congestion management and optimal placement of tcsc...

154

CHAPTER 5

CONGESTION MANAGEMENT AND OPTIMAL

PLACEMENT OF TCSC FOR TRANSFER

CAPABILITY ENHANCEMENT

5.1 INTRODUCTION

In the previous chapter the importance and computation of available

transfer capability are discussed. This chapter gives an introduction to

deregulated energy market, congestion and methods of congestion

management. A method to determine the optimal location of TCSC is also

proposed to enhance the transfer capability which is needed to meet the

rapidly changing demand of competitive markets.

Power system load growth is increasing at a faster rate as

compared to the increase in transmission capability. In the last decade

the increase in transmission capacity is approximately 50% of the

increased generation capacity. This has forced the system to move large

amount of power over transmission system and created challenges

associated with it. In the present open access system where anybody can

buy or sell energy there is a heavy transmission utilization in certain

areas which was not planned initially. This has increased the need of

improvements in the transfer capability of the system while maintaining

the system security and reliability.

155

The other reasons for short fall of transmission capability are

Difficulty in getting right of way permissions due to property

devaluation.

Health hazards due to electromagnetic effects.

Large impact on land use.

Ecological system effects.

Capital cost involved in construction and maintenance of new

lines and lack of investors for the proposed projects.

All the above mentioned factors and rapid growth of the load lead to

congestion of the system.

The challenge of transferring power over the existing grids has

created an interest among the researches in recent years for developing a

more robust power system applying new technologies. The concept and

application of flexible AC transmission systems (FACTS) to power system

was first initiated by Hingorani [127]. FACTS devices have the ability to

allow power systems to operate in a more economic, secure, flexible, and

sophisticated way. This chapter presents a brief discussion on different

methods and proposes an innovative technique for location of FACTs

device using complex valued neural network. This technique aims

towards effect of incorporating FACTS device TCSC on the real and

reactive power flows of that particular line and also on the transfer

capability of the system.

156

5.2 METHODS OF INCREASING TRANSFER CAPABILITY

The system transmission capability can be improved by using

primary methods involving system upgrading. Some traditional methods

are

Reconductoring transmission lines with larger size conductor of

higher power carrying capability and replacing terminal

equipment.

Voltage upgrade i.e. by increasing operating voltage of a

transmission line. This method requires up gradation of towers,

substations, circuit breakers, transformers and other equipment.

Installation of new transmission lines to alleviate overloads by

providing additional path.

Conversion from single line to double circuit by modifying the

tower structure.

By series compensation using a series capacitor in long distance

transmission lines.

Installing phase angle regulators.

Using small inertia generators and dispersed generation.

Insertion of switching stations along the transmission line.

Installing FACTS devices.

157

5.3 FLEXIBLE AC TRANSMISSION SYSTEMS (FACTS)

With the advent of flexible ac transmission system (FACTS) devices

power utilities all over the world are able to improve the system stability

limit, control the power flow, improve the transmission system security

and provide strategic benefits for better utilization of the existing power

system. The operation of FACTS devices is based on power electronic

controllers. These devices are also used to enhance transfer capability

and to minimize the total power loss of a system thereby improving the

system efficiency. In a competitive electric power system the most

important aspect is better utilization of existing lines in the context of

growing demand and outgrowth of energy trading markets. In the

context of restructuring the existing power systems FACTS devices have

assumed an importance since they can expand the usage potential of

transmission systems by controlling power flows in the network. FACTS

devices are operated in a manner so as to ensure that the contractual

requirements are fulfilled as far as possible by minimizing line

congestion. The main device considered here is Thyristor controlled

series capacitor (TCSC) for enhancement of transfer capability. Series

compensation is usually a preferable altenrative for increasing power flow

capability of lines compared to shunt compensators as the ratings

required for series compensators are significantly smaller.

158

5.3.1 TYPES OF FACTS CONTROLLERS

FACTS controllers are classified as series controllers, shunt

controllers, combined series–series controllers and combined series-

shunt controllers.

5.3.2 SERIES CONTROLLERS

These devices are connected in series with the lines to control the

reactive and capacitive impedance there by controlling or damping

various oscillations in a power system. The effect of these controllers is

equivalent to injecting voltage phasor in series with the line to produce or

absorb reactive power. Examples are Static Synchronous Series

Compensator (SSSC), Thyristor controlled Series Capacitor (TCSC),

Thyristor-Controlled Series Reactor (TCSR). . They can be effectively used

to control current and power flow in the system and to damp system’s

oscillations.

5.3.3 SHUNT CONTROLLERS

Shunt controllers inject current in to the system at the point of

connection. The reactive power injected can be varied by varying the

phase of the current. The examples are Static Synchronous Generator

(SSG), Static VAR Compensator (SVC).

5.3.4 COMBINED SERIES-SERIES CONTROLLERS

This controller may have two configurations consisting of series

controllers in a coordinated manner in a transmission system with multi

lines or an independent reactive power controller for each line of a multi

159

line system. An example of this type of controller is the Interline Power

Flow Controller (IPFC), which helps in balancing both the real and

reactive power flows on the lines.

5.3.5 COMBINED SERIES-SHUNT CONTROLLERS

In this type of controller there are two unified controllers a shunt

controller to inject current in to the system and a series controller to

inject series voltage. Examples of such controllers are UPFC and

Thyristor- Controlled Phase-Shifting Transformer (TCPST).

