International Journal of Applied Power Engineering (IJAPE)

Vol. 13, No. 4, December 2024, pp. 982~988

ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i4.pp982-988  982

Power quality enhancement using fully informed particle

swarm optimization based DSTATCOM in distribution systems

Moguthala Shankar, R. Senthil Kumar

Department of Electrical and Electronics Engineering, Saveetha Engineering College (Autonomous), Anna University,
Chennai, Inda

Article Info ABSTRACT

Article history: To compensate for the reactive power, inverter-based conditioners have been
utilized in recent years due to their faster response. Distribution static
Received Dec 26, 2023 synchronous compensator (DSTATCOM) has been utilized to enhance power
Revised Jul 8, 2024 quality in power system that is an inverter-based device that is widely utilized.
Accepted Aug 15, 2024 To control this type of equipment, a proportional integrated (PI) controller has
been utilized to control most of the equipment with respect to certain
parameters. The performance of the controller basically does not meet the
Keywords: expectations because of the dynamics and nonlinearity of a system
parameters. In this present paper, a probabilistic neural network has been used
DSTATCOM in a controller with a fully informed particle swarm optimization (FIPSO)
Fully informed particle swarm algorithm to generate a suitable weight for controlling the axes of various
optimization parameters of DSTATCOMs. Using MATLAB/Simulink software,
Neural network simulations were performed, and the responses were monitored with particular
Power quality regard to the reference reactive parameter. The results are compared.
Probabilistic neural network DSTATCOM improves power system damping.
Proportional-integral This is an open access article under the CC BY-SA license.

Corresponding Author:
Moguthala Shankar
Department of Electrical and Electronics Engineering, Saveetha Engineering College (Autonomous)
Anna University
Chennai, Tamil Nadu, India
Email: [email protected]

1. INTRODUCTION
Numerous researchers have looked into power quality concerns such as neutral current compensation,
voltage sag, swelling, and voltage instability. These problems lead to a slower response, lower power flow
limits, and the system collapse. All these things were a condition for the development of new equipment to
improve the quality of electricity, especially on the customer's side. Power electronic device distribution static
synchronous compensator (DSTATCOM) attempts to restore power quality on the distribution system's source
and load sides [1]. When compared to other devices used methods such as dynamic voltage restorer (DVR),
unified power-quality conditioner (UPQC), interline power-flow controller (IPFC), DSTATCOM offers
numerous advantages like low cost, low loss, size is compact, and less harmonic generation. One of
DSTATCOM's unique benefits is its ability to provide quick, continuous, and uninterrupted inductive or
capacitive compensation. Based on the specified load, the DSTATCOM injects a significant amount of the
leading along with the lagging compensation current to match the utility connection's total demand [2], [3].
The DSTATCOM began to function dynamically in the radial distribution system as a result of this type of
injection. Reactive compensation has been the primary use of this kind of DSTATCOM controller in induction
generators in recent years. The control algorithm typically determines DSTATCOM compliance in each given
application [4], [5] explained DSTATCOM control using a fixed-parameter proportional integral (PI) controller

