 for some n. A typical size for n is 1024 bits, or 309 decimal digits. It uses larger key sizes, because the continuing increases in computing power, and the continuing refinement of factoring algorithms. Performance is degraded by constant exponentiation time. A larger number of bits in RSA mean wider buses, which adds to the area and power consumption of the design. So I have selected other types of public key algorithm as Elliptic Curve Cryptography. Because it gives security for smaller key sizes, thereby reducing processing overhead. It uses 160 bit sizes. It provides power consumption and acceptable performance.

Fig.4. Public Key Cryptosystem: Secrecy

5. ELLIPTIC CURVE CRYPTOGRAPHY

In 1985, Neil Koblitz and Victor Miller independently proposed the ECC, whose security depends on the difficulty of mathematical problem for their security, actually the discrete logarithm problem, over the points on an elliptic curve. ECC can be used to provide both a digital signature scheme and an encryption scheme. How secure are RSA and Diffie Hellman? There are known sub exponential algorithms for breaking RSA and Diffie Hellman based on modular arithmetic. ECC is important because the mathematicians do not have sub exponential algorithms for breaking it. Therefore, it is believed to be secure with much smaller key sizes, which is important for performance. ECC is a candidate replacement for public key cryptography schemes. For some of these cryptographic schemes, one can replace modular multiplication by elliptic curve multiplication directly, resulting in algorithms referred to as ECC Diffie Hellman [4].An elliptic curve is a set of points on the coordinate plane satisfying an equation of the form Y2+aXY+bY=X3+cX2+dX+e. In order to use elliptic curve for, say Diffie Hellman, there needs to be some mathematical operation on two points in the set that will always produce a point also in the set. Let’s call that operation multiplication, although in ECC it will not look like the multiplication you are used to. And the operation has to be associative, so that you can use the repeated squaring trick to raise a number to a large power in time linear with the length of the exponent. And it is also important that doing discrete logs is hard. ECC has the following benefits are ECC has storage efficiencies, bandwidth savings

and computational efficiencies. Therefore an implementation of ECC is particularly beneficial in applications where bandwidths, processing capacity, power availability (or) storages are constrained.

ECC crypto module requires less computational overhead than RSA and also number of gates of ECC is much less than RSA. That is, ECC is more efficient and less memory.

Consequently, ECC is more appropriate for use on secure devices with constrained power consumption and memory resources.

The private key in a public key pair must be kept secret. The private key must be completely inaccessible to all other parties.

However, with ECC, the time needed to generate a key pair is so short that even a device with the very limited computing power.

ECC has the following benefits that make it particularly suitable for embedded applications.

ECC hardware implementations use fewer transistors. As an example, a VLSI implementation of a 155 bit ECC processor has been reported that uses only 11,000 transistors, compared with an equivalent strength 512 bit RSA processor that used 50,000 transistors.

ECC offers the highest security per bit of any known public key cryptosystem so a smaller memory can be used.

Hardware requirement is saved and speed is improved by ECC.

In recent years, ECC has gained widespread exposure and acceptance and has already been included in many security standards. Engineering of ECC is a complex, interdisciplinary research field encompassing such fields as mathematics, computer science and electrical engineering. ECC is being used to secure many applications in wireless communications, classified government communications, digital rights management, and digital post marks and check clearing [5].

A. Arithmetic In An Elliptic Curve Over ZP

1. Negative of point PP= (XP, YP)-P= (XP, -YP)

2. Adding distinct points P and Q

If P (XP, YP) and Q (XQ, YQ) are distinct points such that P ‘“ -Q, thenP+Q = R (XR, YR) is determined by the following rules:

XR = (λ2- XP- XQ) mod P (4)YR = ( (XP – XR) - YP) mod P (5)

λ= (YQ – YQ) / (XQ – XP) mod P (6)Where, λ is the slope of the line through P and Q.

3. Doubling the point PIf P and Q are equal, provided that YP is not

0.2P=RXR = (λ2- 2XP) mod P(7)YR = ( (XP – XR) - YP) mod P(8)

λ= (3 XP 2 +a) / (2YP) mod P (9)

For determining the security of various elliptic curve ciphers, it is of some interest to know the number of points in a finite abelian group defined over an elliptic curve. In the case of the finite group EP (a, b), the number of points N is bounded by

P+1-2"P d” N d” P+1+2"P (10)

So that foe large P, the number of points in EP (a, b) is approximately equal to the number of elements in Zp.

4. Curve MultiplicationMultiplication is defined by repeated

addition, i.e.Q = k P (11) = P+P+P…..+P K times

5. Discrete logarithm problem

Elliptic curve cryptography is based on the discrete logarithm problem applied to elliptic curves over a finite field. In particular, for an elliptic curve EP (a, b), it relies on the fact that it is easy to compute Q = k PP, Q ε EP (a, b). However, there is currently no known sub exponential time algorithm to compute k given P and Q. It can be used to build cryptosystems with any finite abelian group. It is the main strength of elliptic curve cryptosystems [1].ECC Key Exchange

Pick a prime number p, elliptic curve parameters a and b for equation Y2 =x2 +ax +b

(12)

This defines the elliptic group of b points Ep (a, b). Next, pick a base point G = (x, y) in Ep (a, b) whose order is a very large value n. such that nG=O

(13)Ep (a, b) and G are parameters of the cryptosystem known to all participants. ECC key exchange is shown in Fig.5.

ECC Secret Key Exchange

ECC secret key exchange is shown in Fig.6. Both users A, B agrees on the elliptic curve E, a point P on E. Each user chooses a secret key from GF (2^m), nA, nB and calculate own public key PA*nB,

PB*nA and sends it to other user. At this point both users can calculate the secret point S (xs, ys).

Fig.5.ECC Key Exchange

S (xs, ys) = nA*nB*PA = nB*nA*PB

(14)

Both xs, ys are available, only one of them should be used for higher security.

A. ECC ENCRYPTION/DECRYPTION

Fig.7. shows the ECC encryption/Decryption process. To encode the plain text message m to be sent as an x-y point Pm. It is the point Pm that will be encrypted as a cipher text and subsequently decrypted. Note that we cannot simply encode the message as the x or y coordinates of a point, because not all such co ordinates are in Ep (a, b). As with the key exchange system, an encryption/decryption system requires a point G and an elliptic group Ep (a, b) as parameters. Each user A selects a private key nA and generates a public key

PA =nA*G (15)

Encryption

To encrypt and send a message Pm to B, A chooses a random positive integer k and produces the cipher text Cm consisting of the pair of points

Cm= k*G, Pm+kPB (16)

A has used B’s public key PB.

Fig.7. ECC Encryption/decryption process

Decryption

To decrypt the cipher text, B multiplies the first point in the pair by B’s secret key and subtracts the result from the second point.

Pm+(k*PB)-nB (k*G) =Pm+k (nB*G)-nB (k*G) =Pm (17)A has masked the message Pm by adding k*

PB to it. Nobody but A knows the value of k, so even though PB is a public key, nobody can remove the mask k*PB. However, A also includes a ’clue’, which is enough to remove the mask if one knows the private key nB. For an attacker to recover the message, the attacker would have to compute k given G and k*G, which is assumed hard.

