1 s2.0 S2314717218300114 Main RahulAgrawal PDF

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/322999058
Population Based Evolutionary Optimization Techniques for Optimal

allocation and Sizing of Thyristor Controlled Series Capacitor
Article · February 2018

DOI: 10.1016/j.jesit.2017.04.004
CITATIONS READS
12 488
3 authors:
Rahul Agrawal Sk Bharadwaj

Sandip Univesity,Nashik Maulana Azad National Institute of Technology, Bhopal
22 PUBLICATIONS 60 CITATIONS 11 PUBLICATIONS 38 CITATIONS
SEE PROFILE SEE PROFILE
D. P. Kothari
govt engg college nagpur
16 PUBLICATIONS 38 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
An Algorithmic Approach for Model Order Reduction of Discrete The Time Interval Systems. View project
Multilevel Inverters View project
All content following this page was uploaded by Rahul Agrawal on 18 April 2018.
The user has requested enhancement of the downloaded file.

+Model
JESIT-186; No. of Pages 18 ARTICLE IN PRESS
Available online at www.sciencedirect.com
ScienceDirect
Journal of Electrical Systems and Information Technology xxx (2017) xxx–xxx
Population based evolutionary optimization techniques for optimal

allocation and sizing of Thyristor Controlled Series Capacitor
Rahul Agrawal a,∗ , S.K. Bharadwaj b,1 , D.P. Kothari c,2