Journal homepage: http://ijape.iaescore.com

Int J Appl Power Eng ISSN: 2252-8792  983

and offered three different control methods such as P, PI, proportional integral derivative (PID) for producing
reference current components [6]. A DSTATCOM connected to a distributed generation distribution network
features a nonlinear controller to control the grid voltage used self-tuning filters and instantaneous reactive
power theory to construct an algorithm for DSTATCOM. Some researchers used it to dampen the system in
addition to controlling the exchange of reactive power [7], [8]. Reference values for DSTATCOM control are
often acquired via PI controllers, primarily the ‘d’ and ‘q’ axis currents. Data mining is typically done in a
linear controller, which necessitates the use of a mathematical model. In this model, the various parameters
have been tuned to produce the optimum outcomes for a certain region under constrained conditions. These
controllers, which were created using a linear mathematical model, perform poorly in nonlinear system
dynamics, mostly due to load disturbance and parameter changes [9]. Until now, linearized models utilized for
DSTATCOM control were also reported. Some authors created a nonlinear controller for basic power system
models using an intricate Lyapunov technique, which is not applicable to all situations [10], [11].
Recently, fuzzy logic and an artificial neural network controller were utilized to control DSTATCOM
as a substitute for a general controller. The necessity and advantages of fuzzy logic along with the neural
network-based controllers in a grid system were covered in. One significant benefit that will demonstrate good
response times is the controllers created with the expert system do not need a mathematical model [12], [13].
This kind of controller is different from the traditional linear controller in that it allows for a broad variety of
system operating circumstances, regardless of task complexity. Selecting the controller's inputs that are truly
important might improve performance even further; this topic is covered in further detail in this article [14].
Considering all these problems, fully informed particle swarm optimization (FIPSO) is suggested along with a
probabilistic neural network (PNN)-PI controller to control the DSTATCOM 'd' and q axes. PNNs appear to
provide more effective solutions for a range of control issues [15], [16]. The fact that PNNs can solve intricate,
nonlinear, and mathematical problems in real-world dynamical systems certainly draws researchers to employ
them in controller design. Gaining the right weight and losing weight, however, are difficult tasks. Weight
adjustment applications are primarily built by human professionals and rely on many controllers [17].
However, it takes a great deal of expertise, time, and patience, and the tuning and design weight might not be
ideal. There has been a lot of research on estimating the appropriate weight needed for controller design fully
informed particle swarm optimization (FIPSO) is used to extract the weight, this approach has been utilized to
predetermine the weight function on the input properties. Initial settings for these functions are set in order to
accommodate changing parameters. When compared to other algorithms FIPSO algorithm give better accuracy
and takes less time, which can be applicable for all the networks maintaining unique algorithm. Fully informed
PSO requires less terminal parameter data compared to other algorithms. The suggested controller is put into
practice, simulated, and found to have superior qualities than the conventional controller [18], [19]. The work
addresses PNN-PI, PNN FIPSO controller design, and mathematical modeling of DSTATCOM in order of
precedence. The significant initiatives that have fueled and guided particle swarm research, along with some
significant new uses and avenues. The two-step PSO and the PSO-support vector machine (PSO-SVM),
demonstrate the quick adaptation of PSO [20]. Stunning outcomes have also been produced by the PSO's
integration and practical application with the industry-standard method. The inertia weight, mutation operators,
and swarm startup [21]. This overview's primary benefit was emphasizing how crucial it is to introduce several
mutation operators and the inertia weight parameter in order to enhance PSO performance [22]. The PSO
algorithm, as well as doing a theoretical examination of the system. Then, they examined its current use and
research in engineering applications, discrete and parallel PSO algorithms, parameter selection, multi-objective
optimization, algorithm structure, and topological structure. Despite the successful application of PSO in
various domains, there are still challenges that need to be addressed and considered as future research directions
[23], [24]. Over the past few years, PSO has garnered significant interest from researchers and has been applied
in diverse areas [25]. However, there are still critical problems and issues that persist [26], [27]. Therefore, it
is imperative for scholars and researchers to dedicate more research efforts towards overcoming these
challenges and issues that could potentially hinder the future application of PSO. The research community
should focus on addressing new methodologies for complex problems [28], [29].

2. DSTATCOM MODELLING
A 3 MVAR DSTATCOM had been connected in the parallel with load on B3 to raising a load voltage,
as depicted in the given Figure 1. The sources of voltage are directly connected to B1, B2, and B3 through a
network of 21 km feeders, while the generated energy travels through B1, B2, and B3. In order to connect B2
and B3, a 2 km feeder is utilized in this. They have been connected in the B3 via a 25/0.6 step-down distribution
transformer so that both variable and fixed loads can be studied. Connecting a load especially a variable load
is meant to alter the voltage and current flowing through bus B3. The DSTATCOM absorbs or generates
reactive power on the B3 bus to regulate voltage. Using the coupling transformer's leaky reactance to create
a secondary voltage in the phase having the primary voltage, reactive power will be sent to the grid.
Power quality enhancement using fully informed particle swarm optimization … (Moguthala Shankar)
984  ISSN: 2252-8792

The DSTATCOM's pulse-width modulated (PWM) voltage inverter provides voltage support. When the
secondary voltage drops below the system bus voltage, DSATCOM functions as an inductor by absorbing
reactive power conversely, when the secondary voltage rises above the bus voltage, it functions as a capacitor
by producing reactive power. Table 1 gives the parameters and values used for the study of source voltage, line
voltage, and constant and variable load.