5.4 MODELING OF FLEXIBLE AC TRANSMISSION SYSTEM (FACTS)

DEVICES

For the enhancement of available transfer capability using FACTS

controllers it is assumed that the time constant of these devices is

negligible hence only static models are considered. The static models of

some of FACTS devices are explained below. Thyristor Controlled Series

Compensator (TCSC) is one such device which offers smooth and flexible

control for security enhancement with much faster response compared to

the traditional control devices [96,104,118]. Among the various FACTS

controllers, TCSC is considered for the proposed method in this chapter.

The detailed model of TCSC and brief discussion is given below.

5.4.1 THYRISTOR CONTROLLED SERIES COMPENSATOR (TCSC)

Thyristor controlled series compensator (TCSC) are connected in

series with transmission lines. It is equivalent to a controllable reactance

inserted in series with a line to compensate the effect of the line

160

inductance. The net transfer reactance is reduced and leads to an

increase in power transfer capability. The voltage profile as also improved

due to the insertion of series capacitance in the line. Series

compensation is usually a preferable alternative for increasing power flow

capability of lines as compared to shunt compensators as the ratings

required for series compensators are significantly smaller.

The transmission line model with a TCSC connected between the

two buses i and j is shown in Figure 5.1. Equivalent pi model is used to

represent the transmission line. TCSC can be considered as a static

reactance of magnitude equivalent to -jXc. The controllable reactance Xc

is directly used as control variable to be implemented in power flow

equation.

Fig. 5.1 Model of transmission line

161

Fig. 5.2 Model of TCSC

Fig. 5.3 Injection model of TCSC

The following equations are used to model TCSC.

Let the voltages at bus i and bus j are represented by iiV and

jjV

The complex power from bus i to j is

iijiijijij IVQPS ** (5.1)

)()(*

ciijjii jBVYVVV (5.2)

)()]([ **

ijijjicijiji jBGVVBBjGV (5.3)

Where

)(1

CLLijij jXjXR

jBG

(5.4)

162

From the above equations the real and reactive power equations can be

written as

)sin()cos(2

jiijjijiijjiijiij BVVGVVGVP (5.5)

)cos()sin()(2

jiijjijiijjicijiij BVVGVVBBVQ (5.6)

Similarly the real and reactive powers from bus j to i can also be

represented replacing Vi by Vj.

The real and reactive power loss in a line are represented by

equations (5.7) and (5.8)

PL = Pij + Pji (5.7)

QL = Qij + Qji (5.8)

The world’s first three phase TCSC was developed by ABB and

installed at Kayenta substation, Arizona in 1992, that raises the capacity

of a transmission line by almost 30%. By the end of 2004, seven TCSCs

have been installed worldwide. Table 5.1[115] shows the details of the

TCSC installations worldwide. The cost comparison of various FACTS

controllers is given in Table 5.2[115].

163

Table 5.1 List of TCSC Installations

S.No. Year

Installed Country

Voltage

Level(kV) Purpose Place

1 1992 USA 230 To increase power

transfer capability

Kayenta Substation,

Arizona

2 1993 USA 500 Controlling line flow

and increased loading

C.J.Slatt substation,

Northern Oregon

3 1998 Sweden 400 Sub-synchronous

resonance mitigation Stode

4 1999 Brazil 500 To damp inter-area low

frequency oscillation

Imperatrz and Sarra de

Mesa

5 2002 China 500

Stability improvement,

low-frequency

oscillation mitigation

Pinguo substation,

Guangzhou

6 2004 India 400

Compensation,

Damping inter regional

power oscillation

Raipur substation

7 2004 China 220

Increase stability

margin, suppress low

frequency oscillation

North-West China power

system

Table 5.2 Cost of conventional and FACTS Controllers

S. No. FACTS Controllers

Cost (US $)

1 Shunt Capacitor 8/kVAr

2 Series Capacitor 20/kVAr

3 SVC 40/kVAr Controlled portions

4 TCSC 40/kVAr Controlled portions

5 STATCOM 50/kVAr

6 UPFC Series Portions 50/kVAr through power

7 UPFC Shunt Portions 50/kVAr controlled

164

5.5 APPLICATIONS OF FACTS DEVICES

Facts controllers or devices can be used in various power

applications for its performance enhancement. When compared to

conventional devices these devices can be used an all three states namely

steady state, transient state and post transient state of the power

system.

5.5.1 STEADY STATE APPLICATIONS

These applications include steady state voltage control, increase of

thermal loading, post contingency voltage control, loop flow and power

flow control. SVC and STATCOM are preferred for voltage control where

as TCSC is used for loop control and power flow control. The other steady

state applications are

Congestion management: Congestion can increase the price and

may become an obstruction for the free electricity trade in the

present deregulated environment. FACTS devices like TCSC,

TCPAR and UPFC can help to reduce congestion and smoothen

location marginal price (LMP) by redirecting the power from

congested path to other path which is underutilized.

ATC improvement: ATC is the basis for a power transaction

between the buyer and seller in a deregulated market. A low value

of ATC implies the inability of the path for further transaction and

may hinder the free competition. TCSC, TCPAR and UPFC can help

in ATC enhancement y allowing more power transactions.

