Multi-stage voltage control in high photovoltaic based distributed generation penetrated distribution system considering smart inverter reactive power capability

This research aims to investigate the impact of using the reactive power capability of PV smart inverters, which can function as distributed static compensators (DSTATCOMs) during non-feed-in hours, to address voltage control issues in high photovoltaic based distributed generation penetrated distribution system.

Multi-stage voltage control in high photovoltaic based distributed
generation penetrated distribution system considering smart inverter
reactive power capability
Vinay Kumar Tatikayalaa,⇑
, Shishir Dixitb
Department of Electrical Engineering,Madhav Institute of Technology and Science,Gwalior,India
a r t i c l e i n f o
Article history:
Received 7 September 2022
Revised 28 February 2023
Accepted 27 March 2023
Available online xxxx
Keywords:
Distribution system
Photovoltaic
Reactive power
Smart inverter voltage control
Power loss
a b s t r a c t
The intermittent nature ofphotovoltaic (PV) based distributed generation can cause voltage control
issues.This research aims to investigate the impact of using the reactive power capability of PV smart
inverters, which can function as distributed static compensators (DSTATCOMs) during non-feed-in hours,
to address this problem. In other words, the suggested PV-DSTATCOM can be used to provide voltage con-
trol whenever there is a high demand placed on the system around the clock. This study presents a coor-
dinated multi-stage voltage control (CMSVC) strategy that utilizes both PV-DSTATCOMs and traditional
voltage control devices through a hybrid of local and centralized control algorithms.The goal is to min-
imize energy waste while maintaining a voltage that is within acceptable limits.To achieve the best
results,an improved whale optimization algorithm has been proposed for optimal optimization.To test
the proposed method, the IEEE 33 bus radial distribution system and IEEE 69 bus radial distribution sys-
tem were evaluated. According to the findings, the solution offered in this research significantly reduces
energy losses and voltage variations,demonstrating the effectiveness of the proposed method
Ó 2023 THE AUTHORS.Published by Elsevier BV on behalf of Faculty of Engineering,Ain Shams Uni-
versity. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/license
by-nc-nd/4.0/).
1. Introduction
The increasing use of photovoltaic (PV) based distributed gener-
ation (DGs) in low voltage (LV) grids has the potential to signifi-
cantly impact the distribution system’s operation [1–3]. To
address these challenges,Volt/VAR control (VVC) utilizing voltage
control devices presents itself as a viable solution [4–5]. Traditional
voltage controldevices,including capacitor banks (CBs) and on-
load tap changer (OLTC) transformers,have long been utilized to
achieve conventional VVC.However,their frequent use results in
a shortened lifespan [6].In contrast,smart inverter interfaced PV
generation is gaining traction due to its adaptable modes of oper-
ation, such as volt/var mode and volt/watt mode [7–9]. As a result,
utilizing smart inverters for VVC presents itselfas a promising
alternative.By effectively utilizing these inverters,the LV grid
can be stabilized without causing damage to the voltage control
devices. Additionally,their adaptable modes of operation can help
lessen the negative impact of high diffusion levels of PV-based DGs
on the LV grid.
The article addresses the subject of high DG allocation and volt
VAR control.It offers helpful insights into the issues that are con-
nected with integrating large volumes ofdistributed generation
into electrical network as well as potential solutions to those chal-
lenges.The authors of [10] utilised the Manta Ray Foraging opti-
mization algorithm (MRFO) to optimise the capacity and
allocation of distributed generation (DG) Type I in order to reduce
power losses in radial distribution networks (RDNs).The study
provided conclusive evidence that MRFO is an effective method
for solving the issue of scattered generators.Chaotic Maps Inte-
grated Stochastic Fractal Search (CMSFS) is a revolutionary
approach that was developed by authors in the article [11] to
address the optimum distributed energy resources placement
problem in radial distribution networks. It was demonstrated that
the technique that was provided was successful in determining the
best possible solutions for the issue.The authors of paper [12]
explored the use of four bio-inspired optimization algorithms to
optimise the placement of three distributed generation (DG) units
in a power system under load uncertainties.These algorithms
included Grey Wolf Optimization (GWO), Manta Ray Foraging
Optimization (MRFO),Satin Bower Bird Optimization (SBO),and
https://doi.org/10.1016/j.asej.2023.102265
2090-4479/Ó 2023 THE AUTHORS.Published by Elsevier BV on behalf of Faculty of Engineering,Ain Shams University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑ Corresponding author at: Department of Electrical Engineering, Madhav
Institute of Technology and Science,Gwalior, Madhya Pradesh 474005,India.
E-mail addresses:[e-mail] (V.Kumar Tatikayala),shishir.
[e-mail] (S.Dixit).
Ain Shams Engineering Journal xxx (xxxx) xxx
Contents lists available at ScienceDirect
Ain Shams Engineering Journal
j o u r n a lhomepage: w w w . s c i e n c e d i r e c t . c o m
Please cite this article as: V.Kumar Tatikayala and S. Dixit,Multi-stage voltage control in high photovoltaic based distributed generation penetrated dis-
tribution system considering smart inverter reactive power capability,Ain Shams Engineering Journal,https://doi.org/10.1016/j.asej.2023.102265

