Electrical Power and Energy Systems 78 (2016) 158–164

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

journal homepage: www.elsevier.com/locate/ijepes

Swarm intelligence based algorithms for reactive power planning with

Flexible AC transmission system devices
Biplab Bhattacharyya ⇑, Saurav Raj
Dept. of Electrical Engineering, Indian School of Mines, Dhanbad, Jharkhand, India

a r t i c l e i n f o a b s t r a c t

Article history: In the proposed work, authors have applied swarm intelligence based algorithms for the effective
Received 29 August 2015 Co-ordination of Flexible AC transmission system (FACTS) devices with other existing Var sources present
Received in revised form 18 November 2015 in the network. IEEE 30 and IEEE 57 bus systems are taken as standard test systems. SPSO (Simple Particle
Accepted 25 November 2015
Swarm Optimization) and other two swarm based intelligence approaches like APSO (Adaptive Particle
Swarm Optimization) and EPSO (Evolutionary Particle Swarm Optimization) are used for the optimal
setting of the Var sources and FACTS devices. The result obtained with the proposed approach is
Keywords:
compared with the result found by the conventional RPP (Reactive power planning) approach where
Swarm intelligence based algorithms
Reactive power planning
shunt capacitors, transformer tap setting arrangements and reactive generations of generators are used
FACTS devices as planning variables. It is observed that reactive power planning with FACTS devices yields much better
Active power loss result in terms of reducing active power loss and total operating cost of the system even considering the
Operating cost investment costs of FACTS devices.
Ó 2015 Elsevier Ltd. All rights reserved.

Introduction
Effective planning of reactive power sources has significant
effect in secure and economic operation of a power system. In
the present day scenario, growing power demand, restriction in
construction of new lines, unscheduled power flow in lines cause
congestion in the transmission network and increases transmission
loss.
Reactive power planning problems are solved by optimal alloca-
tion of Var sources at definite locations. In optimal Var planning,
first the weak nodes or buses are identified as capacitor placement
positions, then compensation to the entire power network is pro-
vided by proper co-ordination of other Var sources (i.e., trans-
former tap setting arrangements, generators) with the capacitors
placed at the weak nodes of the system. It is observed that by
proper planning of Var sources, voltage profile improves, active
power loss reduces and system security enhances. There are
numerous researches on RPP using both classical and Evolutionary
approaches.
Optimal planning of Var sources using linear programming is
discussed in [1]. Solution of reactive power problem by optimal
placement of capacitor is presented in [2]. Hybrid expert system
⇑ Corresponding author.
E-mail addresses: biplabdgp1@rediffmail.com (B. Bhattacharyya), sauravsonusahu@
gmail.com (S. Raj).
and simulated annealing technique method for optimal Var plan-
ning is presented in [3]. In [4], weak buses are detected first and
Var support is provided at the buses to improve system security.
Reactive power planning problem based on heuristics is presented
in [5]. Binary search technique and special heuristics are applied to
solve reactive power planning problem for a large scale electrical
power network in [6]. A method for the determination of reactive
power margin is presented in [7] for optimal Var planning. Evolu-
tionary programming techniques are applied to solve RPP (Reactive
power planning) problem in [8]. Comparative analysis of some evo-
lutionary algorithms on reactive power planning for IEEE 30 bus
system is presented in [9]. Simulated Annealing based algorithm
for the optimal placement of capacitors in a connected power net-
work is presented by the authors in [10].
Chance constrained programming method is used in [11] for
reactive power planning. A simulated annealing based optimiza-
tion algorithm is developed in [12] for active and reactive power
dispatch of a connected power network using transformer taps
and capacitor banks. Operation strategy for the improvement of
voltage profile and simultaneous reduction of active power loss
is described in [13]. PSO (Particle swarm optimization) based
optimization algorithm is developed for optimal reactive power
planning considering static voltage stability in [14]. Dynamic PSO
based optimization technique is applied for minimization of real
power loss in [15]. An AC model of transmission expansion plan-
ning is presented in [16] by optimal planning of reactive power

http://dx.doi.org/10.1016/j.ijepes.2015.11.086
0142-0615/Ó 2015 Elsevier Ltd. All rights reserved.
 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164 159

sources. An iterative optimization algorithm is developed in [17] C Total ¼ C E þ C Sh ð2Þ