165

Reactive power and Voltage control: SVC, STATCOM can be use

for this purpose.

Loading Margin Improvement: Voltage collapse occurring at the

maximum loadability (nose point) is the main cause of recent world

wide block outs. The maximum transfer capability of a power

system can be improved by using shunt compensators efficiently.

Power flow and balancing control: TCSC, SSSC, UPFC can be

used to enable the load flow through parallel lines and there by

efficient utilization of lines can be made possible.

5.5.2 DYNAMIC APPLICATIONS

In these applications FACTS devices are used to enhance transient

stability by providing fast and rapid response, oscillation damping

dynamic control of voltage during contingencies to alleviate the system

from voltage collapse and sub synchronous resonance mitigation. These

devices are also used for inter connecting power systems for exchanging

the power between the regions over a long distance.

Congestion management

FACTS devices are used for relieving the system from congestion.

They can be used in a line such that least cost generators can be

dispatched more thereby reducing the price.

5.6 ELECTRICITY MARKET DEREGULATION AND CONGESTION

Deregulation and privatization of energy market has a wide range

of impact on the present day power systems around the world. The main

166

objective of the deregulation of power industry is to introduce

competition among the power producers and prevent monopolies.

Deregulation has increased complexity of the system as any body can

participate in the transactions to sell or buy electricity. As market

participants can produce and consume energy in amounts, transmission

lines are operated beyond their capacities causing congestion. It may

occur due to lack of coordination between generation and transmission

utilities. A congestion force higher cost generation in a network and has

direct impact on the economics therefore congestion management has

become very essential in a deregulated power system.

5.6.1 METHODS OF CONGESTION MANAGEMENT

The recent restructuring of energy system with existing generation

and transmission resources requires new methods for congestion

management and it also provides an opportunity for these approaches.

Allocation of transmission resources to support the competitive electricity

market is the current area of interest. Various new approaches are

proposed and being implemented which will provide feedback about their

suitability and contribution in improving the efficiency and reliability.

The two methods used for congestion management are

1. Cost free methods: Outaging of congested lines, adjusting

transformer taps, phase shifters or FACTS devices. The marginal

costs involved in their usage are nominal.

167

2. Non-cost free methods: Re-dispatch of generation, load or

transaction curtailment.

5.6.2 NETWORK CONGESTION MANAGEMENT USING TCSC

APPROACH

In recent years, deregulation of electric industry in the world has

created competitive markets to trade electricity. For deregulated

transmission network, one of the major consequences of the

nondiscriminatory open access requirement is substantial increase of

power transfers. Congestion management of deregulated transmission

network is important to accomplish non discriminative network access.

In this example congestion management approach using TCSC is

demonstrated considering a 5 bus system (Annexure – 1). This approach

aims at maximization of transmission margin without changing

contracted power. In this approach TCSC modeled as a variable

reactance is used. In order to check the validity of the proposed method

of congestion management, numerical results for a 5-bus system [5] are

shown in Figures 5.4, 5.5 and 5.6.

168

Fig. 5.4 Transmission margins without TCSC

Fig. 5.4 shows the transmission line flows without TCSC. It is

observed that Lines 1-2 and 2-5 are over loaded compared to other lines.

The efficient utilization of existing transmission lines installing TCSC in

line 1-3 is shown in Fig 5.5. The power flows in both real and reactive

with and without TCSC are shown in Fig 5.6. But it is important to

ascertain the location of such devices because they are expensive. In the

following section the optimal location of FACTS devices has been

discussed.

169

Fig. 5.5 Transmission margins with TCSC

Fig.5.6 Comparison of power flows with and without TCSC

170

5.6.3 ENERGY PRICING

The three different methods of energy pricing are

1. Uniform marginal pricing (UMP)

2. Zonal marginal pricing (ZMP)

3. Location marginal pricing (LMP)

In UMP method one price is set for whole region ignoring the line

limits, losses and the physics of the power flows. In ZMP a uniform price

is set for zone or a graphical region considering the transmission limits

on the paths between the zones.

5.6.4 LOCATION MARGINAL PRICING (LMP)

Location marginal pricing (LMP) is defined as the marginal cost of

supplying the next increment of electric demand at a specific location on

the electric power network taking in to account both generation marginal

cost and the physical aspects of the transmission system it calculates an

optimal dispatch considering all the above aspects. In this approach a

price is set for each node in the transmission system. This method is

considered to be the most efficient and it is also known as nodal price.

Optimal power flow solution yields the LMPs at various nodes and the

solution can be used to for different inputs.

In a highly congested system negative LMPs occur at one or more

buses which is demonstrated by taking a 5 bus system. Usually this

occurs when the LMPs at other congested buses are very high. In Fig. 5.7

171

[120] the various line flows, area costs and location marginal prices are

shown without system congestion.

Fig. 5.7 Five bus system with no constraints

Fig. 5.8 Congested system with high LMP

172

Fig. 5.8 shows a congested system and the nodal prices are high at bus

3, bus 4 and bus 5 which is due to increase in load at bus 5. High value

of LMP at some buses is mainly due to the congestion.

5.6.4.1 NEGATIVE LMP

Negative LMPs may also occur in a highly congested system. A case

of negative LMP is considered here taking the same 5 bus example. Fig.