a Department of Energy, MANIT, Bhopal, MP, India
b Department of Electrical Engineering, MANIT, Bhopal, MP, India
c SBJIT, Nagpur, MH, India
Received 30 December 2016; received in revised form 25 February 2017; accepted 17 April 2017
Abstract
During the last decade, the application of Flexible AC Transmission System (FACTS) devices in electrical power system increased
due to their numerous advantages such as power transfer capability enhancement, reliability and security improvement and reactive
power compensation. Although to obtain maximum advantage, their size and location should be optimal. Optimal position and size
of Thyristor Controlled Series Capacitor (TCSC) in power system using evolutionary optimization techniques such as, Teaching
Learning Based Optimization (TLBO), Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO) are presented in this
paper. The minimization of transmission loss, the installation cost of TCSC and voltage deviation is considered as an objective
function. To demonstrate the viability of the algorithms, it is validated on the IEEE 14 bus, IEEE 30 bus and Indian 75 bus systems.
The results accomplish from TLBO are compared with the ABC and PSO to show the capability of the TLBO and its superiority
over ABC and PSO.
© 2018 Electronics Research Institute (ERI). Production and hosting by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: ABC; FACTS; PSO; TCSC; TLBO; Transmission loss
1. Introduction
In recent years, the size and complexity of electric power system are increased enormously due to exponential rise in
demand of electricity. Therefore, a stable, reliable and the uninterrupted power quality with the minimum transmission
loss is a Herculean task and demand stern constraints to be satisfied. To accomplish the above objectives, the most
∗ Corresponding author.
1 Member IEEE.
2 Fellow IEEE.
Peer review under the responsibility of Electronics Research Institute (ERI).
https://doi.org/10.1016/j.jesit.2017.04.004
2314-7172/© 2018 Electronics Research Institute (ERI). Production and hosting by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Agrawal, R., et al., Population based evolutionary optimization techniques for
optimal allocation and sizing of Thyristor Controlled Series Capacitor. J. Electr. Syst. Inform. Technol. (2017),
https://dx.doi.org/10.1016/j.jesit.2017.04.004
+Model
2 R. Agrawal et al. / Journal of Electrical Systems and Information Technology xxx (2017) xxx–xxx
reliable and feasible alternative is the application of Flexible AC Transmission System (FACTS) devices, which provide
better power flow control, voltage control, flexibility of operation and superior quality of the existing transmission line
capacity (Hingorani and Gyugyi, 2000; Mathur and Varma, 2002). In this process the size and location of the FACTS
devices in the wide electricity transmission network play a pivotal role. FACTS controller can be classified into three
types, Series Controllers, Shunt Controllers and Combined Series and Shunt Controllers, based on the type of connection
in the transmission line (Hingorani and Gyugyi, 2000). Static Synchronous Compensator (STATCOM) and Static VAR
Compensator (SVC) are classified as shunt controllers whereas Static Synchronous Series Compensator, Interphase
Power Controller (IPC) and Thyristor Controlled Series Capacitor are classified as series controllers. Unified Power
Flow Control (UPFC), Interline Power Flow Control (IPFC) and Thyristor Controlled Phase Shifting Transformer
(TCPST) are examples of combined series-shunt controllers. FACTS devices can also be categorized as, thyristor
based FACTS controller and VSC based FACTS controller, based on the power electronics switch used.
The Thyristor Controlled Series Capacitor (TCSC) is a series connected FACTS device, used for the control of power
transfer between the lines by varying the impedance of the line. The principal applications of TCSC are mentioned in
the literatures as to improve power system stability, to enhance power and to damp oscillations presented in the power
system. The purpose of the present work is to minimize the transmission loss and voltage deviation of power systems.
In conceiving the suggested objectives, the location and ratings of the TCSC must be brought off. The optimization
methods, namely Teaching Learning Based Optimization (TLBO), Artificial Bee Colony (ABC) and Particle Swarm
Optimization (PSO) are used for the same and discussed briefly in the section 4 of the paper. The location and degree
of compensation of TCSC are considered as the optimization variables.
The size and location of the FACTS devices in the wide electricity transmission network play a pivot role. As
FACTS devices are expensive and need to be utilized effectively to get the maximum benefit, these devices must be
placed optimally. Also, it should be ensured that there will be no negative impact in the deregulated environment.
The optimal location and size of FACTS devices are helpful to enhances loadability, available transfer capability, total
transfer capability, voltage stability, transient stability, damping, security, etc. besides reduction of losses, minimization
of generation cost and removal of congestion. The Optimal location and size of the FACTS devices in a power system
will be helpful for utilities in the decision making of investment costs and further expansion plans. This research
work proposes the optimal location and value of FACTS devices in power system using the evolutionary optimization
techniques.
To find the suitable location of TCSC to enhance security and minimize the generation cost of the IEEE 30, IEEE
57 and IEEE 118 bus systems using fuzzy logic with Harmony search algorithm were examined by Pandiarajan and
Babulal (2016). The optimal power flow with TCSC to minimize transmission loss, fuel cost and emission cost using
Symbiotic Organism Search (SOS) algorithm are witnessed in (Prasad and Mukherjee, 2016). It was simulated on
the MATLAB and tested on the standard IEEE 57 bus and IEEE 30 bus systems. The position and range of TCSC
devices to mitigate system losses and investment cost using Genetic Algorithm and Particle Swarm Optimization is
presented in (Rashed et al., 2007, 2011). Duong et al. (2014) proposed a Min cut algorithm, to find the best location of
TCSC for enhancing loadability and minimization of transmission losses. Nagalakshmi and Kamaraj (2012) proposed
the Particle Swarm Optimization (PSO), Differential Evolution (DE) and Composite Differential Evolution (CoDE)
to determine the location and control of FACTS devices to improve the loadability of the power system. It is shown
that the computational time required for CoDE is less than Differential Evolution and Genetic Algorithm. To minimize
the generation, emission and real power transmission loss of IEEE 30 bus and IEEE 57 bus systems using TCSC with
Non-dominated Sorting Genetic Algorithm-II (NSGA-2) is reported in (Basu, 2011). It is revealing that the outcome
obtained from Differential Evolution is improved compared to Genetic Algorithm. Naresh et al. (2016) has used the
Harmony Search Algorithm (HSA) to damp out the power system oscillation. It is shown that the result obtained from
the HSA is better compared to Bacteria Swarm Optimization. In Rezaee Jordehi et al. (2015) Enhanced leader PSO
(ELPSO) was used for optimal location of TCSC to minimize the voltage deviation and power losses under contingency.
The Real Genetic Algorithm (RGA) to determine the proper location and sizing of TCSC to enhance the voltage profile
and Available Transfer Capability (ATC) has been proposed by Rashidinejad et al. (2008).
Going through the research survey, we observed that a numerous evolutionary optimization technique has been
developed to investigate the transmission loss, voltage profile, security, ATC, loadability and installation cost of
TCSC. The examination of literature review also reveals that the solution of the above problem along with TCSC
cannot be solved without optimization technique. The outcome also shows that the multi-objective problem consisting
transmission loss minimization, installation cost minimization and voltage profile improvement is not presented in
+Model
R. Agrawal et al. / Journal of Electrical Systems and Information Technology xxx (2017) xxx–xxx 3
Fig. 1. The variable reactance model of TCSC.
the existing literature. The Newton Raphson power flow incorporating TCSC with modern optimization techniques is
presented in the present work. The multi-objective optimization problem considering the above objective is transformed
into a single objective optimization problem using the weighting factor method.
The major contribution of this research work includes the following
i. Implementation of Newton Raphson (NR) static load flow algorithms with and without Flexible AC Transmission
System (FACTS) devices using evolutionary optimization techniques.
ii. Apart from the determination of the complex transmission losses, active power transmission line loss and reactive
power transmission line loss of the IEEE 14 system are also presented in the research work.
iii. To present the variable reactance model of TCSC incorporated in the load flow algorithms to find the optimal
location of FACTS devices.
iv. The methodology demonstrated on the three standard test systems and simulation results are presented, which
authenticate the effectiveness of the proposed algorithms.
The Teaching Learning Based Optimization (TLBO) is a heuristic optimization technique developed by Rao et al.
(2012) and Rao and Kalyankar (2011). It is a population based iterative method; uses generation of population for
obtaining global solution. As compared to other evolutionary algorithms like GA, DE and PSO, with TLBO require
less number of design parameters and fast convergence time (Satapathy and Naik, 2013).
2. Modelling of TCSC
TCSC contains a series capacitor shunted with a thyristor controlled variable inductor. By varying the firing angle,
the reactance of the TCSC is controlled. The different types of modelling of TCSC are presented in the literature by the
researcher such as a variable reactance model, power flow model, power injection model and current injection model.
The capacitive characteristics and inductive characteristics are two possible characteristics of TCSC, to increase or
decrease the transmission line reactance. By helping these characteristics TCSC are used to increase the transmission
power capacity of the lines, static voltage security margin enhancement, voltage profile improvement, and decreasing
power loss. When it injects reactive power, transmission losses are reduced (negative value of XTCSC ) and when it
absorbs reactive power, transmission losses are increasing (positive value of XTCSC ). In order to use TCSC to satisfy
above mentioned objective, the variable reactance model shown in Fig. 1 has been considered.
The overall reactance of the transmission line Xline consists, TCSC reactance XTCSC and transmission line reactance
Xft as:
Xline = XTCSC + Xft (1)
In terms of degree of compensation of the line ‘k’ the total reactance of the transmission line is modified as
Xline = (k + 1)Xft (2)
where,
XTCSC = k × Xft (3)
The range of ‘k’ for the TCSC is considered between −80% and +20%. The negative value of ‘k’ shows the capacitive
reactance and positive value of ‘k’ shows the inductive reactance of TCSC.
+Model
3. Problem formulation
The problems that are to be faced in planning stage are appropriate type, location, size and setting of TCSC controllers
for various applications. The Newton Raphson Load Flow algorithms incorporating TCSC and optimization methods
have been developed for the solution of the objective function along with constraints as given below.
3.1. Minimize the active power loss
The loss minimization is a useful tool in the reduction of cost and current flow between the lines. Loss minimization
is also important in planning for future expansion as well as for the operation of the existing power system. In active
power loss minimization the voltage of the buses and conductance between the buses is selected as a control variable.
The objective of active power loss minimization subjected to power system constraints is determined by the proper
selection of control variables. The total active power loss in an electric power system is communicated scientifically as