Figure 1. System configuration of a distribution system

Table 1. System parameters

System quantities Values
Line voltage 25000 V
Source voltage 25000 V
Feeder resistance R 0.1155 Ω/ km
Feeder capacitance C 11.36 nF/km
Feeder inductance L 1.049 mh/km
Frequency 50 hz
Variable load 1.8 MVA, pf=0.9
1.2 MVA having the Mod. freq of 5 Hz
Fixed load at bus 2 3 MVA, pf=0.998
Fixed load 1 MVA, pf=1
The reference voltage of the DC bus 2400 V
Capacitor of DC bus 10000 microfarad

3. PROBABILISTIC NEURAL NETWORKS (PNNs)

PNNs use a Bayesian decision strategy in a distribution system. A PNN has been guaranteed to
converge with enough training data. It does not require learning and initial network weights. This had been one
of the supervised learning networks. Figure 2 gives the detailed architecture structure for working out with the
different weights (inputs). This network is suitable for any real-time problems because it learns quickly. Input
units without much deviation distribute their values to the unit of the next layer, which is the sample unit. In
each training unit, there has been one pattern unit for every pattern. In every pattern unit, the weight vectors
are duplicates of the corresponding example. During training, the addition of the selected weight vector is
performed. In this work, the training example is differentiated based on the probability density function, which
has been the basic function of a PNN.
A simple partial differential function is as follows.
𝑁𝑘 ‖𝑋−𝑋𝑘𝑗 ‖
1
𝑓𝑘 (𝑋) = ∑ 𝑒𝑥𝑝 ( ) (1)
𝑁𝑘 𝑗=1 2𝜎 2

By adjusting and applying the above equation to output vector H of the hidden layer in the architecture.
∑𝑖(𝑋−𝑋𝑘𝑗 )2
𝐻ℎ = 𝑒𝑥𝑝 ( ) (2)
2𝜎 2

1
Netj = ∑ℎ 𝑁𝑛ℎ 𝑗𝑦 𝐻ℎ and Nj = ∑ℎ 𝑊𝑛ℎ𝑗𝑦
𝑁𝑘

Netj 𝑚𝑎𝑥𝑘 (𝑛𝑒𝑡𝑘 ) then yi = 1, else yj = 0

Here i is the number of input layers, h is the number of hidden layers, j is the number of output layers, k is the
number of training exemplars, N is the number of classifications, 𝜎 is the smoothing parameters, and X is input

Int J Appl Power Eng, Vol. 13, No. 4, December 2024: 982-988
Int J Appl Power Eng ISSN: 2252-8792  985

vector. ‖𝑋 − 𝑋𝑘𝑗 ‖ is the Euclidean distance among the vectors 𝑋 and 𝑋𝑘𝑗 i.e., ‖𝑋 − 𝑋𝑘𝑗 ‖= ∑𝑖 (𝑋 − 𝑋𝑘𝑗 )2 , 𝑊𝑖𝑥ℎℎ
ℎ𝑦
is the connection weight between the input layer 𝑋 and 𝑀 and 𝑊𝑛 𝑗 as connection weight among hidden layer
and the output layers.