Whale Optimization (WOA).The research showed that optimiza-
tion techniques are effective in solving the problem of distributed
generators,as evidenced by the study. In the article [13], the
authors advocated for the utilisation of the Bald Eagle Search
(BES) optimization method for the purpose of allocating shunt
reactive compensators (SRC) and distributed generation (DG) units
with higher usage capacities into distribution systems in order to
reduce power loss.The study provided conclusive evidence that
BES is an effective solution to the challenge posed by distributed
generators.The authors of [14] used the Cuckoo Search algorithm
to enhance the benefits of the Static Var Compensator (SVC) and
synchronous generator (SG) controller at the port of the induction
generator (IG) in order to minimise frequency and terminal voltage
variations. This was accomplished by improving the gains of the SG
controller and the SVC.The results of the investigation revealed
that the method is effective in resolving the DNR issue. The authors
of the article [15] suggested using the Artificial Hummingbird
Algorithm (AHA) to optimise the location and capacities of renew-
able distributed generators sources within a power system.The
strategy that was suggested took into account a number of differ-
ent goals, such as lowering total cost and emissions, reducing volt-
age deviation, and enhancing voltage stability, all while taking into
account the uncertainties connected with the loading and RDG
output power.To solve the problem of DNR,the authors of [16]
suggested applying the stochastic fractalsearch (SFS) algorithm.
It was demonstrated thatthe SFS algorithm was successful in
determining the best possible solutions for the problem.In [17]
and [18],researchers created a stochastic model with the purpose
of scheduling distributed generation (DG) systems that are pow-
ered by renewable energy sources and managing energy consump-
tion. A MILP solver was used in order to find an answer to the
optimization problem, which led to a solution that was both effec-
tive and efficient. In order to achieve optimal scheduling of renew-
able energy generation and energy management in modern power
systems,this study demonstrates the significance of advanced
modelling and optimization approaches.
One effective technique for managing voltage fluctuations in
distribution systems is through the use of Volt Var Control (VVC)
strategies.Some of the metaheuristic techniques used forVVC
include honey bee mating optimization based on enhanced chaotic
scheme,genetic algorithm,non-dominated sorting genetic algo-
rithm, enhanced grey wolf optimization,and particle swarm opti-
mization. While these techniques have been effective in
minimizing energy loss and optimizing VVC devices in distribution
systems,they have not considered the smart inverter’s reactive
power capability in relation to voltage management.In [19], a
honey bee mating optimization based on an enhanced chaotic
scheme was presented to estimate the active and reactive power
dispatch of DGs, reactive power compensation from capacitor
banks,and tap positions of OLTC transformers.Daily volt/var con-
trol was performed in [20], and a genetic algorithm (GA) was used
in [21] to ensure optimal VVC device operation in active distribu-
tion systems. Similarly, a non-dominated sorting genetic algorithm
(NSGA-II) was used in [22] to simultaneously optimize peak load
reduction and energy loss minimization in the distribution system.
An enhanced grey wolf optimization (IGWO) was devised in [23] to
study the combined effect of VVC control devices and distribution
network reconfiguration for loss minimization.A particle swarm
evolutionary algorithm was adopted in [24] to ensure that voltage
control devices in distribution systems operate efficiently,and an
ideal coordination of OLTCs and static reactive power compen-
sators was proposed in [25] to minimize overall line losses. Particle
swarm optimization (PSO) was used in [26] to set up VVC devices
in the best possible way while taking dispersed energy into
account to minimize energy loss. However, according to the litera-
ture reported in [19]-[26],the reactive power injection capability
of smart inverters has not been fully explored for voltage manage-
ment, with most studies focusing only on active power injection
from PV-based DGs. Additionally, PV-based DGs are often operated
individually [24]-[29], with little consideration given to multi-
stage coordinated voltage control strategies for loss minimization
and voltage regulation.To address these issues,optimal reactive
power dispatching from PV inverters,capacitor banks,and OLTCs
in a distribution system was accomplished in [27] to minimize loss
and voltage deviation.In [28],the energy savings from coordinat-
ing VVC devices with a solar PV inverter were calculated.In [29],
a voltage controlloop was implemented in PV inverters to keep
the voltage within acceptable bounds by absorbing or supplying
reactive power.A smart inverter controlstrategy was suggested
in [30] for high photovoltaic (PV) penetration in distribution sys-
tems,while [31] demonstrated the significant impact of cascaded
voltage regulators in high PV-penetrated distribution networks.
In [32], benefits of VVC have been validated in realtime digital
systems.
In summary, while various VVC techniques have been proposed
to estimate active and reactive power dispatch of DGs, compensate
reactive power from capacitor banks,and adjust tap positions of
OLTC transformers,little attention has been given to the reactive
Nomenclature
Indices
i notation for Bus
h notation for hour
T notation for time
Nb number of buses in the network
XCB; XPV set of capacitor banks installed buses and PV-
DSTATCOM installed buses
Parameters
Vmin, Vmax voltage magnitude minimum and maximum limits
respectively
DqCB
i fixed step change of capacitor bank
Smax
PV;i apparent power of PV smart inverter at ith bus
Qmax
PVDST;i maximum PV-DSTATCOM reactive power at ith bus
Variables
Qh
loss;i; Ph
loss;i reactive power loss and active power respectively in
a branch connected to the ith bus at hth hour
Qh
grid; Ph
grid reactive power and active power taken from grid
respectively at hth hour
Qh
dem;i; Ph
dem;i reactive power and active power demand at ith bus
respectively at hth hour
Qh
CB;i reactive power injected by capacitor banks at ith bus
Qh
PVDST;i; Ph
PV;i reactive power and active power from PV-
DSTATCOM
taph step change of on load tap changer (OLTC) transformer
at hth hour
steph
i variable Step change of capacitor bank at hth hour
Vh
i voltage at ith bus at at hth hour
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
2

