pinch analysis a tool for efficient use of energy

1

Staff Development Programme One week Jan 05th

-09th

, 2009

Under the aegis of

All India Council of Technical Education

Pinch Analysis: A Tool for Efficient Use of Energy

Department of Chemical Engineering offers a one week Training Course on Pinch Technology which provides a basic grounding in Process Integration.

Learn the fundamental concepts of Pinch Technology and how these can be applied to reduce energy and capital costs and increase capacity. It is appropriate for candidates who wish to gain an understanding of how to achieve process energy efficiency improvements and

Reduce utility costs

Remove energy & capacity bottlenecks

Avoid or reduce capital expenditure

Optimum practical design

Department of Chemical Engineering National Institute of Technology Rourkela Rourkela – 769 008, Orissa

Coordinated by: Dr. Shabina Khanam

Co-coordinated by: Prof. K. C. Biswal

http://www.terragalleria.com/california/picture.usca9425.html

http://www.terragalleria.com/california/picture.usca9431.html

2

ACKNOWLEDGEMENT

The AICTE sponsored short term course on “Pinch Analysis: A Tool for Efficient Use

of Energy” is a culmination of Research and Teaching efforts of the Process Engineering

group of the Chemical Engineering Department, NIT Rourkela.

Thanks are also due to Prof. Sunil Kumar Sarangi, Director, NIT Rourkela for his

consistent support and encouragement.

We would like to express my sincere gratitude to Prof. S. K. Jena, Dean (SRICCE), NIT

Rourkela for his excellent cooperative attitude. I thank the staff of SRICCE to provide

invaluable help.

We extend our thanks to all faculty members of Department of Chemical Engineering for

their cooperation and continuous encouragement.

We are also grateful to the Prof. Bikash Mohanty, Resource person from Chemical

Engineering Department, IIT Roorkee for his significant contribution in delivering this

course.

We take this opportunity to express our appreciation to the Post-Graduate students and

Non-Teaching staff of Chemical Engineering Department for their support and

assistance. Without their support this course could not have reached to this stage.

We also realize that without a high level of receptivity, active involvement, and

cooperation from the participants, this course would not have accomplished its

objectives.

Last but not the least; we thank all those who have directly or indirectly contributed

towards the success of this course.

Dr. Shabina Khanam Prof. K.C. Biswal

Coordinator Co-Coordinator

HOD, CH

3

COURSE OVERVIEW

Process Integration (PI) is a general approach for the design of energy efficient process

systems and Pinch Technology is a tool to achieve it. In the late 1970s Pinch Technology

emerged as a tool for the design of heat exchanger networks against the backdrop of

energy crisis. Its key contribution was to provide the engineers with simple concept of

heat, power and thermodynamics, which can be used interactively in each stage of design.

In 1980s, Pinch Technology received prime attention as a heat exchanger network design

tool and it was found that this technology could save around 20–40% of energy bills of

the industry. Since then, the method has become broad based. However, its

thermodynamic principles, heuristic rules and its key strategy to set targets before design

remain intact. With time it has emerged as a powerful, matured integrated design and

retrofitting tool for overall process design.

The present course deals with concept to implementation of Pinch Technology in

integrated process design.

The one week duration course was specially tailored to provide enough resources to the

attending candidates to start a course on Process Integration in their respective

departments or to use it in the industries. This course is expected to be popular in near

future as it holds considerable promise for the conservation of energy in Chemical and

allied industries.

4

LIST OF PARTICIPANTS

S. No. Participant Photograph

1 Dr. RaviShankar R

Prof. and Head

Chemical Engineering Department

Dayananda Sagar College of Engineering

Shavige Malleswara Hills.

Kumaraswamy Layout, Bangalore 560078

Ph: 9448327476

Email:- [email protected]

2 Prof. Mahadeva Raju, G. K.

Assistant Professor


Dayananda Sagar College of Engineering

Shavige Malleswara Hills.

Kumaraswamy Layout, Bangalore 560078

Ph: 9845772214

Email:- [email protected]

3 Mr. Jagadish H Patil

Assistant Professor


R V College of Engineering,

Mysore Road

Bangalore 59

Ph: 080-67178046/67178109

Email: [email protected]

4 Mr. Anil Kumar Prasad

Lecturer,

Deptt. of Applied Mechanics,

NIT Jamshedpur – 831014

Jharkhand

Ph: 9835314761


5 Mrs. Dipa Das

Lecturer

Chemical Engineering Dept.

Indira Gandhi Institute of Technology

Sarang, (Parjang) - 759146

Dist – Dhenkanal


5

6 Mr. Karthik S. P.

Senior Engineer

Sanden Vikas India Ltd.

Plot No. 65, Sector – 27A

Faridabad – 121003, Haryana

Ph: 9958855998


7 Mr. Bhabani Prasanna Pattnaik

Assis. Professor

Mechanical Engineering Dept.

KIIT University

Bhubaneswar, Orissa

Ph: 9437169040


8 Dr. Shib Sankar Saha

Sr. Lecturer

Electrical Engineering

Govt. Gollege of Engg. and Textile Technology

Berhampore – 742 101 W.B.

Ph: 9434315226


9 Dr. S. K. Agarwal

Professor


NIT Rourkela – 769 008,

Orissa

Ph: 9861386942


10 Dr. Basudeb Munshi

Asstt. Professor


NIT Rourkela – 769 008, Orissa

Ph: 0661-2462265


6

11 Dr. Arvind Kumar

Assistant Professor

Chemical Engineering Department.

National Institute of Technology Rourkela –

769 008, Orissa

Ph: 9438348807

e-mail:[email protected]

https://sites.google.com/site/arvindkumarnitr/

12 Dr. Mithilesh Kumar

Asstt. Professor

Department of Met. & Materials Eng.

NIT Rourkela – 769 008, Orissa

Ph: 0661-2463554


13 Mr. Binod Kumar Singh

Research Scholar

Department of Met. & Materials Eng.

NIT Rourkela – 769 008,

Orissa


14 Mr Achyut Kumar Panda

Sr. Lecturer

Department of Chemistry

Jagannath Institute for Tech. & Mgmt.

Gajapati – 761211, Orissa

Ph: 9437132916


15 Mrs Hemalata Patra

Lecturer, G.I.E.T.,

Gunupur

Ph: 9437646933


7

16 Dr. Sunil Kumar Maity

Assit. Professor


NIT Rourkela – 769 008

Ph: 0661-2462266


17 Mr. Akshaya Kumar Rout

Senior Lecturer

Mechanical Engineering Dept.

C.V. Raman College of Engineering

Janla, Bhubaneswar – 752054

Ph: 9437756207


8

CONTENTS

S.No. Lecture(s) Speaker Page

1. Process Intensification Dr. B. Mohanty 9

2. Process Integration Dr. B. Mohanty 19

3. Pinch Technology – An Overview Dr. S. Khanam 28

4. Basic Elements of Pinch Technology Dr. B. Mohanty 37

5. Area Targeting Dr. B. Mohanty 49

6. Number of Unit, Shell and Cost Targeting Dr. S. Khanam 61

7. Pinch Design Methods – Heuristic Rules Dr. B. Mohanty 74

8. Design of HEN for Maximum Energy Recovery, Loop

Breaking & Path Relaxation

Dr. B. Mohanty 80

9. Driving Force Plot and Remaining Problem Analysis Dr. B. Mohanty 97

11. References 106

9

Lectures 1 & 2

PROCESS INTENSIFICATION

Bikash Mohanty

Professor

Department of Chemical Engineering

Indian Institute of Technology Roorkee, Roorkee – 247 667

Today, we are witnessing new developments that go beyond “traditional” chemical

engineering. Investigators at many universities and industrial research centers are

investigating on novel equipment and techniques that could transform our concept of

chemical plants and lead to compact, safe, energy-efficient and environment-friendly

sustainable processes. These developments share a common focus on “Process

Intensification (PI)” – an approach that has been around for quite some time but has truly

emerged only in the past few years as a special and interesting discipline of Chemical

Engineering.

PI refers to the technologies and strategies that enable the physical sizes of conventional

process engineering unit operations to be significantly reduced.

The concept of PI was pioneered in late 70’s by Colin Ramshaw, when the primary goal

was to reduce the capital cost of a production system. The virtue of PI approaches will be

recognized when it is appreciated that the main plant items involved in the process (i.e.

reactors, heat exchangers, separators etc.) only contribute around 20% of the cost of the

given plant. The balance is incurred by installation costs, which involve pipe work,

structural support, civil engineering and so on. A major reduction in equipment size,

coupled preferably with a degree of telescoping of equipment function – for example

reactor / heat exchangers or combined condenser/distillation/re-boilers - could generate

10

very significant cost savings by eliminating support structure, expensive column

foundations and long pipe runs.

Mercer, in 1993, defined the PI as “PI is a design philosophy aiming at radical reduction

of physical size of the process equipment”. These reductions can be a factor three to four

in magnitude. Because energy efficiency of a process is determined by the ability to

transfer the heat in a cost effective way, the design of processes using a small amount of

heat exchanger (using PI studies) or using innovative heat exchanger design (i.e compact

heat exchanger) can save energy.

In 1995, Ramshaw defined PI as a strategy for making dramatic reductions in the size of

a chemical plant so as to reach a given production objective. These reductions can come

from shrinking the size of individual pieces of equipment and also from cutting the

number of unit operations or apparatuses involved. Ramshaw speaks about volume

reduction of the order of 10 to 1000, which is quite a challenging number since then the

definition of PI has been enlarged considerably.

Benefits of Process Intensification

PI has a potential to deliver major benefits to the process industry and many other sectors,

by accelerating the response to market changes, facilitating scale-up and providing the

basis for rapid development of new products and processes. Additional benefits of PI

include reduced capital cost, improved intrinsic safety and reduce environmental impact.

Process Intensification and Its Components

11

Process Intensification, being driven by the need for break through changes in operations,

focuses mainly on novel methods and equipments. PI consists of the development of

novel apparatuses and techniques that are compared to those commonly used today and

are expected to bring dramatic improvements in manufacturing and processing,

substantially decreasing equipment-size/production- capacity ratio, energy consumption

or waste production and ultimately resulting in cheaper, sustainable techniques.

As evident in Fig. 1, the whole field can generally be divided into two areas:

Process-Intensification equipments

Process-Intensification methods

Many industries offer emerging technologies that are designed for various segments of

the process industries having one common feature - Process Intensification. PI is the

miniaturization of unit operations and processes whereas a smaller compact piece of

equipment takes the place of a larger one at the same given capacity and mass flow rate.

Process Intensification Equipments

Static Mixer Reactor (SMR)

SMR has mixing elements made of heat transfer tubes (Fig. 2), can successfully be

applied in processes in which simultaneous mixing and intensive heat removal or supply

are necessary, such as in nitration or neutralization reactions.

The main disadvantage of SMRs is their relative high sensitivity to clogging by solids.

Therefore, their utility for reactions involving slurry catalysis is limited.

12

Examples

Spinning disk reactor Static mixers Reverse-flow

reactors

Membrane

absorption

Centrifugal

fields

Supercritical

fluids

Static mixer reactor

(SMR)

Compact heat

exchanger

Reactive

distillation

Membrane

distillation

Ultrasound Dynamic

(periodic)

reactor

operation

Static mixing catalysts

(KATAPAKs)

Microchannel heat

exchangers

Reactive extraction Adsorptive

distillation

Solar energy

Monolithic reactors Rotor/Stator mixers Reactive

crystallization

Microwaves

Microreactors Rotating packed beds Chromatographic

reactors

Electric

fields

Heat exchanger reactors

(HEX)

Centrifugal adsorber Periodic separating

reactors

Plasma

technology

Supersonic gas/liquid

reactor

Membrane reactors

Jet-impingement reactor Reactive extrusion

Rotating packed-bed

reactor

Reactive

comminution

Fuel cells

Equipment Methods

Equipment for

carrying out

chemical

reactions

Equipment for

operations not

involving chemical

reactions

Multifunctional

reactors

Hybrid

separations

Alternative

energy

sources

Other

methods

Process Intensification

Fig. 1. Process Intensification and its components

13

Monolithic Catalysis

Monolithic substrates used for today’s catalytic applications are metallic or nonmetallic

bodies providing a multitude of straight narrow channels of defined uniform cross-

sectional shapes.