Figure 2. Architecture of PNN

4. FULLY INFORMED PARTICLE SWARM OPTIMIZATION

The global version of the particle swarm optimization is the fully informed particle swarm (FIPS).
Like normal PSO, the FIPSO also adapts the best position of every neighbourhood particle as depicted in (3).
𝜑𝑚𝑎𝑥
𝑣𝑖 𝑡+1 = 𝑘 [𝑣𝑖 𝑡 + ∑𝑘=𝑁𝑖 𝑈 ⊗ (𝑝𝑏𝑘 𝑡 − 𝑥𝑖 𝑡 )]
∣𝑁𝑖 ∣

𝑥𝑖 𝑡+1 = 𝑥𝑖 𝑡 +𝑣𝑖 𝑡+1 (3)

Where 𝑥𝑖 𝑡 denoted as the position of the ith particle in iteration ‘t’, 𝑣𝑖 𝑡 denoted as the velocity of the ith particle in
iteration ‘t’, 𝑝𝑏𝑘 𝑡 is the best position of kth particle in iteration ‘t’, k is the constriction coefficient, Ni is denoted
as the set of the neighborhood particles of the ith particle, ∣ 𝑁𝑖 ∣ denoted as the number of the neighborhood particle
of the jth particle, 𝜑𝑚𝑎𝑥 is the learning coefficient, U is the random vector in the interval (0,1), and ⊗ is the
element-by-element vector multiplication. The learning coefficient 𝜑𝑚𝑎𝑥 and constriction coefficient k must be
fixed according to requirements respectively. The population size has to be assumed suitably. The initial position
of the particles is formed randomly. A greater number of iterations is needed for high-dimensional optimization
problems. Here, to solve two locations of the particles were altered to decrease the number of iterations

5. RESULTS AND DISCUSSION

5.1. PNN–PI controller
All that was involved with the well-known classic control system which is frequently employed in
place of other controller types was a linear PI controller. Due to the fact that this controller is linear, it is
unsuitable for a system that deals with nonlinearity. For the nonlinear system, a PNN controller is utilized as
an alternative. Because this PNN controller does not require a mathematical model to control any type of
system, it has been broadly utilized in systems having complex structures. To execute the process. The input
signals for training are determined randomly. Learning is set to a learning rate of 0.01 and a target error of
0.001. 30 data points from every class are taken for testing and 15 data points from each class are used to train
the network. Every time the network repeats, the weights are updated, providing the network with a fresh
training input. 30 randomly chosen samples are used to evaluate the PNN.

5.2. PNN controller by fully informed particle swarm optimization

This method is carried out using a set of error-free, two-dimensional inputs, and outputs that are
produced by the model operation. Sampling of the input character, which is an error, and changes in the errors
were performed periodically on the basis of output. This method created the weights automatically when the
data was generated. The number of retrieved weights for input and output is constrained to predetermined
criteria in this case by means of FIPSO. With this information, PNN analysis was performed as usual by
creating input and output weights using the MATLAB toolkit. The ‘d’ and ‘q’ axes' current is regulated by a
PI controller, which takes the role of the outputs that this controller estimated. Figures 3 and 4 display the
effects of the PI-controller, and PNN- FIPSO controller simulation results for the change of Iq and Iqref.
According to the data, the PNN controller constructed with FIPSO has a reduced overshoot between
Iq and Iqref than the PI controller. With the improved PNN with FIPSO controller time processing is achieved

Power quality enhancement using fully informed particle swarm optimization … (Moguthala Shankar)
986  ISSN: 2252-8792

better output than the conventional controllers. The well-known conventional classic control system can be
used only for linear systems but by using the PNN controller constructed with FIPSO which can be used for
both the linear and non-linear systems. So, by considering all the features of the PSO algorithm FIPSO can be
applied to a power system to improve the performance for many problems related to power quality
improvement. Figures 5 and 6 show the differences in the average DC voltage that have been depicted in, for
a given parameter, the peak time & rise time of FIPSO were greater in comparison to other approaches.