for optimal Var planning. An hybrid heuristic method is applied
where
for the ORPF (Optimal reactive power problem) in the expansion
of a transmission network in [18]. C FACTS ¼ C TCSC þ C SVC ð3Þ
With the advent of FACTS (Flexible AC transmission system)
devices a new horizon is created for the researchers in RPP issue. where CFACTS is the cost due to FACTS devices and CTCSC & CSVC is the
FACTS can provide benefits in increasing transmission capacity, cost due to TCSC and SVC respectively.
flexible power flow control, improvement of steady state and Again,
dynamic stability and significant amount of transmission loss min-
imization. Flexible AC transmission devices are modeled and their C TCSC ¼ 0:0015ðTCSCvalue Þ2  0:7130ðTCSCvalue Þ
optimal allocation strategy is presented in [20]. Authors in [21] þ 153:75ðUS $=KVARÞ ð4Þ
have shown the effects of FACTS devices in the power flow of the
interconnected power system. Authors in [22] have presented a and
guideline for the choice and allocation of FACTS devices in a large
scale power system. A case of congestion relief by the proper place- C SVC ¼ 0:0003ðSVCvalue Þ2  0:3051ðSVCvalue Þ
ment of TCSC and improvement of power transfer capacity is þ 127:38ðUS $=KVARÞ ð5Þ
addressed in [23]. TCSC location is determined by evolutionary
programming in [24] under single line contingency. Authors have Source of Eqs. (4) and (5) is [22].
shown how system loadability can be increased by optimal setting Cost part CE, due to energy loss arises from the transmission loss
of TCSC using Genetic Algorithm in [25]. Placement of multi type is given as:
FACTS devices by GA and improvement of transfer capacity is dis-
X
n h i
cussed in [26]. Authors have presented OPF model with FACTS PLoss ¼ g k V 2i þ V 2j  2V i  V j  Cosðdi  dj Þ ð6Þ
devices in [27]. Differential evolution based optimization algo- k¼1
rithm is developed in [28] where authors have shown the impact
of FACTS devices in optimal power flow. Loss sensitivity of buses where gk is the conductance of the kth line connected between ith
is calculated in [29] by the authors to recognize weak buses for and jth bus of the power system. Vi and Vj are the voltage magnitude
capacitor Var injection, then evolutionary algorithms are devel- and di and dj are the voltage phase angle of the ith and jth bus
oped for optimal Var settings by the generators, capacitors and respectively. And n is the total number of lines.
transformer taps. The following constraints are to be satisfied while minimizing
In the present work, authors have used FACTS devices for solv- the objective function for the optimal planning of reactive power
ing RPP problem, where the main objective is to reduce the active sources as given in Eq. (1):
power loss and operating cost of the system. Here two types of Voltage magnitude constraints:
FACTS devices are used; first one is SVC (Static Var Compensators)
V min
i 6 V i 6 V max
i ð7Þ
and other type is TCSC (Thyristor controlled Switched Capacitors).
Reactive generation limit of the generator’s:
Proposed approach
Q min
gi 6 Q gi 6 Q max
gi ð8Þ
In the present work, SPSO (Simple Particle Swarm Optimiza- Transformer tap setting arrangements:
tion) and other two swarm based intelligence approaches like
APSO (Adaptive Particle Swarm Optimization) and EPSO (Evolu- T min
i 6 T i 6 T max
i ð9Þ
tionary Particle Swarm Optimization) are applied for reactive
power optimization. The main objective of the present work is to Var output of shunt capacitors:
minimize the overall operating cost of the system. In the present Q min max
Ci 6 Q Ci 6 Q Ci ð10Þ
work, firstly, the optimal planning of Var sources is done with
shunt capacitors, transformer tap settings and reactive power gen- Now for reactive power planning with FACTS devices the fol-
eration of the generators present in the system. Capacitor place- lowing two inequality constraints are to be satisfied by objective
ment positions are determined by weak node analysis. As the function represented by Eq. (2) in addition to the satisfaction of
settings of transformer tap positions and reactive generations of inequality constraints shown by Eqs. (7)–(9).
the generators within the specified limit are independent on the
system cost only, cost of shunt capacitors is to be considered only.
Q min max
SVCi 6 Q SVCi 6 Q SVCi ð11Þ
Hence, the objective function is to minimize the overall operating
and
cost has two parts. One is the cost due to energy loss attributed by
active power loss of the system and the other is the cost of the Q min max
TCSCi 6 Q TCSCi 6 Q TCSCi ð12Þ
shunt capacitors. The objective function is expressed as
where min and max are the minimum and maximum values of the
C Total ¼ C E þ C Sh ð1Þ
variables. The minimum and maximum limit of each variables are
where CE is cost due to energy loss (in $) and CSh is the cost of the given below:
shunt capacitor (in $). CE arises due to the overall transmission loss
in the system.
Installation cost of shunt capacitor = 3 ($/kVar). Fixed installed
cost of shunt capacitor = 1000 ($). Cost due to energy loss = 0.06 Transformer tap Reactive Generation of Shunt Capacitors
positions Generators
($/KW h). The objective function represented by Eq. (1) is calcu-
lated with the following data as obtained from [10]. T1, T2,…….,Tn Qg1, Qg2,…..Qgn Csh1, Csh2,….Cshn
Now for reactive power planning with FACTS devices, the objec-
tive function becomes, Fig. 1. String representing the control variables.
160 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164