To ensure sufficient porosity and enhance the catalytically active surface, the inner walls

of the monolith channels usually are covered with a thin layer of wash coat, which acts as

the support for the catalytically active species.

The most important features of monoliths are:

Very low pressure drop in single & two-phase flow

Fig. 2. Proprietary reactor-mixer is a classic example of

process intensifying equipment.

Fig. 3. Monolithic catalyst

14

High geometric areas per reactor volume

High catalytic efficiency, practically 100 %

Exceptionally good performance in processes in which selectivity is hampered by

mass transfer resistances

Microchannel Heat Exchangers

The geometrical configuration of Microchannel heat exchangers given in Fig. 4

resembles that of the cross-flow monoliths, although the materials and fabrication

methods used differ.

The Microchannel heat exchangers exhibit high heat fluxes and convective-heat-transfer

coefficients. The reported values of heat transfer coefficients in Microchannel heat

exchangers range from 10000 to 35000 W/m2K.

Process Intensifying Methods

Reverse Flow Reactor

For exothermic processes, the periodic flow reversal in such units allows for almost

perfect utilization of the heat reaction by keeping it within the catalyst bed and after

reversion of the flow direction, using it for preheating the cold reactant gases.

Fig. 4. Microchannel heat exchanger

15

These reactors are used in following industrial processes: SO2 oxidation, total oxidation

of hydrocarbons in off-gases, and NOx reduction.

Reactive Distillation

Reactive distillation is one of the better-known examples of integrating reaction and

separation, and is used commercially.

In the column, reactants are converted on the catalyst while reaction products are

continuously separated by fractionation (thus overcoming equilibrium limitations).

The catalyst used for reactive distillation usually is incorporated into a fiberglass and

wire-mesh supporting structure, which also provides liquid redistribution and

disengagement of vapor.

Reactive Extrusion

Reactive extruders are being increasingly used in polymer industries.

They enable reactive processing of highly viscous materials without requiring the large

amounts of solvents that stirred-tank reactors do.

Particularly popular are twin-screw extruders, which offer effective mixing, the

possibility of operation at high pressure and temperatures, plug-flow characteristics, and

capability of multistaging.

Most of the reactions carried out in extruders are single- or two-phase reactions.

Sonochemistry

Sonochemistry is the use of ultrasound as a source of energy for chemical processing and

appears to be the most advanced. Formation of microbubbles (cavities) in the liquid

reaction medium via the action of ultrasound waves has opened new possibilities for

chemical syntheses. These cavities can be thought of as high-energy microreactors.

16

Supercritical Fluids

Supercritical fluids (SCFs) are used industrially for the processing of natural products.

Because of their unique properties, SCFs are attractive media for mass transfer

operations, such as extraction and chemical reactions.

Many of the physical and transport properties of a SCF are intermediate between those of

a liquid and a gas. Diffusivity in an SCF, for example, falls between that in a liquid and a

gas; this suggests that reactions that are diffusion limited in the liquid phase could

become faster in a SCF phase.

SCFs already have been investigated for a number of systems, including enzymes

reactions, Diels-Alder reactions, organometallic reactions, heterogeneously catalyzed

reactions, oxidations and polymerizations.

Case Study of Process Intensification

The isomerisation of pinene oxide to campholenic aldehyde (Equation 1) is an important

reaction for the fragrance industry.

The reaction is complex, and a simplified schematic (Equation 2) shows how there are

four other major products, besides the desired campholenic aldehyde, are formed and

how this product (campholenic aldehyde) itself can react further to another five by-

products. A novel silica-supported zinc triflate catalyst was selected for the work. A

number of other heterogeneous catalysts are used for the reaction, though homogeneous

zinc halides are used commercially.

The results described below are expressed as percentage disappearance of the pinene

oxide (conversion), and efficiency of conversion of the disappeared pinene oxide to

desired campholenic aldehyde product (selectivity).

17

An amount of optimization work was performed in stirred flasks. When stirring a 1 %

solution of pinene oxide in 1.2.dichloroethane solvent with catalyst at 85 C, conversion

achieved was 100 % after 5 min, with selectivity on a plateau of 63-65 % at between 3

18

and 10 min reaction time, and a peak of 65 % at 5 min. Multiplying conversion and

selectivity together to give a product yield, we see a peak of 63-65 % at between 5 and 10

min.

Spinning disc work was first performed as a series of 15 two-minute runs at 25 to test

catalyst stability. The catalyst was attached to the disc surface with adhesive, the

conversion remained constant for each run at 20 %. Three consecutive passes with the

same material gave 55 % conversion at 84 % selectivity – which was encouraging.

A range of runs was now performed at 85, varying disc. The disc residence times was

thus explored in the range 0.5 to 5 sec. The optimum condition proved to be using the

highest tested feed rate of 6 ml/s and a spin-speed of 1000-1200 rpm. Lower spin speeds

gave 100 % conversion and much reduced selectivity – presumably attributable to over-

reaction of product with the longer residence time on the disc surface at lower rpm. Even

higher spin speeds gave up to 62 % selectivity – but at a reduced conversion of 75 %. The

optimum yield (conversion*selectivity) was stable at approximately 55 % at between

1000 and 1500 rpm.

The results show that the catalyst can be extremely effective on the disc, whilst avoiding

the filtration and recovery step required in stirred ranks. A bonus – which is often

experienced with intensified devices – is that the rate of data collection and process

optimization was enormous compared with stirred flask development. A new data point

could be generated every few minutes, a the generous equilibration time given after

adjusting feed-rate or spin-speed was a matter of 10-20 sec, and sample collection time

was only one minute.

19

Lecture-3

PROCESS INTEGRATION

Bikash Mohanty

Professor



Process integration, a part of Process Intensification, is a fairly new term that emerged in

80’s and has been extensively used in the 90’s to describe certain systems oriented

activities related primarily to process design. It has incorrectly been interpreted as Heat

Integration by a lot of people, probably caused by the fact that Heat Recovery studies

inspired by Pinch Concept initiated the field and is still core elements of Process

Integration. It appears to be a rather dynamic field, with new method and application

areas emerging constantly. The Process Integration is defined as “systematic and general

methods for designing integrated production systems, ranging from individual processes

to total sites, with special emphasis on the efficient use of energy and reducing

environmental effects”.

This definition brings Process Integration very close to Process Synthesis, which is

another systems oriented technology. Process Integration and synthesis belongs to

process systems engineering. Process Integration has evolved from a heat recovery

methodology in the 80’s to become what a number of leading industrial companies in

90’s regarded as a “major strategic design and planning technology”. With this

technology, it is possible to significantly reduce the operating cost of existing plants,

while new processes often can be designed with reduction in both investment and

operating costs.

20

Definition of Process Integration as per International Energy Agency (IEA)

• Process Integration is the common term used for the application of methodologies

developed for System- oriented and Integrated approaches to industrial process

plant design for both new and retrofit applications.

• Process Integration refers to Optimal Design; examples of aspects are: capital

investment, energy efficiency, emissions, operability, flexibility, controllability,

safety and yields. Process Integration also refers to some aspects of operation and

maintenance.

• Process integration, combined with other tools such as process simulation, is a

powerful approach that allows engineers to systematically analyze an industrial

process and the interactions between its various parts.

Current Status of Process Integration

Process Integration is a strongly growing field of Process Engineering. It is now standard

curriculum for process engineers in both Chemical and Mechanical Engineering at most

universities around the world, either as a separate topic or as part of a Process Design or

Synthesis course. Research at UMIST has for 25 years been supported by a large number

of industrial companies through a Consortium that was established in 1984. As part of the

International Energy Agency (IEA) project on Process Integration, more than 50 other

universities around the world involved in research in this field have been identified.

From History to the Future

Process Design has evolved through distinct "generations". Originally (first generation),

inventions that were based on experiments in the laboratory by the chemists, were tested

in pilot plants before plant construction.

21

The second generation of Process Design was based on the concept of Unit Operations,

which founded Chemical Engineering as a discipline. Unit Operations acted as building

blocks for the engineer in the design process.

The third generation considered integration between these units; for example heat

recovery between related process streams to save energy.

A strong trend today (fourth generation) is to move away from Unit Operations and focus

on Phenomena. Processes based on the Unit Operations concept tend to have many

process units with significant and complex piping arrangements between the units. By

allowing more than one phenomena (reaction, heat transfer, mass transfer, etc.) to take

place within the same piece of equipment, significant savings have been observed both in

investment cost and in operating cost (energy and raw materials).

Different Schools of Thoughts in Process Integration

The three major features of Process Integration methods are the use heuristics (insight),

about design and economy, the use of thermodynamics and the use of optimization

techniques. There is significant overlap between the various methods and the trend today

is strongly towards methods using all three features mentioned above. The large number

of structural alternatives in Process Design (and Integration) is significantly reduced by

the use of insight, heuristics and thermodynamics, and it then becomes feasible to address

the remaining problem and its multiple economic trade-offs with optimization techniques.

Despite the merging trend mentioned above, it is still valid to say that Pinch Analysis and

Exergy Analysis are methods with a particular focus on Thermodynamics. Hierarchical

Analysis and Knowledge Based Systems are rule-based approaches with the ability to

handle qualitative (or fuzzy) knowledge. Finally, Optimization techniques can be divided

22

into deterministic (Mathematical Programming) and non-deterministic methods

(stochastic search methods such as Simulated Annealing and Genetic Algorithms). One

possible classification of Process Integration methods is to use the two-dimensional

(automatic vs. interactive and quantitative vs. qualitative) representation in Fig. 1.

Application of Process Integration

Process Integration can be applied in following fields of chemical engineering such as:

1. Heat integration – heat exchange network

2. Distillation column targeting

3. Cogeneration and total site targeting

4. Batch process targeting

5. Emission targeting

6. Mass exchange network (water and wastes water management & recovery of

valuable materials)

7. Hydrogen management in refineries

Hierarchical

Analysis

Heuristic

Rules Knowledge

Based Systems

Thermodynamic

Methods Optimization

Methods

qualitative

quantitative

interactive automatic

Fig. 1 One possible Classification of Process Integration

Methods

23

Techniques Available for Process Integration

1. Pinch Technology Approach

2. MILP/MINLP Approach

3. State-Space Approach

4. Genetic Algorithm Approach

5. Process Graph Theory Approach

Concept of Pinch Technology

The term "Pinch Technology" was introduced by Linnhoff and Vredeveld to represent a

new set of thermodynamically based methods that guarantee minimum energy levels in

design of heat exchanger networks. Over the last two decades it has emerged as an

unconventional development in process design and energy conservation. The term ‘Pinch

Technology’ is often used to represent the application of the tools and algorithms of

Pinch Technology for studying industrial processes.

Reactor

Separator

Heat exchange network

Utilities

The heat and material

balance is at this

boundary

Site-Wide Utilities

Fig. 2 Onion Diagram

1

2

3

4

24

Pinch technology provides a systematic methodology for energy saving in processes and

total sites. Fig. 2 illustrates the role of Pinch Technology in the overall process design.

The process design hierarchy can be represented by the “onion diagram” as shown below.

The design of a process starts with the reactors (in the “core” of the onion). Once feeds,

products, recycle concentrations and flow rates are known, the separators (the second

layer of the onion) can be designed. The network (the third layer) can be designed. The

remaining heating and cooling duties are handled by the utility system (the fourth layer).

The process utility system may be a part of a centralized site-wide utility system.

A Pinch Analysis starts with the heat and material balance for the process. Using Pinch

Technology, it is possible to identify appropriate changes in the core process conditions

that can have an impact on energy savings (onion layers one and two). After the heat and

material balance is established, targets for energy saving can be set prior to the design of

the heat exchanger network.

Data Extraction

The most time consuming and often most critical step is the identification of the need for

heating, cooling, boiling and condensation in the process. This task is more art than

science, and if not carried out properly, the final design will not be the best possible. It is

quite easy to accept too many feature of the proposed flow sheet, which inevitably results

in the situation where many good opportunities are excluded from the analysis.

In practice, there are a number of situations where heat integration is not desirable.

Examples include long distances (costly piping), safety (heat exchange between

hydrocarbon streams and oxygen rich streams), product purity (potential leakage in heat

exchangers), operability (start-up and shut-down), controllability and flexibility. A

25

reasonable strategy is, however, to start by including all process streams and keep the

degrees of freedom open. Later, practical considerations can be used to exclude some of

these streams and degrees of freedom, and the engineer will then at any time be able to

establish the consequences with respect to energy consumption and total annual cost. A

central part of data extraction is the identification of heating and cooling requirements in

the process. The necessary data for each process stream are the following:

m = mass flowrate (kg/s, tons/h, etc.)