VI RESULT

Given P=751, EP (-1,188), curve is Y2 = X3- X +188 G = (0,376), Pm = (562, 201), k = 386, PB = (201, 5)

Encryption

Cm = kG, Pm+kPB

Calculation procedure:

Cm = 386 (0, 376), (562, 201) + 386(201, 5)

= (676, 558), (562, 201) + (239, 377) = (676, 558), (562, 201) +431 (562,

201)

= (676, 558), 432 (562, 201)Cm = (676, 558), (385, 328)

To find kG, kPB calculation uses the doubling point calculation first, then use the addition of two distinct points. The values of kG, kPB are (676, 558), (239, 377). For the Pm+kPB calculation, one point should be converted into other point. (239, 377) is converted to 431(562,201). Then use ordinary addition, the value is 432 (562, 201). Then use doubling concept calculation, next addition of distinct point calculation. The cipher text value as (676, 558), (385, 328).

Decryption

Pm + k PB - nB (kG)

In this I should first find nB by using PB=nB*G. This is the key exchange algorithm which I have used in the description of key exchange.

To find nB:(201, 5)= nB (0, 376)

This is the discrete logarithm problem. It is hard to calculate nB. An nB (0, 376) should match the value of (201, 5). The nB value is 58.

Decryption = (385, 328)-58 (386 (0, 376))

= (385, 328)-58 (676, 558)

= (385, 328)-(239, 377) = 588 (239, 377)-(239,

377) = 587 (239, 377) Pm = (562, 201)

In decryption first found the nB value as 58, the first value of cipher text (676, 558) is multiplied by 58. It is subtracted from the second value of cipher text (385, 328). In the subtraction calculation use the negative calculation of one point i.e., - (239, 377) is formed as (239, -377).

Next calculate addition of two different points (385, 328) with (239, -377). One

point is converted to another point as 587 (239, 377). Finally I got the original plain text as (562, 201).

VII CONCLUSION

The security of ECC depends on how difficult it is to determine k given kG and G, PB = nB* G -to find nB. This is referred to as the elliptic curve logarithm problem. As can be seen, a considerably smaller key size can be used for ECC compared to RSA. Furthermore; there is a computational advantage to using ECC with a shorter key length than a comparably secure RSA. The results show that ECC is efficient in terms of the size of Data files and Encrypted files. The above information is useful for wireless communication due to low data rate transmission and for constrained devices because of low power requirements.

REFERENCES

[1] W.Stallings,”Cryptography and Network Security”, Prentice Hall, third Edition, reprint, 2003, pp.258-262, 297-308.

[2] James Goodman and Anantha P.Chandrakasan,”An Energy Efficient Reconfigurable Public key cryptography processor”,IEEE Journal of solid state circuits,Vol.36,No.11.November 2001.

[3] www.certicom.com.

[4] Yadollah Eslami , Ali Sheikholeslami,P. Glenn Gulak and Shoichi Masui “An Area –Efficient Universal Cryptography Processor for Smart Cards” , IEEE Transactions on Very Large Scale Integration (VLSI) Systems,Vol.,14,No.1.Jan.

[5] Philip H.W.Leong, Senior Member, IEEE, and Ivan K.H.Leung “A Microcoded Elliptic Curve Processor Using FPGA Technology”, IEEE

http://www.certicom.com/

Transactions on Very Large Scale Integration (VLSI) Systems, Vol.10, No.5, OCTOBER 2002.

International Journal of Wind and Renewable

Energy

Power of Sharing Knowledge ISSN: 2277-3975

Online Journal (www.ijwre.org)

Notification of Acceptance

Dear : M. P. Poonia & Y. B. Mathur

Paper ID : IJWRE-27

Paper Title: Effect of Intake Charge Throttling and Rate of Injection on the Performance and

Combustion Characteristics of an LPG Diesel Dual Fuel Engine

The review process for International Journal of wind and Renewable Energy (IJWRE) has been

completed. Based on the recommendations of the reviewers and the Technical Program

Committee, we are pleased to inform you that your paper identified above has been accepted for

online publication. This notification email serves as our formal acceptance of your paper as well

as consideration for online publication provided you complete the formalities and update the

paper as per reviewer’s comments.

(Important) So in order to publish your paper successfully, you must finish following

steps.

1. Send your final paper (updated according to reviewers comments) in .pdf and .doc

format. mail it to [email protected] , [email protected]

2. Mail singed and completely filled copyright (in .jpg format), at [email protected]

, [email protected].

You can find all the required formats and details from Web Site.

If the above requirements are met by 30/08/2012, the paper will be published in the IJWRE. Finally, we would like to further extend our congratulations to you.

Feel free to contact us for any query…

With Regards

Editor-in-Chief

IJWRE





1

Journal of Environmental Research And Development Vol. 6 No. 2, October-December, 2011

SPECIMEN COPY

EFFECT OF VEGETABLE OILS AND BIODIESEL ONENGINE PERFORMANCE, COMBUSTION, AND

EXHAUST EMISSION CHARACTERISTICS : A REVIEW

Y.B.Mathur1, M. P. Poonia2 and A. S. Jethoo*3

1. Government Polytechnic College, Bikaner, Rajasthan (INDIA)2. Department of Mechanical Engineering , Malaviya National Institute of Technology, Jaipur (INDIA)3. Department of Civil Engineering, Malaviya National Institute of Technology, Jaipur (INDIA)

*E-mail : [email protected]

Received September 28, 2011 Accepted October 07, 2011

ABSTRACTEnergy is the most important component of human life. There has been phenomenal increase in theconsumption of energy in the world in the last few decades. It is because of the rapid increase inworld’s population and movement of the population from agricultural pursuit to industrialization.Petroleum fuels like diesel and gasoline are the primary sources of energy for sustainable developmentof any country in the world. With the increase in the energy demands the use of the petroleumproducts increase tremendously in last few years. Due to limited resources of fossil fuels and theenvironmental concern, there has been improved focus on vegetable oils and biodiesel fuel as analternative source of energy. Vegetable oils and biodiesel have the potential to reduce the level ofpollution and global warming. Due to scarcity of edible vegetable oils for biodiesel production, non-edible oils seeds such as karanja, jatropha, mahua, sal, neem, rubber, thumba (Citrullus Colocynthis),jojoba, caster, cotton seeds etc, are being tapped for biodiesel production. The major problemassociated with use of straight vegetable oil includes high fuel viscosity than that of diesel. Thus,although short term tests using neat vegetable oils showed promising results but problems appearlike injector coking and piston oil ring sticking after the engine is operated for longer periods. Tosolve these problems associated with high viscosity, neat vegetable oils are blended in small ratiowith diesel or transesterifies to convert it into less viscous biodiesel. In recent time use of vegetableoils and biodiesel obtained from different seeds have been increasingly fuelled in diesel engines tostudy its effect on engine combustion, performance and emission characteristics but these studieshave been rarely reviewed to understand and popularise biodiesel engines. In the present work, theresearch papers about vegetable oils and biodiesel engine performances, combustion and emissionbehavior, were cited preferentially since 2001 year and experiences of researchers on variety of seedoils suitability for engine operation and its environmental impacts have been understood and compiledin detail.