power injection capability of smart inverters for voltage manage-
ment. Moreover, coordinated voltage controlstrategies for loss
minimization and voltage regulation in PV-based DGs have yet to
be fully explored.The contributions of this research paper can be
summarized as follows:
Development of a time series model: A time series modelof
synchronized VVC scheme has been developed to minimize
energy loss and voltage variations in active distribution net-
works. This model provides a framework for the coordinated
control of both conventional and cutting-edge VVC devices.
Introduction of coordinated multi-stage voltage control
methodology: A coordinated multi-stage voltage control
(CMSVC) methodology has been suggested,which takes into
account both traditional and advanced VVC devices. The CMSVC
approach provides a hybrid of local and centralized control
algorithms to improve the effectiveness of voltage control.
Enhanced Grey Wolf Optimization: Grey wolf optimization
(GWO) has been improved and applied to the scheduling of
the mixed-integer nonlinear programming (MINLP)problem,
without relaxation or linearization. This optimization technique
enhances the efficiency and effectiveness of the proposed
approach.
Proposed PV smart inverter control approach: A PV smart inver-
ter control approach for local reactive power voltage (Q-V) has
been proposed, which enhances the voltage control capability of
PV-based DG systems.
Investigation of the impact of high PV penetration: The impact
of high penetration ofPV-based DG on the voltage profile in
active distribution networks has been explored.This analysis
provides insights into the effect of high PV penetration on volt-
age control and highlights the need for advanced control
techniques.
Validation of an autonomous volt/VAR droop controller: An
autonomous volt/VAR droop controller has been validated for
use in dynamic voltage controlunder cloud cover conditions.
This validation provides evidence ofthe effectiveness ofthe
proposed approach in real-world scenarios.
Verification on established distribution systems: The proposed
coordinated approach has been verified on established 33 bus
and 69 bus distribution systems. This verification demonstrates
the practicality and effectiveness of the proposed approach in
real-world scenarios
The paper is organized as follows.In Section 2, we introduce
the proposed architecture for the multistage coordinated voltage
control methodology.This architecture lays the foundation for
the rest of the paper and provides a clear understanding ofthe
proposed methodology.In Section 3,we provide a mathematical
formulation of the research objective.This is an important sec-
tion as it outlines the objectives of our research and provides
the necessary background for the implementation of the method-
ology. Section 4 is the core of the paper where we detail the
implementation ofthe proposed multistage coordinated voltage
control methodology.We describe the execution ofthe central-
ized algorithm using an improved whale optimization technique,
as well as the proposed local control. This section provides a
comprehensive overview ofthe methodology and the technical
details of its implementation. In Section 5, we summarize the
conclusions and conversations ofthe paper. This section high-
lights the significance of our research and the implications of
our methodology for future research.Finally, in Section 6, we
provide a closing section that covers the conclusion of the paper.
This section provides a summary of the main findings and high-
lights the contributions of our research.
2. Proposed coordinated multi stage voltage control scheme
The proposed multistage coordinated voltage control architec-
ture is a novel solution that addresses the challenges of voltage
regulation in active distribution networks.As illustrated in Fig.1,
the architecture incorporates both centralized and localcontrol
techniques for efficient voltage control. At an hourly level, the cen-
tralized control technique assigns the slowly controlled voltage
devices,such as capacitor banks (CBs) and on-load tap changers
(OLTCs),as well as quickly controlled voltage devices like photo-
voltaic (PV)-distribution static synchronous compensators (DSTAT-
COM). However, CBs and OLTCs are not suitable for handling
sudden changes in renewable generation or load dynamics due to
their slow reaction times. Therefore, the local control scheme han-
dles the scheduling of fast-acting devices within each hour, such as
PV-DSTATCOM,which provides assistance against abrupt changes
in voltage and responds more quickly.The proposed multi-stage
coordinated voltage controlstrategy offers an effective solution
to the challenges of coordinated volt-VAR regulation in active dis-
tribution networks. The suggested approach ensures efficient volt-
age control by incorporating both conventionaland cutting-edge
VVC devices. Moreover, it provides a better choice for practitioners
who are seeking a coordinated volt-VAR regulation strategy that
reduces energy losses and eliminates voltage violations.The effi-
cacy of the proposed architecture was verified through extensive
simulations on the established 33 bus and 69 bus distribution sys-
tem. Overall,the study presents an important contribution to the
field of active distribution network research by offering a new
and effective approach to coordinated volt-VAR regulation.
3. Problem formulation
In this work, our primary objective is to minimize active and
reactive energy losses,as defined in Eq.(1)
OF ¼X T
h¼1
X h
i¼1 Ph
loss;iþ Qh
loss;i ð1Þ
3.1.System operational limits
Active and reactive power equilibrium limits
Ph
grid
XNb
i¼1
Ph
loss;i
XNb
i¼1
Ph
dem;iþ X
i XPV
Ph
PV ;i¼ 0 ð2Þ
Qh
grid
XNb
i¼1
Qh
loss;i
XNb
i¼1
Qh
dem;iþ X
i XCB
Qh
CB;iþ X
i XPV
Qh
PVDST;i¼ 0 ð3Þ
Distribution network voltage magnitude limits
Vmin V h
i V max ð4Þ
Onload tap changing (OLTC) transformer settings limits
ah ¼ 1 þ taph Dtapstep
100 ð5Þ
here,.taph tapmin
; ::::; 1; 0; 1; :::::tapmax
Capacitor banks (CBs) limits
Qh
CB;i ¼ steph
i DqCB
i ; i XCB ð6Þ
steph
i 0; 1; :::::stepmax
f g
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
3

Reactive power limit of PV smart inverter
Qh
PVDST;i¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Smax
PV ;i
2
P h
PV ;i
2
r
; i XPV ð7Þ
Q max
PVDST;i Q h
PVDST;i Q max
PVDST;i ð8Þ
4. Implementation of proposed multi stage coordinated voltage
control methodology
To ensure optimal performance of the distribution system,
appropriate settings for VVC devices and PV-DSTATCOM are com-
puted hourly using a centralised control technique,as detailed in
section II. However, despite these efforts, voltage breaches can still
occur due to unforeseen changes in load or generation. To address
this issue,a local control scheme was developed that relies on the
Volt/VAr droop characteristics of PV-DSTATCOM and obtains reac-
tive power every five minutes. Fig. 2 provides a visual representa-
tion of the proposed coordinated approach thatleverages both
centralised and local control techniques to ensure effective voltage
regulation and loss minimization.By incorporating these strate-
gies, the distribution system can operate at its highest level of effi-
ciency,even in the face of unexpected fluctuations in voltage.
4.1.Centralised control algorithm: Improved whale optimization
algorithm (IWOA)
The article utilizes an improved whale optimization algorithm
(IWOA) as a metaheuristic algorithm for a centralized control algo-
rithm. Metaheuristic algorithms are often employed when tradi-
tional analytic methods are not feasible,or when the problem at
hand is highly complex,nonlinear,or involves a large number of
variables and constraints.Unlike analytic methods thatrely on
mathematical equations and assumptions, metaheuristic algo-
rithms are specifically designed to search for global optima, which
may not be attainable through analytic methods.Furthermore,
metaheuristic algorithms tend to be more resilient and adaptable
than analytic methods,as they can effectively handle incomplete
or noisy data and are applicable to a broad range of problems.
In this study,IWOA is utilized to implement a centralized con-
trol algorithm,whereby each humpback whale’s position in the
algorithm functions as a search agent [33]. By continuously updat-
ing these search agents,the whale optimization algorithm can
identify the best solution to the global optimization problem.
i. Encircling prey: Humpback whales have the ability to detect
and circle their prey when they are hunting.Because the
precise position of the best possible design in the search
region is not known, the WOA algorithm operates under
the presumption that the current best candidate solution is
either the target prey or is extremely near to the optimum.
These behaviours are represented by the equations that
follow
D
! ¼ E:
! XpðitÞ X
! ðiterÞ ð9Þ
X
! ðiter þ 1Þ ¼ Xp
! ðitÞ R
! : D
! ð10Þ
Where it is the current iteration. The vectors X
! and X p
! , which
reflect the location of the whale and its prey,respectively. It is fea-
sible to derive the coefficient vectors R
! and E
! using the following
formula
R
! ¼ 2e:r1
! e ð11Þ
E
! ¼ 2r2
! ð12Þ
Because of the iterations involved in the process of managing
exploitation and exploration,the exploration rate ‘e’ drops from
two to zero throughout the course of the procedure.The factor is
expressed ase = 2–2.iert/itermax
.
ii. Spiral updating position and bubble-netassaultapproach:A
spiral-shaped path and a diminishing circle are both used
by humpback whales to swim around their prey.The likeli-
hood of selecting either the spiralmodel or the shrinking
encircling mechanism to update the position of whales dur-
ing optimization is taken to be 50% in order to model this
simultaneous behaviour.
X
! ðiter þ 1Þ ¼
Xp
! ðiterÞ R
! : D
! E:
! X
! pðiterÞ X
! ðiterÞ ifprob < 0:5
D
! ehk cosð2pkÞ þ X
! pðiterÞ ifprob 0:5
8
<
: ð13Þ
Where prob is a random number in [0,1].
Fig. 1. Framework of proposed coordinated multistage voltage control methodology.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
4