Cp = specific heat capacity (kJ/kgC)

Ts = supply temperature (C)

Tt = target temperature (C)

Hvap = heat of vaporization for streams with a phase change (kJ/kg)

Additionally, the following information must be collected on utilities and existing heat

exchangers for retrofit:

Existing heat exchanger area (m2)

Heat transfer coefficient for cold and hot sides of heat exchangers (kW / m2 C).

Utilities available in the process (water temperature, steam pressure levels, etc),

Marginal utility costs, as opposed to average utility costs.

Data extraction must be preformed carefully as the results strongly depend on this step. A

key objective of data extraction is to recognize which parts of the flowsheet are subject to

change during the analysis (e.g. possibility of making modifications to the piping, or

adding new heat exchangers, possibility of making temperature changes in the process or

modifying the utility that heats a given piece of equipment (MP steam instead of HP

26

steam for example), etc). If, during extraction, all features of the flowsheet are considered

to be fixed, there will clearly be no scope for improvement.

At the beginning of a project it is recommended that all process stream be included in the

data extraction. Constraints regarding issues such as distance between operations,

operability, control and safety concerns can be incorporated later on. By proceeding in

such a fashion, it is possible to have an objective evaluation of the costs of imposing such

constraints. PI specialists generally include some constraints form the beginning of the

data extraction procedure. This can speed up the overall analysis, but a lot of experience

is required to ensure that potentially interesting heat-recovery projects are not excluded.

There are a lot of sector specifics for data extraction. However, heuristic rules have been

developed as guidelines. The following are the most relevant:

Do not mix streams at different temperatures. Direct non-isothermal mixing acts as a heat

exchanger. Such mixing may involve cross-pinch heat transfer, and should not become a

fixed feature of the design. For example, if the pinch is located at 70C, mixing a stream

at 90C with a stream at 50C creates a cross pinch, and will increase the energy targets.

The way to extract these streams is to consider them independently, i.e., one stream with

a supply temperature of 50C and the required target temperature, and the other stream

with a supply temperature of 90C and the required temperature.

Do not include utility streams (stream, flu gas, cooling water, refrigerant, cooling air,

etc.) in the process data unless they are involved directly in the process or they cannot be

replaced. One of the goals of using pinch analysis is to reduce the usage of utilities.

Therefore, if utility streams are extracted in a similar way to process streams, they will be

considered as fixed requirements and no opportunities of reduction in utility use will be

27

identified. In some cases, utility streams can be included because it is not practical to

replace them by any form of heat recovery. For example, this is often the case for stream

dryers, ejectors and turbine drives.

Do not consider the existing plant layout. When selecting the inlet and outlet parameters

for a process stream, existing heat exchange equipment and plant topology should not be

taken into account at first. True utility targets (for cooling and heating) should be set

regardless of the existing plant layout. Current plant energy consumption can then be

compared with minimum energy targets. In retrofit of existing facilities, once these

targets have been determined, plant layout (existing heat exchangers and piping,

distances, etc) needs to be taken into account in order to identify practical and cost-

effective projects to reach or approach these targets.

Identify hard and soft constraints on temperature levels. For example, a hard constraint

would be the inlet temperature of a reactor that cannot be changed in any way, while a

soft constraint would be the discharged temperature of a product going to storage, for

which the target temperature is often flexible.

Data extraction is a complex issue, and a significant part of the pinch specialist’s

expertise is related to building a good pinch model during the data extraction phase.

28

Lecture – 4

PINCH TECHNOLOGY – AN OVERVIEW

Shabina Khanam

Lecturer


National Institute of Technology Rourkela, Rourkela – 769 008

One of the most practical tools to emerge in the field of process integration in the past 20

years has been pinch analysis, which may be used to improve the efficient use of energy,

hydrogen and water in industrial processes. Pinch analysis is a recognized and well-

proven method in each of the following industry sectors:

Chemical

Petrochemical

Oil refinery

Pulp and paper

Steel and metallurgy

Food and drink

Over the past 20 years, pinch analysis has evolved and its techniques perfected. It

provides tools that allow us to investigate the energy flows within a process, and to

identify the most economical ways of maximizing heat recovery and of minimizing the

demand for external utilities (e.g., steam and cooling water). The approach may be used

to identify energy-saving projects within a process or utility systems.

Pinch technology analyses process utilities (particularly energy and water) to find the

optimum way to use them, resulting in financial savings. Pinch Technology does this by

making an inventory of all producers and consumers of these utilities and then

29

systematically designing an optimal scheme of utility exchange between them. Energy &

water re-use are at the heart of pinch technology. With the application of pinch

technology, both capital investment and operating cost can be reduced. Emissions can be

minimised and throughput maximised.

The Pinch Concept

Pinch analysis (or pinch technology) is a rigorous, structured approach that may be used

to tackle a wide range of improvements related to process and site utility. This includes

opportunities such as reducing operating costs, debottlenecking processes, improving

efficiency, and reducing and planning capital investment.

Major reasons for the success of pinch analysis are the simplicity of the concepts behind

the approach, and the impressive results it has been obtained worldwide. It analyzes a

commodity, principally energy (energy pinch) hydrogen (hydrogen pinch), or water

(water pinch), in terms of its quality and quantity, recognizing the fact that the cost of

using that commodity will be a function of both.

In general, we are using high-value utilities in our process and rejecting waste at a low

value. For example, if we consider energy, we may be burning expensive natural gas to

provide the process with high temperatures heat, and are rejecting heat at low

temperatures to cooling water or air.

Pinch analysis now has an establishment track record in energy saving, water reduction,

and hydrogen system optimization. In all cases, the fundamental principle, behind the

approach is the ability to match individual demand for a commodity with suitable supply.

The suitability of the match depends on the quality required and the quality offered. In

the context of utility management, the commodity may be heat, with its quality measured

30

as temperature. By maximizing the match between supplies and demands, we minimize

the import of purchased utilities (Fig. 1).

Pinch Technology Versus Process Engineering

Pinch Technology is a vital subdivision of process engineering.

WASTE

(a)

Process

HIGH QUALITY UTILITY

QUANTITY

QU

AL

ITIY

(b)

Process

HIGH QUALITY UTILITY

WASTE

QUANTITY

QU

AL

ITY

Pinch Technology

Pinch

ENERGY: WATER: HYDROGEN

MINIMISE

MINIMISE

Fig.1 Schematic process utility use

31

Carrying out a process engineering project without the input of a pinch study will

lead to a less efficient design.

Our engineers have specialized knowledge of thermodynamics and computer

analysis tools. They can communicate effectively with clients and undertake

conceptual designs. This explains why we are uniquely qualified to help you get

the most out of your pinch projects.

How is Pinch technology different from other energy audits?

Pinch technology reveals all the possible savings and their corresponding Financial

benefits.

• It defines the maximum possible savings.

• It looks at the overall site.

• It does not bench-mark but takes into account all specific mill factors, age,

location, process equipment, operating preferences, product, etc.

• It reveals the maximum cogeneration potential.

Role of Thermodynamic Laws in Pinch Technology

Pinch technology presents a simple methodology for systematically analyzing chemical

processes and the surrounding utility systems with the help of the First and Second Laws

of Thermodynamics. The First Law of Thermodynamics provides the energy equation for

calculating the enthalpy changes (dH) in the streams passing through a heat exchanger.

The Second Law determines the direction of heat flow. That is, heat energy may only

flow in the direction of hot to cold. This prohibits ‘temperature crossovers’ of the hot and

32

cold stream profiles through the exchanger unit. In a heat exchanger unit neither a hot

stream can be cooled below cold stream supply temperature nor a cold stream can be

heated to a temperature more than the supply temperature of hot stream. In practice the

hot stream can only be cooled to a temperature defined by the ‘temperature approach’ of

the heat exchanger. The temperature approach is the minimum allowable temperature

difference Tmin) in the stream temperature profiles, for the heat exchanger unit. The

temperature level at which Tmin is observed in the process is referred to as "pinch point"

or "pinch condition". The pinch defines the minimum driving force allowed in the

exchanger unit.

What Processes does Pinch Apply to?

Pinch applies to a wide range of processes. Pinch originated in the petrochemical sector

and is now widely accepted in mainstream chemical engineering. With a wealth of

applications experience, benefits can now be realized in many other process industries.

Wherever heating and cooling of process materials takes places there is a potential

opportunity. A realistic approach addresses the practical problems specific to each and

every site, leading to:

• Meaningful targets

• Feasible projects

• Real savings

• Essential strategic insights

Benefits of Pinch Technology

Pinch tells the best that can be achieved in a given system.

33

Pinch gives the practical target to aim for that is less than this theoretical

maximum.

Both of the above are done before any detailed design. This target then set the

basis for the design. Most importantly, it gives clear rules about how to construct

a design to achieve the targets. It will also show where the inefficiency lie in the

existing design.

Pinch takes a system-wide view of the problem. This allows one to see interaction

that would be difficult to spot on a process flow diagram or a flow sheet of site

utility system.

Pinch can work with incomplete data. One can refine the data in the areas where

accuracy is most important. This is in the area around the pinch.

Pinch Technology is in contrast to other design tools, which require detailed

information about geometry, flow sheet structure, etc. Pinch technology is one of

the few tools that really can be used in conceptual design.

Problem Addressed by Pinch Technology

Generally two types of problem are addressed:

Creating new designs

This is related to the design of HEN for a new plant, which is in design stage.

The ideal time to apply pinch analysis is during the planning of process

modifications that will require major investments, and before the finalization of

process design. Maximum improvements in energy efficiency; along with reduced

34

investments can be obtained in a new plant design, since many plant layout and

process constraints can be overcome by redesign.

Retrofit – Revamping existing designs

This is related to the retrofitting of an already existing HEN in a plant to improve

its exchange efficiency.

However, in retrofit projects, energy efficiency improvements usually require

some capital expenditure. In this case, pinch analysis can be specifically aimed at

maximizing the return of investment. Pinch analysis techniques allow us to

evaluate combinations of project ideas simultaneously, in order to avoid “double –

counting” savings, as well as conflicting projects. Indeed, the final investment

strategy for the available opportunities will ensure that site development is

consistent and synergistic.

Typical Savings

• BASF AG (Ludwigshafen, Germany), for example, has completed more than 150

retrofit using pinch technology, achieving over 25 % in energy savings site wide.

• In natural gas sweetening, for example, The Ralph M. Parsons Co. (Pasadena,

Calif.) says that pinch technology led to a 10% drop in capital costs and energy

use in its amine absorption column.

• GE plastics was faced with a requirement of invest $15 million in doubling the

capacity of the wastewater handling system of its Silicones Production Facilities

in Netherlands. Linnhoff March aimed to avoid this investment cost by reducing

wastewater flow by 50 %.

• The following benefits have been obtained for refinery retrofits:

35

a) Energy reduced by 15-35 % through revamping of HENs based on

paybacks of 1.5-3 years.

b) Units debottlenecked by 10-20% without modifying fired heaters or major

pumps.

c) Lower fouling from improved understanding of the system dynamics.

d) Improved flexibility giving the lowest cost design for different operating

cases.

e) Reduced emissions at the source.

The potential energy and water consumption savings in major industries sectors

are given in Fig. 2 & 3.

Fig. 2 Potential energy savings

36

Fig. 3 Potential water consumption savings

37

Lecture – 5, 6, 7

BASIC ELEMENTS OF PINCH TECHNOLOGY – PART I, II & III

Bikash Mohanty

Professor



KEY STEPS OF PINCH TECHNOLOGY

There are four key steps of pinch analysis in the design of heat recovery systems for both

new and existing processes:

1) Data Extraction, which involves collecting data for the process and the utility

system.

2) Targeting, which establishes figures for best performance in various respects.

3) Design, where an initial Heat Exchanger Network is established.

4) Optimization, where the initial design is simplified and improved economically.

Data Extraction

The details of data extraction are discussed in Lecture 3.

Targeting

An important feature of Process Integration is the ability to identify Performance Targets

before the design step is started. For heat recovery systems with a specified value for the

minimum allowable approach temperature (Tmin), targets can be established for

Minimum Energy Consumption (external heating and cooling), Fewest Number of Units

(process/process heat exchangers, heaters and coolers) and Minimum Total Heat Transfer

38

Area. In addition, the corresponding calculations will also identify the Heat Recovery

Pinch, which acts as a bottleneck for heat recovery.