Variables Minimum Maximum
value value
(p.u) (p.u)
Voltage (V) 0.95 1.05
Transformer tap setting (T) for IEEE 0.9 1.1
30 bus system
Transformer tap setting (T) for IEEE 0.85 0.95
57 bus system
Shunt capacitor (Qc) 0.0 0.15
Static Var compensators (QSVC) 0.0 0.5
Thyristor controlled switched 0.0 0.6
capacitors (QTCSC)
Q min and Q max Values of each generator are specified in the standard
g g
IEEE-30 and IEEE-57 bus system data. The control variables that are
to be optimized for reactive power planning without FACTS devices
is shown in Fig. 1.
And the control variables for reactive power planning with
FACTS devices are represented by Fig. 2.
The control variables that are represented by a string will be
used as a solution vector for the objective function in PSO and
PSO based algorithms. Initially a population vector is generated
with these control variables and objective function is evaluated
with these control variables. Minimization of active power loss is
highly dependent upon the voltage profile of the entire power sys-
tem, minimum and maximum limit of the voltage at a particular
bus is checked at each time while evaluating the objective function.
Determination of locations of capacitors positions and FACTS devices
At first, RPP problem solved by optimal-co-ordination of capac-
itors and other existing Var sources present in the network. The
capacitor placement positions are determined by weak bus detec-
tion or modal analysis technique applied in [4]. The capacitor
placement positions determined for the IEEE 30 bus test system
are bus nos. 7, 15, 17 and 21. Similarly 25th, 38th and 49th bus
of the IEEE 57 bus system are found as weak buses and therefore
identified as candidate locations for shunt capacitor placement.
For the RPP problem with FACTS devices, SVC (Static Var com-
pensator) positions are kept same as those of the shunt capacitors
positions used in the RPP problem without FACTS devices.
To determine TCSC (Thyristor controlled Switched Capacitors)
placement positions, reactive power flow is calculated for all the
lines and the lines those carry significant amount of reactive power
are selected for TCSC positions. Placement of Thyristor controlled
capacitors at these lines will reduce line reactances of the lines and
have direct impact in reducing reactive power flow through the lines.
Lines 5, 25, 28 and 41 are selected for TCSC placement in case of
IEEE 30 bus test system and 13th, 37th, 57th and 61st lines are
selected for the placement of Thyristor controlled capacitors in
case of IEEE 57 bus test system.
Different techniques in brief
PSO approach
This technique is originally given by two researchers named as
James Kennedy and Russell Eberhart [19] and technique was first
proposed in 1995. Each particle updates its velocity using the fol-
lowing equation:
v kþ1
i ¼ xv ki þ C 1 rand  ðpbest i
 Ski Þ þ C 2 rand  ðg best  Ski Þ ð13Þ
where
v ki ? current velocity of agent i at iteration k,
rand ? is the random number between 0 and 1,
Ski ? current position of agent i at iteration k,
Ci ? weight coefficient for each term,
P besti ? best agent or the best solution vector in the population
for the current generation.
gbest ? group best or the best solution vector considering all the
generations up to the present generation
x ? weight function for velocity of agent i.
where x, the weight factor is updated by the following equation
at each iteration
xmax  xmin
x ¼ xmax   iter
itermax
Here xmax ¼ 0:9, xmin ¼ 0:4, itermax ¼ 500 for IEEE 30 test system
and 1000 for IEEE 57 test system, and iter = current iteration, C1
and C2 are set to 2.0.
Initially strings are generated randomly and each string may be
a potential solution. In PSO, each potential solution, called particles
is assigned a velocity. The particles of the population always adjust
their velocity depending upon their positions with respect to the
position of the pbest (the particle having the best fitness in the cur-
rent generation) and the gbest (the particle having the best fitness
upto the present generation). While adjusting their velocities and
positions, particles adjust their fitness value as well. The particle
having the best fitness among all is selected as the pbest for the
current generation, and if this pbest has better fitness than the
gbest, it takes the position of the gbest as well. In PSO, therefore,
the gbest particle always improves its position and finds the opti-
mum solution and the rest of the population follows it. The string
length depends upon the problem and the control variables within
the string.
Adaptive PSO algorithm (APSO) in brief
In APSO, search trajectory is controlled by two parameters P1
and P2 such that
2 2
C2 ¼ ; C1 ¼  C2 ð14Þ
P1 P2
For initial searching points; P1 and P2 of each agent are set to 0.5
or higher. Then each agent may move close to the position of pbest
and gbest in the following iteration.
For evaluation of searching points: When the agent becomes
gbest, it is perturbed. The parameters P1 and P2 are adjusted to
0.5 or lower, so that the each agent may move away from the posi-
tion of pbest & gbest and the weighting factor can be modified as
 
