Vol 19, No.

10;Oct 2012

Voltage Stability Limit Enhancement and Active Power

Loss Minimization Comparison between Two Different
Combinations of FACTS Devices under Stressed Conditions

L.Jebaraj
Department of Electrical and Electronics Engineering
St.Anne’s College of Engineering and Technology, Panruti, Tamil Nadu, India
Tel: +91 97876 66641 E-mail: [email protected]

C.Christober Asir Rajan

Department of Electrical and Electronics Engineering
Pondicherry Engineering College, Puducherry, India
Tel: +91 94437 13846 E-mail: [email protected]

N.Muralikrishnan
Department of Electrical and Electronics Engineering
Mailam Engineering College, Villupuram, India
Tel: +91 97914 47535 E-mail: [email protected]

ABSTRACT

Modern power system networks are having several thousand of buses and hundreds of generators are
being operated under highly stressed conditions and there is a risk of voltage instability problems
owing to increased load demand. The only way to counteract this problem is by reducing the reactive
power load in the system or by adding new reactive power generation systems in the weakest points of
the system thereby increasing the voltage at those points. A power system needs to be with sufficient
voltage stability margin for secured operation. In this paper, the combination of SVC with TCSC and
SVC with SSSC devices are used separately and compare the results behind the voltage stability limit
improvement and active power loss minimization scenario. The line based LQP voltage stability
indicator is used for voltage stability assessment. Differential Evolution (DE) algorithm is exploited
for optimization of the voltage stability limit, active power loss and location and size of FACTS
devices. The effectiveness of the proposed algorithm is validated by carrying out on the standard IEEE
30 bus system under stressed conditions.

Keywords: Differential Evolution Algorithm (DEA), Voltage Stability, Thyristor Controlled Series
Compensator (TCSC), Static Synchronous Series Compensator (SSSC), Static Var Compensator
(SVC), Line Stability Index, FACTS Devices, Load Flow.

369 [email protected]
 Vol 19, No. 10;Oct 2012

1. INTRODUCTION

Now a days power system are undergoing numerous changes and becoming more complex from
operation, control and stability maintenance standpoints when they meet ever-increasing load demand
(Kundur, 1994). Voltage stability is concerned with the ability of a power system to maintain
acceptable voltage at all buses in the system under normal conditions and after being subjected to a
disturbance. A system enters a state of voltage instability when a disturbance, increase in load demand,
or change in system condition causes a progressive and uncontrollable decline in voltage.

The main factor causing voltage instability is the inability of the power system to meet the demand for
reactive power (Van Cutsem, 2000). Excessive voltage decline can occur following some severe
system contingencies and this situation could be aggravated, possibly leading to voltage collapse, by
further tripping of more transmission facilities, var sources or generating units due to overloading.
Many large interconnected power systems are increasingly experiencing abnormally high or low
voltages or voltage collapse. Abnormal voltages and voltage collapse pose a primary threat to power
system stability, security and reliability. Moreover, with the fast development of restructuring, the
problem of voltage stability has become a major concern in deregulated power systems. To maintain
security of such systems, it is desirable to plan suitable measures to improve power system security
and increase voltage stability margins. (Dobson and Chiang, 1989). Voltage instability is one of the
phenomena that resulted in major blackouts. Recently, several network blackouts have been related to
voltage collapses (Technical Analysis of August 14, 2003: Blackout – A report by NERC).

Flexible AC Transmission System (FACTS) controllers are capable of supplying or absorption of

reactive power at faster rates. The introduction of Flexible AC Transmission System (FACTS)
controllers are increasingly used to provide voltage and power flow controls. Insertion of FACTS
devices is found to be highly effective in preventing voltage instability (Hingorani and Gyugyi,
2000).Series and shunt compensating devices are used to enhance the Static voltage stability margin.

Voltage stability assessment with appropriate representations of FACTS devices are investigated and
compared under base case of study (Musunuri and Dehnavi, 2010; Sode Yome et al., 2005; Canizares
and Faur, 1999). One of the shortcomings of those methods only considered the normal state of the
system. However voltage collapses are mostly initiated by a disturbance like line outages. Voltage
stability limit improvement needs to be addressed during network contingencies. So to locate FACTS
devices consideration of contingency conditions is more important than consideration of normal state
of system and some approaches are proposed to locate of facts devices with considerations of
contingencies too(Maysam Jafari and Saeed Afsharnia, 2007).