Key Words : Diesel Engine, Vegetable Oils, Biodiesel, Transesterification, Performance,Combustion, Emission, Aromatics

INTRODUCTIONPerhaps after food, energy is the most importantcomponent for economic development in acountry and for improving the quality of life ofthe people. The growth in energy demand isexpected to continue unabated owing to increasing

industrialization, increasing number of automobiles,increasing standard of living and expandingpopulation. The entire world is confronted withcrises of fossil fuel depletion and environmentaldegradation due to rapid increase in demand offossil fuel. Hence, much emphasis has to be givenfor exploitation of sustainable, long lasting,renewable sources of energy. Among the various*Author for correspondence


2

SPECIMEN COPY

renewable energy options, biodiesel from seed oilcrops have a great potential for meeting the futureincreasing demands of petroleum and its products.Biodiesel is defined as the Mono-alkyl esters oflong chain fatty acids. Biodiesel, derived from theoils and fats of plants like soybean, cotton,sunflower, jojoba, rapeseed, canola, jatrophacurcas, thumba and animal fat by trans-esterification process can be used as a completesubstitute or an additive to diesel up to maximumpossible extent. The major advantage of biodieselis of zero sulphur content and about 10% oxygenavailability in fuel. As an alternative fuel, biodieselcan provide same power output as of diesel. Theconversion process of vegetable oil into biodieselis quite efficient, in nearly one-to-one ratio.Biodiesel is gaining worldwide acceptance as anenvironment-friendly solution to the energy andenvironment degradation crisis. It is an acceptedoption for achieving energy security, reduction inimports, rural employment, and for improving theagricultural economy. The use of biodiesel willnot only reduce the burden on the foreign exchangeby reducing the imports of oil, but will also be aless polluting renewable sources of energy for theprotection of the environment and fulfilling thefuture energy requirement. Biodiesel results insubstantial reduction of un-burnt hydrocarbons,carbon monoxide, and particulate matter. It hashigher cetane number which improves thecombustion characteristic. Among the manyadvantages of biodiesel fuel which includes its safeuse in all conventional diesel engines, offers almostthe same performance and engine durability asconventional diesel fuel, non-toxic and reducestailpipe emissions, visible black smoke and noxiousfumes and odours. However, NOX emission ofbiodiesel increases because of rapid combustionand other fuel characteristics. Biodiesel can beused in diesel engines with few or nomodifications. Diesel fuel blends with biodieselhave superior lubricating properties with reducedwear and tear on the diesel engine components.The most common diesel fuel blends are B10containing 10 percent biodiesel and B20 containing20 percent biodiesel.Lot of research work is being done to analyse theengine combustion and performance using straightvegetable oils as well as biodiesel but fewer

researchers have analysed and compiled them inrecent past. In this paper, a detail survey ofliterature is therefore undertaken tocomprehensively review the different recentlypublished research papers on vegetable oils andbiodiesel as alternative fuel to compression ignitionengines, their utilization in diesel engines, engineperformance, emissions and combustioninvestigation of vegetable oils and biodiesel in dieselengines.Engine performancePerformance of diesel engine with differentvegetable oils and biodiesel has been evaluatedby many researchers to establish suitability andfeasibility of vegetable oils and biodiesel asalternative or substitute fuel to diesel engines.Vegetable oils produced from numerous oil seedcrops have high energy content, most of themrequire some processing to assure safe use ininternal combustion engines1. They observed littlepower loss, higher particulate emissions and lessNOx emissions with neat vegetable oils. Theyconcluded that as compared to raw vegetable oils,methyl ester of vegetable oils is more acceptablesubstitutes for diesel fuel. Fuel heating wasbeneficial at low speed and part-load operations2.The properties and use of Jatropha curcas oil anddiesel fuel blends in compression ignition engines.It was found that the fuel consumption wasincrease with a higher proportion of the Jatrophacurcas oil in the blends. Acceptable thermalefficiencies of the engine were obtained withblends containing up to 50% (by volume) ofJatropha oil3.The tests were conducted on a single-cylinderdirect-injection engine operated with diesel fuel,Jatropha oil and blends of diesel and Jatropha oilin proportions of 97.4%/ 2.6%; 80%/ 20%, and50%/ 50% by volume. The test results showedthat Jatropha oil can be conveniently used as adiesel substitute in a diesel engine. The test resultsshowed increase in brake thermal efficiency, brakepower output and reduction in specific fuelconsumption for Jatropha oil and it blends withdiesel4. The characterization and effect of usingrubber seed oil, as fuel in compression ignitionengine and found that up to 50% of rubber seed

3


SPECIMEN COPY

oil can be substituted for diesel easily in thecompression ignition engines without any majormodification and operational difficulties withunaltered engine output5,6.The performance evaluation tests on linseed oil,mahua oil and rice bran oil blends. They reportedthat linseed oil blends showed comparable thermalefficiency at lower loads; 50% linseed oil blendswere found to be more efficient than other blends.Blending of 30% mahua oil with diesel fuelshowed marginally better brake specific energyconsumption than other blends. Brake specificenergy consumption was better at lower loads andit was much closer with diesel at higher engineloads. Rice bran oil blends showed comparablethermal efficiency for all levels of blends. Ricebran oil blend of 20% diesel showed minimumbrake specific energy consumption and improvedperformance7.Operation of the test engine with pure coconut oiland coconut oil diesel blends for a wide range ofengine load conditions was successful without anyengine modifications. Increase of coconut oil inthe coconut oil diesel blends resulted in increasein the brake specific fuel consumption8.Performance analysis of DI engine using crudesunflower oil at variable load conditions, observedthat the engine operation with vegetable oilsshowed slightly inferior performance9. Theperformance of diesel engine with differentvegetable oils and found that the vegetable oil canbe used as blending component to diesel fuelwithout any loss in power10-12. Non-edible oilssuch as Pongamia and Jatropha in low heatrejection diesel engine. Esterification, preheatingand increase in injection pressure have been triedfor utilization of the vegetable oils in diesel engine.Performance parameters such as the brakespecific energy consumption and exhaust gastemperature have been reported comparable todiesel performance for varying load for differentnon-edible oils13.The performance tests on diesel engine withKaranja oil and its blends with diesel. The calorificvalue for Karanja oil was found to be around 92.95% of diesel and it was also observed that Karanjaoil could be easily blended up to 40% (by volume)in diesel engine without any significant difference