fC 1 ðpbest  xÞ þ C 2 ðgbest  xÞg
w ¼ gbest  ð15Þ
2þx
In APSO, velocity of the improved PSO can be expressed as
V kþ1
i ¼ wi þ C 1 rand1  ðpbesti  Ski Þ þ C 2 rand2  ðgbest  Ski Þ ð16Þ

Transformer tap Reactive Static Var Thyristor controlled

positions Generation of Compensators (SVC) Switched Capacitors
Generators (TCSC)
T1, T2,…….,Tn Qg1, Qg2,…..,Qgn QSVC1, QSVC2, ….,QSVCn QTCSC1, QTCSC2,….,QTCSCn

Fig. 2. String representing the control variables with FACTS devices.

 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164 161

Table 1 Evolutionary Particle Swarm Optimization (EPSO) in brief

Optimum magnitudes of reactive power sources for IEEE 30 bus system.

Methods Reactive generation of Transformer tap Shunt The idea behind EPSO is to grant a PSO scheme with an explicit
generators (p.u) settings (p.u) capacitors (p. selection procedure and with self adapting properties for its
u) parameters
SPSO 0.1705 (2) 0.9 (11) 0.0330 (7) Position and velocity of an agent is modified in the following
0.2655 (5) 0.9 (12) 0.0527 (15) way:-
0.3028 (8) 0.9019 (15) 0.0 (17), 0.0
(21)
0.0121 (11) 0.9 (36) SKþ1
i ¼ SKi þ V Kþ1
i ð17Þ
0.2467 (13)

APSO 0.1629 (2) 0.9 (11) 0.0174 (7) V Kþ1
i ¼ wi0 V K þ wi1 ðpbesti  SKi Þ þ wi2 ðgbest  SKi Þ ð18Þ
0.2671 (5) 0.9133 (12) 0.0468 (15)
0.2964 (8) 0.9 (15) 0.0025 (17) Here, in EPSO weights as well as gbest undergo mutation in the fol-
0.0563 (11) 0.9010 (36) 0.0209 (21) lowing manner
0.1797 (13)
wiK ¼ wik þ sNð0; 1Þ ð19Þ
EPSO 0.1462 (2) 0.9026 (11) 0.0 (7)

0.2642 (5) 0.9 (12) 0.0 (15)
gbest ¼ gbest þ s0 Nð0; 1Þ ð20Þ
0.2977 (8) 0.9 (15) 0.0295 (17)
0.1061 (11) 0.9 (36) 0.0 (21)
where N(0, 1) is a random variable with Gaussian distribution. s and
0.2213 (13)
s0 are learning parameters.

