139

CHAPTER 7
STATIC SECURITY ENHANCEMENT USING
MULTI TYPE FACTS DEVICES FOR
MULTIPLE CONTINGENCIES

7.1

INTRODUCTION
In the proposed work, Particle Swarm Optimization (PSO)

algorithm is used to optimize the various parameters involved in introduction

of multi type FACTS devices in a power system. The various parameters
taken into consideration are the location of the devices in transmission lines,
their types, settings and installation cost of the devices. The simulation is
performed on a 6 bus and IEEE 30 bus systems with two types of FACTS
devices modelled for steady state studies.

In this problem two cases are considered and they are single and
double contingency. Combination of TCSC and UPFC are used to reduce line
overloads and installation cost of FACTS devices. To evaluate the suitability
of a given branch for placing UPFC and TCSC, an index called Contingency
Severity Index (CSI) is calculated for each branch. This index is used for
installing the UPFC and TCSC. Once located, the optimal settings of these
devices with respect to single and double contingency can be obtained by
solving the optimization problem.

140

7.2

PROBLEM FORMULATION
The objectives used in this problem are eliminating or alleviating

the line overloads and minimizing the installation cost of the multi type
FACTS devices. The problem can be formulated as follows and it involves
three stages:

i)

FACTS devices

ii)

Single contingency

iii) Double contingency

7.2.1

Unified Power Flow Controller (UPFC)

Connecting a STATCOM, which is a shunt connected device, with

a series branch in the transmission line via its DC circuit results in a UPFC, is
shown in Figure 7.1. The UPFC combines the benefits of a STATCOM and a
TCSC.

Figure 7.1 Schematic Diagram of UPFC

141

7.2.1.1

Operation of UPFC
The UPFC consists of two voltage-sourced converters. The two

converters are operated from a common dc link provided by a dc storage

capacitor. An ideal ac-to-ac power converter in which the real power can
freely flow in either direction between the ac terminals of the two converters,
and each converter can independently generate (or absorb) reactive power at
its own ac output terminal.

Converter 2 provides the main function of the UPFC by injecting a

voltage and phase angle in series with the line via an insertion transformer.
This injected voltage acts essentially as a synchronous ac voltage source. The
transmission line current flows through this voltage source resulting in
reactive and real power exchange between it and the ac system. The reactive
power exchanged at the ac terminal (i.e., at the terminal of the series insertion
transformer) is generated internally by the converter. The real power
exchanged at the ac terminal is converted into the dc power which appears at
the dc link as a positive or negative real power demand.

The basic function of Converter 1 is to supply or absorb the real

power demanded by Converter 2 at the common dc link to support the real
power exchange resulting from the series voltage injection. This dc link
power demand of Converter 2 is converted back to ac by Converter 1 and
coupled to the transmission line bus via a shunt connected transformer. In
addition to the real power need of Converter 2, Converter 1 can also generate
or absorb controllable reactive power, if it is desired, and thereby provide
independent shunt reactive compensation for the line. It is important to note
that there is a closed direct path for the real power negotiated by the action of
series voltage injection through Converters 1 and 2 back to the line. The
corresponding reactive power exchange is supplied or absorbed locally by

142

Converter 2 and therefore does not have to be transmitted by the line. Thus,
Converter 1 can be operated at a unity power factor or can be controlled to
have a reactive power exchange with the line independent of the reactive
power exchanged by Converter 2. Obviously, there can be no reactive power
flow through the UPFC dc link.

7.3

OPTIMAL PLACEMENT OF FACTS DEVICES

The essential idea of the proposed multi type FACTS devices,

UPFC and TCSC placement approaches are to determine a branch which is
most sensitive for the large list of single and double contingencies. This
section will describe the optimal placement procedure for the UPFC and
TCSC based on contingency severity index (CSI). The contingency severity
index is computed by using Equation from (5.4) to (5.6).