The generalized power injection model of SSSC needs modification of the jacobian matrix and makes
quiet complex in coding. In (Zhang and Zhang, 2006), the SSSC control parameters, voltage
magnitude and angle of the series converters, are presented as independent variables and their values
are found through the traditional load flow iterative process. In this case, the size of the Jacobian
matrix increases to incorporate the additional independent variables. The new model of the SSSC
changes only the bus admittance matrix and consequently reduces the coding of load flow problem
incorporating SSSC simple. The SSSC control parameters, voltage magnitude and angle of the series
converters, are presented as independent variables and their values are found through the traditional
load flow iterative process. In this case, the size of the jacobian matrix increases to incorporate the
additional independent variables. Hence a simple and easy to implement SSSC model based on the
circuit elements is used in this paper (Ali Akbar Motie birjandi and kauomars Sabzawari, 2010). The
SVC is modeled as a variable reactive power source/sink at the connected bus. The author (S.Sakthivel
and D.Mary) makes use of the same model of SSSC incorporation through particle swarm optimization

370 [email protected]
 Vol 19, No. 10;Oct 2012

algorithm under line outage contingency condition. Line stability indices provided important
information about the proximity of the system to voltage instability and can be used to identify the
weakest bus as well the critical line with respect to the bus of the system (Claudia Reis et al, 2009).

From the family of evolutionary computation, Differential Evolution Algorithm is used to solve a
problem of real power loss minimization and Voltage stability maximization of the system. The DE
algorithm is a population based algorithm like genetic algorithms using the similar operators;
crossover, mutation and selection. Several transformer tap positions along with numbers of reactive
power injections at some selected buses in a power system are simultaneously optimized as control
variables, so that the multiple objectives are fulfilled, keeping an eye to all specified constraints (K.
Price and R. Storn, 1995). Owing to higher capital cost of the SSSC and SVC, the installation is not
recommended to all possible line outages. Hence line outage contingency screening and ranking
carried out to identify the most critical line during whose outage FACTS controllers can be positioned
and system can be operated under stable condition (Vaahedi et al, 1999; Ejebe et al, 1995; Reppen et
al, 1993).

The prime objective of this paper is to compare the voltage stability limit, real power loss and optimal
location and size of two different combinations of FACTS devices such as TCSC with SVC and SSSC
with SVC through DE algorithm under normal, critical and single line outage contingency conditions.

2. CRITICAL CONDITIONS

Voltage collapse is a process in which the appearance of sequential events together with the instability
in a large area of system can lead to the case of unacceptable low voltage condition in the network, if
no preventive action is committed. Occurrence of disturbance or load increasing leads to excessive
demand of reactive power. Therefore system will show voltage instability. If additional sources
provide sufficient reactive power support, the system will be established in a stable voltage level.
However, sometimes there are not sufficient reactive power resources and excessive demand of
reactive power can leads to voltage collapse.

Voltage collapse is initiated due to small changes of system condition (load increasing) as well as large
disturbances (line or generator unit outage) under these conditions FACTS devices can improve the
system security with fast and controlled injection of reactive power to the system. However when the
voltage collapse is due to excessive load increasing, FACTS devices cannot prevent the voltage
collapse and only postpone it until they reach to their maximum limits. Under these situations the only
way to prevent the voltage collapse is load curtailment or load shedding. So critical loading and
contingencies are should be considered in voltage stability analysis. Recent days, the increase in peak
load demand and power transfer between utilities has an important issue on power system voltage
stability. Voltage stability has been highly responsible for several major disturbances in power system.
When load increases, some of the lines may get overloaded beyond their rated capacity and there is
possibility to outage of lines. The system should able to maintain the voltage stability even under such
a disturbed condition.

3. MODELLING OF FACTS DEVICES

3.1 Static Model of SVC

SVC is a shunt connected FACTS device capable of exchanging reactive power with the power system
through the bus at which it is connected. Amount of reactive power injected by the device is varied by

371 [email protected]
 Vol 19, No. 10;Oct 2012

varying the susceptance [4]. A variable susceptance BSVC represents the fundamental frequency
equivalent susceptance of all shunt modules making up the SVC. This model is an improved version of
SVC models. Figure 1 shows the variable susceptance model of SVC which is used to derive its
nonlinear power equations and the linearised equations required by Newton's load flow method.

Vi Vj
Xline
QSVC

BSVC

Fig. 1: Variable susceptance model of SVC

In general, the transfer admittance equation for the variable shunt compensator is
I  jB V 1
The reactive power exchanged is
Q  V  B 2
In SVC susceptance model the total susceptance BSVC is taken to be the state variable, therefore the
linearised equation of the SVC is given by
∆Pj 0 0 ∆θj
   0 θ    3
∆Qj ∆Bsvc/Bsvc
At the end of iteration i the variable shunt susceptance BSVC is updated according to
  !  !
B  B " ∆B /B  B 4
This changing susceptance value represents the total SVC susceptance which is necessary to maintain
the nodal voltage magnitude at the specified value (1.0 p.u. in this paper).