in power output, brake specific fuel consumptionand brake thermal efficiency14. The performanceof engine with Karanja oil blends improved withthe increase in compression ratio from 16:1 to 20:1.Cooking oil originated biodiesel reported that thetorque and brake power output obtained with theused cooking oil derived biodiesel were 3-5% lessthan diesel fuel15.The biodiesel development and characterizationfor use as a fuel in compression ignition engineand observed that 20% biodiesel blend was foundto be optimum, which improved the thermalefficiency of the engine by 2.5%16. Studies havebeen carried out variable speed tests on a singlecylinder direct injection diesel engine using estersof waste vegetable oils to compare the blends withdiesel fuel operation in terms of engineperformance. It was found that with biodieselblends, the engine operated smoothly withoutsignificant problems. The blends burnt moreefficiently with better fuel economy Overall, 100%biodiesel and 75/25 biodiesel/diesel blends werefound to give the best results in terms ofperformance17. The esters derived from coconutoil have many characteristics similar to diesel fuelwith little performance and emission differences18.The cotton seed methyl ester as partial substitutefor diesel oil for single cylinder diesel engines andobserved that the mixture of 30% Cotton seedmethyl ester and 70% diesel oil was suitable tofor engine operation19. Biodiesel derived fromdifferent feed stocks and their blendes with dieselwere investigated for the performancecharacteristics by many researchers. Theyconcluded that biodiesel and its blendes with dieselcan be used successfully in diesel engines withoutmuch modification and with almost sameefficiency as compared to diesel20-32.The experiments has been conducted on fourcylinder, direct injection water cooled diesel enginewith diesel and biodiesel blends of Thumba(Citrullus colocynthis) oil produced throughhydrodynamic cavitation technique. Thumba oilbiodiesel and diesel fuel were blended in 30% /70% ratio while performing the experiments. Theresults observed showed relatively higher brakepower, improved brake thermal efficiency, andreduced brake specific fuel consumption with


4

SPECIMEN COPY

favourable p-è diagram as compared to dieselfuel21.Combustion investigationsHeat release analysis of engine pressure data isa means of indirectly depicting the combustionprocess occurring in the engine. Thermalefficiency and peak cylinder pressure is very muchinfluenced by the heat release pattern. Therapeseed oil methyl ester, heat release alwaystakes place earlier than diesel, because fuelinjection starts earlier for biodiesel blends due totheir higher density, leading to higher peak cylindertemperature. The heat release pattern from theactual pressure crank angle diagram with soyabean oil methyl ester as a fuel in a caterpillarindirect injection diesel engine and concluded thatthe overall combustion characteristics were quitesimilar to diesel operation except shorter ignitiondelay for soybean Methyl Ester.20-21

Experiments] performed on Wisconsin WDI-350single cylinder air-cooled DI diesel engine usingsoya bean oil as the fuel. The heat release ratewas observed to be lowered with thermallycracked soya bean oil as compared to diesel. Theysuggested that by advancing the injection timing,combustion temperatures can be increased and ahigh rate of cylinder pressure rise and higher levelsof premixed burning with the oil can be achieved.The crude palm oil had a 6% higher peak pressurethan diesel. It has been observed that crude palmoil had a 2.6° shorter ignition delay, but lowermaximum heat release rate compared with diesel.Palm oil consists of roughly 50% saturated and50% unsaturated fatty acids. Chemical reactions,such as cracking of the double bonds of carbonchain, could have produced light volatilecompounds which result in a shorter ignition delaycompared with diesel. Due to shorter ignition delay,less fuel was injected during the delay periodresulting in lower heat release rates. This alsoresulted in less intense premixed combustion, andusually translates into lower tendency to knock.Crude palm oil had a longer combustion periodthan diesel. This is due to the fact that anotherchemical reaction, polymerization of vegetable oilat the high temperature spray core, could haveproduced heavy low-volatility compounds. Theseheavy compounds are difficult to combust and

could not completely burn in the main combustionphase, and subsequently continued to burn in thelate combustion phase.27

The studies has been carried out on the dual fueloperation of Karanj oil and its biodiesel with LPGas the inducted fuel in C.I. engine. The series ofexperiments to obtain combustion characteristicsof a 4-stroke C.I. engine operated with neem, ricebran, honge oil and Methyl Ester when directlyinjected as pilot fuel and dual fuelled with producergas induction.

RESULTS AND DISCUSSIONEmissions characteristicsDiesel engines are verdict of polluting environmentby emitting CO2, NOx, particulate emissions, SO2,un-burnt Hydrocarbons (HC) in significant leveldepending on engine load, speed and otheroperating parameters. Tests on diesel engineconducted and compared performance of mineraldiesel with vegetable oil and its blends. HigherCO emissions, Lower CO2 emissions, Lower HCemissions, except 50% blend due to higheroxygen content in molecule of vegetable oil andlower NOx emissions were reported comparedto mineral diesel due to lower calorific value ofvegetable oil.Studied the Karanja esterified oil found that theblends of esterified Karanja oil with diesel up to40% by volume could replace diesel for gettingless emissions without sacrificing the power outputand will thus help in controlling air pollution to agreat extent.The experiments on a single cylinder directinjection diesel engine using olive oil and sunfloweroils as fuels in different proportions with marinediesel. They reported lower un-burnthydrocarbon, carbon monoxide, nitrogen oxide andparticulate with blends compared to neat vegetableoils. The emission characteristics of diesel engineoperating on rapeseed methyl ester and found thatrapeseed methyl ester and its blends with dieselfuel emitted high carbon dioxide compared to testresult in diesel fuel. A very significant reductionin emission of hydrocarbon was recorded whenrunning on rapeseed Methyl Ester. Hydrocarbonemissions were noted to increase with increasedamount of diesel fuel in the blend. The emissions

5


SPECIMEN COPY

and performance of cottonseed oil, soybean oil,sunflower oil and their blends with diesel on adiesel engine, it was reported that the smokedensity significantly increased with the use ofvegetable oil blends of various origins. NOXemissions were slightly reduced with the use ofvegetable oil blends of venous origins with respectto mineral diesel due to lower cetane number(larger premixed combustion part) and theabsence of aromatics. The CO emissions wereincreased and un-burnt hydrocarbons emissionsshowed indefinite trends.Slight increase in NOX emissions for biodiesel isquite obvious, that with biodiesel, due to improvedcombustion, the temperature in the combustionchamber can be expected to be higher and higheramount of oxygen is also present, leading toformation of higher quantity of NOX in biodiesel-fuelled engines. However, biodiesel’s lower sulfurcontent allows the use of NOX control technologiesthat cannot be otherwise used with conventionaldiesel. Hence, NOX emissions can be effectivelymanaged and eliminated by engine optimization.