Figure 3. Simulation outcomes of PI controller of Iq and Iqref

Figure 4. Simulation result of PNN-FIPSO controller for variation of Iq and Iqref

Figure 5. Variation of the average DC voltage with PI controller

Figure 6. Variation of average DC voltage with PNN-FIPSO controller

Int J Appl Power Eng, Vol. 13, No. 4, December 2024: 982-988
Int J Appl Power Eng ISSN: 2252-8792  987

6. CONCLUSION
This paper represented a PNN controller design by FIPSO for DSTATCOM to increase the power
quality of the distribution system. The MATLAB/Simulink platform has been utilized to test the complete system.
The PNN-FIPSO controller and the conventional PI controller are used to compare and show better outcomes. In
contrast to other conventional approaches, the comparison results demonstrate that the PNN controller constructed
using FIPSO delivers better reactions to the change in reference reactive current. With the suggested controller,
Q34, the parameters rise and peak times are longer. From this, the FIPSO approach proves to be the best substitute
for another traditional method for solving complex tasks such as complex mathematical modeling tasks. This
method can be applied to a power system to increase performance for any problems related to power quality
improvement. This paper presents a novel information-sharing-oriented PSO algorithm named FIPSO. Unlike the
conventional PSO algorithm, enhancements have been implemented in two key aspects. Firstly, a new
information-sharing mechanism has been introduced where the worst location of each particle and the worst
location of the entire swarm are both tracked. This innovative approach directs the particles to move away from
the worst particle and swarm locations, thereby expanding the global search space and minimizing the risk of
particles getting trapped in local optima. Additionally, it presents a particle competitive strategy that encourages
the worst information to challenge the best. This approach not only lowers the chances of getting stuck in a
suboptimal solution but also enhances the likelihood of reaching the best solution.
Moreover, the time complexity is only twice that of the conventional PSO algorithm. To summarize,
this paper initially examined the effectiveness of PSO in comparison to GA and concluded that PSO
outperforms GA. Subsequently, the limitations of the standard PSO were analyzed. Based on this analysis, a
novel PSO algorithm called FIPSO was introduced. The experimental findings demonstrated that FIPSO
surpasses both the traditional PSO and the GA algorithm when applied to benchmark functions. Notably,
FIPSO exhibits remarkable enhancements in successfully identifying optimal solutions, particularly for
challenging functions.