Table 2 Result and discussion

Optimum magnitudes of the reactive power sources and FACTS devices for IEEE 30
bus system. Here SPSO (Simple Particle Swarm Optimization) and other two
Methods Reactive generation Transformer SVC TCSC PSO based optimization algorithms APSO (Adaptive Particle Swarm
of generators (p.u) tap settings (p. amount (p.u) Optimization) and EPSO (Evolutionary Particle Swarm Optimiza-
u) (p.u) tion) are applied for the planning of reactive power sources
SPSO_FACTS 0.6 (2) 0.9 (11) 0.0 (7) 0.1463
(25)
0 (5) 0.9 (12) 0.0 (15) 0.0419 Table 4
(41) Optimum magnitude of reactive power sources for IEEE 57 bus systems.
0 (8) 0.9 (15) 0.0 (17) 0.1049
Methods Reactive Transformer tap settings (p.u) Shunt
(28)
0.4 (11) 0.9223 (36) 0.0840 0.1388 generation of Capacitors
(21) (5) generators (p.u) (p.u)
0 (13) SPSO 0.4250 (2) 0.85 (19), 0.85 (20), 0.85 (41), 0.0 (49)
0.51 (3) 0.85 (80), 0.85 (65), 0.85 (54), 0.0 (25)
APSO_FACTS 0 (2) 0.9 (11) 0.0 (7) 0.1463 0.2125 (6) 0.85 (71), 0.85 (66), 0.85 (59), 0.0 (28)
(25) 0.5543 (8) 0.85 (58), 0.95 (31), 0.85 (35),
0 (5) 0.9501 (12) 0.0 (15) 0.0419 0.0765 (9) 0.85 (36), 0.95 (37), 0.85 (46),
(41) 1.3175 (12) 0.85 (76), 0.95 (73)
0 (8) 0.9180 (15) 0.0 (17) 0.1049
(28) APSO 0.4250 (2) 0.85 (19), 0.85 (20), 0.85 (41), 0.15 (49)
0.4 (11) 0.9330 (36) 0.0768 0.1388 0.51 (3) 0.8534 (80), 0.9273 (65), 0.85 0.0001 (25)
(21) (5) 0.2125 (6) (54), 0.8520 (71), 0.8855 (66), 0.0 (28)
0 (13) 1.1637 (8) 0.9116 (59), 0.9092 (58), 0.95
0.0 (9) (31), 0.8518 (35), 0.85 (36),
EPSO_FACTS 0.6 (2) 0.9439 (11) 0.0 (7) 0.1463 0.6944 (12) 0.9053 (37), 0.9022 (46), 0.85
(25) (76), 0.9150 (73)
0 (5) 0.9 (12) 0.0 (15) 0.0419
(41) EPSO 0.4250 (2) 0.85 (19), 0.95 (20), 0.85 (41), 0.15 (49)
0 (8) 0.9 (15) 0.0 (17) 0.1049 0.51 (3) 0.85 (80), 0.95 (65), 0.85 (54), 0.15 (25)
(28) 0.2125 (6) 0.95 (71), 0.95 (66), 0.95 (59), 0.0 (28)
0.4 (11) 0.9326 (36) 0.0 (21) 0.1368 0.3737 (8) 0.95 (58), 0.95 (31), 0.95 (35),
(5) 0.0765 (9) 0.95 (36), 0.8797 (37), 0.95 (46),
0 (13) 1.3175 (12) 0.95 (76), 0.95 (73)

Table 3
Active power loss and operating cost in IEEE 30 bus system for different methods.

Active power loss Operating cost due to energy Methods Active power loss Total operating cost Decrease in active Decrease in
without planning in p.u loss without planning in $ (B) after RPP in p.u (A1) after RPP in $ (B1) power loss in p.u operating cost in $
(A) (A1  A) (B1  B)
0.0711 3737016 SPSO 0.0684 3:5951  106 0.0027 1:41916  105
APSO 0.0684 3:5966  106 0.0027 1:40416  105
EPSO 0.0685 3:6000  106 0.0026 1:37016  105
SPSO_FACTS 0.0435 2:3622  106 0.0276 1:374816  106
APSO_FACTS 0.0434 2:3558  106 0.0277 1:381216  106
EPSO_FACTS 0.0438 2:3671  106 0.0273 1:369916  106
162 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164

Table 5
Optimum magnitude of reactive power sources and FACTS devices for IEEE 57 bus system.