iii. Search for prey: In the exploration phase,as opposed to the
exploitation phase, we update the position of a search agent
according to a randomly selected search agent rather than
the best search agent discovered thus far.This technique
and |A| > 1 emphasise exploration and enable a worldwide
search to be carried out via the WOA algorithm.
D
! ¼ E:
! X
! randðiterÞ X
! ðiterÞ ð14Þ
X
! ðiter þ 1Þ ¼ Xrand
! ðiterÞ R
! : D
! ð15Þ
4.2.Improvements in whale optimization algorithm
(a) The exploration rate ‘e’ varies linearly in standard WOA,
however this causes a lack of updating between the agents
responsible for exploration and exploitation with regards
to current iteration, which results in the local optimal. In this
paper,in order to incorporate the active variation of ‘e’. The
subsequent improvement can be made to (16)
e¼ 1 iter
itermax 1 d: iter
itermax
1
ð16Þ
Where, d is the nonlinear adjustment index in the range (0,1).
(b) Due to the fact that WOA is a memory-less algorithm,the
individual data from the previous generation of agents has
not been exploited by the new generation. This is the situa-
tion as a direct result of the technique’s blatant trans-
parency. Due to the lack of variety among agents,it is
more probable that obtained revised locations will trapped
in localised optima.This is because localised optimals tend
to be more efficient. The updating of the grey wolf positions
has been altered as follows in order to take advantage of the
data that was supplied at the prior personal best and global
best positions.
X
! ðiter þ 1Þ ¼ X
! ðiterÞ þ n
! ðiter þ 1Þ ð17Þ
n
! ðiter þ 1Þ ¼
q:v! ðitÞ þ c1:r2: Xpbest
! ðiterÞ X
! ðiterÞ
þc2:r3: Xgbest
! ðiterÞ X
! ðiterÞ
8
><
>:
9
>=
>; ð18Þ
Where c1 denotes the particular coefficient and c2 denotes the com-
munication coefficient. The notation Xpbest
! denotes the best individ-
ual position in the region,whereas the notation Xgbest
! denotes the
historically best position on a global scale in the area.The inertia
weight is denoted byq, and it may be found in (19)
qðiterÞ ¼ itermax
itermax qinitial qfinal þ qfinal
ð19Þ
The location of the finest whale is described by the initial term
of (18). This gives the agents in the search area the fundamental
push that they need. The second term relates to the distinct
thoughts that are going through the heads of each search agent
as they get closer to the greatest spot that has been found so far.
The third term is a reflection of the cooperative influence that
the whale agents had in the process of identifying the global
optima.
To implement the improved WOA,the paper provides a flow-
chart for the centralised control algorithm, which includes the
steps for executing the IWOA technique as portrayed in Fig.3.
Overall, the improved WOA technique has the potential to enhance
the optimization process and lead to better globaloptimization
solutions.
4.3.Pseudo code and discussion of IWOA algorithm
The Improved Whale Optimization Algorithm (IWOA) is an opti-
mization algorithm based on the behavior ofhumpback whales.
The algorithm is designed to find the optimal solution of a given
problem by simulating the hunting behavior of humpback whales.
Fig. 2. Execution of proposed coordinated multi stage voltage control scheme.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
5

The pseudo code of IWOA outlines the step-by-step process that
the algorithm follows to search for the optimal solution.
Initialize data set and the whale’s population Xi (i = 1,2, 3 . . .
n)
Calculate the fitness of each search agent using (1) for the
current optimal value
iter = 1
While (t < maximum number of iterations)
For each search agent
Update R,E, b, k, and prob values
If (prob < 0.5)
If (A < 1)
Perform search for encircling prey by using (9) and (10),
corresponding e by using (16)
Else If (A 1)
Perform search for prey by using (14) and (15),
corresponding e by using (16)
End If
Else If (prob 0.5)
Perform bubble net attack by using (13), corresponding e by
using (16)
⇑(continued)
Initialize data set and the whale’s population Xi (i = 1,2, 3 . . .
n)
End If
Perform proposed update for including the historical and
personal best by using (17),corresponding e by using (16)
End For
Check to see if any search agents wander outside of the
search space,and make the necessary adjustments.
Calculate the fitness of each search agent using (1) for the
current optimal value
Update Yp if there is a better solution
iter = iter + 1
End While
Output: Optimal settings of traditional VVC devices as well as
reactive power settings of PV-STATCOM
Both the pseudo code and the mathematical model of WOA high-
light a number of the benefits that the method offers.The WOA
algorithm offers numerous benefits,including its understandable
structure and notion, use of the most recent optimal solution as
prey, remembering optimal solutions from previous iterations,
Fig. 3. Flow chart of proposed improved WOA algorithm on centralized algorithm.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
6