Designing

Design of Heat Exchanger Networks in various industries is primarily carried out using

the now classical Pinch Design Method (Linnhoff and Hindmarsh, 1983). While the

original method focused on minimum energy consumption and the fewest number of

units, later graphical and numerical additions made it possible also to consider heat

transfer area and total annual cost during design.

The basic Pinch Design Method respects the decomposition at Process and Utility Pinch

points and provides a strategy and matching rules that enable the engineer to obtain an

initial network, which achieves the minimum energy target.

The Pinch Design Method also indicates situations where stream splitting is required to

reach the minimum energy target. Stream splitting is also important in area

considerations and the optimal use of temperature driving forces.

The design strategy mentioned above is simply to start design at the Pinch, where driving

forces are limited and the critical matches for maximum heat recovery must be selected.

Optimization

Heat exchange network for maximum energy recovery established by pinch design

method, should only be regarded as initial designs and some final optimization is

required. The matches in the initial network depend on pinch location and since the pinch

point depends on the value of Tmin, this becomes a key parameter in the pinch design

method. By repeating all calculations, for synthesis of HEN, for different values of Tmin,

39

it is possible to identify a good starting value for the level of heat recovery. This exercise

of pre-optimization has been referred to as “Supertargeting”. For a typical Problem, the

minimum total annual cost is obtained to be 240.42103 $/yr (Fig. 1). Thus, the optimum

ΔTmin is 13 °C.

BASIC ELEMENTS OF PINCH TECHNOLOGY

Grid Representation

The grid is used to represent heat exchange network more conveniently. The important

features of grid representations are:

Hot streams (streams which require cooling) are drawn at the top running let to

right.

Cold streams (streams which require heating) are drawn at the bottom running

right to left.

The Total Annual Cost Profile

0

50

100

150

200

250

300

350

400

0 20 40 60

Minimum temperature difference

TA

C (

1000 $

/yr)

Δ Tmin Optimum = 13 °

Fig. 1 The total annual cost profile

40

A heat exchanger is represented by a vertical line joining two open circles on the

streams being matched. The heat exchanger load can conveniently be written

under the lower open circle.

Heaters (H) and coolers (C) can be represented in an open circle on the stream

being heated or cooled.

Temperatures can be put on the grid as shown to allow an easy check on the

terminal approach temperature for each unit.

The stream data for the typical process is shown in Table 1. The grid representation for

this process, which includes two hot, H1 & H2, and two cold, C3 & C4, streams, are

shown in Fig.2.

Table 1 The Stream Data for the Process

Stream Ts (oC) Tt (

oC) MCp (kW/ C)

H1 175 45 10

H2 125 65 40

C3 20 155 20

C4 40 112 15

H

2

3

C

175

125

155

112

1400

98

85

1080

1300

1320

45

65

20

40

MCp (kW/ C)

10

40

20

15

Stream

H1

H2

C3

C4 2

Fig. 2 The grid representation of the process

3

41

Composite Curve

The Composite Curves (CCs) are constructed from ‘stream data’ representing a process

heat and material balance. The CCs allow the designer to predict-optimized-hot and cold

utility targets ahead of design, to understand driving forces for heat transfer, and to locate

the heat recovery ‘Pinch’. CCs consist of temperature-enthalpy (T-H) profiles of heat

availability in the process (the “hot composite curves”) and heat demands in the process

(the “cold composite curves”) together in a graphical representation. CCs also provide the

minimum requirement of hot and cold utilities in the process.

The construction of the hot composite curves (as shown in Fig.3) simply involves the

addition of the enthalpy changes of the streams in the respective temperature intervals.

The CCs for the stream data, given in Table 1, are shown in Fig.3. The QHmin and QCmin

are minimum hot and cold utilities.

0

50

100

150

200

0 1000 2000 3000 4000 5000

Heat Content Q (kW)

T (oC)

HCC CCC

Region of heat recovery by

process to process exchange

QHmin

QCmin

Tmin

Above

pinch Below

pinch

Fig. 3 The hot composite curves (HCC) and cold composite curves (CCC) respectively

show the heat availability and heat requirement for the overall process.

42

Problem Table Algorithm

This graphical manipulation of composite curves to generate minimum targets is time

consuming and clumsy. An alternative procedure is entirely based on simply arithmetic

and involves no trial and error.

The procedure is known as the problem table and is broken down into three stages.

1. Set up shifted temperature intervals from the stream supply and target

temperatures by subtracting ΔTmin /2 from the hot streams and adding ΔTmin /2 to

the cols streams.

It is important to note that shifting the curves vertically does not alter the

horizontal overlap between the curves. It therefore does not alter the amount by

which the cold composite curve extends beyond the start of hot composite curve

at the hot end of problem. Also, it does not alter the amount by which hot

composite curve extends beyond the start of cold composite curve at the cold end.

2. In each shifted temperature interval, calculate a simple energy balance from:

(1)

Where ΔHi = heat balance for shifted temperature interval i and ΔHi is the

temperature difference across it

CPc = specific heat capacity of a cold stream (MW/oC)

CPh = specific heat capacity of a hot stream (MW/oC).

If the cold streams dominate the hot streams in a temperature interval, then the

interval has a net deficit of heat, and ΔH is positive. If hot streams dominate cold

streams, the interval has a net surplus of heat, and ΔH is negative.

43

3. Now, cascade any surplus heat down the temperature scale from interval to

interval. This is possible because any excess heat available from the hot streams

in an interval is hot enough to supply a deficit in the cold streams in the next

interval down. First, assume no heat is supplied to the first interval from hot

utility. As a consequence of it some of the heat flows are negative, which is

infeasible. Heat cannot be transferred up the temperature scale. To make the

cascade feasible, sufficient heat must be added from hot utility to make the heat

flows to be at least zero. The smallest amount of heat needed from hot utility is

the largest negative heat flow.

Example

The problem table algorithm is explained using the stream data of a typical process given

in Table 2. The minimum approach temperature is 10 °C. The shifted temperatures for

each stream are detailed in Table 3.

Table 2 Stream data

Stream

Heat capacity flow rate

(MW/°C)

Ts (°C) Tt (°C)

Cold (C1) 0.2 20 180

Hot (H1) 0.15 250 40

Cold (C2) 0.3 140 230

Hot (H2) 0.25 200 80

44

Table 3 Stream Data with Shifted Temperature

Stream

Heat capacity flow rate

(MW/°C)

T*s (°C) T*t (°C)

Cold (C1) 0.2 25 185

Hot (H1) 0.15 245 35

Cold (C2) 0.3 145 235

Hot (H2) 0.25 195 75

The shifted temperatures are arranged in decreasing order. The stream population is

shown in Fig. 4 with a vertical temperature scale. The interval temperatures shown in Fig.

4 are set to ΔTmin /2 below hot stream temperatures and ΔTmin /2 above cold stream

temperatures.

Fig. 4 The stream population for stream

data shown in Table 2

45

Then a heat balance is carried out within each shifted temperature interval according to

Eq. 1. The result is given in Fig. 5, in which some of the shifted intervals are seen to have

a surplus of heat and some have a deficit.

Fig. 5 The temperature interval heat balances

Now, cascade any surplus heat down the temperature scale from interval to interval

assuming no heat is supplied to the first interval from hot utility (Fig. 6). The first interval

has a surplus of 1.5 MW, which is cascaded to the next interval. This second interval has

a deficit of 6 MW, which leaves the heat cascaded from this interval to be -4.5 MW and

so on. Some of the heat flows are negative, which is infeasible. To make the cascade

feasible, largest negative heat flow from Fig. 6 that is 7.5 MW is added from hot utility to

make the heat flows to be at least zero. The revised cascade is shown in Fig. 7 which

gives one heat flow of just zero at an interval temperature of 145 °C.

46

More than 7.5 MW could be added from hot utility to the first interval, but the objective

is to find minimum hot and cold utility. Thus, from Fig. 7 minimum hot and cold utilities

are 7.5 MW and 10 MW, respectively. The point where the heat flow goes to zero at

shifted temperature 145°C corresponds to the pinch. Thus, the actual hot and cold stream

pinch temperatures are 150 °C and 140 °C, respectively.

The composite curves are useful in providing conceptual understanding of the process but

the problem table algorithm is a more convenient calculation tool.

Fig. 6 Cascaded surplus heat from high to

low temperature

Fig. 7 Add heat from hot utility to make

all heat flows zero or positive

47

Grand Composite Curve

The grand composite curve (GCC) is a graphical representation of the heat cascade. GCC

is based on the same process stream data as Composite Curves. GCCs highlight the

process/utility interface. It gives clear visualization of hot and cold utility and provides an

easy approach to use multiple utilities in the process. For the stream data, shown in Table

1, the GCC is represented in Fig. 8.

Maximum Energy Recovery

The overlap between the hot and cold composite curves represents the maximum amount

of heat recovery possible within the process. The source/sink characteristics of process

heat exchange systems give five concepts.

Targets: Once the composite curves are known, we know exactly how much external

heating/cooling is required. Near-optimal processes are confirmed as such and non-

optimal processes are identified with great speed and confidence.

Pinch High temperature process

sink profile

Low temperature process

source profile

Hot utility

Cold Utility

Process to process

heat exchange

Above Pinch

Below Pinch

48

The pinch: The process needs external heating above the pinch and external cooling

below the pinch. This tells us where to place furnaces, steam heaters, coolers etc.

More in, more out: An inefficient process requires more than the minimum external

heating and therefore more than the minimum external cooling. For every units of excess

external heat in a process one has to provide heat transfer equipment twice. This insight

helps us to improve both energy and capital cost.

Freedom of choice: The “heat sink” and the “heat source” in Fig. 8 are separate. This

constraint helps the designer to choose plant-layouts, control arrangements etc. If

designer violates this constraint, he can evaluate the pinch heat flow and therefore predict

what overall penalties will be involved.

Trade-offs: A simple relationship exists between the number of streams (process streams

plus utilities) in a problem and the minimum number of heat exchange units (i.e. heaters,

coolers and interchangers).

Thus if designer goes for best energy recovery, designing the “heat source” and “heat

sink” section separately, he or she will incur the need for more units than if the pinch

division had been ignored. Hence a new type of trade-off has been identified, between

energy recovery and number of units. This insight adds to the traditional concept of a

trade-off between energy and surface area.

49

Lecture – 8

AREA TARGETING

Bikash Mohanty

Professor



Area is important in determining heat exchanger network capital cost. Before explaining

the complete procedure to computation of area it is necessary to discuss the principles for

minimum area in heat exchanger networks.

Start by considering the example in Fig. 1a, where two hot streams exchange heat against

a single cold stream. If we assume the overall heat transfer coefficient U is constant for

all exchangers and these exchangers are countercurrent units then the network has an area

of 88 m2.

Fig. 1b shows a different network with stream splitting. Its area is 84 m2. The reason is

that it has better countercurrent behavior in terms of the overall network. In Fig. 1a the

matches are in temperature sequence whereas in Fig. 1b the matches share more of the

available temperature differences by splitting the cold stream. Fig. 1c shows that we can

do better still. The network area is now 77 m2. This is the minimum area for the stream

set as defined. The network has been developed by stream-splitting only where streams

compete for the same driving forces by overlap in temperature.

The composite curve of the data for example, shown through Fig. 1, is drawn in Fig. 2.

Overall countercurrent heat exchange now appears as vertical heat transfer on the

composite curves. Partitioning of the stream data to follow the temperatures of the

vertical model then leads to the minimum area design for this example.

50

Fig. 1 (a) network with exchangers in temperature sequence on cold stream; (b) network

with exchangers sharing temperature span of cold stream; and (c) network with

exchangers showing correct distribution of temperatures for minimum area.

51

Fig. 2 Resolving temperature contention using the composite curves: (a) overall

countercurrent heat exchange appears as vertical heat transfer on the composites; (b) the

temperatures of enthalpy intervals show where stream-splitting will be required, (c) these

temperatures can be marked on the grid; and (d) used to guide design for temperature

contention.