REFERENCES
[1] S. Devi and M. Geethanjali, “Optimal location and sizing determination of Distributed Generation and DSTATCOM using Particle
Swarm Optimization algorithm,” International Journal of Electrical Power and Energy Systems, vol. 62, pp. 562–570, 2014, doi:
10.1016/j.ijepes.2014.05.015.
[2] D. K. Sambariya and R. Prasad, “Robust tuning of power system stabilizer for small signal stability enhancement using metaheuristic bat
algorithm,” International Journal of Electrical Power and Energy Systems, vol. 61, pp. 229–238, 2014, doi: 10.1016/j.ijepes.2014.03.050.
[3] S. Jazebi, S. H. Hosseinian, and B. Vahidi, “DSTATCOM allocation in distribution networks considering reconfiguration using differential
evolution algorithm,” Energy Conversion and Management, vol. 52, no. 7, pp. 2777–2783, 2011, doi: 10.1016/j.enconman.2011.01.006.
[4] O. P. Mahela and A. G. Shaik, “A review of distribution static compensator,” Renewable and Sustainable Energy Reviews, vol. 50,
pp. 531–546, 2015, doi: 10.1016/j.rser.2015.05.018.
[5] T. Zaveri, B. Bhalja, and N. Zaveri, “Comparison of control strategies for DSTATCOM in three-phase, four-wire distribution system
for power quality improvement under various source voltage and load conditions,” International Journal of Electrical Power and
Energy Systems, vol. 43, no. 1, pp. 582–594, 2012, doi: 10.1016/j.ijepes.2012.06.044.
[6] M. A. Mahmud, H. R. Pota, and M. J. Hossain, “Nonlinear DSTATCOM controller design for distribution network with distributed
generation to enhance voltage stability,” International Journal of Electrical Power and Energy Systems, vol. 53, pp. 974–979, 2013,
doi: 10.1016/j.ijepes.2013.06.017.
[7] B. Singh, S. K. Dube, and S. R. Arya, “An improved control algorithm of DSTATCOM for power quality improvement,”
International Journal of Electrical Power and Energy Systems, vol. 64, pp. 493–504, 2015, doi: 10.1016/j.ijepes.2014.07.055.
[8] N. G. Hingorani, L. Gyugyi, and M. E. El-Hawary, Understanding FACTS: Concepts and technology of flexible AC transmission
systems. New York, US: IEEE, 1999.
[9] S. Moguthala and R. S. Kumar, “Review on Fuzzy Logic Based Power Quality Improvement By Using Dstatcom Based Cascaded
Multilevel Inverter,” in Proceedings on Engineering Sciences, vol. 4, no. 3, pp. 263–270, 2022, doi: 10.24874/PES04.03.004.
[10] B. Singh, A. Adya, A. P. Mittal, and J. R. P. Gupta, “Modeling and control of DSTATCOM for three-phase, four-wire distribution
systems,” in Fourtieth IAS Annual Meeting. Conference Record of the 2005 Industry Applications Conference, 2005., 2005, pp.
2428-2434 Vol. 4, doi: 10.1109/IAS.2005.1518801.
[11] G. K. Venayagamoorthy and R. G. Harley, “Computational Intelligence Techniques for Control of FACTS Devices,” Applied
Mathematics for Restructured Electric Power Systems, pp. 201–237, 2006, doi: 10.1007/0-387-23471-3_10.
[12] D. Amoozegar, “DSTATCOM modelling for voltage stability with fuzzy logic PI current controller,” International Journal of
Electrical Power and Energy Systems, vol. 76, pp. 129–135, 2016, doi: 10.1016/j.ijepes.2015.09.017.
[13] L. A. Zadeh, “Fuzzy sets,” Information and control, vol. 8, no. 3, pp. 338–353, 1965, doi: 10.1016/S0019-9958(65)90241-X.
[14] P. Levy, Simulating Power Systems using MATLAB and Simulink. North Charleston, SC, USA: CreateSpace, 2016.
[15] F. Klawonn and R. Kruse, “Constructing a fuzzy controller from data,” Fuzzy Sets and Systems, vol. 85, no. 2, pp. 177–193, 1997,
doi: 10.1016/0165-0114(95)00350-9.
[16] S. L. Chiu, “Extracting fuzzy rules from data for function approximation and pattern classification,” Fuzzy Information Engineering:
A guided tour of applications, vol. 9, pp. 1–10, 1997.
[17] R. Mendes, J. Kennedy, and J. Neves, “The fully informed particle swarm: Simpler, maybe better,” IEEE Transactions on
Evolutionary Computation, vol. 8, no. 3, pp. 204–210, 2004, doi: 10.1109/TEVC.2004.826074.
[18] J. Kennedy and R. Mendes, “Neighborhood topologies in fully informed and best-of-neighborhood particle swarms,” IEEE
Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 36, no. 4, pp. 515–519, 2006, doi:
10.1109/TSMCC.2006.875410.
[19] A. Bhattacharya, C. Chakraborty, and S. Bhattacharya, “Parallel-Connected Shunt Hybrid Active Power Filters Operating at
Different Switching Frequencies for Improved Performance,” IEEE Transactions on Industrial Electronics, vol. 59, no. 11, pp.

Power quality enhancement using fully informed particle swarm optimization … (Moguthala Shankar)
988  ISSN: 2252-8792

4007–4019, 2012, doi: 10.1109/TIE.2011.2173893.