After calculating CSI values, the branches are ranked according to

their corresponding CSI values. A branch with high value of CSI will be more
sensitive for security system margin. The branch with the largest CSI is
considered as the best location for FACTS device.

7.3.1

Optimal Settings of FACTS Devices

In this thesis, UPFC is modelled as a combination of TCSC in

series with the line and SVC connected across the corresponding buses
between which the line is connected as shown in Figure 7.2.

143

Figure 7.2 UPFC Model

After fixing the location, to determine the best possible settings of
FACTS devices for all possible single and double contingencies, the
optimization problem will have to be solved using PSO technique. The
objective function for this work is,

Objective function = minimize { SOL and IC}

m

SOL =
c =1 k =1

Pk
a k max

Pk

(7.1)

Installation cost includes the sum of installation cost of all the

devices and it can be calculated using the cost function given by,

CTCSC = 0.0015S2 0.71S +153.75 (US$ / KVAR)

(7.2)

C UPFC = 0.0003S2 0.2691S + 188.22 (US$ / KVAR)

(7.3)

where, S - Operating range of UPFC in MVAR

S = Q2 Q1

(7.4)

The objective function is solved with the following constraints:

144

7.3.1.1

Voltage Stability Constraints

VS includes voltage stability constraints in the objective function

and is given by,

if 0.9 < Vb < 1.1

0

VS = 0.9 - Vb if Vb < 0.9

Vb - 1.1 if Vb > 1.1

(7.5)

Vb- voltage at bus b

7.3.1.2

FACTS Devices Constraints

The FACTS device limit is given by,

0.5X L < X TCSC < 0.5X L

200 MVAR QSVC 200 MVAR

7.3.1.3

(7.6)

Power Balance Constraints

While solving the optimization problem, power balance equations

are taken as equality constraints. The power balance equations are given by,

= PD + PL

(7.7)

Pi = E i E k [G ik cos (i k ) + Bik sin (i k )]

(7.8)

Qi = E i E k [G ik sin (i k ) + Bik cos (i k )]

(7.9)

145

7.4

IMPLEMENTATION OF PSO FOR OPTIMAL LOCATION

OF FACTS DEVICES

PSO is initialized with a group of random particles and searches for

optima by updating generations. In every iteration, each particle is updated by
following two best values. The first one is the best solution (fitness value) it
has been achieved so far. This value is called Pbest. Another best value that is
tracked by the particle swarm optimizer is the best value obtained so far by
any particle in the population. This best value is the global best, called Gbest.
After finding the best values, the particles update its velocity and position.
Calculation of fitness function:

Fitness function = SOL + ( 1 VS ) + ( 2 IC )

(7.10)

Algorithm:

Step 1.

The bus data, line data, and number of FACTS devices

are given as inputs

Step 2.

The initial population of individuals is created in

normalized form so as to satisfy the FACTS devices
constraints given by (7.6)

Step 3.

For each individual in the population, the fitness function

given by (7.10) is evaluated in denormalized form after
simulating all possible single and double contingencies
by using AC Load flow

Step 4.

The velocity is updated and new population is created.

Step 5.

If maximum iteration number is reached, then go to next

step else go to step 3

Step 6.

Print the best individuals settings.

146

7.5

RESULTS AND DISCUSSION

The optimal location of FACTS devices, in order to minimize the

installation cost of devices and overloads for 6 bus and IEEE 30 bus systems,
are obtained and discussed in this section. The simulation studies are carried
out on Intel Pentium IV Processor computer with 3GHZ, 256MB RAM,
40GB Hard drive using MATLAB 7.0 version.
7.5.1

6 Bus System

The bus data and line data of the six bus test system are given in
Appendix 3. This system is analyzed for both single and double
contingencies.
7.5.1.1

Single Contingency

The location of FACTS devices depends upon the CSI values

which are calculated for 11 branches by considering all single contingencies.
Then, the branches are ranked according to their values of CSI which are
given in Table 7.1.
Table 7.1 Ranking of Branches for 6 Bus System Single Contingency
Rank
1
2

Branch
1-2
2-6

CSI
0.0445
0.0383

1-4

0.0329

4
5

3-5
2-5

0.0286
0.0279

2-4

0.0252

2-3

0.0205

*Remaining Branches having Less CSI Value

147

7.5.1.2

Double Contingency

Considering, two branches outaged at a time for 11 branches 55

double contingency combinations are available. Considering all the double
contingency combinations, the 11 branches are ranked based on their CSI
values. They are given in Table 7.2.

