International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 6 Issue 6, September-October 2022 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Particle Swarm Optimization (PSO) Algorithm Based MPPT

for PV System Operating under Partial Shading Condition
Sarad Kumar1, Pramod Kumar Rathore2
1
Student, 2Assistant Professor,
1,2
RKDF College of Engineering, Bhopal, Madhya Pradesh, India

ABSTRACT How to cite this paper: Sarad Kumar |

With ever-increasing demand, solar energy has emerged as one of the Pramod Kumar Rathore "Particle Swarm
most significant renewable energy sources; consequently, operating Optimization (PSO) Algorithm Based
photovoltaic panels to maximize efficiency is vital. However, solar MPPT for PV System Operating under
panels have difficulties, such as the Partial Shading Condition, which Partial Shading Condition" Published in
International Journal
has a substantial influence on the efficiency of the PV system. MPPT
of Trend in
plays a critical role in improving the power of PV systems. PV Scientific Research
trajectories will be diverse and exceedingly complicated due to the and Development
numerous peaks accessible during PSC. Traditional MPPT (ijtsrd), ISSN: 2456-
approaches will fail to achieve the Global Maximum Power Point and 6470, Volume-6 |
instead will linger at the Local Maximum Peak Point, decreasing the Issue-6, October IJTSRD52182
PV module's efficiency and performance. The Particle Swarm 2022, pp.1870-
Optimization technique is used in this work to determine the GM PP 1875, URL:
by measuring peak power. In terms of Maximum Power Point, this www.ijtsrd.com/papers/ijtsrd52182.pdf
approach offers a high degree of reliability, robustness, and
proficiency. MATLAB/Simulink is used to verify the suggested Copyright © 2022 by author(s) and
International Journal of Trend in
approach.
Scientific Research and Development
KEYWORDS: Photovoltaic, Maximum Power Point MPP, Maximum Journal. This is an
Open Access article
Power Point Tracking MPPT, Global PowerPoint GMP, Particle
distributed under the
Swarm Optimization (PSO), DC-DC Boost Converter, and Local terms of the Creative Commons
Maximum Power Point are some of the terms used to describe partial Attribution License (CC BY 4.0)
shading conditions (http://creativecommons.org/licenses/by/4.0)

1. INTRODUCTION
With ever-increasing power usage, electricity bypass diodes that are coupled across. The partial
expenses must be addressed, which may be reduced shading situation will result in a loss of power from
by employing PV panels. Furthermore, renewable the PV module due to the bypass diodes employed,
energy from the sun may be the ideal method to the placement of the PV array, and the shading
generate power bills, allowing producers to sell patterns. The bypass diodes will produce complicated
excess electricity to local electricity providers through forms and several peaks in the PV trajectory. The
PV panels. As demand grows, PV technology will standard algorithm becomes confused between the
become more efficient and provide more advantages Global and Local Maxima points and is unable to
in the near future. Because solar panels have a greater discriminate between the two, reducing the efficiency
initial cost, it is preferred that the maximum quantity of maximum power extraction. For PV systems, the
of electricity be recovered from the PV panels. PV P&O and INC maximum power point tracking
production is purposefully dependent on the strength approaches are often utilized. The allocated voltage is
of solar light and temperature. Because these limits either deducted or added on a regular basis in the
alter over time, is mandatory to propose a MPPT way P&O algorithm while examining the latest power
for tracing the highest power from the PV One of the pattern value of MPP. In the INC method to MPP, we
most significant considerations for getting the most differentiate two values, power against voltage, and
electricity out of a PV system is partial shading. This maintain the resulting value equal to zero. When
circumstance occurs as a result of impediments such partial shade occurs, traditional algorithms may
as buildings, trees, clouds, and towers. The partial identify just the Local Maximum Peak Power,
shading scenario causes a hotspot to form in the solar reducing overall power. In addition, the researchers
module. In order to avoid self-heating, the array has are working on the ANN and direct search algorithm