and balancing exploration and exploitation.It is a swarm intelli-
gence algorithm that continually enhances the initialization ran-
dom solution, giving it superior global exploration capabilities
over individual-based algorithms.
To further improve the WOA algorithm, two enhancements
have been proposed.The first involves dynamically varyinge by
modulating the exploration rate with a nonlinear modulation
index l in the range (0,3) [23].The second enhancement involves
updating grey wolf positions using data from the previous genera-
tion of agents to prevent new locations from getting trapped in
localised optima. This is done by updating positions with a combi-
nation of the best individual and historically best position on a glo-
bal scale in the area.
4.4.Local voltage control algorithm:
To address voltage violations that occur during disturbances in
load or generation,a local control algorithm has been proposed.
The algorithm utilizes the volt/VAR droop features [32],which
are depicted in Fig. 4. The droop feature is proportional to the volt-
age in a piecewise manner, consisting of four points (P1, P2, P3, and
P4). Before the point of connection (P1),the PV-DSTATCOM can
contribute the most current reactive power to the connecting
point. From P1 to P2, the PV-DSTATCOM has the ability to provide
additional reactive power. The region between P2 and P3 is known
as the dead band (DB), where the inverter does not inject or absorb
reactive power.The POC supplies additional reactive power to
power the inverter from P3 to P4,and the inverter consumes all
the reactive power from P4.The mathematical formulation of the
local control algorithm is given by equation (18).
Qt
PVDST;i¼
Qmax
PVDST;i Vt
i < VP1
1
VV P2
2
VP1
1 V P2
2
QmaxPVDST;i
PVDST;i VP1
1 < Vt
i V P2
2
0 VP2
2 < Vt
i V P3
3
VV P3
3
VP4
4 V P3
3
QmaxPVDST;i
PVDST;i VP3
3 < Vt
i V P4
4
Q max
PVDST;i Vt
i > VP4
4
8
>>>>>>>>>>><
>>>>>>>>>>>:
9
>>>>>>>>>>>=
>>>>>>>>>>>;
ð20Þ
5. Outcomes and discussions
In this section, we present the outcomes and discussions of the
research study. The proposed method of multi-coordinated voltage
control has been implemented using the MATLAB environment,
and its efficacy has been analyzed.The control mechanism has
been applied to the IEEE 33 bus [35] and 69 bus [36] radial distri-
bution systems.
Table 1 summarizes the details of the bus systems, considering
mounted OLTC, CBs, and PV-based DGs. Pictorial representations of
the IEEE 33 bus and 69 bus radial distribution systems are shown
in Figs. 5 and 6, respectively.Additionally,the forecasted PV and
load data have been plotted in Fig.7.
To assess the effectiveness of the centralized control procedure,
the suggested improved whale optimization algorithm (IWOA) was
utilized to minimize the overall loss of active and reactive power.
To achieve this, the population size and maximum iterations were
set at 30 and 100 respectively.Additionally,the IWOA’s parame-
ters, including c1 and c2, were established as 0.5,while x final
and x initial were determined to be 0.9 and 0.1,respectively.By
implementing these parameters,the IWOA was able to effectively
optimize the centralized controlprocedure and minimize power
loss.
Table 2 presents the outcomes of four different scenarios: Case
1, which serves as a starting point and does not take PV integration
into account; Case 2, which includes a high penetration of PV; Case
3, which is similar to Case 2 but also incorporates the use of stan-
dard VVC devices (OLTC and CBs); and Case 4, which includes typ-
ical VVC devices in addition to a PV inverter that functions as a PV-
DSTATCOM.
5.1.For test system 1 (IEEE 33 bus system):
Fig. 5 shows the single diagram of modified IEEE 33 bus system.
Capacitor banks have been installed at buses 3, 5, 10, and 24 with a
capacity of 600 kVAR. To handle high PV penetration, it is assumed
that PV is installed at buses 15, 17, and 33 with capacities of
0.8 MW, 1.0 MW, and 0.8 MW,respectively.The permissible volt-
age limit is taken as 0.95 pu to 1.05 pu.
The outcomes of the various cases are summarized in Table 3.
Active power losses and reactive power losses in the system have
been reduced by 9.7% and 5.32%,respectively,in Case 2 compared
to Case 1. However, it was discovered that the minimum and max-
imum voltage magnitudes violated the upper and lower voltage
permissible limits,measuring 0.9069 pu and 1.0545 pu,respec-
tively, due to uncontrolled excessive PV penetration. In Case 3, typ-
ical legacy VVC devices (i.e., CBs and OLTCs) continue to govern the
system, resulting in a reduction of 20.09% in real power losses and
21.79% in reactive power losses compared to Case 1. In the last sce-
nario, Case 4,active power loss has been reduced by 28.87%,and
reactive power losses have decreased by 22.4% compared to Case
Fig. 4. PV-DSTATCOM Volt/VAR droop characteristics for local control algorithm.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
7

1 due to the functioning of typical legacy VVC devices in addition
to the reactive power correction provided by PV-DSTATCOM.
Fig. 8 presents the optimalreactive power dispatch from PV-
DSTATCOM, which shows zero reactive absorption/injection at
12:00 h since the total capacity of the inverter has been utilized
for active power generation only.In the remaining hours,there is
a scope of injection/absorption of reactive power that depends on
the availability of the capacity of the PV smart inverter.
5.2.For test system 2 (IEEE 69 bus system):
Fig. 6 shows the single diagram of modified IEEE 69 bus system.
Capacitor banks have been installed at buses 9,18, 22, 57, and 62
with a capacity of 600 kVAR.To handle high PV penetration,it is
assumed that PV is installed at buses 15, 27, and 65 with capacities
of 0.8 MW, 1.0 MW, and 0.4 MW, respectively.The permissible
voltage limit is taken as 0.95 pu to 1.05 pu.
Table 4 summarizes the outcomes under different cases. In Case
2, a reduction of 3.58% in real power losses and 7.41% in reactive
power losses was achieved.However,it was noted that the mini-
mum and maximum voltage magnitudes exceeded the permissible
limits of 0.95 and 1.05 pu,with values of 0.9099 pu and 1.054 pu,
respectively.This is attributed to the high PV penetration without
proper control.
In Case 3,a significant reduction of 40.48% in real power losses
and 42.88% in reactive power losses was observed due to the con-
trol of traditional legacy VVC devices,such as shunt capacitor
banks and OLTCs.Case 4 resulted in the highest reduction of real
power losses and reactive power losses,with values of 41.26%
and 43.49%,respectively,achieved through the combined opera-
tion of traditional legacy VVC devices and reactive power compen-
sation from PV-DSTATCOM.
The optimal reactive power dispatch from PV-DSTATCOM is
illustrated in Fig. 9, which shows zero reactive absorption/injection
during the 12:00 h since the total capacity of the inverter was uti-
lized for active power generation only. During the remaining hours,
the injection/absorption of reactive power was found to depend on
the availability of capacity of the PV smart inverter, as indicated by
the positive and negative values in the figure denoting the injec-
tion and absorption of reactive power into the distribution system.
Overall, the results obtained from the analysis demonstrate the
effectiveness of the proposed approach in reducing power losses
and improving voltage stability in the IEEE 33 bus and 69 bus sys-
tem,highlighting the importance of proper control and utilization
of legacy VVC devices and PV-DSTATCOM in achieving optimal
performance.
5.3.Impact on active and reactive power losses:
The active and reactive power losses of the IEEE 33 bus system
were examined in four different cases over the course of a typical
day, and the results were plotted in Figs. 10 and 11. It can be seen
that losses were highest in Case 1,when there was no integration
of photovoltaic (PV) energy.However, losses were significantly
reduced during the hours of 13:00 to 19:00 with the integration
of PV, as shown in Case 2. In Case 3, the losses were further reduced
by the use of traditional voltage and var control (VVC) devices in
association with PV. Finally, in Case 4, the implementation of smart
inverters in addition to traditional VVC devices led to a significant
reduction in losses.Similar results were observed for the IEEE 69
bus system,as shown in Figs.12 and 13. These findings suggest
that the use of PV energy,along with traditionaland smart VVC
Table 1
Details of test systems.
Parameters Test system 1 Test system 2
Bus system IEEE 33 bus IEEE 69 bus
voltage 12.66 KV 12.66 KV
active power demand (MW) 3.715 MW 3.801 MW
reactive power demand (MVAR) 2.30 MVAR 2.693.6 MVAR
Capacitor bank mounted locations 3,5,10,15,24 9, 18, 22, 57, 62
Capacitor bank rating (MVAR) 0 to 0.600 0 to 0.600
PV plant mounted locations 15,17,33 15, 27, 65
PV plant capacity (MVA) 0.8,1.0,0.8 0.8,1.0,0.4
allowable voltage bounds 0.95 pu to1.05 pu 0.95 pu to1.05 pu
Fig. 5. Modified single line diagram of IEEE 33 bus distribution system.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
8

