OPTIMAL ALLOCATION OF FACTS DEVICES IN

POWER SYSTEM USING GENETIC ALGORITHMS

M.M.El Metwally1, A.A.El Emary2, F.M.El Bendary3, and M.I.Mosaad4
1, 2

Electrical Power and Machines Dept. Faculty of Engineering, Cairo University, Egypt.
Electrical Dept. Faculty of Engineering, Shoubra University, Cairo, Egypt.
4
Electrical and Computer Dept. Higher Technological Institute (HTI), Tenth of Ramadan City, Egypt.
3

Abstract This paper presents evolutionary technique to seek

Newtons method, Linear programming method and Interior

point method [7].
Different heuristic methods have been also used to handle the
OPF problem. Network flow algorithm with tie line constrains
was proposed by [8]. Adaptive Hopfield Neural Network has
been applied [9].GAs with fuzzy logic controllers to adjust its
crossover and mutation probabilities are also used [10].
In this paper, the optimal location of SVC in power systems
using GAs is introduced. To test the capability of this method,
the application is made on IEEE14-bus system.

the optimal location of FACTS devices in a power system.

FACTS controllers in the form of SVC type are introduced. This
evolutionary technique is Genetic Algorithms (GAs). Provided
optimal locations, FACTS devices can be used to achieve the
optimal power flow (OPF) without any constraint violation and
thus to increase the utilization of the lowest cost generation in
power system. FACTS location should be reasonably chosen
according to their contribution to the general objective of power
system economic generation and dispatch. The objective cost
function, which consists of the investment costs for this type of
FACTS devices and the generation costs, is minimized.
Numerical examples will be indicated to obtain the optimal
allocation of SVC.

II-PROBLEM FORMULATION
The mathematical formulation of the optimal allocation of
FACTS devices can be expressed as cost function and
constrains.
A-Cost function
The generating cost function has been approximated as a
quadratic function in $/Hour given by:
(1)
f 1 ( P) = aP 2 + bP + c

Index Terms Genetic Algorithms, FACTS, Optimal Location,

Optimal Power Flow and SVC.

I-INTRODUCTION
ith the rapid development of power electronics, FACTS
devices have been proposed and implemented in power
systems. FACTS devices can be used in the operation,
control power systems and enhance system stability. FACTS
controllers such as SVC, TCSC, STATCOM, SSSC and UPFC
are a family of electronic controllers used to enhance power
system performance [1], [2]. These devices are characterized
by fast response, wide operating range and high reliability [3].
Unlike traditional shunt reactive elements, FACTS are able to
rapidly and smoothly supply or absorb reactive power by
controlling the firing delay angles of thyristors (valves). They
have been used in transmission and distribution systems for
voltage and reactive power control [4], [5] .SVC uses power
electronics to control its reactive power output to regulate bus
voltage. Compared to mechanically switched capacitor banks,
SVC reacts very fast and has high reliability. In this paper
,SVC devices optimal location can be used as a proposed
method based on GAs to achieve the optimal power flow and
thus to increase the utilization of the lowest cost generation in
power systems and achieve voltage support.
The Optimal Power Flow problem (OPF) has been discussed
since its introduction by Carpentier [6]. Because the OPF is a
very large, non-linear mathematical programming problem, it
has taken decades to develop efficient algorithms for its
solution. Many different mathematical techniques have been
employed for this purpose. The majority of the techniques
discussed in the literature use one of the following five
methods, Lambda iteration method, Gradient method,

978-1-4244-1933-3/08/$25.00 2008 IEEE

Where P is a vector that represents the active power outputs

of the generators in (MW) a, b and c are the generating cost
coefficients in $/MW2 per hr, $/MW per hr and $ per hr
respectively are given in appendix A.
B-SVC cost function
The SVC cost functions in $/KVar based on [11] is given by:
(2)
f 2 = 0.0003 S 2 0.3051S + 127 .38
Where S is the operating range of the SVC devices in MVar.
The overall cost function F consists of generation costs and
FACTS devices investment costs.
The objective function is to be minimized F is given as
(3)
Min(F) = Min ( f1 ( P) + f 2 ( g ) )
Where g is a vector that represents the variables of FACTS
devices.
Subjected to the following constrains:

E ( g , x) = 0

B1 ( g ), B2 ( x ) 0

(4)

Where
; represents the conventional power flow
equations
B1 ( g ), B2 ( x ) 0
; are the inequality constrains for FACTS
devices and the optimal power flow
respectively.
x
; represents the operating states of the power
system.