NL
Ploss = Gkj [Vk2 + Vj2 − 2Vk Vj cos(δk − δj )] (4)
k=1
j=
/ i
where NL is the overall transmission lines of the system, Gkj is the conductance between the bus k and j, Vk and Vj
is the voltage of the kth and jth bus respectively, δk and δj are the angles of the kth and jth bus respectively (Agrawal
et al., 2016).
After the normalization first objective function is given by
Ploss
min f1 = (5)
Plosso
where Ploss is the active power loss with TCSC and optimization methods and Plosso is the active power loss without
TCSC and optimization method.
3.2. Voltage profile improvement
The low voltages cause the voltage stability problems and poor performance of electrical load, whereas the high
voltages affect the economic operation of power system. The present study considers the voltage deviation minimization
of all the load buses from the reference bus voltage. The bus voltage limit is considered between the ranges 0.95 p.u.
and 1.05 p.u. The voltage deviation can be expressed as

nk

VD = Viref − Vi (6)
i=1
where Viref is the reference voltage at the ith bus, and nk is the number of load buses, Vi is the voltage of the ith bus.
The normalization form of second objective function can be expressed as
VD
min f2 = (7)
VDo
where VDo is the voltage deviation without TCSC and optimization method.
3.3. Minimize the installation costs
FACTS devices are very costly due to its complex circuitry and modelling, therefore the location of FACTS devices
should be selected wisely. The minimization of the installation cost of the TCSC device is chosen as the third objective
function and written by (8)
ICTCSC = CTCSC × STCSC × 1000 (8)
+Model
where ICTCSC is the installation cost of TCSC [in US$], CTCSC is the cost of TCSC devices [in US$/kVAR] and
STCSC is the Operating range of TCSC [in MVAR].
The normalization form of installation cost is given by the Eq. (9)
ICTCSC
min f3 = (9)
ICTCSCo
where ICTCSCo installation cost of TCSC without optimization method.
The multi objective optimization function is transformed and mathematically formulated into a single objective
optimization function and given by Agrawal et al. (2016)
F = α1 × f1 + α2 × f2 + α3 × f3 (10)
where α1 ,α2 and α3 are the weighting factors. The weighting factor is defined as the (Kim and Weck, 2006)
α 1 = w1 × w2 (11)
α2 = (1 − w1 ) × w2 (12)
α3 = (1 − w2 ) (13)
α1 + α2 + α3 = 1 (14)
The w1 and w2 are selected between 0 and 1. In present research the value of the factor w1 = 0.8, w2 = 0.6,
α1 = 0.4800, α2 = 0.1200 and α3 = 0.4000 are chosen based upon the number of initial divisions for the objective
function.
Eq. (10) is subjected to the following equality and inequality constraints.
3.3.1. Equality constraints

The real and reactive power balance equations for n-bus power system are
PiG − PiD − Pic (δ1 , δ2 . . . δn , |V1 | |V2 | . . . |Vn |) = 0 (15)
i − Qi − Qi (δ1 , δ2 . . .δn , |V1 | |V2 | . . . |Vn |) = 0

QG D c
(16)
where Pic and Qci are the calculated active and reactive power of ith bus and can be expressed as

n

Pic = |Vi | Vj {Gij cos(δi − δj ) + Bij sin(δi − δj )} (17)
j=1

n

Qci = |Vi | Vj {Gij sin(δi − δj ) − Bij cos(δi − δj )} (18)
j=1
where PiG and PiD are the real power generation and real power demand at the ith bus, QG D
i and Qi are the reactive
power generation and reactive power demand at the ith bus, Bij the susceptance between the ith and jth bus, Gij the
conductance between the ith and jth bus, Vi and Vj is the voltage of the ith and jth bus, δi and δj is the phase angles of
the ith and jth bus respectively.
3.3.2. Inequality constraints

Real power and reactive power generation limit at the ith bus
|PiG | ≤ ε (19)
|QG
i |≤ε (20)
where ‘ε’ is defined as the tolerance limit.
Bus voltage limits constraint at the ith bus

min
V ≤ |Vi | ≤ V max
i i (21)
+Model
where Vimin and Vimax is the minimum and the maximum bus voltage at the ith bus respectively.
Tap setting transformer limits constraint at the ith bus
tapmin
i ≤ tapi ≤ tapmax
i (22)
where tapmin
i and tapmax
i minimum and maximum tap setting at the ith bus respectively.
TCSC reactance limits constraints
−0.8Xft ≤ XTCSC ≤ 0.2Xft (23)
4. Optimization methods
Optimization is a mathematical discipline to carry out optimal value of a physical system in terms of minimum and
maximum of a function, subject to certain system constraints. To overcome the limitation of conventional optimization
techniques such as slow speed and large computation burden, many evolutionary optimization techniques has been
introduced in the literature during past years. This section presents the brief of three heuristic optimization methods,
TLBO, PSO and ABC.
4.1. Implementation of TLBO for the optimal location of TCSC
In TLBO a group of learner (or student) comprises the population P, is generated randomly in the search space
bounded by the number of subject D (or design variable). The number of subject offered to the students indicates
the number of optimization parameter used. The output of the learner gives the objective function of the optimization
problem. TLBO is searched for optimal value through learning of the learner by the teacher. The working of TLBO is
principally divided into two phases, teacher’s phase and learner’s phase. In teacher’s phase all the learners, learn from
the teacher. In learner’s phase, all the learners learn from the interaction amongst them.
The step by step procedure of TLBO is explained as
Step I: Initialization of population-Initialize the population x randomly in the search space bounded by the D and P.
The initial population of the jth parameter of the ith variable is represented by (24)

xi = xjmin + xjmax − xjmin ∗ rand (P, D) (24)
where xmin and xmax is the minimum and maximum value of the jth parameter and rand is the uniformly distributed
random number between the (0, 1)
Step II: Teacher’s phase-Evaluate the objective function ‘f’ for each design variable.
Step III: column wise arranges the mean value of each design variable offered to the learners in the population,
MD = [m1 , m2 , m3 . . .mD ] (25)
Step IV: Set the iteration k

Step V: Choose the learners with the lowest value of the fitness function fmin which is considered as teachers Xteachers
for that subject.
Xteachers,D = fmin,D (26)

The best teachers, Xteachers will effort to change the average from M to new average Mnew as
new
MD = Xteachers,D (27)
Step VI: Compute the difference between the new average Mnew and the current average M with the use of teaching
factor TF according to the equation
+Model
new
Differnce meanD = rand MD − TF ∗ MD (28)
The value of teaching factor TF (1 or 2) is generated randomly by using (29)
TF = round [1 + rand] (29)
where rand is the uniformly distributed random number between the (0, 1)
Step VII: Update the knowledge of ith learner by adding the difference mean and existing population of the learners.
xinew = xi + Diffrence meanD (30)
Step VIII: Update the objective function fnew corresponding to the new value of learner xnew .
Step IX: Learner’s phase-In this phase learners are interacting with each other, the process of interaction increases the
learner’s knowledge. Updating the learner’s knowledge by selecting randomly some another learner xnew such that
xinew =
/ xjnew (31)
new
xi + rand(xinew − xjnew ), if (finew < fjnew )
xinew = (32)
xinew + rand(xjnew − xinew ), if (finew > fjnew )
Step X: Termination-Increase the iteration k = k + 1. Check the termination criterion, otherwise go to step IV.
The pseudo code of the proposed TLBO algorithms with optimal location of TCSC is given in Table 1.
4.2. Implementation of PSO for the optimal placement of TCSC
The working of PSO algorithms is presented as follows