3.2 Static model of TCSC

Zij = Rij + Xij

-jXTCSC

-jBsh -jBsh

Fig.2: Model of TCSC

TCSC is a series compensation component which consists of a series capacitor bank shunted by
thyristor controlled reactor. The basic idea behind power flow control with the TCSC is to decrease or

372 [email protected]
 Vol 19, No. 10;Oct 2012

increase the overall lines effective series transmission impedance, by adding a capacitive or inductive
reactance correspondingly. The TCSC is modeled as variable reactance shown in figure 2. The
equivalent reactance of line Xij is defined as:
X  0.8X'() * X+ * 0.2X'() 5
where, Xline is the transmission line reactance, and XTCSC is the TCSC reactance. The level of the
applied compensation of the TCSC usually varies between 20% inductive and 80% capacitive.

3.3 Static model of SSSC

The static synchronous series compensator (SSSC) can be operated without an external energy source
as reactive power source with and fully controllable independent of transmission line current for the
purpose of increasing or decreasing the overall reactive voltage drop across the transmission line and
thereby controlling the electric power flow. The widely used power injection model of SSSC requires
modification of the jacobian matrix and makes the Newton-Raphson load flow (NRLF) coding more
complex.

Series Coupling
Transformer
Bus 1 Bus 2
VS VR

Voltage Source
Converter
C
_
+

Energy Source
(Optional)

Fig. 3: SSSC Configuration

A new circuit elements based model of SSSC is utilized to control the line power flows and bus
voltage magnitudes for voltage stability limit improvement. The new model of the SSSC changes only
the bus admittance matrix and consequently reduces the coding of load flow problem incorporating
SSSC simple. This converter performs the main function of injecting a controllable series voltage. The
basic configuration of SSSC is depicted in Figure 3. The model of SSSC is also shown in Figure 4.

-. /δ -. /γ -0 /δ
ZL

Fig. 4: Model of SSSC

The real and reactive powers exchanged with the line by the series voltage inserted by SSSC are
modeled as a negative resistance and reactance connected in parallel. The negative resistance
represents injection of real power and the reactance may be either capacitive or inductive depending on

373 [email protected]
 Vol 19, No. 10;Oct 2012

whether reactive power is delivered or absorbed. The complex power exchanged by the series
converter with the line is expressed as
S)  V) I 6
Where Vse is the complex voltage injected by the converter and I the current through the line given by

V /δ " V) /γ  V /δ

I  7
Z4
The active and reactive powers exchanged with the line are modeled as resistance and reactance
associated as in equations (8) and (9).
V)

R 8
P)
-89
7 9
:89
The elements R and X can be calculated directly using the following equations.
V) Z4
R 10
V sin>δ – δ – γ@ " V sinγ
V) Z4
X 11
V cos>δ  δ – γ@  V cosγ " V)
The resistance R and the reactance X representing the effect of series voltage are transformed into their
equivalent series combination. This makes the line simple with only series connected elements of the
line (XL) and the RSSSC and XSSSC denoting the SSSC.
RX 
R   12
R " X 
XR
X  13
R " X

4. LINE VOLTAGE STABILITY INDEX (LQP INDEX)

The line stability index (LQP) based on a power transmission concept is used in this paper. The value
of line index shows the voltage stability of the system. The value close to unity indicates that the
respective line is close to its stability limit and value much close to zero indicates light load in the line.
The formulation begins with the power equation in a power system. Figure 5 illustrates a single line of
a power transmission concept.

Vi Vj

Bus i Si, Pi, Qi Bus j Sj, Pj, Qj

Z = R + jX
Fig. 5: Single line concept of power transmission

374 [email protected]
 Vol 19, No. 10;Oct 2012

The power equation can be derived as;

B B 
Q  Q  " E P " Q F 14
D
C D
C

The line stability factor is obtained by setting the discriminant of the reactive power roots at bus 1 to
be greater than or equal to zero thus defining the line stability factor, LQP as,
X X 
LQP  4 H  I H  P " Q I 15
V V
This indicator is highly sensitive to change in reactive power flow the line. In this work, reactive
power flow is adjusted for voltage stability improvement. Change in reactive power flow affects the
voltage stability limit and it can be assessed suitably by using LQP index.