5.9 demonstrates the negative LMPs with congestion in line connected

between bus 1 and bus 2. The value of LMP at bus 5 is negative with a

value of -22.45$/MWh this may be due to large value of LMP

621.45$/MWh at bus and 700.98$/MWh at bus 4. Serving an additional

load of 1 MW at negative LMP will reduce the operating costs. Congestion

can be mitigated by increasing the flows to the loads at theses buses to

allow counter flows. It can be shown that this method reduces the

operating costs.

Fig. 5.9 Congested system with negative LMP

173

In Fig.5.9 it is shown that by increasing the load by 1MW at bus 5

with LMP has changed from -22.45$/MWh to 10.62$/MWh.

Fig. 5.10 LMPs with increased load at bus 5

It is observed that by reducing load by 1 MW at bus 3 also has the

same effect on LMP thereby allowing dispatch at cheaper generation as

shown in Fig. 5.10. These methods create counter flows to alleviate

congestion. The effect of reduction in load on area cost is shown in Fig.

5.11.

174

Fig. 5.11 LMPs with curtailed load at bus 3

5.6.4.2 TRANSMISSION LINE RELIEF (TLR) SENSITIVITIES

Transmission Line Relief sensitivities can be used for the purpose

of congestion alleviation by load curtailment. TLR sensitivities are

considered as inverse of power distribution factors (PTDFs). PTDFs are

used to determine the sensitivity of transmission line flow to a single

power transfer where as TLR sensitivity of the flow on single

transmission element to various transactions in the system. In the

method of congestion alleviation using load curtailment, TLR sensitivities

at all the load buses for the most overloaded line are considered

175

5.6.4.3 LOAD CURTAILMENT BASED ON TLR SENSITIVITIES

TLR sensitivity at a bus k for a congested line i - j is given by

equation

k

ijk

ijP

PS

(5.9)

Where

∆Pij is the excess power flow on line i - j

ijijij PPP (5.10)

Where

ijP : Actual power flow through line i - j

ijP : Flow limit of transmission line i - j

The new load new

kP at bus k can be obtained by

ijN

i

k

ij

k

ij

k

new

k P

S

SPP

1

(5.11)

Where

new

kP = Load after curtailment at bus k

Pk = Load before curtailment at bus k

k

ijS = sensitivity of power flow on line i - j due to load

change at bus k

N = Total number of load buses.

176

Table 5.3: TLR sensitivities

Bus

Congested Line i - j

1 - 2 2 - 5 3 - 4

1 0 0 0

2 -0.699 0.012 -0.104

3 -0.601 -0.023 0.208

4 -0.653 -0.064 -0.428

5 -0.676 -0.526 -0.266

Table 5.3 represents the TLR values of congested lines of a 5 bus

system. The higher the TLR sensitivity value more the effect of a single

MW power transfer at any bus. Based on these sensitivities load are

curtailed in required amounts at load buses in order to alleviate

congestion on the congested line i - j.

5.7 LOCATION OF FACTS DEVICE FOR TRANSFER CAPABILITY

ENHANCEMENT

The performance of power system can be improved considerably by

incorporating FACTS devices without changing the topology of the system

or generation reschedules [105-107]. These controllable devices are used

to decrease the system transmission congestion and increase the

available transfer capability. In recent years there is an increased

concern in these controllers essentially due to the development of higher

177

rating power electronic devices and secondly the need of control of power

transactions due to the deregulation.

The objective for the location of FACTS device may be one of the

following:

1. To allow a reduction in the real power loss of a particular line.

2. To improve the efficiency by reducing the total system real power

loss.

3. To reduce reactive power loss of the total system.

4. For congestion management and to provide maximum relief of

congestion in the system.

For the first three objectives, methods based on the sensitivity

approach may be used. If the objective of FACTS device placement is to

provide maximum relief of congestion, the devices may be placed in the

most congested lines or, alternatively, in locations determined by trial-

and-error.

A number of methods are proposed for optimal location of these

devices [108-112]. Sensitivity approach [108] based on line loss has been

proposed for placement of series capacitors, phase shifters and static

VAR (Volt Ampere Reactive) compensators. In [109][110] optimization

with different objective functions is discussed for optimal power flow with

FACTS devices. In [113][114], economic dispatch problem including cost

of the FACTS devices is solved to obtain the optimal locations of FACTS

devices. In this method it is assumed that initially these devices are

178

included in all the lines. In a vertically integrated power system several

methods can be used for determining the optimal location of the FACTS

devices. In Reference [108] for location of series capacitors and SVCs loss

sensitivity based approach is proposed. In reference [103] Continuation

Power flow method is used to determine the size and location of the

series compensator to increase the power transfer capability.

Various objective functions are proposed in [109][110] for optimal

power formulation in a deregulated environment using FACTS devices. In

reference [115] performance index is used which incorporates two factors

sensitivity matrix of TCSC with respect to congested line and shadow

pricing corresponding to the congested line for optimal location of TCSC

for reducing congestion cost. Genetic algorithm based approach is

presented in Reference [100] for the optimal location of the devices in a

distribution system. In [116] particle swarm optimization algorithm is

used to determine the optimal location of various FACTS devices in a

power system in order to relieve the lines from over loads. Bees Algorithm

is proposed for the optimal location of FACTS controllers in Ref [101].