Step I: Input the standard data of the test systems.
Step II: Run the Newton Raphson load flow algorithm and determine the bus voltage, active power and reactive
power for all the buses.
Step III: Estimate the losses of the power system without TCSC.
Step IV: Initialize the position and rating of the TCS as an optimization parameter of PSO.
Step V: Randomly generate the position of the ith particle in the dimension D at iteration k as

xik = xi1 k k
, xi2 k
, xi3 k
. . .xiD (33)
Step VI: Randomly generate the velocity of the ith particle as

Vik = Vi1k , Vi2k , Vi3k . . .ViD
k
(34)
Step VII: Run the load flow with TCSC for the ith particle.
Step VIII: Reckon the objective function.
Step IX: Determine pbestik and gbestik of the swarm.
Step X: Modify the particle velocity and position of ith particle according to (35) and (36)
Vik+1 = wVik + c1 rand1 × (pbestik − xik ) + c2 rand2 × (gbestik − xik ) (35)
xik+1 = xik + Vik+1 (36)

Here w is the weighting factor, constants c1 and c2 represent the learning factors, rand1 and rand2 are the random
number in the range 0–1, pbestik the personal best position of the ith particle, gbestik the global best position of the
swarm, xik and xik+1 are the particle position at ‘k’ and ‘k + 1 iteration, Vik and Vik+1 are the particle velocity at ‘k’ and
‘k + 1 iteration respectively.
+Model
Table 1
The pseudo code for TLBO with TCSC.
Begin
Read system data, line data and bus data
Run the Newton Raphson load flow algorithm and determine the bus voltage, active power and reactive power for all the buses.
Initialize the position and rating of the TCSC as an optimization parameter of TLBO
Initialize D, P and termination criterion

Generate an initial
population
xi = xjmin + xjmax − xjmin ∗ rand (P, D)
Calculate objective function f(x)
For iteration = 1: maximum iteration {termination criterion}
{Teacher phase}
Calculate the mean value of each design variable MD = [m1 , m2 , m3 . . .mD ]
Calculate the minimum value of objective function min(f)
For m = 1:P
Calculate the teaching factor TF = round [1 + rand] new
Calculate difference between two means Differnce meanD = rand MD − TF ∗ MD
Update the initial population xinew (m) = xi (m) + Diffrence meanD
Update system data, line data and bus data
Run the Newton Raphson load flow algorithm for each design variable xinew (m)
Update the objective function fnew (m)
If f(m) < = fnew (m)
xinew (m) = xi (m)
fnew (m) = f(m)
End
End of teacher phase
{Student phase}
Select randomly some another student xnew such that xinew = / xjnew
For n = 1:P
if (finew (n) < fjnew (n))
xi (n) = xinew (n) + rand(xinew (n) − xjnew (n))
new
end
if (finew (n) > fjnew (n))
xinew (n) = xinew (n) + rand(xjnew (n) − xinew (n))
end
End of for loop
Update the design variable
If f(m) ≤ fnew (m)
xinew (m) = xi (m)
fnew (m) = f(m)
End
End of termination criterion
Display optimal value of result
The inertia weight factor is varied with iteration as per Eq. (37)
wmax − wmin
w = wmax − × iteration (37)
maximum iteration
Here wmax represent the maximum value of the weight factor, wmin represent the minimum value of the weight
factor.
Step XI: Check the termination criterion, otherwise go to step II.

Step XII: Calculate the transmission loss with TCSC.
Step XIII: End.
+Model
Table 2
Parameter values for PSO, ABC and TLBO.
PSO parameter ABC parameter TLBO parameter
Parameter Value Parameter Value Parameter Value
Number of iterations 35 Maximum number of iterations 35 Termination criterion 35

Number of design variable 2 Number of decision variable 2 Number of subjects 2
Population size 20 Colony Size (CS) 40 Number of students 20
Inertial Weight, w 0.88–0.38 Limit 40 Rand 0–1
Constant, C1 2 Number of employed bees (EB) 20
Constant, C2 2 Rand 0–1
rand1 0–1
rand2 0–1
4.3. Implementation of ABC for the optimal Placement of TCSC
The Artificial Bee Colony (ABC) Algorithm is a heuristic optimization technique developed by Karaboga and
Basturk (2007). It is a population based swarm intelligence method (Karaboga and Akay, 2009). It is based on the
foraging behaviour of the honey bees.
The operation of the ABC method is presented as follows:
Step I to step III are same as PSO algorithm.

Step IV: Introduced the ABC parameter according to Table 2.
Step V: randomly initialize the position of the employed bee in the search area.
Step VI: Run the Newton Raphson (NR) load flow for each employed bee as generated above. Calculate the objective
function for each employed bee. Evaluate the fitness function for each decision variable.
Step VII: Set Cycle = 1.
Step VIII: Update the employed bee position, objective function and fitness function for each decision variable. Apply
greedy selection for employed bee phase; compare the new food source to old food source to decide the good food
source.
Step IX: Select onlooker bee based on the probability value of the food source and update onlooker bee position.
Update the bus data, and run the load flow using the NR method for each decision variable of onlooker bee. Update
objective function and fitness function and apply greedy selection for onlooker bee phase.
Step X: Scout bee phase.
Step XI: Remember the best food so far.
Step XII: Cycle = Cycle + 1
Step XIII: If cycle ≤ max cycle, go to step VI else go to step XII.
Step XIV: Calculate the transmission loss with TCSC.
Step XV: Stop.
The flow chart of ABC algorithms with optimal location of TCSC is shown in Fig. 2.
5. Test results and discussion
In this work, the optimal location and sizing of FACTS devices are determined by MATLAB simulation to minimize
the cost of installation of FACTS devices, transmission loss and to improve the voltage profile using evolutionary
optimization algorithms namely, Artificial Bee Colony algorithm, Teaching Learning Based Optimization and Particle
Swarm Optimization. This section present the test system used, result for real power transmission loss, reactive power
transmission loss, optimal location and sizing of TCSC, line loss and bus voltage.
+Model
Fig. 2. The flow chart of the ABC algorithms with TCSC.
5.1. Tools and test system
The proposed algorithms have been coded in the MATLAB programming environment for ABC, PSO and TLBO
with power flow and TCSC. Simulation is performed on the subsequent test systems
(a) Test system 1 (TS 1) – IEEE 14 bus system,