5. PROBLEM FORMULATION

The objective of this work is to improve the voltage stability limit improvement by minimizing real
power loss, sum of load bus voltages and sum of line voltage stability index. An augmented objective
function is formed with the three objective components and weights.
5.1 Objective function
The objective function of this work is to find the optimal rating and location of TCSC and SVC which
minimizes the real power loss, minimization of voltage deviation and maximizes the voltage stability
limit. Hence, the objective function can be expressed as:
F  KLMNP4 " wVD " 1  w LQPQ 16
Where w is the weighing factor for voltage deviation and LQP index and is set to 10.
5.2 Real power loss minimization (PL)
The total real power of the system can be calculated as follows
VW

P4  R GT U V " V   2V V cos δ  δ Y 17

TX!
Where , NL is the total number of lines in the system; Gk is the conductance of the line ‘k’; Vi and Vj
are the magnitudes of the sending end and receiving end voltages of the line; δi and δj are angles of the
end voltages.
5.3 Load bus voltage profile minimization (VD)
Bus voltage magnitude should be maintained within the allowable range to ensure quality service.
Voltage profile is improved by minimizing the deviation of the load bus voltage from the reference
value (it is taken as 1.0 p.u. in this work).
V]^

VD  R|V  V[)\ | 18
TX!

5.4 Line voltage stability index minimization (LQP)

Voltage stability limit of a power system is increased by minimizing voltage stability index value. The
indicator takes values between 0 (no-load) and 1 (full load). The line based stability index (LPQ) is
given as:
VW

LQP  R LQP 19
X!

375 [email protected]
 Vol 19, No. 10;Oct 2012

5.5 Constraints
The minimization problem is subject to the following equality and inequality constraints
5.5.1 Equality constraints:
Load Flow Constraints:
Vd

P_  P`  R V V Y cos bδ " γ  γ c  0 20

X!
Vd

Q_  Q `  R V V Y sin bδ " γ  γ c  0 21

X!

5.5.2 Inequality constraints:

Reactive Power Generation Limit of SVCs:
Qe(
 * Q  * Q  ; i i N
efg
22
Reactance Limits of TCSCs:
0.8X * X+ T * 0.2X ; k i N+ 23
Voltage Constraints:
Ve( * V * Vefg ; i i Nl 24
Transmission line flow limit:
S * Sefg ; i i N' 25

6. DIFFERENTIAL EVOLUTION ALGORITHM – AN OVER VIEW

Differential Evolution (DE) is a population based evolutionary algorithm, capable of handling non-
differentiable, nonlinear and multi-modal objectives functions. DE generates new offspring by forming
a trial vector of each parent individual of the population. The population is improved iteratively, by
three basic operations namely mutation, crossover and selection. A brief description of different steps
of DE algorithm is given below.
6.1 Initialization
The population is initialized by randomly generating individuals within the boundary constraints
7.0m  70n.o " pqMr>70nst  70n.o @; L  1,2,3, … wx, y  1,2,3, … z 26
where “rand” function generates random values uniformly in the interval (0, 1); NP is the size of the
population; D is the number of decision variables. Xjmin and Xjmax are the lower and upper bound of the
jth decision variable, respectively.
6.2 Mutation
As a step of generating offspring, the operations of “Mutation” are applied. “Mutation” occupies quite
an important role in the reproduction cycle. The mutation operation creates mutant vectors Vik by
perturbing a randomly selected vector Xak with the difference of two other randomly selected vectors
Xbk and Xck at the kth iteration as per the following equation:
-.{  7s{  7|{  7}{ ; L  1,2,3 … . . wx 27

376 [email protected]
 Vol 19, No. 10;Oct 2012

Xak, Xbk and Xck are randomly chosen vectors at the Kth iteration and a ≠ b ≠ c ≠ i and are selected a
new for each parent vector. F is the scaling constant that controls the amount of perturbation in the
mutation process and improves convergence.
6.3 Crossover
Crossover represents a typical case of a “genes” exchange. The trial one inherits genes with some
probability. The parent vector is mixed with the mutated vector to create a trial vector, according to the
following equation:
-.0{ , L€ pqMr  ‚ƒ „p y  …
~.0{   28 Œ
7.0{ , †‡ˆ‰pŠL‹‰
Where i=1, 2, 3……………NP; j=1, 2, 3…………..D. Xij k , Vij k Uij k are the jth individual of target
vector, mutant vector, and trial vector at kth iteration, respectively. q is a randomly chosen index in the
range (1,D) that guarantees that the trial vector gets at least one parameter from the mutant vector. CR
is the cross over constant that lies between 0 and 1.
6.4 Selection
Selection procedure is used among the set of trial vector and the updated target vector to choose the
best one. Selection is realized by comparing the fitness function values of target vector and trial vector.
Selection operation is performed as per the following equation:
~.{ , L€ >~.{ @ * €>7.{ @; L  1,2,3 … . . wx
7.{!   29 Œ
7.{ , †‡ˆ‰pŠL‹‰
The relative changes in the fitness of the global frog within a number of consecutive shuffling
iterations are less than a pre-specified tolerance. The maximum predefined number of shuffling
iteration has been obtained. The optimal parameter values of DE are shown in table 1.
Table 1. Optimal values of DE parameters
Optimal
Parameters
value
Number of individuals 50
Cross over constant 0.6
Scaling constant 0.3
No of iterations 100

7. IMPLEMENTATION OF DIFFERENTIAL EVOLUTION ALGORITHM

7.1 Representing an individual

Each individual in the population is defined as a vector containing the values of control parameters
including the size of the TCSC, SSSC and SVC.
7.2 Number of individuals
There is a trade-off between the number of individuals and the number of iterations of the population
and each individual fitness value has to be evaluated using a power flow solution at each iteration,
thus the number of individuals should not be large because computational effort could increase
dramatically. Individuals of 10, 20 and 50 are chosen as an appropriate population sizes.