CONCLUSIONMany energy fuels are being recently investigatedas potential substitutes for the current high-pollutant diesel fuel derived from diminishingcommercial sources. In the past 15 years,biodiesel has progressed from the research stageto a full-production scale in many developedcountries. In Indian context, non-edible oils areemerging as preferred feedstock, and several trialshave been made using non-edible oil species forthe production of biodiesel. Vegetable oil eitherfrom seasonal plant crops or from perennial foresttrees origin, after being formulated, have beenfound suitable for utilization in diesel engines. Manytraditional seed oils like Pongamia glabra,Jatropha (Jatropha curcas), Mallousphilippines, Garcinia indica, Thumba, Karanja andMadhuca indica etc. are available in the country,which can be exploited for biodiesel productionpurposes. Biodiesel is a renewable dieselsubstitute that can be prepared by chemicallycombining any natural oil or fat with an alcohol(usually methanol). It can be used neat or as adiesel additive, and is typically used as a fueladditive with petroleum diesel in compression

ignition engines. Some of the salient conclusionsare drawn on the basis of available literature :1. Non-edible oils seeds such as karanja,

jatropha, mahua, sal, neem, rubber, thumba (Citrullus colocyntis), jojoba, caster, cottonseeds etc, are being tapped for biodieselproduction

2. Biodiesel can be used in diesel engines withfew or no modifications.

3. As compared to raw vegetable oils, methylester of vegetable oils is more acceptablesubstitutes for diesel fuel. The major problemassociated with use of straight vegetable oilincludes high fuel viscosity than that of diesel.

4. The most common diesel fuel blends are B10containing 10 percent biodiesel and B20containing 20 percent biodiesel.

5. Biodiesel results in substantial reduction of unburnt hydrocarbons, carbon monoxide, andparticulate matter. However, NOX emissionof Biodiesel increases because of rapidcombustion and other fuel characteristics

6. Brake specific energy consumption was betterat lower loads and it was much closer withdiesel at higher engine loads

7. It is concluded that the overall combustioncharacteristics are quite similar to dieseloperation except shorter ignition delay forbiodiesel fuels but longer combustion duration.

REFERENCES1. Altin R., Cetinkaya S. and Yucesu H. S., The

Potential of Using Vegetable Oil Fuels as Fuelfor Diesel Engines, J. Energy Conv. Manag.,42(5), 529–538, (2001).

2. Nwafor O. M. I., Rice G. and Ogbonna A. I.,Effect of Advanced Injection Timing on thePerformance of Rapeseed Oil in DieselEngines, J. Ren. Energy, 21(4), 433-444,(2001).

3. Parmanik K., Properties and Use of JatrophaCurcas Oil and Diesel Fuel Blends inCompression Ignition Engine, J. Ren. Energy,28(2), 239-248, (2003).

4. Forson F. K., Oduro E. K. and DonkohHammond E., Performance of jatropha oilblends in a diesel engine, J. Ren. Ener., 29(7), 1135-1145, (2004).


6

SPECIMEN COPY

5. Ramadhas A. S., Jayaraj S. andMuraleedharan C., Use of vegetable oils asI.C. engine fuels : A review, J. Ren. Energy,29(1), 727–769, (2004).

6. Ramadhas A. S., Jayaraj S. andMuraleedharan C., Characterization andeffect of using rubber seed oil as fuel in C.I.Engines, J. Ren. Energy, 30(5), 795-803,(2005).

7. Agarwal D. K., Agarwal L. and Kumar A.,Performance evaluation of vegetable oilfuelled compression engine, J. Ren. Energy,33(6), 1147-1156, (2008).

8. Herchel T. C., Machacon, Shiga S.,Karasawa T. and Nakamura H., Performanceand Emission characteristics of a diesel Enginefuelled with coconut oil-diesel fuel blends, J.Biomass Bioener., 20(1), 63-69, (2001).

9. Mustafa C., Ahmet N. O., Ali T., Combustionanalysis of preheated crude sunflower oil inan IDI diesel engine, J. Biomass andBioenergy, 33(5), 760-767, (2009).

10. Haldar S. K., Ghosh B. B. and Nag A., studieson the comparison of performance andemission characteristics of a diesel engineusing three degummed non-edible vegetableoils, J. Biomass and Bioener., 33(4), 1013–1018, (2009).

11. Bajpai S., Sahoo P. K. and Das L. M.,Feasibility of blending karanj vegetable oil inpetro-diesel and utilization in a direct injectiondiesel engine, fuel, J. Biomass and Bioener.,88(4), 705–711, (2009).

12. Srivastava A. and Prasad R., Triglycerides-based diesel fuels, J. Ren. and Sust. EnergyReview, 4(2), 111–144, (2000).

13. Prasad C. M. V., Krishna M. V. S. M., ReddyC. P. and Mohan K. R., Performanceevaluation of non-edible vegetable oils assubstitute fuels in low heat rejection dieselengines, Pro. Inst. Mech. Engg., 21(D), 181-187, (2001).

14. Batt Y. C., Murthy N. S. and Dutta R. K.,Karanja (Pomgamia glabra) oil as a fuel indiesel engines, J. Agr. Engg. Today, 25(1),45-57, (2001).

15. Merve C., Yahya U., Yucel T. and Filiz K.,Engine and winter road test performances ofused cooking oil originated biodiesel, J.Energy Con. and Manag.,5(4), 441-445,(2004).

16. Agarwal A. K. and Das L. M., BiodieselDevelopment and Characterization for use asa fuel in compression ignition engine, J. Engg.and Gas Turb. Power, 123(1), 440–447,(2001).

17. AI-Widyan M. I., Tashtoush G. and Abu-Quadais M., Utilization of Ethyl Ester ofwaste vegetable oil as fuel in diesel engines,J. Fuel Proce. Tech., 76(2), 91-103, (2002).

18. Swain E. and Shaheed, An Experimental studyto evaluate the use of Coconut Based Fuelsas Alternatives to Diesel Oil, J. The Inst. ofEnergy, 73(1), 100-105, (2001).

19. He Y. and Bao Y. D., Study on cotton seed oilas partial substitute for diesel oil in fuel forsingle cylinder diesel engine, J. Ren. Ener.,30, 805-813, (2005).

20. Zhu Lei, Cheung C.S., Zhang W.G. andHuang Z., Emissions characteristics of adiesel engine operating on biodiesel andbiodiesel blended with ethanol and methanol,J. Science of the Total Environ., 408, 914–921, (2010).

21. Radu R., Petru C., Edward R. and GheorgheM., Fuelling an D.I. agricultural diesel enginewith waste oil biodiesel, Effects over injection,combustion and engine characteristics, J.Energy Conve. and Manag., 50(1), 2158–2166, (2009).

22. Nurun N. M., Mustaûzur R. M.and ShamimA. M., Biodiesel from cotton seed oil and itseffect on engine performance and exhaustemissions, J. App. Therm. Engg., 29(1),2265–2270, (2009).

23. Gumus M. and Kasifoglu S., Performance andemission evaluation of a compression ignitionengine using a biodiesel (Apricot seed kerneloil Methyl Ester) and its Blends with DieselFuel, J. Biomass Bioener., 34(1), 134–139,(2010).

7


SPECIMEN COPY

24. Zafer U. and Mevlut S. K., The Effect ofBiodiesel fuel obtained from waste frying oilon direct injection diesel engine performanceand exhaust emissions, J. Ren. Energy,33(12), 1936–1941, (2008).

25. Rodriguez C. C., Joao Andrade de CarvalhoJr., Tiyoko Y . J. and Luz S. J., Review,determination of ecological efficiency ininternal combustion engines, The Use ofBiodiesel, J. App. Therm. Engg., 29(1),1887–1892, (2009).