(b) Test system 2 (TS 2) – IEEE 30 bus system and
(c) Test system 3 (TS 3) – UPSEB 75 bus system.
The system data (at 100 MVA base) has taken from (Joshi, 1995) and (Verma, 2005). The parameter values of the
PSO, ABC and TLBO optimization are listed in Table 2. In all the three optimization methods the maximum number
of iterations is taken 35, number of design variable is 2 and population size of 20.
5.2. Results for real power transmission loss with TCSC
The active power transmission loss, including TCSC and without considering TCSC is given in Table 3. The active
power transmission losses are 11.165 MW for the IEEE 14 bus system, 6.8212 MW for the IEEE 30 bus system and
224.3341 MW for Indian 75 bus system when TCSC not included. From Tables 3 and 4 it is evident that if TCSC is
+Model
Table 3
Real power transmission loss in MW with PSO, ABC and TLBO optimization technique.
S.n. Bus system Without TCSC PSO ABC TLBO
1 IEEE-14 bus system 11.1650 7.8779 7.3701 7.2610

2 IEEE-30 bus system 6.8212 4.7883 4.7883 4.7143
3 UPSEB-75 bus system 224.3341 168.4428 167.0246 167.0246
Fig. 3. Active power loss of test system 1 with optimization methods.
introduced in the line 9 to the test system 1, the real power transmission loss is reduced from 11.165 MW to 7.261 with
TLBO, though with PSO and ABC it is discovered 7.8779 MW and 7.3701 MW individually.
It is also seen from Tables 3 and 4 that if TCSC is placed in line 6 and 33 for the test systems 2 and 3, the real
power losses are diminished to 4.7143 MW and 167.0246 MW respectively with TLBO. This outcome demonstrates
that power transfer capacity of the considered test system has made strides. It is evident from the results that after
placement of TCSC, the real power losses of the all the three test systems are reduced, with all the three optimization
techniques. From Table 3 it is clear that the real power loss of the test system 2 with ABC and PSO is same whereas
for the test system 1 and 3 the result of ABC is better as compared to PSO. Table 3 additionally demonstrates that
outcome acquired from TLBO is better as compared with those obtained from PSO and ABC. Table 3 also shows that
the result obtained from ABC is better than PSO. The real power loss for the test system 1, 2 and 3 with iteration is
sketched in Figs. 3–5.
It is observed from Fig. 3 that during the first iteration, the real power loss for the test system 1 is minimum with
ABC while maximum with TLBO. With the iteration the results of TLBO are improved and converge faster and after
30 iterations it gives best results. Similarly from Figs. 4 and 5 it is observed that the real power loss obtained from the
TLBO for the test systems 2 and 3 is found minimum after 20 iterations. It also detects that the results converge faster
with TLBO as compared to ABC and PSO is also depicted in the figures.
Table 4
Optimal location and sizing of TCSC with PSO, ABC and TLBO optimization technique.
S.n. Bus system PSO ABC TLBO
k XTCSC IC Line k XTCSC IC Line k XTCSC IC Line
(pu) (US$) number (pu) (US$) number (pu) (US$) number
1 TS-1 −0.8 −0.178 27,457 9 −0.8 −0.1584 24,368 9 −0.5556 −0.1180 16,920 9
2 TS-2 −0.8 −0.148 22,795 6 −0.7703 −0.1427 21,900 6 −0.7649 −0.1417 21,793 6
3 TS-3 −0.8 −0.067 10,304 48 −0.7845 −0.0374 5754.2 27 −0.6057 −0.0182 2791.2 33
+Model
5.3. Results for optimal location and sizing of TCSC
The position, value and cost of TCSC obtained from TLBO are compared with those obtained from PSO and ABC
is exhibited in Table 4. Table 4 shows that if TCSC with TLBO of capacitive reactance 11.8%, 14.17% and 1.8% of
the line, for test system 1, 2 and 3 installed in the line number 9, 6 and 33 respectively, the real power transmission loss
and reactive power transmission loss is decreased. The total cost of the TCSC via TLBO for the IEEE 14, IEEE 30 and
UPSEB 75 bus systems is 16,920 US$, 21,793 US$ and 2791.2 US$ respectively, with ABC it is 24,368 US$, 21,900
US$ and 5754.2 US$ respectively and whereas by PSO it is 27,457 US$, 22,795 US$ and 10,304 US$ respectively.
The installation cost and size of TCSC got with TLBO is better as compared to PSO and ABC as displayed in Table 4.
Also the installation cost of the TCSC using ABC is less against those obtained from PSO as appeared in Table 4.
It is seen from Tables 3 and 4 that if TCSC is placed in line 6 for the IEEE 30 bus test systems the real power losses
are diminished to 4.7883 MW, 4.7883 MW and 4.7143 MW respectively with PSO, ABC and TLBO. This outcome
demonstrates that power transfer capacity of the considered test system has made strides.
5.4. Results for reactive power transmission loss with TCSC
The reactive power loss with TCSC using PSO, ABC and TLBO is presented in Table 5. The reactive power losses
in test system 1, 2 and 3 are 29.2705 MVAR, 20.3095 MVAR and 319.5196 MVAR respectively, when TCSC is not
considered in the systems, and with TCSC using TLBO it is discovered 15.2473 MVAR, 12.7712 MVAR and 280.4618
MVAR respectively. It is obvious from Table 5 that in the wake of setting TCSC on the considered system the reactive
power losses of the test system 1, 2 and 3 are reduced. It is seen from Table 5 that the reactive power losses in the test
+Model
Table 5
Reactive power transmission loss in MVAR with PSO, ABC and TLBO optimization technique.
S.n. Bus system Without TCSC PSO ABC TLBO
1 IEEE-14 bus system 29.2705 19.4431 16.6165 15.2473