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1870
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
fuzzy logic control [3-7]. However, these
technologies are time-consuming, complicated in
hardware design, and very costly. Swarm intelligence
and evolutionary algorithms have been presented by
researchers as naturally encouraged solutions for
coping with multi-dimensional and multimodal Fig. 1 Block Diagram of PV System with MPPT
challenges [8], [9]. The MPPT system employs
genetic algorithms, however they take a long time to 3. MAXIMUM POWER POINT TRACKING
execute since the position of restrictions is The most extreme Power Point Tracking (MPPT)
determined through trial and error. Swarm device is useful in PV applications. The key factors
intelligence is also employed in optimization, such as that influence the electric power generated by a
ant optimization and particle swarm optimization. photovoltaic framework are solar radiation and
ANO helps in the tracking of GMPP, however it takes temperature. The voltage at which a PV module can
a long time and is user dependant. [10],[11],[12]. This produce the highest power is referred to as the' most
research employs the Particles Swarm Optimization extreme power point' (pinnacle control voltage). The
approach to follow the maximum power point in fundamental rule of MPPT is in charge of isolating
order to locate GMPP while simultaneously grappling the most possible power from the solar and feeding it
with the complex and non-linearity concerns of PV to the heap via a dc to dc converter that steps
Modules. This PSO method delivers great resilience, up/down the voltage to the desired size [5]. There are
reliability, and efficiency in achieving PV system several strategies for maximizing power in
GMP. PowerPoint. Two MPPT methods are among these in
this research, ANN and perturb and observe (P&O)
2. STAND -ALONE SOLAR POWER SYSTEM have been chosen for comparison.
A PV module, a dc/dc boost converter, the maximum
power point tracking algorithm, and a load comprise 3.1. Meta heuristic MPPT algorithm for Partial
the solar PV system. Radiation (R) strikes the PV shaded solar PV system
module. It produces a voltage (V) and current (I) that 3.1.1. Partial Shaded PV Model
is supplied into the load [3]. Because of variations in The total of 60 solar PV cells is split into three
atmospheric conditions, the voltage power models, each with 20 cells. The open circuit voltage
characteristic of a photovoltaic (PV) array is is 12.64 volts, the voltage at maximum power is 10.32
nonlinear and time changing. When the amount of volts, the short circuit current is 8.62 A, and the
solar energy and temperature change, the output current at maximum power point is 8.07 and the
power of the The PV module is also altered. To maximum power point is 83.2824Watts for the 20 cell
achieve optimal efficiency, the PV module must be type. This three models are linked in series, and each
operated at the highest point of the PV characteristic. features a bypass diode. Figure 1 depicts the simulink
The most severe power point is dependent on model of the three series-connected solar PV system.
temperature and irradiance, both of which are indirect
in nature. The most powerful point following control
framework is used, as well as work viability on non-
straight variations in factors such as temperature and
radiations [4]. A MPPT is used to extract the most
power from a solar PV module and transmit it to a
load. A dc/dc converter (boost converter) transfers the
maximum power from the solar PV module to the
load. A dc/dc converter serves as a conduit between
the load and the module. Figure 1 depicts a dc/dc
converter with a maximum power point tracking
method and a load. By varying the duty cycle, the
load impedance as viewed by the source is adjusted
and matched at the time of maximum power transfer
with the source. As a result, MPPT procedures are
required to keep the PV array running at its MPP [3].
In this research, the two most prevalent MPPT Figure 1 Series Connected three PV model
techniques (Perturb and Observe (P&O) methods and Figure 2 depicts the current-voltage and power-
artificial neural network (ANN) approaches, as well voltage characteristics of the solar PV model under
as a dc-dc converter, will be compared. normal test conditions.

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1871
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Figure 4 Simulink model of DC-DC boost

converter
3.2. Particle swarm optimization
PSO, a novel swarm intelligence optimization method
published by Kennedy and Eberhart in 1995 [28], has
grown quickly over the last 20 years. It is patterned
by the I of bird flocks. Its benefits include easy
implementation and quick convergence, and it is
ideally suited for determining the global optimum
solution in a nonlinear, discontinuous, non-
differentiable curve. This approach employs many
Figure 2 I-V and P-V characteristics of single cooperating particles in an n-dimensional space. Each
solar PV model particle has its own location Pi (distributed at
random) and velocity Vi (Vi = 0 at the start). A
particle's location is impacted by its best position so
far, Pbest, as well as the best position of all particles
so far, Gbest. Particle velocity and position are
computed by

Figure 3 I-V and P-V characteristics of partial

shaded solar PV model
The partly shadowed state of the solar PV model is
achieved by setting the irradiance of each model to a
different number, such as 1000 W/m2 for panel 1,
300 W/m2 for panel 2, and 600 W/m2 for panel 3.
Figure 3 depicts the IV and PV characteristics of the
partially shadowed situations. 104.5 Watts is the
maximum power point. The maximum power point
tracking method should be used to obtain this point
from the solar PV panel. PSO, Cuckoo search
algorithm, flower pollination algorithm, and grey
wolf algorithm are used in this application. To extract
the most power from the partially shaded solar PV
model, a boost converter is connected between the
solar PV model and the load. Figure 4 shows the
boost converter simulink model.
Figure 4 PSO – Particle swarm optimization
Algorithm Flow Chart

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1872
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
4. SIMULATION AND RESULTS
This paper describes a technique for maximum power point tracking (MPPT) that uses particle swarm
optimization (PSO) with variable step size to avoid steady-state oscillations. This will eliminate the effect of
partial shadowing on the efficiency of photovoltaic (PV) systems. A DC-DC boost converter is used to maximize
the power output of the solar panels. The examination of Maximum Power Point Tracking Algorithms follows
the modeling of the Stand-Alone PV System. Maximum Power Point Tracking and PSO simulation models
Algorithm are executed Version R2021A of MATLAB/Simulink The next sections exhibit the simulation results
of Maximum Power Point Tracking using PSO Algorithm.