26. Murayama T., Fujiwara Y. and Noto T.,Evaluating waste vegetable oil as a diesel fuel,J. Mech. Engg., 5(1) 141–149 (2001)

27. Hazar H., Effects of Biodiesel on a low heatloss diesel engine, J. Renew. Ene., 34(1),1533–1537 (2009)

28. Kinney A. J. and Clemente T. E., Modifyingsoybean oil for enhanced performance inbiodiesel blends, J. Fuel Pro. Tech., 86(1),1137–1147, (2005).

29. Hansen A.C., Zhang Q. and Lyne P. W. L.,Ethanol-diesel fuel blends : A Review, J.Bioreso. Tech., 96(1), 277–285, (2002).

30. Seyin A. H. and Bayindir H., Performanceand emission analysis of cottonseed oil MethylEster in a diesel engine, J. Ren. Energy,35(1), 588–592, (2010).

International Journal of Wind and Renewable Energy Volume 1 Issue 2 (Page, 108-113), ISSN: 2277-3975

Copyright © 2012 IJWRE, All right reserved 108

Performance and Emission Characteristics of Diesel

Engine Using Low Concentration Thumba Oil Diesel

Blends

Y. B. Mathur1, M. P. Poonia

2, U. Pandel

3 and R. Singh

4

1. Ph.D. Scholar, Suresh Gyan Vihar University, Jaipur (INDIA), 2. Professor, Deptt. of Mechanical Engg., Malaviya National Institute of Technology, Jaipur (INDIA),

3. Professor, Deptt. of Metallurgy, Malaviya National Institute of Technology, Jaipur (INDIA), 4. Asst. Professor, Deptt. of Mechanical Engg., Engineering College, Bikaner (INDIA)

ABSTRACT

The world is focusing on the impending energy

crisis and each country needs to concentrate on

energy security as well as growing pollution and

environmental concerns. The rapid depletion of

fossil fuel reserves with increasing demand and

uncertainty in their supply as well as the rapid rise

in petroleum prices has stimulated the search for

other alternatives to fossil fuels. Although there are

wide range of alternative fuels available, vegetable

oils like jatropha oil, karanja oil, castor oil, jojoba

oil, cotton seed oil, neem oil, mahua oil, palm oil,

soybean oil, and sunflower oil, thumba oil etc. are

being investigated as potential substitute for

current high polluting fuels obtained from

conventional sources to meet fuel crises.

In the present paper, attempt has been made to

explore possibility and potentiality of low

concentration thumba oil diesel blends as substitute

fuel for compression ignition engines.

The thumba (Citrullus Colocynthis) plant is a

native of arid soils and grows in large quantities in

hot regions of India in the form of a creeper. This

plant occupies the vast area in the Rajasthan and

Gujarat. The potentiality of thumba seed oil as

diesel engine fuel is not still assessed extensively.

The results of experimental study show that the low

concentration thumba oil blends with diesel can be

used successfully in diesel engines without any

hardware modifications and without posing any

problems related to combustion and engine knock.

Thumba oil blended with diesel up to 30% levels

can be used successfully in diesel engines with no

major reduction in engine performance and

reduced tail pipe emissions.

Index Terms: Thumba Oil, Diesel Engine, Vegetable

Oil, Performance & Emission

1. INTRODUCTON

Energy is the prime mover of economic growth and is

vital to the sustenance of a modern economy. Future

economic growth crucially depends on the long-term

availability of energy from sources that are affordable,

accessible and environmental friendly [1], [2].

Increasing industrialization, growing energy demands,

limited reserves of fossil fuel and increasing

environmental pollution have joined necessitated

exploring some alternative of conventional petroleum

fuels.

Biofuels are strongly emerging as partial substitutes for

fossil fuel from the economic as well as environmental

angle. Among the biofuels, vegetable oils like jatropha

oil, karanja oil, castor oil, jojoba oil, cotton seed oil,

neem oil, mahua oil, thumba oil, palm oil, soybean oil,

sunflower oil etc. are being explored as promising

alternative to hydrocarbon based fuels to full fill the

future energy needs. [3]. Vegetable oils can be used as

alternative fuels because they are biodegradable, non-

toxic and significantly reduce pollution. Vegetable oils

and their derivatives as diesel engine fuels lead to

substantial reductions in carbon monoxide, smoke and

particulate emissions. Vegetable oils have

approximately 90% heating value of mineral diesel due

to higher oxygen content [4]. One of the main

problems of vegetable oil use in diesel engines is high

viscosity than that of mineral diesel due to which

problems occur in pumping and atomization, ring-

sticking, carbon deposits on the piston, cylinder head,

ring grooves, etc. Therefore, a reduction in viscosity is

of prime importance to make vegetable oils a suitable

alternative fuel for diesel engines. The problem of high

viscosity of vegetable oils can be resolved in several

ways, such as preheating the oils, blending or dilution

with other fuels, transesterification and thermal

cracking or pyrolysis. One of the possible methods to

overcome the problem of high viscosity is blending of

vegetable oil in proper proportions with diesel. In the

present study, blending of straight vegetable oil with

diesel fuel in small proportions has been adopted to

overcome higher viscosity related issues [5] - [7].

In the present investigation non-edible vegetable oil,

thumba oil (Citrullus Colocynthis oil) has been chosen

to explore its possibility and potentiality as diesel

engine fuel. Thumba (Citrullus Colocynthis) is known

as Indrayan in Hindi and Bitter Apple in English, is a

native of Turkey and also found in many parts of Asia

and Africa. In India it mainly grows in almost all parts

of Rajasthan and Gujarat. Thumba (Citrullus

Colocynthis), that grows as creeper in sandy soil

within a six month crop cycle has enormous potential

for biodiesel production as well as use as diesel engine

fuel in conventional diesel engines. The thumba plant

is mainly used as cattle feed by farmers and raw

thumba oil is currently utilized by the soap

manufacturing units in large quantities [8] – [12].

Figure [1] shows the thumba fruits and thumba seeds.



The important chemical and physical properties of

thumba oil is determined and compared with diesel.

Some of the measured properties of pure thumba oil

are presented in Table [I].

Figure [1] Thumba Fruits and Thumba Seeds

Table [I] Some Important Fuel Properties of

Thumba oil

Properties Diesel Thumba Oil

Specific Gravity at 20° C 0.835 0.905

Viscosity, cSt at 40° C 2.75 31.52

Calorific Value, MJ/Kg 42.25 39.78

Flash Point, ° C 66 201

Pour Point, ° C -20 -5

Free Fatty Acid, % - > 1

Cetane Number 47 45

2. EXPERIMENTAL SETUP

The experiments were conducted on naturally aspirated

4-stroke single cylinder, water-cooled, multi fuel,

variable compression ratio, direct injection diesel

engine test rig having rated power output of 3 to 5 HP.