To calculate the heat exchanger network area from composite curve, utility streams must

be included with the process streams in the composite curves to obtain the balanced

composite curves (BCC). The resulting BCC (Fig. 3a) should have no residual demand

for utilities. The BCC are divided into vertical “enthalpy intervals”. The intervals are

defined whenever a change in slope occurs in either balanced composite profile. Next, a

network design is considered within each enthalpy interval, which can satisfy vertical

52

heat transfer. Fig. 3b demonstrates this for an interval, which contains two hot streams

and three cold streams. Each hot stream is split into the same number of branches as the

number of cold streams in that interval. Similarly, each cold stream is split into the same

number of branches as the number of hot streams in that interval. Hence, each hot stream

can be matched with each cold stream such that every match occurs between the corner

temperatures of the enthalpy interval. The heat exchanger of these matches must

therefore appear as vertical on the BCC.

Fig. 3. Example of general stream splitting and matching scheme for vertical heat

transfer in an enthalpy interval of the balanced composite curves.

The minimum total area could be taken as the sum of the areas of all such exchangers

from all enthalpy intervals. However, this is not necessary if U = constant. From the

composite curves, the area from vertical heat transfer in interval i is simply:

(1)

53

where ΔHi is the enthalpy width of interval i and ΔTLM,i is the logarithmic mean

temperature difference of interval i.

Hence, the total minimum network area is given by:

(2)

This shows that in order to derive an area target based on U = constant no design is

required.

Different heat transfer coefficients in the model for minimum area

Consider again the design in Fig. 3 for vertical heat transfer in enthalpy interval i of the

composite curves. If the heat transfer coefficients differ then the total area of these

exchangers is:

(3)

Where, Q13 is the duty of the match between streams 1 and 3, U13 its overall heat transfer

coefficient, etc.

Now,

(4)

where h1 is the heat transfer coefficient of stream 1 (including film, wall and fouling

resistances), etc.

54

So,

(5)

But

(6)

where (qj)i is the enthalpy change of stream j in enthalpy interval i.

so,

(7)

The argument applies in general for other enthalpy intervals. Summing up over all

intervals on the composite curves gives:

(8)

This simple formula incorporates stream individual heat transfer coefficients and allows a

“target” for the minimum heat exchange area to be calculated from the composite curves.

Further, within ith

enthalpy interval, all hot streams undergo the same temperature change

(dTh)i as do all the cold streams (dTc)i. As q = MCpdT, then

55

(9)

Example:

Stream Data of a typical process with Tmin = 20˚ C is given in following table.

Stream(s) Ts (C) Tt (C) MCp (kW/ C) h (kW/m2 C)

H1 175 45 10 0.2

C1 20 155 20 0.2

H2 125 65 40 0.2

C2 40 112 15 0.2

Steam (HU) 180 179 - 0.2

Cold Water (CU) 15 25 - 0.2

The step wise procedure is described below:

Calculation of minimum hot and cold utilities

Minimum hot and cold utilities are calculated by Problem Table Algorithm which are as

follows:

Hot utility, Qhu,min = 605 kW

Cold utility, Qcu,min = 525 kW

Calculation of utility flow rates

The MCp values of hot utility (hu) and cold utility (cu) are given as:

(MCp)hu = Qhu,min/(Tin-Tout)hu = 605/(180-179) = 605 kW/° C

(MCp)cu = Qcu,min/(Tout-Tin)cu = 525/(25-15) = 52.5 kW/° C

jc jc

p

ic

jh jh

p

ih

ervals

i

iLMh

MCdT

h

MCdTTA )()(/1

int

min

56

Plotting the Balanced Composite Curves

The procedure for plotting the Balanced Hot Composite Curve and Balanced Cold

Composite Curve is the same as the Hot Composite Curve and Cold Composite Curve,

except that the utilities are also considered as additional streams.

Balanced Hot composite Curve (BHCC)

For BHCC the temperatures of hot streams and hot utility are arranged in ascending order

(Fig. 4). The sum of the MCP values of hot streams and utility present in each interval is

calculated. Then this sum is multiplied by the temperature difference of each interval.

After that a cumulative enthalpy is calculated using the formula:

CumQhb, i = CumQhb, i-1 + Qint, hbi (10)

Fig. 4 Data for balanced hot composite curve

Now, BHCC is obtained by plotting temperature and CumQhb as shown in Fig. 5.

Similarly Balanced cold composite curve can be drawn. The two curves are

superimposed on each other to get BCC as shown in Fig. 6. The BCC are divided into

vertical “enthalpy intervals”. The intervals are defined whenever a change in slope

occurs in either balanced hot composite curve (BHCC) and balanced cold composite

CumQhb Qhb

45

65

125

175

179

180

1

2

3

4

5

H1

H2

H

U

10

40

605

10

50

10

0

605

MCp,hb

3000

500

0

605

200 200

3200

3700

3700

4305

57

curve (BCCC) profiles. The BCC on being divided into enthalpy intervals, allow

calculation of the area target based on a model of vertical heat transfer.

Fig. 5 Data for balanced hot composite curve

Fig. 6 The balanced composite curve for the example

0

20

40

60

80

100

120

140

160

180

200

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

Heat Content Q (kW)

T (

C)

BHCC

BCCC

Interval i Th,i.-1

Tc,i.-1

Th,i.

Tc,i.

0

20

40

60

80

100

120

140

160

180

200

0 1000 2000 3000 4000 5000

Heat content Q, kW

Te

mp

era

ture

, D

eg

C

58

Determination of enthalpies for intervals

CumQhb and CumQcb (for BCCC) are merged by omitting cumulative enthalpies common

to both values and the entries are then sorted in ascending order. This identifies all points

where composite curve has a vertex (change in slope).

Calculation of interval temperatures on BHCC

The following formula is used for calculation of interval temperature:

Th3 = Thb,row r – (CumQhb,row r- CumQ3)/MCp,hb row r

Where, Thb,row r and CumQhb,row r are temperature and CumQ in the row r (in which the

temperature is available), In this case, row r = 6

For CumQi = 262.5 kW, Thi = 125˚ - (3200-262.5)/50 = 66.25˚C.

For CumQi = 200 kW, Tci = 20˚ - (262.5-200)/52.5 = 18.81˚C.

Similarly other temperature intervals are found and shown in Fig. 7.

Fig. 7 Determination of the enthalpy intervals

Calculation of (MCp/h)h and (MCp/h)c for each interval

66.25

73.5

79.5

149.5

18.81

105

124.5

124.5

59

These are calculated in a manner similar to MCp,hb of Fig. 4. For example, consider first

interval of Fig. 7 where only stream H1 exists, therefore (MCp/h)h = 10/0.2 = 50. Next

four interval contain streams, H1 and H2, thus, (MCp/h)h = 50/0.2 = 50. These data are

shown in Table 1.

Calculation of (Q/h)

For first interval, (Q/h) = (65˚ - 45˚)50 + (18.81˚ - 15˚)262.5 = 2000

The complete data are shown in Table 1.

Calculation of log mean temperature difference, TLM

This is done by the following formula:

For first interval:

TLM, 1 = [(65-18.81)-(45-15)]/[ln(65-8.81)/(45-15) = 37.51˚ C.


Calculation of countercurrent exchanger area in each interval

This is calculated by dividing the (Q/h) by the corresponding TLM in for the interval.

For first interval: A1=2000/37.51 = 53.31 m2


Based on above calculation the minimum area is found as 1312.57 m2

for the example

undertaken.

1,1,

,,

1,1,,,

ln

)()(

icih

icih

icihicih

LM

TT

TT

TTTTT

60

Table 1 Calculation of countercurrent exchanger area

int Thi Tci (MCp/h)h (MCp/h)c (Q/h) TLM, i Ai

0 45 15 0 0 0 0 0

1 65 18.81 50 262.5 2000 37.51 53.31

2 66.25 20 250 262.5 625 46.22 13.52

3 73.5 25 250 362.5 3625 47.37 76.53

4 79.5 40 250 100 3000 43.85 68.42

5 125 105 250 175 22750 28.65 794

6 149.5 112 50 175 2450 27.84 88.01

7 175 124.75 50 100 2550 43.56 58.53

8 179 124.75 0 100 0 52.22 0

9 180 155 3025 100 6050 37.76 160.23

61

Lecture – 9 & 10

NUMBER OF UNIT, SHELL AND COST TARGETING

Shabina Khanam

Lecturer


National Institute of Technology Rourkela, Rourkela – 769 008

Number of unit targeting

The capital cost of chemical processes tends to be dominated by the number of items on

the flowsheet. This is certainly true of heat exchanger networks and there is a strong

incentive to reduce the number of matches between hot and cold streams.

To understand the minimum number of matches or units in a heat exchanger network,

Fig. 1 is considered which shows the heat loads on one hot stream and three cold streams

written within the circles representing the streams. The predicted hot utility load is shown

similarly. In this process only hot utility is required but no cold utility. The total system is

in enthalpy balance i.e. the total hot plus utility is equal to the total cold.

Matching Steam with Cold1 and maximizing the load completely satisfies or “tick off”

Steam, leaving 1165 units of heating required by Cold1. Matching Cold1 with Hot and

Steam

1068 Hot

2570

Cold1

2233

Cold2

413

Cold3

992

Fig. 1 Illustration of minimum number of units design.

1068

1165 413 992

62

maximizing the load on this match so that it “ticks off” the 1165 residual requirement on

Cold1, leaves 1405 residual heat available from Hot.

So following the principle of maximizing loads, i.e. “ticking off” stream or utility loads

or residuals, leads to a design with a total of four matches. This is in fact the minimum

for this problem.

Thus, Umin = N – 1

Where, Umin = minimum number of units (including heaters and coolers)

N = total number of streams (including utilities)

Another problem, Fig. 2(a) having two hot streams and two cold streams. Both hot and

cold utility are required. For this problem 5 (N-1) [Where, N = 6.0] units are required

which is obtained by putting the matches using ticking off loads or residuals loads to a

design.

Fig. 2(a). Number of unit is one less than the number of streams included utilities

63

Fig. 2(b). Same principle for separate components – “Subset Equality”

Fig. 2(c). One unit more for every loop

Fig. 2(b) shows a design having one unit less than previous design. The subset of streams

H2, C1 and CW is in enthalpy balance. Similarly, ST, H1 and C2 are in enthalpy balance

(which they must be if the total problem is in balance). What this means is that for the

given data set we can design two completely separate networks, with the formula Umin =

N – 1 applying to each individually. The total for the overall system is therefore (3-1)+(3-

1) = 4 units. This situation is termed “subset equality”

The new unit is placed between ST and C2 as shown in Fig. 2(c). The extra units

introduces what is known as a “loop” into a system. At the hot utility ST, the loop can be

64

traced through the connection to C1, from C1 to H1, from H1 to C2, and from C2 back to

ST.

Suppose the new match, which is between ST and C2, is given a load of X units. Then by

enthalpy balance the load on the match between ST and C1 is 30-X, between C1 and H1,

10 + X, and between H1 and C2, 60-X.

The features discussed above are described by a theorem from graph theory in

mathematics, known as Euler’s general network theorem. This theorem translates into the

terminology of HEN, states that

Umin = N + L – s

Where, Umin = minimum number of units (including heaters and coolers)

N = total number of streams (including utilities)

L = number of loops

s = number of separate components.

Normally we want to avoid extra units, and so design for L=0. Also, if there will be no

subset equality in the data set and then minimum number of unit targets is

Umin = N – 1

Since the pinch divides the problem into two thermodynamically independent regions, the

targeting formula must applied to each separately.

Shell Targeting

The shell and tube heat exchanger (SHE) is most common type of heat transfer

equipments used in heat exchanger networks (HENs) of chemical process industries.

Generally multipass SHE is employed in these industries because of its following

advantages: (1) the configuration gives a large surface area in a small volume, (2) good

65

mechanical layout: a good shape for pressure operation, (3) uses well established design

procedures and fabrication techniques, (4) can be constructed from a wide range of

materials and (5) easily cleaned.

Many HEN design methods described in literature make the simplifying assumption of

counter current exchanger. It has been seen that an optimal solution of the HEN problem

based on purely counter current heat exchanger only will remain optimal in practice if

each unit can be realized by one exchanger with single shell. However, it rarely occurs in

industry as multipass construction of SHE is used here. Therefore, it is practically

feasible to target number of shells than the units at the synthesis stage of HEN.