[20] R. L. De Araujo Ribeiro, C. C. De Azevedo, and R. M. I. De Sousa, “A robust adaptive control strategy of active power filters for
power-factor correction, harmonic compensation, and balancing of nonlinear loads,” IEEE Transactions on Power Electronics, vol.
27, no. 2, pp. 718–730, 2012, doi: 10.1109/TPEL.2011.2161334.
[21] M. Depenbrock, V. Staudt, and H. Wrede, “A Theoretical Investigation of Original and Modified Instantaneous Power Theory
Applied to Four-Wire Systems,” IEEE Transactions on Industry Applications, vol. 39, no. 4, pp. 1160–1168, 2003, doi:
10.1109/TIA.2003.814553.
[22] E. F. Fuchs and M. A. S. Masoum, Power Quality in Power Systems and Electrical Machines. London, UK: Academic Press, 2012.
[23] H. Hu, W. Shi, Y. Lu, and Y. Xing, “Design considerations for DSP-controlled 400 Hz shunt active power filter in an aircraft power
system,” IEEE Transactions on Industrial Electronics, vol. 59, no. 9, pp. 3624–3634, 2012, doi: 10.1109/TIE.2011.2165452.
[24] P. Kanjiya, V. Khadkikar, and H. H. Zeineldin, “A noniterative optimized algorithm for shunt active power filter under distorted
and unbalanced supply voltages,” IEEE Transactions on Industrial Electronics, vol. 60, no. 12, pp. 5376–5390, 2013, doi:
10.1109/TIE.2012.2235394.
[25] R. Poli, J. Kennedy, and T. Blackwell, “Particle swarm optimization,” Swarm Intell, vol. 1, pp. 33–57, 2007, doi: 10.1007/s11721-
007-0002-0.
[26] V. Kothari, J. Anuradha, S. Shah, and P. Mittal, “A survey on particle swarm optimization in feature selection,” International
Conference on Computing and Communication Systems, 2012, vol. 7, pp. 192–201, doi: 10.1007/978-3-642-29216-3_22.
[27] M. Imran, R. Hashim, and N. E. A. Khalid, “An overview of particle swarm optimization variants,” Procedia Engineering, vol. 53,
pp. 491–496, 2013, doi: 10.1016/j.proeng.2013.02.063.
[28] D. Wang, D. Tan, and L. Liu, “Particle swarm optimization algorithm: an overview,” Soft Computing, vol. 22, no. 2, pp. 387–408,
Jan. 2018, doi: 10.1007/s00500-016-2474-6.
[29] X. Rao and X. Yan, “Particle Swarm Optimization Algorithm Based on Information Sharing in Industry 4.0,” Wireless
Communications and Mobile Computing, vol. 2022, no. 1, 2022, doi: 10.1155/2022/4328185.

BIOGRAPHIES OF AUTHORS

Moguthala Shankar has received a B.Tech. and M.Tech. from JNTU

Hyderabad, India, in 2012 and 2015 respectively, and pursuing a Ph.D. at Anna University,
Chennai, Tamil Nadu, India. He has presented technical papers in various national and
international journals and conferences. His areas of interest include power electronics, power
quality, multi-level inverters neural networks, and fuzzy logic. He can be contacted at email:
[email protected].

Dr. R. Senthil Kumar currently is working as a professor in the Electrical and

Electronics Engineering Department of Saveetha Engineering College (Autonomous),
Chennai, and Tamil Nadu. He has done B.E. in electrical engineering from the University
College of Engineering (guindy) and an M.E. and Ph.D. from Anna University, Chennai,
Tamil Nadu, India. His main research directions include artificial intelligence (neural
networks, fuzzy logic, genetic algorithm, particle swarm optimization, and imperialist
competitive algorithm) gravitational search algorithm, power quality, and proportional-
integral controller. He has published 25 international journal papers, 15 international
conference papers, and 2 National Conference papers and also published a book named
Special Electrical Machines in the year 2009. Reviewer of the book - Basic Electrical and
Electronics Engineering – Charulatha Publications and reviewer of the Journal in
International Journal of Electrical Engineering and Technology and the Journal of IET in
Power Electronics. His areas of research interest are renewable and non-renewable energy
sources, power converters, and electrical machines. He is a professional member of IEI,
IEEE, ISTE, and UACEE. He is one of the members of the advisory committee for one of
the leading engineering colleges. He can be contacted at email: [email protected].

Int J Appl Power Eng, Vol. 13, No. 4, December 2024: 982-988