Peerapol Jirapong, Weerakorn Ongsakul [117] presented Hybrid

Evolutionary Algorithm for optimal location of multi type FACTS devices.

In Reference [118] a sensitivity factor based approach for the optimal

placement of the TCSC to minimize the congestion cost is presented. The

sensitivity factor is based on the ratio of change in real power flow to the

base case power flow in the most congested line.

179

5.8 INDICES USED FOR THE DETERMINATION OF LOCATION OF

FACTS DEVICES

There are different sensitivities used for finding the optimal

location of FACTS devices. The main criterion may be reduction in total

power loss or increase in maximum power transmittable or voltage

stability. The power flow between the buses is given by

SinX

VVP

ij

ji

ij (5.12)

Where Vi and Vj are the voltages of buses i and j

δ is the angle and

Xij is the reactance of the line

From the above equation it is clear that the transient stability of

the system can be controlled by varying the reactance.

In following section determination of reactive power loss sensitivity

factor is discussed. This factor and the proposed method of finding the

location of TCSC is compared in the subsequent sections.

5.8.1 REACTIVE POWER LOSS SENSITIVITY

The reactive power loss is affected by change in reactance of the

line by using compensators such as facts devices. This effect can be

utilized to devise reactive power sensitivity index. This index is basically

the change in reactive power loss with respect to line reactance.

The reactive power loss of Kth line between buses i and j is

jijijikkloss CosVVVVBQ 222

)( (5.13)

180

The total reactive power loss is

nl

k

jijijikT CosVVVVBQ1

22 2 (5.14)

Where Bk is the line susceptance and nl is number of lines

The reactive power loss sensitivity of kth line is given by

jijiji

kk

kk

k

T CosVVVVXR

XR

X

Q

222

222

22

(5.15)

5.9 DETERMINATION OF OPTIMAL LOCATION OF TCSC USING

COMPLEX BACK PROPAGATION ALGORITHM

FACTS devices are used to change the system parameters which

derive different results on above said objective functions. The location of

facts devices also influence the objectives of a system. During steady

state operation TCSC can be considered as a controllable reactance

connected in series with the line. The effects of insertion of TCSC in a

line are as follows:

It reduces transfer reactance of the connected between the two

buses.

The transfer capability of the line increases.

There is a reduction in effective reactive power loss.

It also improves the voltage profiles.

A complex valued neural network has been developed to determine

the changes in the transfer capability of the line, reactive power loss and

the available transfer capability (ATC) between the areas of a system. In

this new approach a neural network specifically designed to manipulate

181

the complex numbers in electrical power engineering is described. This

newly developed approach requires lesser number of input neurons or

nodes, lesser converging time and less prone to local minima problems

when compared to the conventional neural networks using real numbers.

5.9.1 OPTIMAL LOCATION OF TCSC IN A 14 BUS SYSTEM

The performance of a network can be improved considerably

without topological changes or generation rescheduling by controlling

power flows using FACTS devices. The insertion of such devices can

alleviate the system congestion and improve transfer capability. To

demonstrate the effectiveness of the proposed complex valued neural

network approach a 14 bus and 30 bus systems are considered.

IEEE 14 bus system [120] consists of 5 generators, 20 lines

(Annexure – 1) as shown in Fig. 5.12. Among these lines 15 lines shown

in Table 5.4 are considered for the location of TCSC as the real power

loss of the other lines is zero. For a smaller power system with less

number of buses, optimal location can be determined taking one line at a

time. Repeated power flow (RPF) method is used to compute the transfer

capability of each line with and without incorporating TCSC. The various

objectives considered for the location of FACTS devices are explained in

section 5.6. Here transfer capability, real and reactive power losses are

considered to be the main objectives that are influenced by the value and

the location of the FACTS device.

182

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Fig.5.12 IEEE 14 Bus system

183

TABLE 5.4 Location of TCSC

S. No. Line S. No. Line

1 Line 1-2 9 Line 6-12

2 Line 1-5 10 Line 6-13

3 Line 2-3 11 Line 9-10

4 Line 2-4 12 Line 9-14

5 Line 2-5 13 Line 11-10

6 Line 3-4 14 Line 12- 13

7 Line 5-4 15 Line 13-14

8 Line 6-11

5.9.2 TRAINING PHASE

In this phase a total of 105 training samples are generated and

used to train the proposed complex valued neural network. If the number

of training patterns are increased more accurate predictions are obtained

but over training may lead to memorization. RPF method is used for this

purpose. As the effect of TCSC inserted in a transmission line is

equivalent to a controllable reactance in series with the line which

changes the system parameters, line admittances in complex form are

taken as the inputs to the neural network. It is found that a learning rate

of 0.011 is adequate for achieving a mean squared error of 0.01. The

convergence of error is shown in Figure 5.13.

184

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Iterations

Error

Fig.5.13 convergence of error

185

Figure.5.14 Line Real Power loss with TCSC

As discussed earlier the real power loss varies with amount of

compensation. There are 15 lines and each line reactance is varied from

20% to 80% in steps. Figure 5.14 shows the change in real power loss in

all the 15 lines with various amounts of compensation. The first bar of

each line represents the real power loss without any compensation.

186

Fig.5.15 Line Reactive Power loss with TCSC

Reactive power depends on the line reactance as shown in Eq.