377 [email protected]
 Vol 19, No. 10;Oct 2012

7.3 Feasible region Definition

There are several constraints in this problem regarding the characteristics of the power system and the
desired voltage profile. Each of these constraints represents a limit in the search space. Therefore the
DE algorithm has to be programmed so that the individual can only move over the feasible region. For
instance, the network in figure 7 has 4 transmission lines with tap changer transformer. These lines are
not considered for locating series compensating devices, leaving 37 other possible locations for the
devices. In terms of the algorithm, each time that an individual’s new position includes a line with tap
setting transformer, the position is changed to the geographically closest line (line without
transformer). Finally, in order to limit the sizes of the series compensating device units, the restrictions
of level of compensation is applied to the individuals.
7.4 Integer DE
For this particular application, the position of individuals is determined by an integer number (line
number). Therefore the individuals’ movement is approximated to the nearest integer numbers.
Additionally, the location number must not be a line with tap setting transformer. If the location is line
with tap setting transformer, then the individual component regarding position is changed to the
geographically closest line without a tap setting transformer.

8. RESULTS AND DISCUSSION

29 28
27
26 25
30
24
23

15 18 19
17 20

21
14 16 22

11
13 12 10
G 9
G
6 8
1 3 4
7 G
G
2 5
G G

Fig. 6: One line diagram of IEEE 30 Bus Test System

378 [email protected]
 Vol 19, No. 10;Oct 2012

The proposed work is coded in MATLAB 7.6 platform using 2.8 GHz Intel Core 2 Duo processor
based PC. The method is tested in the IEEE 30 bus test system shown in figure 6. The line data and
bus data are taken from the standard power system test case archive. The system has 6 generator buses,
24 load buses and 41 transmission lines. System data and results are based on 100 MVA and bus no 1
is the reference bus. In order to verify the presented models and illustrate the impacts of TCSC, SSSC
and SVC study, three different operating conditions are considered as mentioned below.
Case 1: The system with normal load in all the load buses is considered as normal condition and the
Newton-Raphson load flow is carried out with loading factor value equal to 1.
Case 2: The system with 50 % increased load in all the load buses is considered as a critical condition.
Loading of the system beyond this level, results in poor voltage profile in the load buses and
unacceptable real power loss level.
Case 3: Contingency is imposed by considering the most critical line outage in the system. This is the
most suitable condition for voltage stability analysis of a power system as voltage stability is
usually triggered by line outages. The loading factor is taken as 1.5 for this case.
Newton – Raphson program is repeatedly run with the presence of FACTS devices through DE
algorithm. The voltage stability limit improvement is assessed by the value of LQP index. The LQP
index values after insertion of two combinations of FACTS devices in all cases are shown in table 2. It
is evident from the table that LQP values of 16 lines are unaltered and reduced in 14 lines through
SSSC with SVC insertion under normal loading condition. In critical loading condition the LQP values
of 11 lines are unaltered and reduced in 25 lines through TCSC with SVC insertion. But line 5 having
more stress relief than other lines under SSSC with SVC insertion. The LQP values of 25 lines are
reduced and 12 lines are unaltered through the insertion of SSSC with SVC devices.