devices, can effectively reduce power losses in distribution
systems.
5.4. Impact on voltage profile at high PV generation and high load hour
Fig. 7 shows that the highest photovoltaic (PV) generation was
observed at 12:00. To study the voltage behavior of the distribution
system under high PV generation conditions,Figs.14 and 15 pre-
sent the voltage profile at various buses of the IEEE 33 bus system
and IEEE 69 bus system,respectively. It was observed that in Case
1, where there was no PV integration,the voltage magnitude vio-
lated the minimum permissible limit. In Case 2, although the volt-
age magnitude managed to stay above the lower limit,the upper
voltage limit was violated due to the high penetration ofPV in
the distribution system. In Case 3, the operation of traditional volt-
age and var control (VVC) devices helped to keep the voltage mag-
nitude within both the lower and upper permissible limits.
However,the voltage magnitude profile was lower compared to
Case 4. In Case 4, the operation of traditional VVC devices and
PV-STATCOM led to a healthy and effective voltage profile during
the high PV generation hour.
Fig. 7 shows that the highest load was observed at 19:00.To
study the voltage behavior of the distribution system under high
loading conditions,Figs. 16 and 17 present the voltage profile at
various buses of the IEEE 33 bus system and IEEE 69 bus system,
respectively.It was observed that in Cases 1 and 2,where there
was no PV power, the voltage magnitude violated the minimum
permissible limit. In Case 3, the operation of traditional voltage
and var control (VVC) devices helped to keep the voltage magni-
tude within both the lower and upper permissible limits, although
the voltage magnitude profile was lower compared to Case 4.In
Case 4, the operation of traditional VVC devices and PV-
STATCOM led to a healthy and effective voltage profile during
the high loading hour. These findings suggest that the use of tradi-
tional VVC devices and PV-STATCOM can help to maintain a stable
voltage profile in the distribution system under high PV generation
as well as high loading conditions.
To further analyze the voltage behavior of the distribution sys-
tems, heat curve graphs of voltage magnitude profiles correspond-
Fig. 6. Modified single line diagram of IEEE 69 bus distribution system.
Fig. 7. Forecasted Load,and PV generation output.
Table 2
Cases study.
Cases PV OLTC SCBs PV-DSTATCOM
Case 1
Case 2 U
Case 3 U U U
Case 4 U U U U
Udenotes considered,denotes not considered.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
9

ing to all buses at different hours and for different cases have been
plotted in Figs. 18 and 19 for the IEEE 33 bus system. Similarly, an
analysis was conducted for the IEEE 69 bus system and presented
in Figs. 20 and 21. These figures provide a visual representation of
the voltage magnitude profile at each bus under various operating
conditions. The heat curve graphs can be used to identify potential
voltage violations and to assess the effectiveness of voltage control
strategies.
5.5.Efficacy of local control procedure with cloudy condition
Table 5 shows that there will be a sudden change in the amount
of photovoltaic energy produced at 12:15,12:30,and 12:45.As a
result, the total power generated by photovoltaic cells willdrop
from 2600 kW to 0 kW and from 2200 kW to 0 kW in the IEEE
33 bus and IEEE 69 bus systems, respectively.To mitigate the
effects of this sudden change,local control has been implemented
using the recommended droop parameter from Ref.[2]. As shown
in Fig. 22 and Fig. 23 for the IEEE 33 bus and 69 bus systems,
respectively, the voltage of the system drops below the lower per-
missible limit (0.95 pu) between 12:15 PM and 12:45 PM when the
sky is partly overcast or cloudy.This is due to the reduction in PV
active power output under such conditions.To provide the neces-
sary VAR support during this time period,the PV-DSTATCOM is
used as shown in Fig.24 and Fig.25 for the IEEE 33 bus and 69
bus systems, respectively. This device effectively raises the voltage
of the system above the lower permissible limit using available
reactive power handled by equation (18).
5.6.Performance of different metaheuristic algorithms
In this research paper,we have evaluated the performance of
five different metaheuristic algorithms,namely GA, GWO, AHA,
WOA, PSO and IWOA,on a particularly difficult scenario,referred
to as case 4.As shown in Fig.26, the convergence pattern of each
algorithm is depicted,and it is observed that IWOA approaches
its minimal value of 2209.766,while GA, GWO, AHA, WOA, and
PSO converge at 2580.95,2549.2, 2498.9, 2334.9, 2280.95 and
2255.13,respectively for 33 bus system.Similarly for 69 bus sys-
tem, it is observed that IWOA approaches its minimal value of
1701.33,while GA [21],GWO [23],AHA [15],WOA [33],and PSO
[34] converge at 2100.8,2056.89,1921.98,and 1798.98,respec-
tively in Fig. 27. To further compare the performance of these algo-
rithms, we present the most favourable,least favourable,and
average values of aforementioned algorithms in Table 6 and Table 7
for 33 bus system and 69 bus system respectively. As evident from
the table, IWOA outperforms all the other algorithms in every
respect for both test systems.Therefore,it can be concluded that
IWOA has a decisive advantage over the aforementioned algo-
rithms for this particular scenario. To achieve results that are even
closer to optimal, we suggest making adjustments to the
diversification-intensification balance,keeping track ofhistorical
Table 3
Outcomes under various cases for 33 bus system.
Parameters Case 1 Case 2 Case 3 Case 4
Active power losses (kWh) 2497.71 2255.52 1995.99 1776.65
Percentage reduction in active power losses (%) —— 9.70 20.09 28.87
Reactive power losses (kVARh) 1693.28 1603.13 1324.40 1314.00
Percentage reduction in reactive power losses (%) ——— 5.32 21.79 22.40
Minimum voltage magnitude (p.u) 0.9037 0.9069 0.95 0.969
Maximum voltage magnitude (p.u) 1 1.0545 1.036 1.05
Fig. 8. Reactive power injection/absorption by PV-DSTATCOM under case 4 for 33
bus system.
Table 4
Outcomes under various cases for 69 bus system.
Parameters Case 1 Case 2 Case 3 Case 4
Active power losses (kWh) 2653.26 2558.2 1579.31 1558.57
Percentage reduction in active power losses (%) 3.58 40.48 41.26
Reactive power losses (kVARh) 1207.23 1117.8 689.56 682.19
Percentage reduction in reactive power losses (%) 7.41 42.88 43.49
Min. voltage (p.u) 0.9091 0.9099 0.966 0.9665
Max. voltage (p.u) 1 1.054 1.05 1.05
Fig. 9. Reactive power injection/absorption by PV-DSTATCOM under case 4 for 69
bus system.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
10