2 IEEE-30 bus system 20.3095 12.9473 12.9182 12.7712
3 UPSEB-75 bus system 319.5196 285.3305 281.4962 280.4618
Table 6
Line losses of the line with PSO, ABC and TLBO optimization technique.
S.n. From bus To bus Real power line losses in MW Reactive power line losses in MVAR
Without TCSC PSO ABC TLBO Without TCSC PSO ABC TLBO
1 8 3 0 0 0 0 1.982 1.829 1.79 2.122

2 9 6 0 0 0 0 0.802 0.778 0.793 0.666
3 9 7 0 0 0 0 1.082 1.024 1.035 0.893
4 1 8 2.015 1.604 0.611 0.862 5.477 4.235 3.743 −0.328
5 2 8 0.618 0.333 0.255 0.418 0.093 −0.567 −0.828 −1.05
6 4 9 0.579 0.005 0.897 0.045 −0.292 −1.689 −1.792 0.498
7 9 8 0.451 0.244 0.873 0.304 0.769 0.289 0.098 2.091
8 1 2 3.243 3.771 1.484 3.62 6.995 8.101 8.561 1.579
9 2 4 2.144 0.409 1.672 0.324 6.722 1.068 −0.633 4.685
10 6 5 0 0 0 0 0.876 0.967 0.964 1.115
11 2 9 1.243 0.661 0.641 0.832 1.811 0.506 −0.011 −0.062
12 6 7 0 0 0 0 1.173 1.15 1.155 1.114
13 7 10 0.010 0.010 0.016 0.01 0.029 0.027 0.026 0.041
14 3 11 0.156 0.155 0.183 0.157 0.326 0.329 0.325 0.384
15 3 12 0.092 0.089 0.093 0.089 0.192 0.186 0.185 0.193
16 3 13 0.303 0.292 0.312 0.294 0.597 0.579 0.575 0.615
17 7 14 0.061 0.064 0.048 0.062 0.130 0.131 0.136 0.102
18 10 11 0.069 0.071 0.087 0.072 0.162 0.169 0.167 0.203
19 12 13 0.014 0.015 0.016 0.015 0.0134 0.013 0.013 0.015
20 13 14 0.158 0.154 0.182 0.157 0.323 0.319 0.314 0.371
system 1, 2 and 3 with ABC are found 16.6165 MVAR, 12.9182 MVAR and 281.4962 MVAR respectively, while with
PSO the reactive power losses are obtained 19.4431 MVAR, 12.9473 MVAR and 285.3305 MVAR respectively, i.e. in
all the three cases the results obtained from ABC are superior as compared to PSO. The reductions in reactive power
loss with TCSC using TLBO are lower as compared to those obtained from PSO and ABC for all the test systems.
5.5. Results for line loss with TCSC
The active power transmission line losses and reactive power transmission line losses, including TCSC and without
considering TCSC for the test system 1 is demonstrated in Table 6. Results obtained from TLBO are compared and
found better with those obtained from PSO and ABC is also presented in Table 6. It is seen from Table 6 that after
placement of TCSC in the line number 9 the real power transmission line losses is reduced from 2.14406 to 0.409 MW,
1.672 MW and 0.324 MW with PSO, ABC and TLBO respectively. Also the reactive power line losses are reduced
from 6.7228 MVAR to 1.068 MVAR, 1.633 MVAR and 4.685 MVAR with PSO, ABC and TLBO respectively.
5.6. Results for the Bus Voltage with TCSC
The IEEE 14 bus system consists 2 generators (bus no. 1 and 2) three condenser (bus no. 3–5), 9 load buses (bus no.
6–14) and 20 transmission lines. The voltage of the buses for the test system 1 with PSO, TLBO and ABC is shown in
Table 7. For the test system 1 the bus voltage without TCSC at bus number 6 is 1.000 per unit (pu), whereas with PSO,
it is 1.0188 pu, with ABC it is 1.0157 pu and with TLBO it is found 1.0112 pu. Table 7 indicates the improvement in
the bus voltage with TCSC. From Table 7 it is clear that, in all the optimization methods voltage of all the load buses
+Model
Table 7
Bus voltage of the IEEE 14 bus system in pu with ABC, PSO and TLBO.
Bus number Specified voltage magnitude Vsp in pu With TCSC
PSO ABC TLBO
1 1.0600 1.0600 1.0600 1.0600

2 1.0450 1.0450 1.0450 1.0450
3 1.0700 1.0600 1.0600 1.0600
4 1.0100 1.0200 1.0200 1.0200
5 1.0900 1.0800 1.0800 1.0800
6 1.0000 1.0188 1.0157 1.0112
7 1.0000 1.0104 1.0102 1.0069
8 1.0000 1.0184 1.0182 1.0101
9 1.0000 1.0121 1.0101 1.0061
10 1.0000 1.0137 1.0126 1.0108
11 1.0000 1.0129 1.0118 1.0013
12 1.0000 1.0124 1.011 1.0012
13 1.0000 1.0149 1.0125 1.0042
14 1.0000 1.0024 1.0013 1.0001
Fig. 6. Voltage of test system 1 with optimization methods.
is improved. From Table 7 it is clear that the outcome obtained from TLBO is prominent over those obtained from
ABC and PSO. The graphical representation of the bus voltage for the test system 1, 2 and 3 is also shown in Figs. 6–8
respectively.
+Model
Table 8
Line flow of the IEEE 14 bus system with TCSC.
LN FB TB Real power line flow in MW Reactive power line flow in MVAR
Without TCSC ABC TLBO PSO Without TCSC ABC TLBO PSO
1 8 3 28.836 28.098 30.705 28.418 6.754 4.685 0.334 4.644