Table 2. Voltage Stability Index Values for all cases

Case 1 Case 2 Case 3
Line LQP values LQP Values LQP Values
No. Without TCSC SSSC Without TCSC SSSC Without TCSC SSSC
FACTS with with FACTS with with FACTS with with
SVC SVC SVC SVC SVC SVC
1 0.0797 0.0755 0.0329 0.0594 0.1117 0.0688 0.0786 0.0616 0.0568
2 0.0076 0.0076 0.0076 0.0114 0.0114 0.0114 0.0076 0.0076 0.0076
3 0.0113 0.0113 0.0115 0.0184 0.0180 0.0184 0.0113 0.0113 0.0115
4 0.0026 0.0026 0.0026 0.0043 0.0042 0.0043 0.0027 0.0027 0.0026
5 0.1284 0.0702 0.0987 0.3596 0.2154 0.0793 0.0000 0.0000 0.0000
6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
8 0.0574 0.0267 0.0585 0.0953 0.0953 0.0953 0.0635 0.0350 0.0635
9 0.0386 0.0179 0.0387 0.0650 0.0642 0.0646 0.0403 0.0219 0.0386
10 0.0012 0.0079 0.0069 0.0522 0.0199 0.0333 0.0077 0.0192 0.0034
11 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
12 0.3691 0.3678 0.3700 0.3899 0.3852 0.3874 0.3851 0.3792 0.3686
13 0.1226 0.1190 0.1169 0.2549 0.2283 0.2351 0.1732 0.1554 0.1152
14 0.0678 0.0676 0.0676 0.0715 0.0694 0.0709 0.0693 0.0688 0.0675
15 0.0749 0.0747 0.0752 0.1255 0.1234 0.1244 0.0775 0.0767 0.0749
16 0.0513 0.0485 0.0461 0.1687 0.1658 0.1767 0.0836 0.0724 0.0475
17 0.0180 0.0180 0.0180 0.0302 0.0291 0.0298 0.0183 0.0182 0.0180
18 0.0147 0.0160 0.0146 0.0246 0.0236 0.0242 0.0149 0.0148 0.0146

379 [email protected]
 Vol 19, No. 10;Oct 2012

19 0.0157 0.0157 0.0156 0.0263 0.0253 0.0259 0.0159 0.0158 0.0157

20 0.0408 0.0408 0.0407 0.0700 0.0672 0.0689 0.0416 0.0413 0.0407
21 0.0484 0.0483 0.0782 0.0830 0.0454 0.0817 0.0494 0.0491 0.0482
22 0.0091 0.0091 0.0090 0.0157 0.0150 0.0154 0.0092 0.0092 0.0090
23 0.0207 0.0207 0.0204 0.0364 0.0348 0.0358 0.0212 0.0210 0.0206
24 0.0023 0.0023 0.0162 0.0040 0.0038 0.0039 0.0023 0.0023 0.0023
25 0.0064 0.0064 0.0468 0.0111 0.0106 0.0110 0.0066 0.0066 0.0064
26 0.0206 0.0206 0.0205 0.0357 0.0197 0.0352 0.0212 0.0210 0.0205
27 0.0374 0.0374 0.0372 0.0647 0.0618 0.0638 0.0384 0.0381 0.0372
28 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
29 0.0018 0.0018 0.0018 0.0031 0.0030 0.0031 0.0018 0.0018 0.0018
30 0.0349 0.0149 0.0148 0.0258 0.0247 0.0254 0.0152 0.0151 0.0149
31 0.0226 0.0226 0.0225 0.0635 0.0607 0.0623 0.0233 0.0230 0.0224
32 0.0303 0.0303 0.0301 0.0855 0.0816 0.0840 0.0311 0.0308 0.0301
33 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0961
34 0.0486 0.0485 0.0485 0.0873 0.0845 0.0837 0.0504 0.0497 0.0469
35 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
36 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
37 0.0182 0.0181 0.0181 0.0322 0.0314 0.0306 0.0189 0.0186 0.0177
38 0.0557 0.0556 0.0557 0.0987 0.0963 0.1121 0.0579 0.0571 0.0543
39 0.0435 0.0434 0.0435 0.0797 0.0776 0.0880 0.0453 0.0446 0.0424
40 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
41 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

In case 3, the line outage is ranked according to the severity and the severity is taken on the basis of
the line stability index values (LQP) and such values are arranged in descending order. The maximum
value of index indicates most critical line for outage. Line outage contingency screening and ranking is
carried out on the test system and the results are shown in table 3. It is clear from the results that
outage of line number 5 is the most critical line outage and this condition is considered for voltage
stability improvement. Outage of other lines has no much impact on the system and therefore they are
not given importance.
Table 3. Contingency ranking
Rank Line No. LQP Values
1 5 0.9495
2 9 0.6050
3 2 0.4993
4 4 0.4968
5 7 0.4693
6 6 0.3965
7 10 0.3960
8 15 0.3943
9 3 0.3940
10 11 0.3917
Load flow is run on the system with line 5 outaged. Outage of this line results in large real power loss
and voltage profile reduction in most of the load buses. The system is under stressed conditions and
needs to be relieved by some means. Installation of FACTS devices at suitable locations can relive the
system much from stressed conditions (reduced line losses).

380 [email protected]
 Vol 19, No. 10;Oct 2012

FACTS devices help the system to maintain acceptable voltage profile in the load buses. The voltage
profile values for both combinations of FACTS devices are compared graphically in figure 7(a) under
case 1. Obviously from the figure the voltage stability improvement in SSSC with SVC is highly
encouraged. In case 2 the voltage stability limit is remarkable in majority of buses through TCSC with
SVC insertion as compared with the other. But the voltage profile improvement is considerably in
remote buses through the SSSC with SVC insertion is shown in figure 7(b).