FT Correction Factor

In case of the simplest multipass SHE, the 1-2 type, the liquid in one tube pass flows in

counter flow while in the other pass flows in parallel relative to shell fluid. To account

counter and parallel flows in 1-2 SHE, a correction factor FT is introduced into the basic

heat exchanger design equation, shown through Eq. 1, to take into account the above

phenomena,

Q = UA (Tln) FT where 0< FT<1 (1)

Where,

Q = heat exchanger duty (kW)

U = overall heat transfer coefficient, (kW/m2 C)

A = Heat exchanger area (m2 )

Tln = log mean temperature difference (C)

66

The FT factor is represented as the ratio of actual mean temperature difference in a 1-2

SHE to counter flow Tln for the same terminal temperatures. FT is a function of

dimensionless ratios, R and P, where

Heat capacity ratio, R = CPH / CPC = ((TCo – TCi ) ((THi – THo ) (2a)

and thermal effectiveness, P = (THi - THO ) / (THi – TCi ) (2b)

where THi = Hot stream inlet temperature (oC)

THo = Hot stream outlet temperature (oC)

TCi = cold stream inlet temperature (oC)

TCo = cold stream outlet temperature (oC)

Based on the value of FT, feasible design of heat exchanger is screened amongst different

alternative designs. For this purpose a rule of thumb i.e. FT > 0.8 is used and each design

with unacceptably low FT value is discarded.

It is well known fact that for multipass exchangers heat recovery is limited by Tln

correction factor, FT. If FT<0.8 one should increase the number of shells till FT becomes

greater than 0.8. For a 1-2 SHE, FT falls sharply with increasing temperature cross. The

ability to accommodate a temperature cross increases rapidly as the number of shell

passes is increased. However, designers often encounter situations where the FT is too

low or the FT slope is too large. If this happens, the designers may be forced to consider

multiple shell arrangements of 1-2 type. Therefore, it is required to compute number of

shells for a HEN.

A method to account for design sensitivity, based on the fact that for any value of R there

is a maximum asymptotic value for P, say Pmax, which is given as FT tends to – , and is

evaluated by

67

)11 2/(R P 2

max R (3)

Practical designs will be limited to some fraction of Pmax that is:

P = XP Pmax 0 < XP < 1 (4)

Where XP is a constant defined by the designer. The value of XP = 0.9 is sufficient to

satisfy FT≥0.75, while also avoiding regions of steep slope and therefore assuring a more

reliable design.

Situations are often encountered where FT is too low (or within the present context the FT

slope too steep) for a single shell. If this happens the designer may be forced to consider

an arrangement of multiple shells in series. If multiple shells are required then the most

common practice is to adopt a trial and error approach in which the number of shells in

series is progressively increased until a satisfactory value of FT is obtained for each shell.

Using the constant XP approach any need for trial and error can be eliminated since an

explicit expression for the number of shells can be derived. This is done by using the

following equation for N number of 1-2 shells in series.

R ≠ 1

N

rP

RPYwhere

YR

YP )

1

1(

1

2

21

(5a)

R = 1

12121

21

PNP

NPP (5b)

P1-2 is the effectiveness of each single 1-2 shell (given by XP * Pmax) whereas P applies

overall to the series of shells. Equations (3) and (4) which together relate P1-2 to XP and

R, can then be used to eliminate P1-2 from equation (5) to give the following expressions:

68

R ≠ 1 N = 1n ((1- RP)/(1-P))/ln W (6a)

Where

)211/()211( 22

PP XRRRXRRW (6b)

R = 1 pP XX /))2/2( P))(1-(P/(1 N (6c)

In terms of R, P and P1-2, the number of shells can be computed using following

equations:

N = ln [(1-RP)/(1-P)]ln[(1-RP12)/(1-P12)] for R ≠ 1 (7a)

And

N = [P/(1-P)]/[P12/(1-P12)] for R = 1 (7b)

XP is chosen to satisfy the minimum allowable FT (for example, for FT ≥ 0.75, XP=0.90 is

used). The application of XP is valid under the same assumptions as those of FT. Eq. 6 or

7 then evaluates explicitly the number of shells required and, at the same time, ensures

that each shell in the design satisfies the required sensitivity criterion given by the

specification for XP .The number of shells predicted by Eq. 6 or 7 is a real (that is,

fractional or non-integer) number and the actual number of shells in practice would

obviously be taken to the next largest integer.

If each match enthalpy interval i requires Ni number of shells using temperatures of

interval i in equation (6) then the maximum shells count for the interval is:

Ni (Si – 1) (7)

Notice the temperatures defining Ni are achieved by a minimum of (Si – 1) matches

where S is the total number of streams present in ith

interval.

69

The real (non-integer) number of shells target is then simply the sum of the real number

of shells from all the enthalpy intervals:

M

i

iishell SNN1

)1( (8)

where M is the total number of enthalpy, intervals on the balanced composite curves.

Furthermore, actual designs will normally observe the pinch division. Hence, Nshell

should be evaluated and taken as the next largest integer for each side of the pinch. The

number of shells target is then:

])[(])[(][ belowpinchshellabovepinchshellshell NNN (9)

Where the symbol [N] represents the next largest integer to the real number N.

Example

The Stream Data, shown through Table 1, is considered for this purpose. Here Tmin =

20˚ C.

Table 1 Stream data for a typical process

Hot utility inlet and outlet temperature are 180 ° C and 179 ° C.

Cold utility inlet and outlet temperature are 15 ° C and 25 ° C.

Calculation of P and R for an interval

Stream

Type

Supply temp. TS (˚C)

Target temp. TT (˚C)

Heat capacity flow rate MCp

(kW/ ˚C)

H1

Hot

175

45

10

H2

Hot

125

65

40

C3

Cold

20

155

20

C4

Cold

40

112

15

70

The temperature effectiveness, P, is defined as the ratio of the temperature change in one

of the streams to the maximum possible temperature difference.

Pi = (Th,i. – Th,i.-1) / (Th,i. – Tc,i.-1)

For i= 1, P1 = (65˚ - 45˚) / (65˚ - 15˚) = 0.4

R is defined as the ratio of the heat capacity flow rates of the hot streams to the cold

streams.

Ri = (Tc,i. – Tc,i.-1) / (Th,i. – Th,i.-1)

For i=1, R1 = (18.81˚ - 15˚) / (65˚ - 45˚) = 0.1905

The complete calculation is shown in Table 2.

Calculation of the temperature effectiveness of an individual 1-2 exchanger

P12 = XP Pmax where )11 2/(R P 2

max R

For i.=1 and XP = 0.9,

P12,i.=1 = 0.9 * 2 / (0.1905+1+(0.19052+1)1/2) = 0.815

Calculation of number of 1-2 shells needed in series

N = ln [(1-RP)/(1-P)]ln[(1-RP12)/(1-P12)] for R ≠ 1

And

N = [P/(1-P)]/[P12/(1-P12)] for R = 1

For i = 1,

N = ln [(1-0.1905*0.4) / ln [(1-0.1905*0.815) / (1-0.815)]

= 0.2841.


71

Table 2 Determination of P and R for non countercurrent flow

Table 3 Determination of number of Shells for each enthalpy interval

Calculation of number of shells in an interval (Ni[Si – 1])

The minimum number of shells in an enthalpy interval, i, is Ni(Si – 1).

For i. = 3, Ni(Si – 1) = 0.0237*2 = 0.0474.


Int.

i.

Col. A

Th,i

Col. B

Tc,i

Col. A

P

Col. B

R

0

1

2

3

4

5

6

7

8

9

45

65

66.25

73.5

79.5

125

149.5

175

179

180

15

18.81

20

25

40

105

112

124.75

124.75

155

0.4000

0.0263

0.1355

0.1101

0.5353

0.5506

0.4048

0.0000

0.0181

0.1905

0.9524

0.6897

2.5000

1.4286

0.2857

0.5000

0.0000

30.250

Int.

i.

Col. A

Th,i

Col. B

Tc,i

Col. A

Pi

Col. B

Ri

Col. C

P12, i

Col. D

Ni

0

1

2

3

4

5

6

7

8

9

45

65

66.25

73.5

79.5

125

149.5

175

179

180

15

18.81

20

25

40

105

112

124.75

124.75

155

0.4000

0.0263

0.1355

0.1101

0.5353

0.5506

0.4048

0.0000

0.0181

0.1905

0.9524

0.6897

2.5000

1.4286

0.2857

0.5000

0.0000

30.250

0

0.8150

0.5400

0.6197

0.2907

0.4314

0.7740

0.6875

0.0000

0.0293

0.2841

0.0237

0.1160

0.2152

1.7304

0.5081

0.3944

0.0000

0.3630

72

Table 4 Number of Shell for present problem

Calculation of estimate of shells targets

The pinch occurs at 125 ˚ C/105 ˚ C. So,

Shells below pinch = 0.2841+0.0474+0.3481+0.4305+5.1912

= 6.3013, (rounded off to 7)

Shells above pinch = 1.0163+0.3944+0.3630

= 1.7737 (rounded off to 2).

Thus total number of shells required is 9.

Cost Targeting

The cost of the network basically comprises the operating cost and capital cost.

Operating cost

The operating cost is the function of energy requirements and is given by:

OC = Chu * Qhu,min + Ccu * Qcu, min (10)

Where Chu & Ccu are the costs of minimum loads of hot and cold utility respectively and

Qhu,min and Q cu,min are the minimum requirements of hot and cold utilities respectively.

0.0237

73

Capital cost

A simple linear cost law for individual heat exchanges is

CC = a + b Ak (11)

The capital cost of a network can then be predicted on the basis of targets for the number

of units for maximum energy recovery (Umin,MER ) and minimum network area (Amin ).

Thus:

.min.minmin,

minmin

bAaUAbaUCCCC MERk

U

k

MERk

U

k

network

MERMER

(12)

Most often, the cost law for individual exchangers takes the nonlinear form as:

CCk = a + b c

kA (13)

If nonlinear cost law is used in targeting, we assume the areas of individual units are all

identical:

MER

kU

AA

,min,

min (14)

This leads to the network capital cost given by:

c

MER

networkU

AbaUCC )(

min,

min

min (15)

Total annual cost (TAC)

TAC is given by:

TAC = OC + CCnetwork * Af (16)

Where Af = (1 + r)t /t

Where Af is the annualization factor, r, is the rate of return of capital interest and t is the

expected plant life.

74

Lecture – 11

PINCH DESIGN METHODS – HEURISTIC RULES

Bikash Mohanty

Professor



The pinch design method incorporates two fundamentally important features. First, it

realizes the pinch is the most temperature constrained region. The design is started at the

pinch and developed moving away. Second, it allows the designer to choose between

options.

Feasibility Criteria at the Pinch

The identification of essential matches at the pinch, of a available design options and of

the need to split streams, is achieved by applying three feasibility criteria to the stream

data at the pinch. In developing these feasibility criteria reference is made to "pinch

exchangers" (sometimes called "pinch matches").

A pinch match

75

The number of process streams and branches

The first feasibility criterion concerns the stream population at the pinch. The population

of hot and cold streams has to be such that it will allow an arrangement of exchangers

compatible with minimum utility usage.

Consider a hot end design as in Fig. 1(a). Utility cooling above the pinch would violate

the minimum utility objective. Therefore, each hot stream has to be cooled to the pinch

temperature by process exchange. This is attempted in Fig. 1(a) by placing pinch matches

between hot stream No. 2 and cold stream No. 4 and hot stream No. 3 and cold stream

No. 5. Notice, however that having made these matches hot stream No. 1 cannot be

matched with either cold stream without violating the minT constraint. Utility cooling

Exchanger 2 is not a

pinch match

Exchanger 3 is not a pinch match

76

would now be required above the pinch to cool stream No. 1 to the pinch temperature. In

such circumstances we say the original stream data at the pinch is not compatible with a

minimum utility design.

When this incompatibility occurs the streams at the pinch need "correcting" by stream

splitting (see Fig. 1(b)). By splitting a cold stream an extra cold "branch" is created,

allowing a pinch match with hot stream No. 1.

To summarize, the hot end stream population at the pinch is compatible with a minimum

utility design only if a pinch match can be found for each hot stream. For this to occur

inequality (1a) must apply

NH NC (1a)

Where NH is the number of hot streams or branches and NC is the number of cold

streams or branches. Stream splitting may be needed to ensure that the inequality is

fulfilled.