The view of the experimental setup and

instrumentation are depicted in the Figure [2.a] and the

schematic block diagram of the experimental set-up is

shown in Figure [2.b]. The detailed technical

specifications of engine are given in Table [II]. An

eddy current dynamometer was used for loading the

engine. Fuel flow rate was measured on volumetric

basis by using a burette equipped with an electronic

sensor. Thermocouples located at various positions

have been used for measuring the temperatures. Multi

gas analyzer was used to measure carbon monoxide,

hydrocarbon and NOX in the exhaust. Smoke level was

obtained by using smoke meter. Diesel fuel and all the

reagents used in the study were purchased from local

market and thumba oil was procured from Udaipur and

Jodhpur regions of Rajasthan.

Figure [2. a] Overall View of Experimental Setup

Figure [2. b] Schematic Layout of Experimental

Setup

1. E

2.1 EXPERIMENTAL PROCEDURE

The various low concentration blends of thumba oil and diesel were prepared on volumetric basis ranging from 10% to 50% thumba oil in petro-diesel. Neat thumba oil has higher specific gravity as compared to petro-diesel, therefore thumba oil was splash blended on the top of diesel for proper mixing of the blend. Experiments were initially carried out on the multi fuel engine using diesel as a fuel in order to provide base line data, subsequently experiments were conducted



with prepared specified blends of thumba oil in diesel. All tests were conducted at 19 compression ratio, 203 barks injector opening pressure and at 23° CA BTDC injection timing for entire load range at rated engine speed of 1500 rev/min. The engine was stabilized before taking all measurements.

3. RESULTS AND DISCUSSION

In the present paper, the performance and emission

behavior of multi fuel diesel engine with low

concentration thumba oil blends in diesel fuel of

different proportions was evaluated and compared with

diesel fuel engine operation. The engine performance

parameters like brake thermal efficiency, brake

specific fuel consumption, exhaust gas temperature etc.

and emission levels of carbon monoxide, hydrocarbon,

smoke opacity and NOX, in exhaust were evaluated /

recorded to assess the performance and emission

characteristics of compression ignition engine running

with different thumba oil diesel blends as well as on

diesel fuel.

4. ENGINE PERFORMANCE

4.1 Brake Thermal Efficiency

Figure [3] shows variation of brake thermal efficiency

with respect to load for different low concentration

thumba oil diesel blends at 19 compression ratio, 203

bars injector opening pressure and at 203⁰ CA BTDC

injection timing and at 1500 rev/min engine speed.

The brake thermal efficiency increases with increasing

of engine load for diesel fuel as well as for all the

blends tested.

There is a clear distinction from the graph, that higher

brake thermal efficiency was observed for diesel fuel

engine operation for entire load range compared to all

thumba oil diesel blends. The maximum brake thermal

efficiency among all thumba oil diesel blends was

observed for 20% thumba oil diesel blend. The

improvement in brake thermal efficiency can be

attributed to the enhanced oxygen content which

improves combustion especially during the phase of

diffusion combustion and to the higher lubricity of

thumba oil which reduces the friction loss. When

blending has been increased to 30% thumba oil in

diesel, brake thermal efficiency was found marginally

lower as compared to blend of 20% thumba oil in

diesel. It can be clearly seen from the Figure [3] that

after 20% thumba oil in blend, reduction in brake

thermal efficiency was observed with the increase in

concentration of thumba oil in the blend. This drop in

thermal efficiency with increase in proportion of

thumba oil can be attributed to the poor combustion

characteristics of the blends due to their relatively high

viscosity, poor volatility, slow burning rate and lower

calorific values that overcomes the excess oxygen

present in the thumba oil. The lowest brake thermal

efficiency was observed while using 50% blend of

thumba oil in diesel fuel.

4.2 Brake Specific Fuel Consumption

The comparison of brake specific fuel consumption

with respect to load for diesel fuel and various blends

of thumba oil with diesel are presented in Figure [4]. It

is clearly seen from the plot that the specific fuel

consumption for various blends of thumba oil in diesel

found marginally higher than diesel at all load

conditions.

The brake specific fuel consumption for different

thumba oil diesel blends initially decreases slightly

with the addition of thumba oil content in the blend

until it reaches 20% by volume and then, increases

with more addition of the thumba oil content owing to

the lower calorific value, poor mixture preparation,

slow burning which deviates the cycle from constant

volume combustion. At maximum load the specific

fuel consumption of engine fuelled with 20% thumba

oil diesel blend was found minimum among all the

tested blends of thumba oil in diesel.

4.3 Exhaust Gas Temperature

The exhaust gas temperature increases with increase in

load for diesel as well as for all thumba oil blends as

indicated in Figure [5]. The exhaust gas temperature

was observed higher for all thumba oil blends

compared to diesel fuel at high loads due to the higher

viscosity then diesel fuel which results in poorer

atomization, poorer evaporation and extended

combustion which goes on up to exhaust stroke. The

lowest exhaust temperature among all the thumba oil

diesel blends was observed for the blend of 20%

thumba oil in diesel at low as well as at high load

conditions. Close values of exhaust gas temperatures

were also observed for 10% and 30% thumba oil blend

in diesel.

Figure [3] Variation of Brake Thermal Efficiency

with Respect to Load at Different Thumba Oil-

Diesel Blends

Figure [4] Comparison of Brake Specific Fuel

Consumption with Respect to Load at Different

Thumba Oil-Diesel Blends



Figure [5] Variation of Exhaust Gas Temperature

with Respect to Load at Different Thumba Oil-

Diesel Blends

5. Emission Characteristics

5.1 Smoke opacity

Smoke opacity in the exhaust depends upon fuel

combustion and carbon content of the fuel. In diesel

engine smoke formation generally occurs in the rich

zone at high temperature, particularly within the core

region of fuel spray. The variation of smoke opacity

with respect to load is shown in Figure [6]. It is

observed that the smoke opacity of the exhaust gas

increases with increase in load for diesel fuel and for

all blends tested. It is also noticed that the smoke

opacity increases with the increase of concentration of

thumba oil in the blends. This is mainly due to the

bigger size fuel molecules and specific physical and

chemical properties of thumba oil. The minimum

smoke level was found for 20% thumba blend in

diesel, followed by 30% and 10% concentration of

thumba oil in blend due to better combustion

efficiencies compared to other blends.

5.2 Carbon Monoxide Emission

Carbon monoxide production was dominated by local

mixture condition rather than overall mixture

condition. Higher oxygen content in the local mixture

favors better combustion producing less carbon

monoxide in case of all thumba oil blends compared to

diesel at moderate and high loads. The minimum

values of carbon monoxide were found for 20%

thumba blend in diesel. Although the high

concentration fuel blends contains less carbon content

in fuel but due to inferior combustion on the account of

poor vaporization and atomization leads to increase in

carbon monoxide in exhaust at high concentration

blends. The 20% blend of thumba oil in diesel showed

about 7.2 % reductions in carbon monoxide emission

compared to diesel at full load, mainly due to

additional amount of oxygen content in blends helps

for the relatively more complete combustion. As

indicated in Figure [7], that the reduction in carbon

monoxide emission is more significant at full load

conditions. The difference may be due to the fact that,

at full load conditions sufficient oxygen is not

available for the combustion in the cylinder, the

additional oxygen content in the thumba oil blends

provides oxygen for the complete combustion which

promotes further oxidation of carbon monoxide during

the engine exhaust process, hence, reduction in the

carbon monoxide emission. But at partial load, enough

oxygen is available due to lean mixture thus there is

little variation in carbon monoxide levels in exhaust

emissions.