Without FACTS With TCSC and SVC With SSSC and SVC
1.09
Voltage in p.u

1.07
1.05
1.03
1.01
0.99
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Bus Number
Fig. 7(a): Voltage profile under case 1

Without FACTS With TCSC and SVC With SSSC and SVC
1.08
1.05
Voltage in p.u

1.02
0.99
0.96
0.93
0.9
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Bus Number
Fig. 7(b): Voltage profile under case 2

Without FACTS With TCSC and SVC With SSSC and SVC
1.1
Voltage in p.u

1.07

1.04

1.01

0.98

0.95
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Bus Number
Fig. 7(c): Voltage profile under case 3
In case 3 the SSSC with SVC based voltage profile dominates in majority of the buses. The voltage
stability limit is superior in this case as compare with the other was exposed in figure 7(c). From the

381 [email protected]
 Vol 19, No. 10;Oct 2012

voltage stability limit improvement point of view, we conclude that the SSSC with SVC devices are
more suitable than TCSC with SVC devices for voltage stability improvement in load buses.
Reduction in reactive power loss indicates that power flow through the heavily loaded lines are
diverted through the under loaded lines and the result is improved voltage profile. In loss minimization
point of view the real power loss reduced by 0.098 MW and 0.758 MW through insertion of TCSC
with SVC and SSSC with SVC respectively compare with absence of FACTS devices under normal
conditions. The real power loss under critical loading is decreased by 1.262 MW and 5.644 MW
through insertion of TCSC with SVC and SSSC with SVC respectively compare with absence of
FACTS devices. Similarly the real power loss decreased by 0.629 MW and 9.252 MW through
insertion of TCSC with SVC and SSSC with SVC respectively compare with absence of FACTS
devices under single line outage contingency conditions. The real power losses under all cases are
shown in table 4.
Table 4. Real power loss values in MW

FACTS Devices Case 1 Case 2 Case 3

Without FACTS 17.514 46.900 32.569
TCSC with SVC 17.416 45.638 31.940
SSSC with SVC 16.756 41.256 23.317

The percentage level of real power loss reduction through insertion of FACTS devices in all three
cases is graphically depicted in figure 8. It is very clear from the real power loss minimization results
the compensation is highly encouraging by SSSC with SVC devices over TCSC with SVC devices.

TCSC with SVC SSSC with SVC

30
28.407
Percentage of Loss

25
20
15 12.034
10
4.327 2.69
5 0.559 1.931
0
Case 1 Case 2 Case 3

Fig. 8: Percentage of Real Power Loss Reduction while using FACTS devices
From Table 5 the most suitable location for TCSC to control power flow is found to be line number 18
for normal loading. For critical loading it is found to be line number 5. In line outage contingency
conditions the location found to be line number 28. Similarly the optimal location of SVC to improve
voltage profile is found to be bus number 7 for both normal loading and single line outage contingency
condition. In critical loading it is found to be bus number 17. The size of TCSC (in degrees of
compensation) with SVC under all cases are also mentioned this table.
From Table 6 the optimal location for SSSC to control power flow is found to be line number 6 for
both normal loading and single line outage contingency conditions. For critical loading it is found to
be line number 5. Similarly the optimal location of SVC to improve voltage profile is found to be bus
number 20 for normal loading and bus number 30 for critical loading. In single line outage

382 [email protected]
 Vol 19, No. 10;Oct 2012

contingency condition it is found to be bus number 25. The size of SSSC (in impedance values of
compensation) with SVC under all cases are also mentioned this table.
Table 5. Optimal parameters of TCSC and SVC combination

FACTS Case 1 Case 2 Case 3

Devices Location Size Location Size Location Size
Between Between Between
buses buses buses
TCSC 0. 1539 -0.2414 -0.4735
12 and 15 2 and 5 10 and 22
(Line No.18) (Line No.5) (Line No 28)
Bus Bus Bus
SVC 15.9749 13.7207 16.9132
No. 7 No. 17 No. 7
Table 6. Optimal parameters of SSSC and SVC combination

Case 1 Case 2 Case 3

FACTS
Devices Size Size Size
Location Location Location
[R+jX] [R+jX] [R+jX]
Between Between Between
0.0239 0.0289 0.0112
buses buses buses
SSSC + + +
2 and 6 2 and 5 2 and 6
j0.2664 j0.0931 j0.0478
(Line No. 6) (Line No. 5) (Line No. 6)
Bus Bus Bus
SVC 5.8131 6.2554 5.8559
No. 20 No. 30 No. 25

Generally DE algorithm helps to reduce the real power loss and increase in voltage magnitudes after
the insertion of FACTS devices proves that SSSC with SVC combination is highly efficient in
relieving a power network from stressed condition and improving voltage stability limit over TCSC
with SVC combination.