Fig. 10. Active power loss under four different cases for 33 bus system.
Fig. 11. Reactive power loss under four different cases for 33 bus system.
Fig. 12. Active power loss under four different cases for 69 bus system.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
11

Fig. 14. Voltage magnitude profile under different cases for 33 bus system at high PV generation at 12:00.
Fig. 15. Voltage magnitude profile under different cases for 69 bus system at high PV generation at 12:00.
Fig. 13. Reactive power loss under four different cases for 69 bus system.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
12

Fig. 17. Voltage magnitude profile under different cases for 69 bus system at peak loading hour at 19:00.
Fig. 18. Voltage magnitude profile for 33 bus system (a) case 1 and (b) case 2.
Fig. 16. Voltage magnitude profile under different cases for 33 bus system at peak loading hour at 19:00.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
13

Fig. 20. Voltage magnitude profile for 69 bus system (a) case 1 and (b) case 2.
Fig. 21. Voltage magnitude profile for 69 bus system (a) case 3 and (b) case 4.
Fig. 19. Voltage magnitude profile for 33 bus system (a) case 3 and (b) case 4.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
14

Fig. 22. Without local control for 33 bus system.
Fig. 23. Without local control for 69 bus system.
Table 5
PV output status: fully cloud and partly day condition.
Time 12:00 12:15 12:30 12:45
PV generation (p.u) @ cloudy day 1 0.6 0.2 0
Aggregated PV production (kW) @ cloudy sky: for 33 bus system 2600 1560 520 0
Aggregated PV production (kW) @ cloudy sky: for 69 bus system 2200 1320 440 0
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
15

Fig. 25. With local control for 69 bus system.
Fig. 26. CONVERGENCE pattern of different algorithms for 33 bus system.
Fig. 24. With local control for 33 bus system.
Fig. 27. Convergence pattern of different algorithms for 69 bus system.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
16