Table 7.2 Ranking of Branches for 6 Bus System Double Contingency

Rank
1
2

Branch
3-6
2-3

CSI
0.872
0.6931

3
4

1-2
1-5

0.4866
0.4100

5
6

2-6
2-5

0.3597
0.3280

7
8

2-4
1-4

0.3064
0.2736

*Remaining Branches having Less CSI Value

Tables 7.1 and 7.2 show that branch number 1-2 and 3-6 are chosen
as the best location to place the first available multi type FACTS devices for
single and double contingencies. Depending on the available budget, the
placement of other FACTS devices can proceed where branch 2-6 and 2-3
will be the second choice, branch 1-4 and 1-2 are the third choice and so on.
Once the location is determined, their type, optimal settings and cost of
installation can be obtained by solving the optimization problem using PSO.
Tables 7.3 and 7.4 show the overloading of branches when different number
of FACTS devices is installed during single and double contingencies.

148

Table 7.3

Overloading of 6 Bus Branches Before and After Placing

Multi Type FACTS Device Single Contingency

No. of
devices

SOL

No. of
overloads

FACTS

Fitness

device Cost
6

(US$) (10 )

(109)

Execution
time
(seconds)

Value

126.0266

23

0.1410

107.8061

19

3.7821

2.0515

231.50

100.0111

18

12.591

1.9555

230.90

96.4418

17

16.353

1.8822

232.50

89.5330

15

15.404

1.7193

232.79

93.4915

17

17.964

1.8334

226.84

Table 7.4

Overloading of 6 Bus Branches Before and After Placing

Multi Type FACTS Devices- Double contingency

No. of
devices

SOL

FACTS
device Cost
overloads
(US$) (106)
No. of

Fitness
Value

Execution
time

(109)

(Seconds)

447.39

188

0.2010

382.71

187

2.9306

6.1280

1404.3

342.36

179

79.347

5.8560

1660.2

318.20

175

9.5265

5.5502

1361.4

258.53

182

28.480

5.6950

1149.9

505.40

198

33.205

6.2356

1415.9

149

Tables 7.3 and 7.4 show that the severity index (SOL) and the
number of overloads are reduced from 23 to 15 when four FACTS devices are
placed for single contingencies and 188 to 175 when three FACTS devices are
placed for double contingencies. Further increase of devices, shows no
improvement in reduction of severity, overloading and cost of installation
rather they start increasing. Hence in this case, four and three number of
FACTS devices are considered for optimal system security for single and
double contingencies. The optimal settings, line number and the type of
device are obtained by solving optimization algorithms using PSO is given in
Tables 7.5 and 7.6.

Figures 7.3 and 7.4 represent the fitness convergence curve for
IEEE 6 bus system for single and double contingencies. Number of
generations / iterations is taken in X axis and Fitness value in Y axis. The
simulation is carried out with multiple runs to get the optimal results of multitype FACTS devices.