Fig. 1. (a) An infeasible hot end design at the pinch. (b) Stream splitting at the pinch

77

The converse arguments apply below the pinch. To avoid utility heating each cold stream

must be brought to the pinch temperature by process exchange. As a result, a pinch

match is required for each cold stream at the pinch and this is possible only if inequality

(1b) holds

NH NC (1b)

Once again stream splitting may be necessary to ensure that the inequality is fulfilled.

The CP inequality for individual matches

The second feasibility criterion is concerned with temperature feasibility. As shown in

Fig. 2, temperature driving force in a pinch match cannot decrease away from the pinch.

For this condition to be fulfilled the following CP inequalities must apply in every pinch

match

Hot end pinch match

CPH CPC (2a)

Cold end pinch match

CPH CPC (2b)

Where CPH is the heat capacity flowrate of a hot stream or stream branch and CPC is the

heat capacity flowrate of a cold stream or stream branch.

If an arrangement of matches fulfilling these inequalities is not possible then it is

necessary to change one or more CPs by stream splitting.

It should be noted that inequalities (2a) and (2b) only apply at the pinch. Away from the

pinch, temperature driving forces may have increased sufficiently to allow matches in

which the CP's of the streams matched violate the inequalities.

78

The CP difference

To understand the third feasibility criterion at the pinch it is convenient to define the "CP

difference". It can be understood by Fig. 3a, 3b and 3c.

For a hot end pinch match

CP difference = CPC-CPH (3a)

For a cold end pinch match

CP difference = CPH - CPC (3b)

Similar equations can be written for differences in the overall sum of hot stream CPs and

cold stream CPs at the pinch.

Immediately above the pinch

Overall CP difference = NHNC

CPHCPC11

(4a)

Fig. 2 (a) A feasible pinch exchanger above the pinch (b) A feasible pinch exchanger

below the pinch

79

Immediately below the pinch

Overall CP difference = NCNH

CPCCPH11

(4b)

Fig. 3a The sum of the match CP

differences equals the overall

difference. All stream at the

pinch are involved in pinch

exchangers.

Fig. 3b The sum of the match CP differences

amount to less than the total. In this

case not all streams at the pinch are

involved in pinch match.

Fig. 3c The sum of the match CP differences

exceeds the total. The pinch match

shown is feasible by itself as it fulfills

CP inequality criterion but it is

incompatible with overall CP

difference. (the pinch match has a CP

difference of 6 whereas the total

available is only 4.) Thus, it is not

possible to complete this design.

80

Lecture – 12 & 13

DESIGN OF HEN FOR MAXIMUM ENERGY RECOVERY, LOOP

BREAKING & PATH RELAXATION

Bikash Mohanty

Professor



Design of HEN for Maximum Energy Recovery

The pinch represents the most constrained region of a design; after all, Tmin exists

between all hot and cold streams at the pinch. As a result the number of feasible matches

in this region is severely restricted. Quite often there is a crucial or "essential" match. If

this match is not made, this will result in heat transfer across the pinch and thus in

increased hot and cold utility usage. The pinch design method, therefore

* recognizes the pinch division

* starts the design at the pinch developing it separately into two remaining problems.

This approach is completely different from the normal intuitive approach of starting the

design at the hot side and developing it towards the cold. When a design is started at the

hot side, initial design decisions may later necessitate follow-up decisions which violate

the pinch. On the other hand, when a design is started at the pinch, initial design

decisions are made in the most constrained part of the problem and are less likely to lead

to difficulties later.

Thus, commencing a design at the pinch has the distinct advantage of allowing the

designer to identify essential matches or topology options in the most constrained region

of the design, which are in keeping with minimum utility usage or maximum energy

81

recovery (MER).

Basic element for Design of HEN

The CP table

Fig. 1(a) and 1(b) show a step-by-step procedure for applying the feasibility criteria such

as: number of process streams & branches, CP inequality for individual matches and CP

difference. By following the sequence, the designer can

* identify essential matches at the pinch.

* identify available match options at the pinch.

* identify the need to split streams and generate stream splitting options at the pinch.

The procedure is aided by the use of another new concept, the "CP table". CP tables for

the hot and cold ends of a typical problem are shown in Figs. 2 and 3 respectively. In

these tables hot and cold stream CPs at the pinch are separately listed in numerical order.

The appropriate feasibility criteria are noted at the top of the table and the CPs

representing streams, which have to be involved in process exchange at the pinch, are

boxed for emphasis. A pinch match is represented in the table by pairing the CPs of a hot

and a cold stream. Stream splits are represented by writing the separate branch flowrate

CPs adjacent to the original CP (see Fig. 3(C)). The step by step procedure from Fig. 1 is

easily followed in the CP table.

82

Fig. 2. (a) The CP table for a typical problem hot end. (b) & (c) Feasible pinch matches

identified in the CP table

Fig. 1. (a) Hot end pinch design procedure. (b) Cold end pinch design procedure.

83

The "tick-off" heuristic

Once a pinch topology has been chosen, the design of both hot and cold ends must be

continued in such a manner as to keep capital costs at a minimum, i.e. the final designs

ought to be steered towards the minimum number of units. This can be achieved by

employing a "tick-off" heuristic to identify the heat loads on the pinch exchangers.

The targeting equation for the minimum number of units is satisfied if every match brings

one stream to its target temperature or exhausts a utility. In this case, the match is said to

"tick-off" the stream or utility, i.e. the stream or utility need no longer be considered part

of the remaining design task.

The pinch exchangers can usually be made to tick-off streams by choosing each

Fig. 3(a) The CP table for a

typical process cold end.

(b) Infeasible pinch

topologies.

(c) Feasible pinch topology

with two stream splits.

(d) Feasible pinch topology

with one stream split

84

exchanger load to equal the smaller heat load of the two streams matched. The CP

inequalities will guarantee the possibility of choosing pinch exchanger loads by ticking-

off streams as long as the stream CP remains constant with varying temperature and as

long as cold and hot stream temperature overlaps do not require an excessive number of

shells for a single pinch match.

The tick-off heuristic is a "heuristic" as it can occasionally penalize the design by

introducing the need for increased utility usage. Temperature driving force, essential

elsewhere, may be used up excessively in pinch exchangers that are extended too far into

the remaining problem. In such cases the designer can choose either to

* reduce the load on the offending pinch match and run the risk of needing more than

the minimum number of units.

* use another pinch topology in which the tick-off heuristic does not cause essential

driving force to be used up.

Design method summary

The pinch design method incorporates five important stages. These are:

1. The HEN problem is divided at the pinch into separate problems.

2. The design for these separate problems is started at the pinch and developed

moving away from the pinch. At the pinch essential matches, match options and stream

splitting requirements are identified by applying the feasibility criteria.

3. When options exist at the pinch, the engineer is free to base his selection to suit

the process requirements.

4. The heat loads of exchangers at the pinch are determined using the stream "tick-

off" heuristic. In case of difficulty (increased utility usage) a different exchanger

85

topology at the pinch can be chosen or the load on the offending match can be reduced.

5. Away from the pinch there is generally a "free choice" of matches. The procedure

does not insist on particular matches but allows the designer to discriminate between

matches based on his judgment and process knowledge.

Example

The stream data is shown below. For this problem Tmin = 10 °C and the hot and cold

utility requirements are 7.5 MW and 10 MW. Hot and cold pinch temperatures are 150

and 140 °C, respectively. Number of units required, including heaters and coolers, are 7

(4 above the pinch and 3 below the pinch). The grid representation of this data is shown

in Fig. 4.

Stream Type TS (°C) TT (°C) H (MW) CP (MW °C-1

)

1 Cold 20 180 32 0.2

2 Hot 250 40 - 31.5 0.15

3 Cold 140 230 27 0.3

4 Hot 200 80 - 30 0.25

Fig. 4 The grid diagram

86

Design above the pinch

Fig. 5a shows the grid diagram with CP-table for design above the pinch. Cold utility

must not be used above the pinch, which means that hot streams must be cooled to pinch

temperature by heat recovery. Hot utility can be used, if necessary, on the cold streams

above the pinch. Thus, it is essential to match hot streams above the pinch with a cold

partner. In addition, if the hot stream is at pinch conditions, the cold stream it is to be

matched with must also be at pinch conditions, otherwise the Tmin constraint will be

violated. Fig. 5a shows a feasible design arrangement above the pinch that does not use

temperature differences smaller than Tmin. Note again that the CP inequality only

applies when a match is made between two streams that are both at the pinch. Away from

the pinch, temperature differences increase, and it is no longer essential to obey the CP

inequalities.

Fig. 5 The CP table for the designs above and below the pinch

NHNC

NHNC

(a) (b)

87

Design below the pinch

Fig. 5b shows the grid diagram with CP-table for the design below the pinch. Hot utility

must not be used below the pinch, which means that cold streams must be heated to pinch

temperature by heat recovery. Cold utility can be used, if necessary, on the hot streams

below the pinch. Thus, it is essential to match cold streams below the pinch with a hot

partner. In addition, if the cold stream is at pinch conditions, the hot stream it is to be

matched with must also be at pinch conditions, otherwise the Tmin constraint will be

violated. Fig. 5b shows a design arrangement below the pinch that does not use

temperature differences smaller than Tmin.

Sizing the units above the pinch using the tick-off heuristic

Once the matches around the pinch have been chosen to satisfy the criteria for minimum

energy, the design should be continued in such a manner as to keep capital costs to a

minimum. One important criterion in the capital cost is the number of units (there are

others, of course, which shall be addressed later). Keeping the number of units to a

minimum can be achieved using the tick-off heuristic. To tick off a stream, individual

units are made as large as possible, that is, the smaller of the two heat duties on the

streams being matched.

Fig. 6a shows the matches around the pinch from Fig. 5a with their duties maximized to

tick off streams. It should be emphasized that the tick-off heuristic is only a heuristic and

can occasionally penalize the design. Methods will be developed later, which allow such

penalties to be identified as the design proceeds.

The design in Fig. 6a can now be completed by satisfying the heating and cooling duties

away from the pinch. Cooling water must not be used above the pinch. Therefore, if there

88

are hot streams above the pinch for which the pinch matches do not satisfy the duties,

additional process-to-process heat recovery is required. Fig. 6b shows an additional

match to satisfy the residual cooling of the hot streams above the pinch. Again, the duty

on the unit is maximized. Finally, above the pinch, the residual heating duty on the cold

streams must be satisfied. Since there are no hot streams left above the pinch, hot utility

must be used as shown in Fig. 6c.

Similarly sizing of units below the pinch can be done as shown in Fig. 7.

Fig. 6 Sizing of units above the pinch

Fig. 7 Sizing of units below the pinch

The complete HEN design for MER

The final design shown in Fig. 8 amalgamates the hot end design from Fig. 6c and cold

end design from Fig. 7c. The duty on hot utility of 7.5 MW agrees with QHmin and the

duty on cold utility of 10.0 MW agrees with QCmin predicted by the composite curves and

the problem table algorithm.

89

Note one further point from Fig. 8 that the number of units is 7 in total (including the

heater and cooler) which is equal to the targeted value. It therefore appears that there was

something in the procedure that naturally steered the design to achieve the target for the

minimum number of units.

Fig. 8 The completed design for the stream data undertaken

Design of HEN with Stream Splitting

The pinch design method developed earlier followed several rules and guidelines to allow

design for minimum utility (or maximum energy recovery) in the minimum number of

units. Occasionally, it appears not to be possible to create the appropriate matches

because one or other of the design criteria cannot be satisfied. In such cases stream

splitting is done. The algorithm of stream splitting is shown in Fig. 9.

90

Fig. 9 Stream splitting algorithm

Example

The grid representation for a high temperature process is shown in Fig. 10a where

Tmin=20 C. The process requires 9.2 MW of hot utility, 6.4 MW of cold utility and the

pinch is located at 520 C for hot streams and 500 C for cold streams.

Stream data

at pinch

NH NC ? CpH CpC for

pinch match

Split cold

stream

Split hot

stream

Place

match

Yes Yes

No

No

Stream data at

pinch

NH NC ? CpH CpC for

pinch match

Split hot

stream

Split cold

stream

Place

match

Yes Yes

No

No

(a) Stream splitting above pinch

(b) Stream splitting below pinch

91

Fig. 10b shows the CP tables for the above- and below-pinch designs. Following the

algorithms in Fig. 9, a hot stream must be split above the pinch to satisfy the CP

inequality, as shown in Fig. 10b.