Methods Reactive generation Transformer tap settings (p.u) SVC amount (p.u) TCSC (p.u)
of generators (p.u)
SPSO_FACTS 0.5 (2) 0.9 (19), 0.9 (20), 0.9 (41), 0.9 (80), 0.9 (65), 0.9 (54), 0.9 (71), 0.9 (66), 0.0 (49) 0.0331 (37)
0.6 (3) 0.9 (59), 0.9 (58), 1.0128 (31), 0.9 (35), 0.9 (36), 1.0203 (37), 0.9 (46), 0.0 (25) 0.0304 (13)
0.25 (6) 0.9 (76), 1.1 (73) 0.3945 (28) 0.0163 (61)
0.2 (8) 0.0410 (57)
0.09 (9)
0.0 (12)

APSO_FACTS 0.1213 (2) 0.9 (19), 0.9152 (20), 1.0892 (31), 0.9 (80), 0.9021 (41), 0.9456 (65), 0.0 (49) 0.0331 (37)
0.5754 (3) 0.9558 (54), 0.9 (66), 0.9274 (71), 0.9 (35), 0.9 (59), 0.9 (58), 0.9474 0.0 (25) 0.0304 (13)
0.25 (6) (36), 0.9 (46), 1.0281 (37), 1.1 (73), 1.0357 (76) 0.5099 (28) 0.0163 (61)
0.2 (8) 0.0410 (57)
0.09 (9)
0.0 (12)

EPSO_FACTS 0.5 (2) 1.1 (19), 0.9 (20), 0.9 (41), 0.9 (80), 0.9 (65), 0.9 (54), 1.1 (71), 0.9 (66), 0.0 (49) 0.0331 (37)
0.6 (3) 0.9 (59), 0.9 (58), 1.1 (31), 1.1 (35), 0.9 (36), 1.1 (46), 0.9 (76), 1.1 (73), 0.0 (25) 0.0334 (13)
0.25 (6) 1.0109 (37) 0.4397 (28) 0.0163 (61)
0.2 (8) 0.0410 (57)
0.09 (9)
0.0 (12)

without FACTS devices and planning of the Var sources with FACTS
devices. Reactive power sources or Var sources are generator’s
reactive power generation, settings of transformer tap positions
and shunt capacitors at the weak nodes. OLTC (On-load tap chan-
ger) positions and generators locations are already defined in IEEE
30 and IEEE 57 bus standard test systems. Once the locations of all
the Var sources are defined then the SPSO, APSO and EPSO algo-
rithms are applied to determine optimal settings of the Var
sources. This approach is used as reactive power planning without
FACTS devices. In the second approach, SPSO, EPSO and APSO algo-
rithms are applied to determine optimal Var injections of reactive
power sources along with the optimal setting of the FACTS devices.
Here two kind of FACTS devices, namely SVC (Static Var compen-
sator) and TCSC (Thyristor controlled series capacitor) are used.
SVC positions are made same as those of shunt capacitor’s loca-
tions determined by modal analysis where as TCSC (Thyristor con-
trolled series capacitor) positions are determined by reactive
power flow analysis in all the lines. Thereafter, SPSO, APSO and
EPSO algorithms are applied for reactive power optimization.
Table 1 shows the setting of different Var sources without FACTS
devices using SPSO, APSO and EPSO algorithms for IEEE 30 bus sys-
Fig. 3. Operating cost characteristics for the allocation of Var sources with FACTS
tem. Table 2 shows the setting of reactive power sources including
and without FACTS devices using SPSO method in IEEE 30 bus system.
FACTS devices by SPSO, APSO and EPSO method of optimization for
IEEE 30 bus system. Table 3 shows active power loss, operating
cost for IEEE 30 bus test system by reactive power planning with- arrangements and reactive power generation of the generators,
out and with FACTS devices. From Table 3, it is observed that when the active power loss and operating cost of the system reduces sig-
FACTS devices are co-ordinated with transformer tap setting nificantly compared to RPP without FACTS devices. It is also

Table 6
Active power loss and operating cost in IEEE 57 bus system for different methods.