5.15. The effect of change in reactance on reactive power loss is

calculated at various values. The total reactive power loss of the system

at maximum loadable point with different line compensations in all 15

lines is shown in Figure 5.15.

187

Fig.5.16 Line Transfer Capability

The line transfer capability is a function of line reactance as

explained by the Equation 5.12. In Figure 5.16 the variation in line

transfer capabilities with different compensations is shown. It can be

observed that the change in transfer capability is very small in lines 9,

11, 12 and 14.

188

Fig.5.17 voltage profile

The above figure represents voltage at bus 14 with and without

TCSC in line 9-14 line reactance is 50%. The stability margin

improvement is very low. The point of collapse after compensation is

shown.

189


The above Figure 5.18 represents voltage at bus 14 with and

without TCSC in line 9-14. The line reactance compensation is 80% in

this case. Stability margin and point of voltage collapse can be improved

by changing the compensation.

190


The above Figure 5.19 represents variation of voltage at bus 14

with and without TCSC in line 9-14 line reactance compensation is 20%.

Comparatively with smaller margin and change in point of collapse.

191

Fig.5.20 Voltage profile at Bus 14


without TCSC in line 13-14 line reactance compensation is 80%. It can

be seen that the there is a noticeable change in stability margin. Point of

collapse is also changed when TCSC is installed in this line.

192

Fig.5.21 Voltage profile at Bus 14


without TCSC in line 13-14 line reactance compensation is 50%. The

improvement in voltage profile at bus 14 with compensation of 50% of

line reactance can be observed from the above figure.

193

5.9.3 RESULTS AND DISCUSSIONS OF 14 BUS SYSTEM

The proposed complex valued neural network method is tested for

different values of compensation in the lines 1-5, 2-3, 2-5, 3-4 and 5-4.

The results are shown in Tables 5.4 to 5.9. All the values are represented

in per unit. It is observed that this method is very efficient to determine

the location of the FACTS devices.

Table 5.5 (a) Results with various compensations in line 1-5

Compensation 25 %

Compensation 30%

CVNN RPF (NR) CVNN RPF (NR)

Line Flow

0.2305 + j0.0568 0.2300+j0.0690 0.2371+j0.0558 0.2390+j0.0680

Total Loss

0.1398 + j0.4837 0.1390+j0.4950i 0.1411+j0.4838 0.1390+j0.4860

Table 5.5 (b) Results with various compensations in line 1-5

Compensation 45 %

Compensation 55%

CVNN RPF (NR)

CVNN RPF (NR)

Line Flow

0.2654+j0.0522 0.2710+j0.0600 0.2944+j0.0494 0.2970+j0.0510

Total Loss

0.1464+j0.4839 0.1390+j0.4600 0.1519+j0.4840 0.1410+j0.4420


Compensation 25 % Compensation 30%

CVNN RPF (NR)

CVNN RPF (NR)

Line Flow

0.1825-j0.0706 0.1730-j0.0560 0.1831-j0.0739 0.1760-j0.0590

Total Loss

0.1409+j0.4990 0.1390+j0.5070 0.1410+j0.4987 0.1390+j0.5030

194


Compensation 45 %

Compensation 55%

CVNN RPF (NR) CVNN

RPF (NR)

Line flow

0.1861-j0.0855 0.1870-j0.0790 0.1895-j0.0945 0.1950-j0.0890

Total loss

0.1414+j0.4977 0.1390+j0.4910 0.1419+j0.4970 0.1390+j0.4840


Compensation 25 %

Compensation 30%

CVNN RPF (NR)

CVNN RPF (NR)

Line flow

0.1811+j0.0871 0.1820+j0.0940 0.1872+j0.0871 0.1890+j0.0930

Total loss

0.1416+ j0.4906 0.1440+j0.5180 0.1427+j0.4907 0.1440+j0.5080



CVNN RPF (NR) CVNN RPF (NR)

Line flow

0.2120+j0.0872 0.2140+j0.0900 0.2354+j0.0872 0.2430+j0.1000

Total loss

0.1473+j0.4910 0.1430+j0.4800 0.1517+j0.4911 0.1540+j0.4930


Compensation 25 %

Compensation 30%

CVNN RPF (NR) CVNN

RPF (NR)

Line flow 0.1526+j0.0724 0.1420+j0.0730 0.1560+j0.0726 0.1470+j0.0730

Total loss

0.1426+j0.4977 0.1390+j0.5010 0.1432+j0.4987 0.1390+j0.4960

195



CVNN RPF (NR) CVNN

RPF (NR)

Line flow 0.1699+j0.0733 0.1700+j0.0750 0.1836+j0.0737 0.1860+j0.0730

Total loss

0.1454+j0.5017 0.1460+j0.5120 0.1476+j0.5036 0.1470+j0.5000


Compensation 25 %

Compensation 30%

CVNN RPF (NR)

CVNN RPF (NR)

Line flow 0.0258+j0.1440 0.0160+j0.1120 0.0260+j0.1449 0.0190+j0.1200

Total loss

0.1512+j0.5309 0.1390+j0.5030 0.1514+j0.5319 0.1460+j0.5280


Compensation 45 %

Compensation 55%

CVNN RPF (NR)

CVNN RPF (NR)