E (g , x)

312

SVC Modeling
The unit for generation cost f1 is $/Hour and for the
investment costs of FACTS devices f2 are $. They must be
unified into $/Hour. Normally, the FACTS devices will be in
service for many years [12], [13]. However, only a part of its
life time will be employed to regulate the power flow. In this
paper, four years is applied to evaluate the average cost
function for SVC devices. This value has been applied only to
unify the unit and has no influence on the global optimization.
Therefore the average value of the investment costs function
of SVC will be:
f2 (g ) =

f2 (g )
4 8760

The SVC consists of a TCR in parallel with a bank of

capacitors. From an operational point of view, the SVC
behaves like a shunt-connected variable reactance, which
either generates or absorbs reactive power in order to regulate
the voltage magnitude at the point of connection to the AC
network. It is used extensively to provide fast reactive power
and voltage regulation support. The firing angle control of the
thyristor enables the SVC to have almost instantaneous speed
of response. A schematic diagram of the SVC is shown in
Figure 1.
The TCR whose reactive impedance is XL, with a bidirectional thyristor valves.
The controllable reactance of the TCR part is Xv, [11], which
is defined by:

(5)
III- FACTS DEVICES

According to IEEE, FACTS, is defined as follows [12]:

Alternating current transmission systems incorporating power
electronics based on other static controllers to enhance
controllability and power transfer capability. Within the basic
system security guidelines, the FACTS devices enable the
transmission system to obtain one or more of the following
benefits [12, 13]:
1. Control of power flow as ordered. This is the main function
of FACTS devices. The use of power flow control may be to
follow a contract, meet the utilities own needs, ensure
optimum power flow, ride through emergency conditions, or
a combination of them.
2. Increase utilization of lowest cost generation.
3. Dynamic stability enhancement. This additional function of
FACTS includes the transient stability improvement, power
oscillation damping and voltage stability control.
4. Increase the loading capability of lines to their thermal
capabilities, including short term and seasonal demands.
5. Reduce reactive power flows, thus allowing the lines to
carry more active power.

XV = X L

(6)
2 2 + sin(2 )
where is the firing angle of the thyristor.
The effective susceptance B of the FC-TCR is given by:
X + XC
B= V
(7)
XV X C
where XC is the capacitive impedance of the fixed capacitor.
Xc

XL
Figure 1. Schematic diagram of FC-TCR

According to the operation principles of FACTS controllers

there are mainly two categories of them [2, 3]:
A-Conventional Thyristor-Controlled Power System Controllers
This type of FACTS devices consists of conventional
capacitor or reactor banks and thyristors. They control on-off
periods of the fixed capacitors and reactors and, thereby
varying the capacitive and inductive effective impedance.

IV-OPTIMIZATION ALGORITHMS
In recent years, optimization algorithms have received
increasing attention by the research community as well as the
industry. In conventional OPF algorithms, the objective is to
minimize the overall generation function, which includes
generation cost of power plants and investment cost of SVC
devices. For a certain network the OPF identifies the various
quantities as parameters, control variables and state variables.
The most common control variables are the generator power
and voltages.
The optimization algorithms simulate an evolutionary process
where the goal is to evolve solutions by means of crossover,
mutation, and selection based on their quality (fitness) with
respect to the optimization problem of interest. Evolutionary
algorithms are highly relevant for industrial applications,
because they are capable of handling problems with non-linear
constraints, multiple objectives, and dynamic components that
frequently appear in real-world problems.

B-Synchronous Voltage Source (SVS) Controllers

This type of FACTS devices can generate capacitive or
inductive reactive power without fixed capacitor or reactor
banks. The SVS is implemented by multi-phase inverter using
a power semiconductor device with switch-off capability (like
IGBT or GTO thyristors). The SVS can generate internally the
reactive power necessary for compensation and with an energy
storage device can also negotiate real power with the system
for a limited amount of time.
Among of theses types of FACTS controllers, SVC is selected
from the Conventional Thyristor-Controlled Power System
Controllers in this search.