PSO parameters used in this work are:

i)

No. of population = 30

ii)

Max Generation

= 150

iii) C1

= C2 =

iv) Vmax

= 0.25

v)

= linearly varied from 0.9 to 0.2 during a

run

150

Table 7.5

Optimal Settings Multi Type FACTS Devices 6 Bus

System Single Contingency

No. of

Reactance

Reactive

XTCSC

Power QSVC

(P.U)

(MVAR)

0.1497

-155.1191

0.1501

-150.4248

-0.0313

-141.3419

0.0515

0.1437

-119.3479

2-4

-0.0335

-146.2072

2-6

0.0401

-74.848

-0.0336

-134.7928

0.1435

-122.7183

2-5

-0.0668

1-2

-0.07451

-136.9343

2-6

0.0764

-131.8085

-0.0532

-36.1739

2-3

-0.0103

-155.3881

3-5

0.1401

-64.1792

Branch

devices Number

1
2

3-5
3-5
2-4

Type of device
TCSC

UPFC

2-6
3

3-5

1-4
3-5

1-4

151

Table 7.6

Optimal Settings Multi Type FACTS Devices 6 Bus

System Double Contingency

Type of device
No. of

Branch

devices Number

1
2

2-6

TCSC

UPFC

Reactive
Power
QSVC

XTCSC
(pu)

(MVAR)

0.0815

0.0750

-85.7757

-0.0296

-101.0229

0.0299

-36.0040

0.0717

20.9624

1-4

-0.0381

3-6

-0.0431

-122.0320

0.0807

-146.1893

-0.0206

-40.6733

1-4

-0.0034

-120.0641

3-6

-0.0485

109.7632

2-3

0.0095

-118.6482

0.0524

-51.9278

2-5

-0.0662

-135.2774

2-4

-0.0277

-131.2981

2-6
2-4

1-5
3

Reactance

2-6

1-5
2-4

1-5

Fitness Value

152

Figure 7.3

Fitness Convergence Curve for 6 Bus System-Single

Fitness Value

Contingency

Figure 7.4

Fitness Convergence Curve for 6 Bus System-Double

Contingency

153

7.5.2

IEEE 30 Bus System

The IEEE 30 bus system consists of 41 branches. Line data and bus
data are given in Appendix 4. This system is also analyzed for both single and
double contingencies.

7.5.2.1

Single Contingency

Among the 41 possible single contingencies, leaving 3 branches

(25-26,9-11,12-13) connected to isolated buses, only 38 single contingencies
are considered. The CSI index is calculated for all the 41 lines and top 9
critical line values are given in Table 7.7.

Table 7.7

Ranking of Branches for IEEE 30 Bus System-Single

Contingency

Rank

Branch

CSI

15-18

0.0511

25-27

0.0505

2-6

0.0460

22-24

0.0424

1-3

0.0389

3-4

0.0384

27-29

0.0245

1-2

0.0232

27-30

0.0185

154

7.5.2.2

Double Contingency

Considering two branches are outaged at a time, for 41 branches

820 double contingency combinations are available. Leaving the branches
connected to isolated buses, the remaining double contingency combinations
are considered in this work. These contingencies are ranked based on CSI
values. They are given in Table 7.8. In this work only top 22 critical lines are
used for detailed analysis.
Table 7.8

Ranking of Branches for IEEE 30 Bus System Double

Contingency

Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

Branch
25-27
22-24
16-17
23-24
2-6
2-4
15-23
15-18
14-15
6-10
4-12
10-21
12-15
1-2
10-17
1-3
3-4
18-19
6-28
27-29
9-10
12-13

CSI
0.8894
0.7306
0.4899
0.2464
0.2060
0.1935
0.1920
0.1631
0.1588
0.1400
0.1048
0.1018
0.0991
0.0883
0.0870
0.0814
0.0802
0.0699
0.0265
0.0262
0.0194
0.0161

155

After ranking of the branches, the PSO algorithm is used to find out
the location of the devices, their types, settings and installation cost to
alleviate the line overloads and to improve the system security margin. They
are given in Tables from 7.9 to 7.12.

Figures 7.5 and 7.6 represent the fitness convergence curve for
IEEE 30 bus system for single and double contingencies. Number of
generations / iterations is taken in X axis and Fitness Value in Y axis.

PSO parameters used in this work are:

i) No. of population

25

ii) Maximum Generation

100

iii) C1 = C2

iv) Vmax

0.25

v) W - linearly varied from 0.9 to 0.2 during a run.