Line flow 0.0269+j0.1472 0.0250+j0.1390 0.0277+j0.1483 0.0300+j0.1560

Total loss

0.1522+j0.5347 0.1500+j0.5340 0.1531+j0.5360 0.1560+j0.5460


Compensation 25 %

Compensation 30%

CVNN RPF (NR) CVNN

RPF (NR)

Line flow 0.1347+j0.0412 0.1300+j0.0470 0.1408+j0.0447 0.1310+j0.0470

Total loss

0.1829+j0.4473 0.1390+j0.5180 0.2011+j0.4584 0.1390+j0.5160

196

Table 5.10(b) Results with various compensations in line 5-4

Compensation 45 %

Compensation 55%

CVNN RPF (NR)

CVNN RPF (NR)

Line flow 0.1402+j0.0447 0.1360+j0.0480 0.1347+j0.0413 0.1400+j0.0470

Total loss

0.1993+j0.4584 0.1400+j0.5160 0.1829+j0.4478 0.1400+j0.5120

Table 5.11 Effect of TCSC on Line Transfer Capability

Transfer

Capability

Without

Compensation

With TCSC % increase in TC

Line 1-2 436 527 20.8%

Line 1-5 190 386 103%

Line 2-3 158 217 37.3%

Line 2-4 150 306 104%

Line 2-5 119 245 105%

Line 3-4 13 50 284%

Line 5-4 123 148 20.3%

Line 6-11 58 77 32.7%

Line 6-12 20 30 50%

Line 6-13 59 82 38.9%

Line 9-10 7 8 14.2%

Line 9-14 17 22 29.4%

Line 11-10 47 63 34%

Line 12-13 13 19 46.2%

Line 13-14 53 91 71.7%

197

The effect of TCSC on line transfer capability is shown in Table

5.11. It is seen that there is a large change in transfer capability of some

lines when considered individually.

5.10 ENHANCEMENT OF ATC IN A MODIFIED IEEE 30 BUS SYSTEM

In this section modified IEEE 30 bus system (Annexure – 1) has

been considered for simulation studies. In an open access deregulated

environment the power transaction can occur from any point of

generation to any point of load. The single line diagram of 30 bus system

is shown in Figure 5.22. Some typical transactions considered are

between area 1 to area 2 and area 1 to area 3 for illustration purpose.

Repeated power flow method is used and in all the transactions

considered here the real and reactive powers are increased at a constant

power factor. To train the network and to test its robustness a number of

patterns are generated with different values of compensations

incorporating TCSC in each of the four tie lines as shown in Figure 5.22.

The 3 areas of the system are shown in Figure 5.23. Line parameters are

considered as inputs to the network and it is assumed that these

transaction paths have a small amount of line resistances, otherwise

product of weighted inputs would become zero. Various locations of

TCSC are also show in the Figure 5.24 [128] and Table 5.10.

198

Fig.5.22 Modified IEEE 30 Bus System

199

Table 5.12: 30 Bus System Tie Lines

Location of TCSC Tie Line between the areas

1 Line 4-12

2 Line 6-10

3 Line 9-10

4 Line 28-27

Fig.5.23. Representation of 3-Area System

200

5.10.1 ALGORITHM FOR THE PROPOSED APPROACH

The procedure followed for the determination of Transfer capability

and available transfer capability is explained in the following steps.

1. Newton Raphson power flow method is used to obtain the base

case results.

2. Area 1 is assumed to be the seller area, Area 2 and Area 3 as

buyer areas.

3. The load and the generation are increased in steps.

4. Nose curve or PV curves are traced using Repeated Power flow

which is explained in section [5.5.3].

5. The sum of power flows through the tie lines at the point of

voltage collapse are taken as Total transfer capability.

6. At this voltage limit the Available Transfer Capability is obtained

from

ATCmn = Pij – Pij0 (5.16)

7. The enhancement of ATC with different compensations can be

obtained incorporating the model of TCSC in the method used.

The following transactions are considered for estimating the ATC of

the system.

1. Transaction 1 between Area 1 and Area 2 in which Area 1

supplies the increase in load in Area 2.

2. Transaction 2 between area 1 and Area 3 in which Area 1

supplies the increased load in area 3.

201

The various effects of incorporating TCSC in each of the 4 tie lines

are simulated using the mathematical model of TCSC and used for

training and validation of proposed method.

Line 23-24 is not considered as it is between area 2 and area 3. That

is not between buyer and seller areas.

5.10.2 RESULTS AND DISCUSSIONS

The proposed method has been applied to IEEE 14 bus system and

a modified IEEE 30 bus system to find the optimum location of the

TCSC. In this work on TCSC is modeled and considered because it

reduces net transfer reactance and enhances power transfer capability.

As discussed earlier the added advantage of series compensator TCSC is

its smaller rating compared to shunt compensators. It is observed that

there is an improvement in voltage profile and also reduction in line loss

and total loss of the system. For optimum location of TCSC the selection

criteria can be one of the parameters explained in section 5.7.

202

Fig.5.24 Line Transfer Capability

Figure 5.24 shows the variation of transfer capability with TCSC

placed in lines 4-12, 6-10, 9-10 and 28-27. Each TCSC is operated with

an inductive reactance varying from 20% to 80% of line reactance.

According to the results obtained it can be stated that the best

candidates for placing the TCSC controllers are those lines which have

largest change in power flow between the base case and the ATC point.