The grid diagram

The splitting of hot stream

92

Fig. 10 Maximum energy recovery design with stream splitting

Identification of Loops & Paths and Loop Breaking and Path Relaxation

There will generally be scope to simplify minimum utility designs by a controlled

reduction in the number of units. By transferring heat across the pinch and therefore

increasing the utility usage the number of capital items can be reduced. There is a trade-

off between units (capital cost) and the utility usage (energy cost).

In order to explore the scope for a controlled reduction in the number of units it is

important to understand the concepts of heat load loops and heat load paths.

Heat Load Loops

A loop is a set of connections that can be traced through a network (via streams and

units) that starts at one exchanger and returns to the same exchanger.

Whenever a design features more than the target minimum number of units for the whole

problem, ignoring the pinch, it is due to the existence of heat load loops. There will be

one loop for each extra unit. As an example, the minimum utility design for a typical

93

problem has two more units than the definite minimum according to Fig. 11. Hence there

must be two loops in the design. Figs. 12(a) and (b) show these loops.

An important feature of every loop is that heat loads can be shifted around the loop from

one unit to another. The load is subtracted from the next and so on around the loop. This

load shift always maintains the correct stream heat loads but the exchanger duties are

changed and may cause a violation of Tmin. However, driving forces can be "restored"

using heat load paths.

(b)

Fig 11. (a) The number of units for

maximum energy recovery.

(b) The overall minimum number of

units.

94

Heat Load Paths

A path is a continuous connection in the grid between a heater, heat exchangers and a

cooler. Fig. 14 shows the simplest form of a path.

Load shifts along paths follow equivalent rules to load shifts around a loop. Load is

added to a heater, subtracted from an exchanger, added to the next exchanger in the path,

subtracted from the next, and so on along the path until it is finally added to a cooler.

Stream enthalpy balance is maintained but exchanger loads and operating temperatures

are changed. This last feature means that a path can be used to restore driving forces.

Loop Breaking and Path Relaxation

We will now illustrate the use of heat load loops and paths to reduce the number of units

of the design in Fig. 12a from seven to six.

Fig. 12 (a,b) A minimum utility design for

a typical problem showing the

two heat load loops

95

It is apparent that load shifts around loops can form the basic mechanism for the

reduction in the number of units. When the load shift around a loop leads to a reduction

in the heat load of a unit, which equals the load on that unit, then the unit is removed

from the design and the number of units is reduced by one.

Consider Fig. 12(a), which shows a minimum utility design with seven units. A good

choice of exchanger to remove is exchanger No. 4 as it has the smallest load and forms

part of the simplest loop. Fig. 13 shows the topology and temperatures after the load of

match No. 4 has been transferred to exchanger No. 1. The heat loads of all other units in

the design are unchanged as they were not part of the original loop. There is now a small

violation in Tmin as reflected by the difference in temperatures T1 and T2.

However, Tmin can be restored using the heat load path shown in Fig. 14. It is apparent

that T1 is fixed at 62C. It is therefore T2 which must be changed to restore Tmin.

Requiring T2 to equal 82C, the heat load of individual units can now be changed while

the stream heat loads are maintained by using the path through exchanger No. 1. It is a

trivial task to calculate the hot and cold utility increase x required. This load is 4 kW. In

other words, by supplying a further 4 kW of utility heating and cooling and by reducing

the heat load on exchanger No. 1 by 4 kW, the solution is brought back in line with the

Tmin.

Fig. 13 A six unit topology that is a result of

breaking Loop 1 (Fig. 12a).

96

The summary of this section is as follows

There is generally scope to reduce the number of units in a pinched problem

starting from a minimum utility design.

This reduction in the number of units can be achieved in a controlled manner. By

this we mean that the utility penalty incurred in reducing the number of units is

minimized.

Not all units exist in a suitable loop or along a suitable path. Thus, the procedure

would not be applicable to the "removal" of such units.

Fig. 14 Identifying a path

97

Lecturer - 14

DRIVING FORCE PLOT AND REMAINING PROBLEM ANALYSIS

Bikash Mohanty

Professor



The Driving Force Plot

Fig. 1 shows two networks having the same CP-ratios for the pinch match. However,

network 2A comes to within 16% of the above pinch area target, while network 2B

requires 108% more area than target. Why is there such a large discrepancy? Examining

the Composite Curves, we suspect network 2B makes poor use of driving forces away

from the pinch. To take this further the concept of “Driving Force Plot” is used.

Fig. 1 Both networks have

identical CP-ratios for

pinch matches. There is,

however, significant

difference in network

areas

98

The area target is based on the vertical temperature differences along the whole balanced

composite curves. Ideally, we need to measure the temperature differences of individual

matches against the vertical driving forces available on the composites. A simple way of

expressing this is firstly to draw the vertical temperature difference T between the

composites as it changes with the temperature of say the cold composite Tcold (Fig. 2).

Equivalently, T=f(Thot) or Thot=f(Tcold) may also be used. The diagram is called the

"Driving force Plot" (Fig. 2).

Fig. 2 Construction of Driving Force Plot

Next, individual matches are shown in these coordinates (Fig. 3). Matches displaying

vertical heat transfer on the composites fit the Driving force Plot exactly, such as the

match shown in Fig. 3. Matches which are not vertical (or which criss-cross) on the

composites show a blatant misfit (Fig. 4, 5). Matches using excessive temperature

differences have less area than if they had been vertical, but cause other (subsequently

placed) matches to have smaller temperature differences. The net result overall is

increased heat exchange area for the network.

99

Fig. 3 The match with “vertical heat transfer”

Fig. 4 The match with “excessive driving force”

Fig. 5 The match under-utilizing driving force

The Driving Force Plot provides a rapid and easy way to use guideline for designing

networks, which are close to minimum area. However, it is only a guideline and does not

provide quantitative information.

100

Networks 2A and 2B are displayed against their Driving Force Plot in Fig. 6. The pinch

matches placed according to the CP-rules follow well the slope of the Driving Force Plot

near the pinch. Away from the pinch, however, network 2B shows a poorer overall fit to

the plot. Its pinch matches are too large to duty and under-utilize driving forces away

from the pinch. These duties were established using the “tick-off’ heuristic for obtaining

minimum number of units in the design.

Fig. 6 Networks 2A & 2B compared on the Driving Force Plot. Network 2B shows a

much poorer overall fit to the plot than network 2A.

The plot shows the tick-off heuristic is inappropriate here for achieving low network area.

Violation of the tick-off rule usually means additional units above target, as in network

2A. The significantly improved area performance in this example gives lower overall

capital cost. Designs achieving a good fit to the Driving Force Plot in minimum number

101

of units or within 10% of this (to the nearest integer number of units) are usually within

10% of the area target.

Remaining Problem Analysis

Suppose a design obtains a good fit to the Driving Force Plot but the final network area is

appreciably above target. Such an occurrence is infrequent considering the plot steers

design towards vertical heat transfer and minimum area. Fig. 7, however, demonstrates

the plot may not always be sufficient for minimum area. Networks 3A and 3B appear

remarkably similar in use of driving forces, but 3B has an area 22% in excess of the

above-pinch target whereas 3A is only 10% above this target.

Fig. 7 Both networks show very similar fit to the Driving Force Plot but differ appreciably

in area

102

The Driving Force Plot works in temperatures only, neglecting the effect of duty on heat

exchanger area. It is possible for matches to appear identical in Driving Force

coordinates, yet have very different duties. Generally, good utilization of driving forces

for matches of large duty is required in regions of small temperature difference.

When a match is placed, the duty needs to be chosen with some quantitative assessment

of the match in the context of the whole network, without having to complete the

network. This can be done by exploiting the powers of targeting using a technique known

as Remaining Problem Analysis.

Consider the design for minimum energy in a more complex problem. If a problem table

analysis (PTA) is performed on the stream data, QHmin and QCmin can be calculated. When

the network is designed and a match is placed, it would be useful to assess whether there

will be any energy penalty caused by some feature of the match without having to

complete the design. This penalty can be determined by performing a PTA on the

remaining problem.

The PTA is simply repeated on the stream data, leaving out those parts of the hot and

cold stream satisfied by the match. One of the two results would then occur:

1. The algorithm may calculate QHmin and QCmin to be unchanged. In this case, the

designer knows that the match will not penalize the design in terms of increased

utility usage.

2. The algorithm may calculate an increase in QHmin and QCmin. This means that the

match is transferring heat across the pinch or that there is some feature of the

design that will cause cross-pinch heat transfer if the design was completed. If the

103

match is not transferring heat across the pinch directly, then the increase in utility

will result from the match being too big as a result of the tick-off heuristic.

The remaining problem analysis (RPA) technique can be applied to any feature of the

network that can be targeted, such as a minimum area.

RPA can be used to approach the area target, as closely as a practical design permits,

using a minimum (or near minimum) number of units. Suppose a match is placed, then its

area requirement can be calculated. A RPA can be carried out by calculating the area

satisfied by the match. The area of the match is now added to the area target for the

remaining problem. Subtraction of original area target for the whole-stream data gives the

area penalty incurred.

Targets for number of shells, capital cost and total cost also can be set. Thus, RPA can be

used on these design parameters also.

The “Remaining Problem Analysis” is explained in Fig. 8. Suppose the minimum total

area possible for a design completed after accepting a match M is Atotal. M. This is the sum

of the match area aM and the area target for the remaining stream data Ar, M. Subtraction

of the original area target for the whole stream data Amin gives the minimum area penalty

incurred.

The analysis can quantify both surplus and deficit use of driving forces. A large T

match incurs area penalty from the small T caused in the remaining problem. A small

T match incurs area penalty from the match itself.

104

Fig. 8 Remaining Problem Analysis for area

Fig. 9 the Remaining Problem Analysis for match 4 in network 38 shows significant

penalty in area for the network.

105

The Remaining Problem Analyses for networks 3A and 3B are shown in Fig. 9. It is now

clear that match 4 in network 3B is not as good as the rest. Surprisingly, it looks similar

on the Driving Force Plot (Fig. 7) to matches 4 and 5 in network 3A, which return much

lower area penalties. The Remaining Problem Analysis improves on the Driving Force

Plot. At present, it is the only known method for quantifying approach to the targets

during design development.

The Remaining Problem Analysis discussed so far treats each match in isolation of the

others when several matches exist at any stage of design (as in Fig. 9). In other words, the

remaining problem is defined as the full stream data excluding only the hot and cold

stream sections of the match being analyzed.

106

REFERENCES

1. Linnhoff, B.; Dunford; H.; and Smit, R.; “Heat Integration of Distillation Columns

into Overall Process”; Chem. Engg. Science; Vol.38; No.8; pp-1175-1189, (1983).

2. Robin Smith, “Chemical Process Design”, McGraw Hill, 1995.

3. Linnhoff B, Townsend D W, Boland D, Hewitt G F, Thomas B E A, Guy A R &

Marsland R H, “User guide on process integration for the efficient use of energy”,

(The Institution of Chemical Engineers, Rugby, U.K.; available in the U.S. through

Pergamon Press, Inc. Elmsord, N.Y.), 1982.

4. Ahmad S and R .Smith, “Targets and design for minimum number of shell in heat

exchangers networks”, Chem. Eng. Res. Des., Vol. 67, Sep 1989, pp 481-494.

5. Linnhoff B. and S. Ahmad, “Cost optimal heat exchanger networks I Minimum

energy and capital cost using simple models for capital cost , Comp Chem Engg.,

Vol. 14, No. 7 1990, pp 729-767.

6. Linnhoff B. and E. Hindmarsh, “The pinch design methods for heat exchanger

networks”, Chem Engg. Sci, Vol. 38, No. 5 1983, pp 745-763.

7. Linnhoff B., “Pinch Analysis-A state-of-the- art overview”, Trans IchemE, Vol. 71,

Part A, Sep 1993, pp 503-522.

8. Linnhoff B. and J. R. Flower, “Synthesis of heat exchanger networks”, AIChE J,

Vol. 24, July 1978, pp 633-642.

9. Smith G. and A. Patel, “Step by step through pinch”, The Chem. Eng., Nov 1987,

pp 26-31.

10. Linnhoff B. and D. R. Vredeveld, “Pinch Technology has come of age”, CEP, July

1984, pp 33-39.

11. Uday V. Shenoy, “Heat Exchange Network Synthesis”, Gulf Publishing Company,

1995.

pinch analysis a tool for efficient use of energy

Education

process engineering

integrated process design

s pinch technology

overall process design

week training course

week duration course

stage of design

conservation of energy