Table 7.9

Overloading of IEEE 30 Bus Branches Before and After

Placing Multi Type FACTS Devices Single Contingency

Fitness
Value
(109)
0
1.8459

Execution
time
(seconds)

SOL

No. of
overloads

0
1

145.7700
144.7901

16
15

FACTS
Device Cost
(US$) (106)
0
18.398

2
3

144.0800
141.7725

14
13

21.050
19.409

1.7985
1.8156

14072.79
13921.73

4
5

140.6269
142.8015

13
14

9.9994
27.950

1.7378
1.8338

14115.35
14223.51

No. of
devices

0.17800
14010.60

156

Table 7.10

Overloading of IEEE 30 Bus Branches Before and After

Placing Multi Type FACTS Devices Double Contingency

Execution
time
(seconds)

0
46.391

Fitness
Value
x (109)
0
7.5111

84

36.900

7.4852

71432.04

78
91
100

23.750
21.813
25.188

7.4673
7.3991
7.7408

71508.06
72083.18
72204.67

SOL

No. of
overloads

FACTS
Device Cost
(US$) x (106)

0
1

546.2114
540.9482

105
91

524.2522

3
4
5

519.8329
528.5074
529.0064

No. of
devices

Table 7.11

0.488150
71184.04

Optimal Settings of Multi Type FACTS Devices IEEE

30 Bus System Single Contingency

No. of
devices

Branch
Number

2-6
2-6
3-4
15-18
2-6
22-24
15-18
2-6
3-4
27-29
15-18
2-6
22-24
3-4
27-30

Type of device
TCSC

UPFC

Reactance
XTCSC (P.u)

-0.0974
0.0819
0.0136
0.1488
-0.0925
-0.0823
0.1068
0.0861
0.0170
0.2538
0.1088
-0.0883
-0.0839
-0.0204
-0.4327

Reactive
Power QSVC
(MVAR)
-54.3545
0
0.3438
0
0
0
32.54
0
0
-56.24
0
-43.9082
-18.4190
64.4621
0

157

Table 7.12

Optimal Settings of Multi Type FACTS Devices IEEE

30 Bus System Double Contingency

Type of device

Reactance

Reactive

No. of

Branch

devices

Number

2-6

-0.0699

26.0844

23-24

-0.1402

-15.0381

-0.0632

-0.0791

-25.3856

14-15

-0.1352

-19.5650

12-15

0.0572

-49.3143

-0.0685

-29.7246

2-6

-0.0726

42.9757

15-23

-0.1047

17.5657

10-17

0.0396

-0.1139

2-6

0.0853

-51.2918

15-18

-0.1091

60.6020

6-10

0.2109

-64.2678

4-12

0.1328

24.2349

22-24
4

25-27

UPFC

2-6
2-6

TCSC

XTCSC (P.u)

Power QSVC
(MVAR)

158

Figure 7.6

Fitness Convergence Curve for IEEE 30 Bus System-Single

Contingency

Figure 7.7

Fitness Convergence Curve for IEEE 30 Bus System-Double

Contingency

159

7.6

CONCLUSION

This research work has presented a PSO based solution for solving
the power system security enhancement problems involving multiple
constraints. The results show that the PSO approach has provided a good
solution to place multi type FACTS devices along the system branches based
on CSI values, to alleviate system overloads and to improve the system
security margin during single and double contingencies. Once good PSO
control parameters are obtained, the solution to the problem can easily be
obtained.

Simulations are performed on 6 and IEEE 30 bus systems. The

location of multi type FACTS devices, the type of device to be placed, their
settings and installation cost are taken as the optimization parameters for
multiple (single and double) contingencies. In multiple contingencies, it is
observed that the system security margin cannot be improved further after
placing certain optimal number of multi type FACTS devices. These settings
can effectively be used to enhance the system security margin without
investing additional transmission resources. The test results indicate that the
PSO approach is promising and efficient in dealing the location of multi type
FACTS devices with multiple constraints, which occur in real life.