From the Figure 5.24 it can be observed that there a parabolic increase

in power flow with TCSC compensation in line 6-10.

203

Fig.5.25 Available Transfer Capability


204



205

In figures 5.25 – 5.28 the change in available transfer capbility

from area 1 to 2 also area 1 to 3 are shown with respect to diffent values

and locations of TCSC. Again it is seen that there is a largest percentage

change (approximately 40%) in ATC1-3 when the TCSC is located in line

6-10.

Table 5.13: ATC with TCSC placed between buses 4 and 12

% of Line reactance

ATC1-2 (MW) ATC1-3 (MW) Reactive power loss

(MVAR)

CVNN RPF CVNN RPF CVNN RPF

65% 149 151 260 257 583 579

85% 148 148 268 264 599 590

110% 136 135 273 275 614 620

TABLE 5.14: ATC with TCSC placed between buses 6 and 10

% of Line reactance

ATC1-2 (MW) ATC1-3 (MW) Reactive power loss

(MVAR)

CVNN RPF CVNN RPF CVNN RPF

65% 140 140 287 289 648 650

85% 142 141 280 279 640 635

110% 143 142 266 271 622 619

The results obtained from the proposed CVNN method are

tabulated in Tables 5.13 and 5.14. Test patterns within and outside

training data are used to extrapolate the capability of this method. Even

though the training is limited to a value of 90% of the line reactance

value the results obtained at 110% of reactance are approximatelely

206

same as that of conventional method. This method is used to generalize

the nonlinear relationship between the system parameters with series

compensation and the transfer capability, line loss and system total real

and reactive power loss.

5.10.3 LOCATION OF TCSC FOR REDUCTION OF SYSTEM TOTAL

REACTIVE POWER LOSS

In this section a method using sensitivity factor is used to find the

optimal location of TCSC. This method is based on the sensitivity of the

total system reactive power loss (QL) with respect to the variable that can

be controlled incorporating the FACTS devices. When a TCSC is placed in

series with the transmission line between the buses i and j it

compensates the line reactanace Xij. The reduction in transfer reactance

leads to increase in maximum power that can be transferred on the line

and also with a reduction in system effective reactive power loss.

The optimal location of TCSC can be found using the reactive

power loss sensitivity factor. With respect to the control variable that is

line reactance Xij incorporating TCSC between the i and j the loss

sensitivity factor is calculated as follows:

222

22

22

)()(2

ijij

ijij

jijiji

ij

Lij

XR

XRCosVVVV

X

Qa

(5.17)

207

TABLE 5.15: Loss Sensitivity Index

Tie

Line

From

Bus

To Bus Sensitivity Index Total Reactive power loss (p.u.)

obtained from proposed method

1 4 12 a4-12 = -0.0377 0.583

2 6 10 a6-10 = -0.0382 0.648

3 9 10 a9-10 = 0.0352 0.489

4 28 27 a28-27 = -0.0229 0.524

To satisfy one of the objectives i.e. reduction of system total

reactive power loss using TCSC, 30 bus system shown in Figure 5.24 is

considere here. To determine the optimal placement of TCSC Loss

sensitivity factor approach explained above is performed. The loss

sensitivity factor aij using Equation 5.17 is computed for each line as

shown in Table 5.15. For optimal location of TCSC the lines with most

positive loss sensitivilty index is selected. In this case Line 9-10 is the

selected for the location of TCSC. It can be observed that the total

reactive power loss obtained usnig CVNN approach is also minimum. A

compensation is 65% of the line reacatnce is considered for obtaining the

results shown.

208

5.11 CONCLUSIONS

The present restructuring of electric power industry has created

most challenging problems with respect to operation and security of the

system. Recent deregulation of the energy market resulted heavy

transmission utilization. As discussed earlier the power system is over

loaded and the delay in new transmission projects has a great impact on

it causing overloading of transmission lines and voltage sags. In the

present open access deregulated environment, market participants can

produce and consume energy in amounts, transmission lines are

operated beyond their capacities causing congestion. Methods of

congestion management using TCSC and load curtailment are shown in

this chapter.

In situations stated above, increasing network security by

controlling power flow i.e. re-dispatching of power and injection of

reactive power play a vital role. There is also a need of improving transfer

capability while maintaining the security of the system. This necessity

has created interest among the researchers to propose cost effective

methods for a robust power system using the latest technologies. The

most suitable means for this purpose are FACTs devices which can

enhance the transfer capability, improve voltage profile and there by

network security. In contrast to the conventional compensators like

series, shunt capacitors and reactors which can be used for improvement

of transfer capability and voltage profile, FACTS devices have added

209

advantages of step less control for fine regulation, capability to increase

or decrease the power flow according to the requirement and their re-

locatability as they can be built in to movable containers.

In this chapter a method using complex valued neural networks is

proposed to determine optimal location of FACTS devices, particularly

TCSC considering requirements, such as reduction of loss, increasing

transfer capability and there by alleviating congestion in deregulated

electricity market. This method is particularly useful for the system

planners during the expansion procedure to determine the size and

location of FACTS devices as they provide most reliable and efficient

solution. It is shown that this method is very effective and easy to apply

in a deregulated system.

congestion management and optimal placement of tcsc...

Documents