2 9 6 19.574 20.037 18.440 19.837 5.050 3.529 1.928 3.545
3 9 7 11.157 11.415 10.492 11.301 9.314 8.762 8.149 8.778
4 1 8 64.049 57.592 33.272 59.719 8.009 2.887 11.333 2.695
5 2 8 34.046 23.997 12.821 27.526 2.730 7.722 17.308 6.791
6 4 9 −26.163 1.826 −33.363 −8.089 13.495 −2.853 15.843 1.096
7 9 8 −58.573 −43.709 −79.216 −48.781 3.955 −1.312 21.458 0.713
8 1 2 137.116 144.366 90.932 141.464 −3.662 −33.457 −19.821 −32.761
9 2 4 70.181 28.530 62.509 42.730 −0.619 11.263 6.006 8.791
10 6 5 0.000 0.000 0.000 0.000 −22.816 −24.303 −26.052 −24.335
11 2 9 47.945 34.383 32.417 39.123 2.443 8.127 12.228 6.857
12 6 7 19.574 20.037 18.440 19.837 27.064 27.039 27.314 27.103
13 7 10 −5.666 −5.207 −6.733 −5.399 −1.371 −2.060 −2.076 −1.999
14 3 11 9.402 8.943 10.519 9.139 9.489 10.178 10.304 10.125
15 3 12 8.447 8.402 8.613 8.426 3.205 3.295 3.283 3.285
16 3 13 19.787 19.553 20.373 19.653 10.325 10.654 10.743 10.629
17 7 14 6.897 7.159 6.165 7.037 0.127 −0.325 −0.330 −0.285
18 10 11 −5.677 −5.217 −6.749 −5.410 −7.200 −7.886 −7.917 −7.827
19 12 13 2.254 2.213 2.420 2.237 1.412 1.510 1.489 1.498
20 13 14 8.223 7.959 8.965 8.081 5.327 5.775 5.803 5.736
5.7. Line flow
The active power line flow and reactive power line flow with TCSC and without considering TCSC for the IEEE
14 bus test system are presented in Table 8. The sketch of active power line flow and reactive power flow of Table 8
is represented by bar graph as shown in Figs. 9 and 10 respectively. Results obtained from TLBO are compared with
those obtained from PSO and ABC is also presented in Table 8. It is seen from the table that after placement of TCSC
the real power transmission line flow and reactive power line flow is reduced.
where LN represents the Line Number, FB represents the From Bus and TB represents the To Bus.
5.8. Comparative study
Under the identical network conditions, i.e. controls, variable limits, equality constraints, inequality constraints, line
data, bus data and system data, the results obtained using the ABC, PSO and TLBO technique reported in this work
+Model
Fig. 9. Active power line flow of the IEEE-14 bus power system with TCSC.
Fig. 10. Reactive power line flow of the IEEE-14 bus power system with TCSC.
are compared to some other techniques reported in the literature as shown in Table 9. From Table 9 it is evident that,
the proposed techniques outperform many techniques used to minimize the transmission losses because the results
obtained using the proposed techniques are either better as comparable to those obtained using other techniques. This
highlights their ability to find a better quality solution.
6. Conclusion
In this paper a new optimization technique, Teaching Learning Based Optimization (TLBO) has been presented in
order to determine the optional location and rating of Thyristor Controlled Series Capacitor (TCSC) of the considered
power system. This work shows a generalized algorithm of optimization technique TLBO, keeping in mind the end
objective to get better, power transfer capacity and voltage profile. From the outcome, it is apparent that the after
placement of TCSC in the power system, active power loss of the line, reactive power loss of the line is reduced
+Model
Table 9
Comparative study.
Sr. no. Optimization technique Active power Bus system FACTS device Ref.
transmission
loss in MW
1 Particle swarm optimization 21.586 IEEE-14 bus system TCSC Rashed et al. (2007)
2 Genetic algorithms 21.688 IEEE-14 bus system TCSC Rashed et al. (2007)
3 Differential evolution 21.688 IEEE-14 bus system TCSC Rashed et al. (2011)
4 Differential evolution 9.32 IEEE-30 bus system TCSC Basu (2008)
5 Evolutionary programming 10.42 IEEE-30 bus system TCSC Basu (2008)
6 Genetic algorithms 10.17 IEEE-30 bus system TCSC Basu (2008)
7 Gravitational search algorithm 7.436 IEEE-14 bus system TCSC Benabid et al. (2009)
8 Non-dominated sorting particle 5.29 IEEE-14 bus system TCSC Benabid et al. (2009)
swarm optimization
9 Gravitational search algorithm 5.086 IEEE-30 bus system TCSC Bhattacharyya and
Kumar (2015)
and voltages of the buses are improved. From the result, it is apparent that the minimization of transmission loss
and installation cost of TCSC implemented using TLBO is promising. The TLBO algorithm gives unrivalled and
preferable results than Particle Swarm Optimization and Artificial Bee Colony for the power system of any size. The
TLBO algorithm is effective, requires a few numbers of optimization parameters and useful for the utilities in future
assessment making.
References
Agrawal, R., Bharadwaj, S.K., Kothari, D.P., 2016. Transmission loss and TCSC cost minimization in power system using particle swarm optimization.
Int. J. Innov. Res. Electr. Electron. Instrum. Control Eng. 4 (March (3)), 226–231.
Basu, M., 2008. Optimal power flow with FACTS devices using differential evolution. Electr. Power Energy Syst. 30, 150–156.
Basu, M., 2011. Multi-objective optimal power flow with FACTS devices. Energy Convers. Manage. 52 (February (2)), 903–910.
Benabid, R., Boudour, M., Abido, M.A., 2009. Optimal location and setting of SVC and TCSC devices using non-dominated sorting particle swarm
optimization. Electr. Power Syst. Res. 79 (December (12)), 1668–1677.
Bhattacharyya, B., Kumar, S., 2015. Reactive power planning with FACTS devices using gravitational search algorithm. Ain Shams Eng. J. 6
(September (3)), 865–871.
Duong, T.L., Gang, Y.J., Truong, V.A., 2014. Application of min cut algorithms for optimal location of FACTS devices considering system loadability
and cost of installation. Int. J. Electr. Power Energy Syst. 63 (December), 979–987.
Hingorani, N.G., Gyugyi, L., 2000. Understanding FACTS, Concepts and Technology of Flexible AC Transmission Systems. IEEE Press.
Joshi, S., (Ph.D. Thesis) 1995. Voltage Stability and Contingency Selection Studies in Electrical Power System. IIT Kanpur.
Karaboga, D., Akay, B., 2009. A comparative study of artificial bee colony algorithm. Appl. Math. Comput. 214 (August (1)), 108–132.
Karaboga, D., Basturk, B., 2007. A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm. J.
Glob. Optim. 39 (November (3)), 459–471.
Kim, I.Y., Weck, O.L., 2006. Adaptive weighted sum methods for multiobjective optimization: a new methods for pareto front generation. Struct.
Multidiscipl. Optim. 31 (February (2)), 105–116.
Mathur, R.M., Varma, R.K., 2002. Thyristor-Based FACTS Controllers for Electrical Transmission Systems. John-Wiley.
Nagalakshmi, S., Kamaraj, N., 2012. Comparison of computational intelligence algorithms for loadability enhancement of restructured power system
with FACTS devices. Swarm Evol. Comput. 5 (August), 17–27.
Naresh, G., Ramalinga Raju, M., Narasimham, S.V.L., 2016. Coordinated design of power system stabilizers and TCSC employing improved
harmony search algorithm. Swarm Evol. Comput. 27 (April), 169–179.
Pandiarajan, K., Babulal, C.K., 2016. Fuzzy harmony search algorithm based optimal power flow for power system security enhancement. Int. J.
Electr. Power Energy Syst. 78 (June), 72–79.
Prasad, D., Mukherjee, V., 2016. A noval symbiotic organisms search algorithms for optimal power of power system with FACTS. Int. J. Eng. Sci.
Technol. 19 (March (1)), 79–89.
Rao, R.V., Kalyankar, V.D., 2011. Parameter optimization of advanced machining processes using TLBO algorithm. In: International Conference
on Engineering, Project, and Production Management, 20–21 September, Singapore, pp. 21–32.
Rao, R.V., Savsani, V.J., Vakharia, D.P., 2012. Teaching–learning-based optimization: an optimization method for continuous non-linear large scale
problems. Inf. Sci. 183 (January), 1–15.
Rashed, G.I., Shaheen, H.I., Cheng, S.J., 2007. Optimal location and parameter setting of TCSC by both genetic algorithms and particle swarm
optimization. In: IEEE Conference on Industrial Electronics and Application, 23–25 May, Singapore, pp. 1141–1147.
+Model
Rashed, G.I., Sun, Y., Shaheen, H.I., 2011. Optimal location of thyristor controlled series compensation in a power system based on differential
evolution algorithm considering transmission loss reduction. In: IEEE Proceedings of the 8th World Congress on Intelligent Control and
Automation Taipei, Taiwan, 21–25 June, Taipei, Taiwan, pp. 610–616.
Rashidinejad, M., Farahmand, H., Fotuhi-Firuzabad, M., Gharaveisi, A.A., 2008. ATC enhancement using TCSC via artificial intelligent technique.
Electr. Power Syst. Res. 78 (January (1)), 11–20.
Rezaee Jordehi, A., Jasni, J., Abd Wahab, N., Kadir, M.Z., Javadi, M.S., 2015. Enhanced leader PSO (ELPSO): a new algorithms for allocating
distributed TCSC’s in power system. Int. J. Electr. Power Energy Syst. 64 (January), 771–784.
Satapathy, S.C., Naik, A., 2013. A teaching learning based optimization on orthogonal design for solving global optimization problems. Decis. Sci.
Lett. 2 (March (1)), 23–34.
Verma, M.K., (Ph.D. Thesis) 2005. Voltage Stability Considered Power System Security Assessment and its Enhancement Using FACTS Controllers.
IIT Kanpur.
View publication stats