313

In this area of computation, heuristic methods may be used to

solve combinatorial optimization problems. These heuristic
methods are called intelligent, because the movement from
one solution to another is done using rules close to the human
reasoning .The most important advantage of heuristic methods
lies in the fact that they are not limited by restrictive
assumptions about the search space like continuity, existence
of derivative of objective function, etc.
Several heuristic methods exist, such as, Tabu Search method
(TS) [11], Simulated Annealing (SA) [12], Genetic
Algorithms (GAs) [14], and Particle Swarm Optimization
(PSO) algorithms [16].In this paper, a proposed method based
on GAs to achieve optimal allocation of FACTS has been
used.

SVC. It can take discrete values contained between 0 and 1; 0

corresponding to the minimum value that the device can take
and 1 to the maximum. According to the model of the SVC,
the real value of the device Vreal is calculated with the
relation:
(8)
Vreal = Vrmin + (Vmax - Vmin ) * V
Where Vmin and Vmax are the minimum and maximum setting
value of the SVC, and V is its normalized value.
The rated values of SVC are between -100 MVar and 100
MVar [10].
The creation of an individual is done in the following stages.
First, a set of buses of the network that can locate SVC at it, in
the first string. The second step consists of the attribution of
the characteristics of the device. Setting values of SVC are
finally randomly chosen among the possible values. Then, the
objective function is computed for every individual of the
population. It has to be elaborated so as to favor the
reproduction of good individuals without preventing
reproduction of interesting others.
The move to a new generation is done from the results
obtained for the old generation according to the values of the
objective function of it. After that, the operators of
reproduction, crossover and mutation are applied successively
to generate the offsprings. These three operations are repeated
until the number of desired offsprings is created. The objective
function is then calculated for every offsprings and the best
individuals among the entire pool, comprising parents and
their offsprings, are kept to constitute the new generation. By
this way, the objective function of the best individual of the
new generation will be the same or higher than the objective
function.

Genetic Algorithms
Based on the mechanisms of natural selection and genetics,
genetic algorithms are global search techniques. They can
search several possible solutions simultaneously and they do
not require any prior knowledge or special properties of the
objective function [14]. Moreover, they always produce high
quality solutions and, therefore, they are excellent methods for
searching optimal solution in a complex problem. GAs starts
with random generation of initial population which represent
possible solutions of the problem. Then the fitness of each
individual is evaluated and new populations are generated by
genetic operators (Reproduction, Crossover and Mutation)
until the maximal number of generation is reached. Figure.4
depicts the Outline of GAs for optimization problems.
The GAs differ from other optimization and search procedures
in four ways:
1- GAs work with a coding of the parameter set, not the
parameters themselves. Therefore GAs can easily handle the
integer or discrete variables.
2- GAs search from a population of points, not a single point.
Therefore GAs can provide globally optimal solutions.
3- GAs use only objective function information, not
derivatives or other auxiliary knowledge. Therefore GAs
can deal with the non-smooth, non-continuous and nondifferentiable functions which are actually existed in a
practical optimization problem.
4- GAs use probabilistic transition rules, not deterministic
rules

VI-SIMULATION RESULTS
In this paper the main objective is to find the best locations of
SVC in a network using GAs. By re-dispatching of the load
flows in power systems, the total generation costs can be
minimized by optimal SVC allocation. Therefore, the
objective is to minimize the overall cost function, which
includes the generation costs of power plants and the
investment costs of SVC devices.
In these proposed methods, Newton Raphson method has been
used for power flow. GAs is used for handling the
optimization problem. The software was written in MATLAB
6.0 package.
In order to verify the effectiveness of the proposed method,
IEEE 14-bus test system is used. After the optimization, for
the considered power system, SVC at bus 5 is the best
location. For implementing the GA in this problem, population
size of 100 was taken and the maximum number of
generations was taken as 100. Table 1 depicts the real power
supplied from each supply.