Active power loss Operating cost due to energy Methods Active power loss Total operating cost Decrease in active Decrease in
without planning in p.u loss without planning in $ (B) after RPP in p.u (A1) after RPP in $ (B1) power loss in p.u operating cost in $
(A) (A  A1) (B  B1)
0.2799 1:471  107 SPSO 0.2522 1:325  107 0.0277 1:46  106
APSO 0.2495 1:311  107 0.0304 1:60  106
EPSO 0.2526 1:327  107 0.0273 1:44  106
SPSO_FACTS 0.2210 1:168  107 0.0589 3:03  106
APSO_FACTS 0.2231 1:179  107 0.0568 2:92  106
EPSO_FACTS 0.2275 1:203  107 0.0524 2:68  106
 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164
Fig. 4. Operating cost characteristics for the allocation of Var sources with FACTS
and without FACTS devices using APSO method in IEEE 30 bus system.
Fig. 5. Operating cost characteristics for the allocation of Var sources with FACTS
and without FACTS devices using EPSO method in IEEE 30 bus system.
Fig. 6. Operating cost characteristics for the allocation of Var sources with FACTS
and without FACTS devices using SPSO method in IEEE 57 bus system.
observed from Table 3, that there is a significant decrease in oper-
ating cost for all the swarm-based optimization techniques.
Optimal values of reactive power sources for IEEE 57 test sys-
tem is shown in Table 4. Similarly, Table 5 shows the setting of
163
Fig. 7. Operating cost characteristics for the allocation of Var sources with FACTS
and without FACTS devices using APSO method in IEEE 57 bus system.
Fig. 8. Operating cost characteristics for the allocation of Var sources with FACTS
and without FACTS devices using EPSO method in IEEE 57 bus system.
the different Var sources along with the FACTS devices for IEEE
57 bus system with different optimization methods. Active power
loss, operating cost for different optimization techniques in case of
IEEE 57 bus standard system is shown in Table 6. Similarly from
Table 6, it is clear that the total operating cost and active power
loss is much less when FACTS devices are taken as planning
variables.
Figs. 3–5 shows the convergence characteristics of SPSO, APSO
and EPSO techniques without and with FACTS devices for IEEE 30
bus system. Similarly Figs. 6–8 shows the convergence characteris-
tics of SPSO, APSO and EPSO methods of optimization approach
without and with FACTS devices for IEEE 57 bus system.
In each figure, two characteristics are shown, one, when RPP
was made using a particular method of optimization technique
not considering FACTS devices and the other where RPP was made
using same optimization technique considering FACTS devices. For
all the figures from 3 to Fig. 8, it is observed that there is a large gap
between two characteristics indicating superiority of the second
characteristics i.e., the characteristics obtained with FACTS devices
over the first characteristics obtained without FACTS devices.
164 B. Bhattacharyya, S. Raj / Electrical Power and Energy Systems 78 (2016) 158–164
Conclusion
Active power loss minimization and minimization of operating
cost by reactive power planning has been a challenging task for
power system Engineers. Reactive power planning by utilization
of existing Var sources is a conventional approach. In the proposed
approach, a method of Co-ordination of FACTS devices with the
existing Var sources is introduced. In the beginning, placement
locations of FACTS devices are determined, then Simple Particle
Swarm Optimization (SPSO) and other two PSO based algorithms
are applied for the optimal settings of FACTS devices and other
Var sources. It is observed that incorporating FACTS devices in
reactive power planning, the total active power loss of the system
reduces to a great extent. Similarly, there is a significant reduction
of total operating cost of the system. This technique is validated by
comparing this method of planning of Var sources with the con-
ventional approach of reactive power planning.
References
[1] Deeb NI, Shahidehpour SM. Cost minimization in allocation of reactive power
source. Int J Elect Power Energy Syst 1990;12(4):263–70.
[2] Yehia M, Ghandour I, Saidy M, Stroev VA. Reactive power optimization in large
scale power system. Int J Elect Power Energy Syst 1992;14(4):276–83.
[3] Jwo WS, Liu CW, Liu CC, Hsiao YT. Hybrid expert system and simulated
annealing approach to optimal reactive power planning. IEE Proc Gener
Transmiss Distrib 1995;142(4):381–5.