bests, and utilizing a binary updating mechanism. By doing so, the
IWOA algorithm can potentially produce even better results than
those obtained in this research.
Overall, our findings indicate that the IWOA algorithm is a
promising approach for solving difficult optimization problems,
and it can be further improved by incorporating the aforemen-
tioned adjustments.These results have important implications
for various applications that require efficient optimization
techniques.
6. Conclusion
In this research paper,a novel coordinated multi-stage voltage
control (CMSVC) methodology for power loss minimization has
been proposed, and its effectiveness has been evaluated. The study
showcasesthe potential of the proposed approach to achieve
greater energy savings than the conventionalVVC technique by
effectively combining both conventional VVC devices and PV smart
inverters. Additionally, the proposed CMSVC methodology can
handle cloud movements without exceeding safe operating voltage
limits. To optimize the proposed approach,an improved whale
optimization algorithm has been developed and tested on the IEEE
33 bus and IEEE 69 bus radial distribution systems.The results
show that the IWOA algorithm outperforms other conventional
optimization algorithms, leading to better global optimization
solutions.The findings of this research suggest that the proposed
CMSVC methodology is a viable and efficient solution for voltage
control and power loss minimization in distribution systems.Fur-
thermore, the future expansion of this methodology to include net-
worked microgrid operations could enhance its practical
applicability.Therefore,the proposed CMSVC methodology could
contribute significantly to the field of power system optimization
and control.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
References
[1] Vijay Babu P, Singh S, Singh SP. Distributed generators allocation in
distribution system. In: In 2017 IEEE Power & Energy Society General
Meeting.p. 1–5.
[2] Pamshetti Vijay Babu, Singh SP. Coordinated allocation of BESS and SOP in high
PV penetrated distribution network incorporating DR and CVR schemes.IEEE
Syst J 2021;16(1):420–30 [March 2022].
[3] Zare M, Niknam T, Azizipanah-Abarghooee R, Amiri B. Multiobjective
probabilistic reactive power and voltage control with wind site correlations.
Energy 2014;66:810–22.
[4] Mataifa H, Krishnamurthy S, Kriger C. Volt/VAR optimization: a survey of
classical and heuristic optimization methods.IEEE Access 2022:13379–99.
[5] Jabr RA. Robust Volt/VAr control with photovoltaics.IEEE Trans Power Syst;
2019.p. 1–1.
[6] Pamshetti VB, Singh S, Singh SP. Reduction of energy demand via conservation
voltage reduction considering network reconfiguration and soft open point. Int
Trans Electr Energy Syst 2020;30(1):e12147.
[7] Smith JW, Sunderman W, Dugan R, Seal B. Smart inverter volt/var control
functions for high penetration of PV on distribution systems. In: In 2011 IEEE/
PES Power Systems Conference and Exposition.p. 1–6.
[8] Ding F, Baggu M. Coordinated use of smart inverters with legacy voltage
regulating devices in distribution systems with high distributed PV
penetration — increase CVR energy savings.IEEE Trans Smart Grid; 2018.p.
1–1.
[9] Varma RK, Siavashi EM. PV-STATCOM: a new smart inverter for voltage control
in distribution systems.IEEE Trans Sustain Energy 2018;9(4):1681–91.
[10] Hemeida MG,Ibrahim AA, Mohamed AAA,Alkhalaf S,El-Dine AMB. Optimal
allocation of distributed generators DG based Manta Ray Foraging
Optimization algorithm (MRFO).Ain Shams Eng J 2021;12(1):609–19.
[11] Duong TL,Nguyen PT,Vo ND, Le MP. A newly effective method to maximize
power loss reduction in distribution networks with highly penetrated
distributed generations.Ain Shams Eng J 2021;12(2):1787–808.
[12] Hemeida MG, Alkhalaf S, Senjyu T, Ibrahim A, Ahmed M, Bahaa-Eldin AM.
Optimal probabilistic location of DGs using Monte Carlo simulation based
different bio-inspired algorithms.Ain Shams Eng J 2021;12(3):2735–62.
[13] Eid A, Kamel S,Zawbaa HM,Dardeer M. Improvement of active distribution
systems with high penetration capacities of shunt reactive compensators and
distributed generators using Bald Eagle Search.Ain Shams Eng J 2022;13
(6):101792.
[14] Mehta P, Bhatt P,Pandya V.Optimized coordinated control of frequency and
voltage for distributed generating system using Cuckoo Search Algorithm. Ain
Shams Eng J 2018;9(4):1855–64.
[15] Ramadan A, Ebeed M, Kamel S, Ahmed EM, Tostado-Véliz M. Optimal
allocation of renewable DGs using artificial hummingbird algorithm under
uncertainty conditions.Ain Shams Eng J 2023;14(2):101872.
[16] Tran TT, Truong KH, Vo DN. Stochastic fractal search algorithm for
reconfiguration of distribution networks with distributed generations. Ain
Shams Eng J 2020;11(2):389–407.
[17] Tostado-Véliz M, Kamel S, Aymen F, Jordehi AR, Jurado F. A Stochastic-IGDT
model for energy management in isolated microgrids considering failures and
demand response.Appl Energy 2022;317:119162.
[18] Tostado-Véliz M, Kamel S, Hasanien HM, Arévalo P, Turky RA, Jurado F. A
stochastic-intervalmodel for optimal scheduling of PV-assisted multi-mode
charging stations.Energy 2022;253:124219.
[19] Niknam T. A new HBMO algorithm for multiobjective daily volt/Var control in
distribution systems considering distributed generators. Appl Energy 2011;88
(3):778–88.
[20] Malekpour AR, Niknam T. A probabilistic multi-objective daily Volt/Var control
at distribution networks including renewable energy sources. Energy 2011;36
(5):3477–88.
[21] Satsangi S,Kumbhar GB. Effect of load models on energy loss reduction using
volt-var optimization.In: Proc. Nat. Power Syst.Conf.; 2016.p. 1–6.
[22] Padilha-Feltrin A, Rodezno DAQ, Mantovani JRS. Volt-VAR multiobjective
optimization to peak-load relief and energy efficiency in distribution
networks.IEEE Trans Power Deliv.2015;30(2):618–6.
[23] Pamshetti VB, Singh S, Singh SP. Combined impact of network reconfiguration
and volt-var control devices on energy savings in the presence of distributed
generation.IEEE Syst J 2019;14(1):995–1006.
Table 6
Comparative Analysis of different algorithms for 33 bus system.
Algorithm GA GWO AHA WOA PSO IWOA
Best (losses in kVA) 2549.2 2498.9 2334.9 2280.95 2255.13 2209.766
Average (losses in kVA) 2786.809 2703.385 2496.436 2503.876 2448.21 2370.196
Worst (losses in kVA) 4720.701 4663.893 4093.255 4318.266 4299.379 3956.08
Standard deviation (losses in kVA) 480.0451 454.7001 379.0104 450.3812 429.341 376.415
Table 7
Comparative Analysis of different algorithms for 69 bus system.
Algorithm GA GWO AHA WOA PSO IWOA
Best (losses in kVA) 2100.8 2056.89 1921.98 1865.78 1798.98 1701.33
Average (losses in kVA) 2786.809 2703.385 2496.436 2503.876 2448.21 2370.196
Worst (losses in kVA) 3890.338 3838.935 3369.375 3532.271 3429.734 3045.842
Standard deviation (losses in kVA) 395.606 374.2719 311.9836 368.4045 342.4973 289.8072
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
17

[24] Madureira AG, Lopes JP. Coordinated voltage support in distribution networks
with distributed generation and microgrids.IET Renew Power Gener 2009;3
(4):439–54.
[25] Daratha N,Das B,Sharma J.Coordination between OLTC and SVC for voltage
regulation in unbalanced distribution system distributed generation. IEEE
Trans Power Syst 2014;29(1):289–99.
[26] Anilkumar R, Devriese G, Srivastava AK. Voltage and reactive power control to
maximize the energy savings in power distribution system with wind energy.
IEEE Trans Ind Appl 2018;54(1):656–64.
[27] Chen Y, Strothers M, Benigni A. All-day coordinated optimal scheduling in
distribution grids with PV penetration. Electr Pow Syst Res 2018;164:112–22.
[28] Singh S, Singh SP. Energy saving estimation in distribution network with smart
grid enabled CVR and solar PV inverter.IET Gener Transm Distrib 2018;12
(6):1346–58.
[29] Jahangiri P, Aliprantis DC.Distributed volt/VAr control by PV inverters.IEEE
Trans Power Syst.2013;28(3):3429–39.
[30] Smith JW, Sunderman W, Dugan R, Seal B. Smart inverter volt/var control
functions for high penetration of PV on distribution systems. In: Power
Systems Conference and Exposition (PSCE),2011 IEEE/PES,IEEE; March 2011.
p. 1-6.
[31] Chamana M, Chowdhury BH. Optimal voltage regulation of distribution
networks with cascaded voltage regulators in the presence of high PV
penetration.IEEE Trans Sustain Energy July 2018;9:1427–36.
[32] Pamshetti VB, Singh S, Thakur AK, Singh SP. Multistage coordination Volt/VAR
control with CVR in active distribution network in presence of inverter-based
DG units and soft open points.IEEE Trans Ind Appl 2021;57(3):2035–47.
[33] Mirjalili S, Lewis A. The whale optimization algorithm. Adv Eng Softw
2016;95:51–67.
[34] Del Valle Y, Venayagamoorthy GK,Mohagheghi S, Hernandez J, Harley RG.
Particle swarm optimization: basic concepts,variants, and applications in
power systems.IEEE Trans Evol Comput Apr.2008;12(2):171–95.
[35] Baran ME, Wu FF. Network reconfiguration in distribution systems for loss
reduction and load balancing.IEEE Trans Power Del Apr.1989;4(2):1401–7.
[36] Sivanagaraju S,Visali N, Sankar V,Ramana T.Enhancing voltage stability of
radial distribution systems by network reconfiguration.Elect Power Compon
and Syst 2005;33(5):539–50.
V. Kumar Tatikayala and S.Dixit Ain Shams Engineering Journal xxx (xxxx) xxx
18