CONCLUSION

In this paper, comparison between two different combinations of FACTS devices for voltage stability
limit enhancement, real power loss minimization and optimal location are demonstrated through
Differential Evolution algorithm. The voltage stability limit improvement and real power loss
minimization are done under normal, critical loading and line outage contingency conditions with DE
algorithm implementation. The LQP index is used for voltage stability assessment. The circuit
element model of TCSC and SSSC are considered to improve the voltage stability limit by controlling
power flows and maintaining voltage profile. This model is easy to incorporate the effect of TCSC
and SSSC into Newton - Raphson load flow program coding. The performance of TCSC with SVC
and SSSC with SVC combination in optimal power flow control for voltage stability limit
improvement is proved in the results by comparing the system real power loss and voltage profile with
Differential Evolution algorithm. It is clear from the numerical results that voltage stability limit
improvement and real power loss minimization are highly encouraging in SSSC with SVC insertion
compare with TCSC with SVC insertion under all cases.

383 [email protected]
 Vol 19, No. 10;Oct 2012

References
Canizares. C.A and Faur. Z.T (1999). Analysis of SVC and TCSC Controllers in Voltage
Collapse. IEEE Transactions on power systems, 14, 1,158-165.
Dobson,I and Chiang.H.D (1989). Towards a theory of voltage collapse in electric power systems,
Systems and Control Letters, 13, 253-262.
Ejebe.G.C, Irisarri.G.D, Mokhtari.S, Obadina.O, Ristanovic.P and Tong.J (1995). Methods for
contingency screening and ranking for voltage stability analysis of power systems. Proceedings of
IEEE Power Industry Computer Application Conference, Salt Lake City, UT, 249-255.
Hingorani.N.G and Gyugyi.L (2000). Understanding FACTS: Concepts and Technology of Flexible
ACTransmission Systems, (2nd Ed) New York, Wiley Inter science, IEEE Press, (Chapter 5 and 6).
Jafari.M, and Afsharnia.S (2007). Voltage Stability Enhancement in Contingency Conditions using
Shunt FACTS Devices. EUROCON – IEEE international conference on computer as a tool, Warsaw,
1660-1665.
Kundur.P (1994). Power System stability and control, (2nd Ed.) New York, McGraw-Hill, (Chapter
14).
Motie birjandi,A.A and Sabzawari.K (2010). The modeling of UPFC based on Circuit Elements in an
Exact Transmission Line Model, International journal of Engineering. 4,2,105-118.
Musunuri, S and Dehnavi, G (2010). Comparison of STATCOM, SVC, TCSC, and SSSC Performance
in Steady State Voltage Stability Improvement. North American Power Symposium (NAPS), 1-7.
Price.K and Storn.R (1995), Differential evolution – A simple and efficient adaptive scheme for global
optimization over continuous spaces, Technical Report, International Computer Science Institute,
Berkley.
Reis.C, Andrade.A and Maciel.F.P (2009). Line Stability Indices for Voltage Collapse Prediction,
IEEE Power Engineering conference, Lisbon, 239-243.
Reppen.N.D, Austria.R.R, Uhrin.J.A, Patel.M.C and Galatic.A (1993). Performance of methods for
ranking a evaluation of voltage collapse contingencies applied to a large-scale network. Athens Power
Tech proceedings: Joint international power conference Athens, Greece, 337-343.
Sakthivel.S and Mary.D (2011). Voltage Stability Limit Improvement Incorporating SSSC and SVC
under Line Outage Contingency Condition by Loss Minimization, European Journal of Scientific
research,59,1, 44-54 .
Sode-Yome.A, Mithulanathan.N and Lee.K.Y (2001). Static Voltage Stability Margin Enhancement
Using STATCOM, TCSC and SSSC. Proceedings of the IEEE/PES Transmission and Distribution
Conference and Exhibition, Asia and Pacific, 1-6 .
Technical Analysis of the August 14, 2003, Blackout: What Happened, Why, and What Did We Learn?
A report by the North American Electrical Reliability Council Steering Group, July 13, 2004.
Vaahedi.E, Fuchs.V, Xu.W, Mansour.Y, Hamadanizadeh.H and Morison.G.K, (1999). Voltage
Stability Contingency Screening and Ranking. IEEE Transactions on power systems, 14, 1, 256 – 265.
Van Cutsem.T (2000). Voltage instability: Phenomena, countermeasures, and analysis methods.
Proceedings of the IEEE, 88, 208–227.
Zhang.Y and Zhang.Y (2006). A Novel Power Injection Model of Embedded SSSC with Multi-
Control Modes for Power Flow Analysis Inclusive of Practical Constraints. Electric Power System
Research, 76, 5, 374–381.

384 [email protected]