5.3 Hydrocarbon Emission

Un-burnt hydrocarbons are also important parameter

for determining the emission behavior of the engine.

Significant reduction in hydrocarbon emission was

observed for all the thumba oil blends compared to

neat diesel engine operation as indicated in Figure [8].

Lowest level of hydrocarbon emission was found for

20% thumba oil blend mainly due to oxygen content in

fuel and lower viscosity which results in almost

complete combustion inside the cylinder. The 20%

blend of thumba oil in diesel showed reduction of

about 51.72 % in hydrocarbon emission compared to

diesel at peak loads.

5.4 NOX Emission

Figure [9] presents comparison of NOX emissions from

engine exhaust with load for diesel fuel and different

blends of thumba oil at 1500 rev/min. It is seen from

the plot that with blends of thumba oil, engine usually

yield higher NOX emissions compared to diesel fuel

operations for entire load range may be due to

additional oxygen content and higher gas temperature

in combustion chamber. However, at low load

conditions when there was much excess of air, then

slight increase in NOX emissions was noticed with

thumba oil blends as compared to diesel. It is also seen

from the plot that the higher NOX emission is noticed

at higher concentration blends and lowest level of NOX

was observed at 20% thumba oil blend.

Figure [6] Smoke Opacity with Respect to Load for

Different Thumba Oil-Diesel Blends



Figure [7] Carbon Monoxide Emission with Respect

to Load for Different Thumba Oil-Diesel Blends

Figure [8] HC Emission with Respect to Load for

Different Thumba Oil-Diesel Blend

Figure [9] NOX Emission with Respect to Load for

Different Thumba Oil-Diesel Blends

6. CONCLUSIONS

The significant conclusions from the present

experimental investigation are summarized below:

No problem was faced at the time of starting the

diesel engine and the engine ran smoothly over the

range of thumba oil percentage in fuel blend.

The blend of 20% thumba oil in diesel was found

to give best brake thermal efficiency with lowest

brake specific fuel consumption compared to other

blends of thumba oil diesel blends. Also the

exhaust emission levels of smoke, carbon

monoxide, NOX and hydrocarbons found lowest

for this blend. Hence, this work establishes that

20% thumba oil in the fuel blend can be

considered as optimum blend and may be used in

diesel engines without any adverse effect.

Lower content of thumba oil in blends,

particularly the blends of 10% and 30% thumba

oil in diesel, also showed satisfactory performance

and emission characteristics, and can also be used

in compression ignition engines without any major

modifications.

It is evident from the experiments that low

concentration blends of thumba oil in diesel

showed satisfactory short term engine operation.

However, some operational problems with high

concentration blends particularly for 50% thumba

diesel blend were observed like coking of injectors

and cylinder walls, valves and piston sticking,

carbon deposits on piston etc. Slight gum

formation around valves and injectors were also

noticed after extended engine operation. The

problems transpire during engine tests can be

reduced to some extent by modifying the

maintenance schedule of engine and regular

cleaning of injectors.

REFERENCES

1. International Energy Outlook 2010-2011, Energy

Information Administration (EIA), Office of

Integrated Analysis and Forecasting, U.S.

Department of Energy, U.S.A., 2010-2011.

2. TERI, Energy Data Directory and Yearbook

2010/2011, Tata Energy Research Institute, New

Delhi, India, 2010/2011.

3. J. Yan and T. Lin, “Biofuels in Asia”, Applied

Energy, Volume 86, 2009, Pages S1–S10.

4. Indian Biofuels Annual Report, Global

Agriculture Information Network, New Delhi,

2011.

5. A. K. Agarwal, “Biofuels (Alcohols and

Biodiesel): Applications as Fuels for Internal

Combustion Engines”, Progress in Energy and

Combustion Science, Volume 33, 2007, Pages

233–271.

6. A. K. Hossain and P. A. Davies, “Plant Oils as

Fuels for Compression Ignition Engines: A

Technical Review and Life-Cycle Analysis”,

Renewable Energy”, Volume 35, 2010, Pages 1–

13.

7. C. D. Rakopoulos, K. A. Antonopoulos, D. C.

Rakopoulos, D. T. Hountalas and E. G.

Giakoumis, “Comparative Performance and

Emissions Study of a Direct Injection Diesel

Engine Using Blends of Diesel Fuel with

Vegetable Oils or Biodiesels of Various Origins”,

Energy Conversion and Management, Volume 47,

2006, Pages 3272–3287.

8. Karnwal Ashish, Kumar Naveen, Hasan M M,

Chaudhary Rajeev, Noor Siddiquee Arshad and

Zahid A. Khan, “Production of Biodiesel from

Thumba Oil: Optimization of Process Parameters”

Iranica Journal of Energy and Environment

(IJEE), Volume 1, Issue 4, 2010, Pages 352-358.

9. Solomon Giwa, Luqman Chuah Abdullah and Nor

Mariah Adam, “Investigating “Egusi” (Citrullus

Colocynthis L.) Seed Oil as Potential Biodiesel

Feedstock”, Energies, Volume 3, 2010, Pages

607–618.

http://www.doaj.org/doaj?func=openurl&issn=20792115&genre=journal

http://www.doaj.org/doaj?func=openurl&issn=20792115&genre=journal



10. Agarwal Deepak Kumar, Lokesh Agarwal and

Avinash Kumar, “Performance Evaluation of

Vegetable Oil Fuelled Compression Engine”,

Renewable Energy, Volume 33, 2008, Pages

1147-1156.

11. Amit Pall, Ashish Verma, S. S. Kachhwaha and S.

Mali, “Biodiesel Production Through

Hydrodynamic Cavitation and Performance

Testing”, Renewable Energy, Volume 35, 2010,

Pages 619-624.

12. Amit Pall, S. S. Kachhwaha, S. Maji and M. K. G.

Babu, “Thumba (Citrullus Colocynthis) Seed Oil:

A Sustainable Source of Renewable Energy for

Biodiesel Production”, Journal of Scientific &

Industrial Research, Volume. 69, May 2010, Pages

384-389.

Table [II] Technical Specifications of Test Engine


Make Legion Brothers, Bangalore

(India)

Type

Single Cylinder, Direct Injection,

Four-Stroke, Vertical, Water-

Cooled, Naturally Aspirated

Variable Compression Ratio

Multi-Fuel Diesel Engine

Power 3 to 5 HP

Rated Speed 1500 rev/min (Governed Speed)

Number of

Cylinders One

Compression

Ratio

5:1 to 20:1 (Variable

Compression Ratio)

Bore 80 mm

Stroke 110 mm

Injector Opening

Pressure 203 bars