[4] Chen YL. Weak bus oriented reactive power planning for system security. IEE
Proc Gener Transmiss Distrib 1996;143(6):541–5.
[5] Mantovani JRS, Garcia AV. A heuristic method for reactive power planning.
IEEE Trans Power Syst 1996;11(1):68–74.
[6] Liu CW, Jwo WS, Liu CC, Hsiao YT. A fast global optimization approach to VAR
planning for the large scale electric power systems. IEEE Trans Power Syst
1997;12(1):437–47.
[7] Vaahedi E, Mansour Y, Li W, Tam J, Sun D, Maratukolam D. Evaluation of
existing optimal VAR planning tools on utility systems. IEE Proc Gener
Transmiss Distrib 1998;145(6):663–8.
[8] Lai LL, Ma JT. Practical application of evolutionary computing to reactive power
planning. IEE Proc Gener Transmiss Distrib 1998;145(6):753–8.
[9] Lee KY, Yang FF. Optimal reactive power planning using evolutionary
algorithms: a comparative study for evolutionary programming, evolutionary
strategy, genetic algorithm, and linear programming. IEEE Trans Power Syst
1998;13(1):101–8.
[10] Chiang Hsiao-Dong, Wang Jin-Cheng, Cockings Orville, Shin Hyon-Duck.
Optimal capacitor placements in distribution systems: Part 1 & 2: Solution,
algorithms and numerical results. IEEE Trans Power Deliv 1990;5(2):643–9.
[11] Yang N, Yu CW, Wen F, Chung CY. An investigation of reactive power planning
based on a chance constrained programming. Int J Elect Power Energy Syst
2007;29(9):650–6.
[12] Gomesa MH, Saraivab JT. A market based active/reactive dispatch including
transformer taps and reactor and capacitor banks using simulated annealing.
Elect Power Syst Res 2009;79(6):959–72.
[13] Zhu J, Cheung K, Hwang D, Sadjadpour A. Operation strategy for improving
voltage profile and reducing system loss. IEEE Trans Power Del 2010;25
(1):390–7.
[14] Zhong LH, Zhong CH, Zheng Y. A novel reactive power planning method based
on improved particle swarm optimization with static voltage stability. Eur
Trans Elect Power 2010;20(8):1129–37.
[15] Badar AQH, Umre BS, Junghare AS. Reactive power control using dynamic
Particle Swarm Optimization for real power loss minimization. Int J Elect
Power Energy Syst 2012;41(1):133–6.
[16] Hooshmand RA, Hemmati R, Parastegari M. Combination of AC transmission
expansion planning and reactive power planning in the restructured power
system. Energy Convers Manage 2012;55:26–35.
[17] Lin CH, Lin SS, Horng SC. Iterative simulation optimization approach for
optimal volt-ampere reactive sources planning. Int J Elect Power Energy Syst
2012;43(1):984–91.
[18] Mahmoudabadi A, Rashidinejad M. An application of hybrid heuristic method
to solve concurrent transmission network expansion and reactive power
planning. Int J Elect Power Energy Syst 2013;45(1):71–7.
[19] Kennedy J, Eberhart R. Particle swarm optimization. In: Proc. IEEE int. conf.
neural networks, vol. IV; 1995. p. 1942–8.
[20] Lie TT, Deng W. Optimal flexible AC transmission systems (FACTS) devices
allocation. Int J Electr Power Energy Syst 1997;19(2):125–34.
[21] Gotham DJ, Heydt GT. Power flow control and power flow studies for system
with FACTS devices. IEEE Trans Power Syst 1998;13(1):60–5.
[22] Cai LJ. Optimal choice and allocation of FACTS devices in deregulated
electricity market using genetic algorithms. In: IEEE, 0–7803-8718-X/04/2,
2004.
[23] Besharat H, Taher SA. Congestion management by determining optimal
location of TCSC in deregulated power systems. Elect Power Energy Syst
2008;30:563–8.
[24] Rashed GI, Shaheen HI, Duan XZ, Cheng SJ. Evolutionary optimization
techniques for optimal location and parameter setting of TCSC under single
line contingency. Appl Math Comput 2008;205(1):133–47.
[25] Abdelaziz AY, El-Sharkawy MA, Attia MA. Optimal location of thyristor-
controlled series compensators in power systems for increasing loadability by
genetic algorithm. Elect Power Comp Syst 2011;39:1373–87.
[26] Gerbex S, Cherkaoui R, Germond AJ. Optimal location of multi-type FACTS
devices in a power system by means of genetic algorithms. IEEE Trans Power
Syst 2001;16(3):537–44.
[27] Xiao Y, Song YH, Liu CC, Sun YZ. Available transfer capability enhancement
using FACTS devices. IEEE Trans Power Syst 2009;18(1):305–12.
[28] Basu M. Optimal power flow with FACTS devices using differential evolution.
Int J Elect Power Energy Syst 2008;30:150–6.
[29] Bhattacharyya B, Goswami SK, Bansal RC. Loss sensitivity approach in
evolutionary algorithms for reactive power planning. Electr Power Comp
Syst 2009;37(3):287–99.