V-DEVELOPMENT OF THE PROPOSED METHOD

In this paper, the process of determining the optimal allocation
and ratings of SVC is introduced using GAs. They are
developed to obtain efficiently a high quality solution within
practical power system operation.
The goal of the optimization is to find the best location of
SVC devices in the power systems. The configuration of SVC
is defined with two parameters, the location and rated values.
In order to take into account the two aforementioned
parameters in the optimization, a particular coding is
developed. An individual is represented with two strings. The
first string corresponds to the location of SVC. It contains the
numbers of buses where the SVC are to be located. The
second string of the individual represents the values of the
314

[8] Streiffert Dan. Multiple Area Economic Dispatch with line constrains
IEEE Transactions on Power Systems, 1995,10(4),pp.1946-1951.
[9] Lee KY,Yome AS,Park JH Adaptive Hopfield Neural Network For
Economic Load Dispatch IEEE Transactions on Power Systems,
1998,13(2),pp.519-526.
[10] Song YH, Wang PV ,Johns AT. Environmental Economic dispatch
using fuzzy logic controller Genetic Algorithms IEE Proc. Gener.
Transm. Distr. 1997 ;144(4),pp.377-382
[11] K. Habur, and D. Oleary, "FACTS - Flexible AC Transmission Systems,
for cost effective and reliable transmission of electrical energy,
"http://www.siemenstd.com/TransSys/pdf/CostEffective ReliabTrans.pdf.
[12] N.G. Hingorani, L. Gyugyi, Understanding FACTS Concepts and
technology of Flexible AC Transmission Systems, IEEE Press, 2000.
ISBN 0-7803-3455-8.
[13] Y. H. Song, A. T. Johns. Flexible AC Transmission Systems (FACTS),
IEE Press, London, 1999. ISBN 0-85296-771-3.
[14] S. Gerbex, R. Cherkaoui, and A. J. Germond, "Optimal Location Of
Multi-Type FACTS Devices In A Power System By Means Of Genetic
Algorithms," IEEE Transaction Power Systems, vol. 16, August. 2001,
pp. 537-544.
[15] M.S.Osman, M.A.A-Sinna ,A.A.Mousa A Solution To The Optimal
Power Flow Using Genetic Algorithm MEPCON2003 Power System,
pp425-432.
[16] J. Kennedy and R. Eberhart, Particle swarm optimization, Proc. IEEE
Int. Conf. Neural Networks, vol. IV, pp. 19421948, 1995.

Table 1
Simulation results of economic dispatch with GAs for IEEE 14 bus system
with SVC at bus 5.
Generating Unit

GAs
Pg (MW)

G1

179.9637

G2

45

G3
Fuel Cost $/hr

41.7851
1064

The rate of the SVC placed at this bus is 51.66 MVar.

VII-CONCLUSIONS
In large power systems, the selection of proper location for
FACTS devices is the first and important step in designing
FACTS controllers. Based on the FACTS locations, the design
and coordination of their controllers can be carried out.
FACTS devices includes improvement of system stability,
enhancement of system reliability and reduction of operation
and transmission investment cost. In practical operation, the
main concern for the allocation of FACTS devices is the
economic effect. For instance, to increase the power transfer
capability and to reduce the generation costs. The proposed
GA is based on the economic criterion and it is a general
method for allocating of SVC from FACTS devices in large
power systems.
APPENDIX A
Generation Cost Function
Generator

0.0060

60

100

0.0050

50

100

300

500

1.000

0.020

15

100

0.030

45

100

REFERENCES
[1] IEEE FACTS working group 15.05.15, FACTS Application, December
1995.
[2] C. A. Canizares and Z. T. Faur, Analysis of SVC and TCSC Controllers
in Voltage Collapse, IEEE Trans. on Power Systems, vol. 14, no. 1,
February 1999, pp. 158165.
[3] L.Gyugui, Dynamic Compensation of AC Transmission Lines by SolidState Synchronous Voltage Converters IEEE Trans. on Power Delivery
, vol. 9, no. 2, pp.904-911 ,1994.
[4] J.H.Chen,W.J.Lee, and M.S.Chen Using a Static Var Compensator to
Balance a Distribution System IEEE Trans. on Industry Applications,
vol. 35, no. 2, pp. 298-304 ,1999.
[5] R.C. Pires Unbalanced Phase-to-Phase Voltage Compensators
Applied to Radial Distribution Feeders IEEE Trans. on Power
Delivery, vol. 19, no. 2, pp.806-812,2004
[6] J. Carpienter, Contribution a letude du Dispatching Economique,
Bulletin Society Franaise Electriciens, Vol. 3, August 1962.
[7] A. J. Wood and B. F. Wollenberg, Power Generation Operation and
Control,New York, NY: John Wiley & Sons, Inc., 1996, pp. 39-517.

315