1 s2.0 S2314717218300114 Main RahulAgrawal PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 s2.0 S2314717218300114 Main RahulAgrawal PDF

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Population Based Evolutionary Optimization Techniques for Optimal

Article · February 2018

Rahul Agrawal Sk Bharadwaj

SEE PROFILE SEE PROFILE

Multilevel Inverters View project

The user has requested enhancement of the downloaded file.

Population based evolutionary optimization techniques for optimal

Rahul Agrawal a,∗ , S.K. Bharadwaj b,1 , D.P. Kothari c,2

Keywords: ABC; FACTS; PSO; TCSC; TLBO; Transmission loss

Fig. 1. The variable reactance model of TCSC.

3.1. Minimize the active power loss

3.2. Voltage profile improvement

3.3. Minimize the installation costs

3.3.1. Equality constraints

i − Qi − Qi (δ1 , δ2 . . .δn , |V1 | |V2 | . . . |Vn |) = 0

3.3.2. Inequality constraints

4.1. Implementation of TLBO for the optimal location of TCSC

MD = [m1 , m2 , m3 . . .mD ] (25)

Step IV: Set the iteration k

Xteachers,D = fmin,D (26)

xinew = xi + Diffrence meanD (30)

4.2. Implementation of PSO for the optimal placement of TCSC

The working of PSO algorithms is presented as follows

Step VI: Randomly generate the velocity of the ith particle as

xik+1 = xik + Vik+1 (36)

Step XI: Check the termination criterion, otherwise go to step II.

Parameter Value Parameter Value Parameter Value

Number of iterations 35 Maximum number of iterations 35 Termination criterion 35

4.3. Implementation of ABC for the optimal Placement of TCSC

Step I to step III are same as PSO algorithm.

5. Test results and discussion

Fig. 2. The flow chart of the ABC algorithms with TCSC.

5.1. Tools and test system

(a) Test system 1 (TS 1) – IEEE 14 bus system,

5.2. Results for real power transmission loss with TCSC

1 IEEE-14 bus system 11.1650 7.8779 7.3701 7.2610

Fig. 3. Active power loss of test system 1 with optimization methods.

Fig. 4. Active power loss of test system 2 with optimization methods.

Fig. 5. Active power loss of test system 3 with optimization methods.

5.3. Results for optimal location and sizing of TCSC

5.4. Results for reactive power transmission loss with TCSC

1 IEEE-14 bus system 29.2705 19.4431 16.6165 15.2473

1 8 3 0 0 0 0 1.982 1.829 1.79 2.122

5.5. Results for line loss with TCSC

5.6. Results for the Bus Voltage with TCSC

PSO ABC TLBO

1 1.0600 1.0600 1.0600 1.0600

Fig. 6. Voltage of test system 1 with optimization methods.

Fig. 7. Voltage of test system 2 with optimization methods.

Fig. 8. Voltage of test system 3 with optimization methods.

1 8 3 28.836 28.098 30.705 28.418 6.754 4.685 0.334 4.644

5.7. Line flow

5.8. Comparative study

View publication stats

You might also like