Figure 5 Simulink Model of the MPPT_PV_SYSTEM Using PSO Algorithm

To analyse the performance of the output voltages, the output currents and the output power are measured as
shown. The PV parameters of the system are shown in Table I.
TABLE I PV PARAMETERS
Parameters Specifications
Maximum power (W) 83.2824 W
Parallel strings 1
Series-connected modules per string 1
Cells per module (Ncell) 60/3
Open circuit voltage Voc (V) 37.92*20/60
Short-circuit current Isc (A) 8.62
Voltage at maximum power point Vmp (V) 30.96*20/60
Current at maximum power point Imp (A) 8.07

Figure 6 PWM Pulses Vs Time in (S)

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1873
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Figure 7 Output Voltage Vs Time in (S) from Boost Converter

Figure 8 Output Current Vs Time in (S) from Boost Converter

Figure 9 Output Power Vs Time in (S) from Boost Converter

5. CONCLUSION tracking approach based on the PSO Algorithm for
In this study, the PV system and boost converter monitoring the PV system's Global Maximum Power
parameters are calculated, and MPPT tracking is Point. Figure.6 depicts the PWM used on the boost
achieved using the PSO Algorithm. The primary goal DC-DC converter. During the PSC condition, the
of this work is to offer a maximum power point suggested PSO algorithm may be correctly

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1874
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
conducted. All other algorithms failed to create simulation of a predictive current control for
numerous peaks due to the PSC condition. photovoltaic system based on the MPPT
differentiate between GMPP and LMPP are used to strategy” 2016 13th International Conference
provide electricity surrounding LMPP. As a result, on Power Electronics (CIEP).
the PSO Algorithm is effective; it has a high [9] Taiyang Wu; Md Shamsul Arefin; Doron
convergence rate, which distinguishes it from many Shmilovitz; Jean-Michel Redoute; Mehmet
traditional algorithms. Rasit Yuce “A flexible and wearable energy
harvester with an efficient and fast-converging
REFERENCES
analog MPPT” 2016 IEEE Biomedical Circuits
[1] Youssef Mallal; Lhoussain El Bahir; Touria
and Systems Conference (BioCAS).
Hassboun “A Comparative Study of
[10] Sean Yang; Kwang-Ming Lin; Wen-Chin Lee;
Photovoltaic Array Configurations effect on the
Wen-Shun Lo; Chia-Hsiang Chen; Jyh-Lih Wu;
Output Power under Partial Shading
Chi-Yu Chiang “Achievement of 16. 5% total
Conditions” 2019 4th World Conference on
area efficiency on 1. 09m2 CIGS modules in
Complex Systems (WCCS).
TSMC solar production line” 2015 IEEE 42nd
[2] Fathima Raziya; Mohamed Afnaz; Stany
Photovoltaic Specialist Conference (PVSC).
Jesudason; Iromi Ranaweera; Harsha Walpita
[11] Kevin Anderson; Maryam Shafahi; Reza
“MPPT Technique Based on Perturb and
Baghaei Lakeh; Sean Monemi; Chris
Observe Method for PV Systems Under Partial
McNamara “CFD study of compost waste heat
Shading Conditions” 2019 Moratuwa
for use in a hybrid solar tower” 2015 IEEE
Engineering Research Conference (MERCon).
Conference on Technologies for Sustainability
[3] Mithun Krishnan M.; K. R. Bharath “A Novel
(SusTech).
Sensorless Hybrid MPPT Method Based on
[12] M. Horváth; S. Hrabovszky-Horváth; T.
FOCV Measurement and P&O MPPT
Csoknyai “Parametric analysis of solar hot
Technique for Solar PV Applications” 2019
water production in “commi-block” buildings”
International Conference on Advances in
2015 5th International Youth Conference on
Computing and Communication Engineering
Energy (IYCE).
(ICACCE).
[13] Bart Vermang; Yi Ren; Jonathan Joel;
[4] Hayder Ali; Hassan A. Khan “Evaluation of
Christopher Frisk; Olivier Donzel-Gargand;
Low-Voltage Loss under Partial Shading
Pedro “Rear surface optimization of CZTS
Conditions in Solar Photovoltaic Systems”
solar cells by use of a passivation layer with
2018 IEEE Energy Conversion Congress and
nano-sized point openings” 2015 IEEE 42nd
Exposition (ECCE).
Photovoltaic Specialist Conference (PVSC).
[5] Jirada Gosumbonggot; Duy-Dinh Nguyen;
[14] Evelyn Butler “The next phase of solar insights
Goro Fujita “Partial Shading and Global
driven by innovation and standards” 2015 IEEE
Maximum Power Point Detections Enhancing
First International Conference on DC
MPPT for Photovoltaic Systems Operated in
Microgrids (ICDCM).
Shading Condition” 2018 53rd International
[15] Salah Larbi; Amor Bouhdjar; Khaled Meliani;
Universities Power Engineering Conference
Abdeldjalil Taghourt; Hakim Semai “Solar
(UPEC).
chimney power plant with heat storage system
[6] Murari Lal Azad; Soumya Das; Pradip Kumar
performance analysis in South Region of
Sadhu; Biplab Satpati; Anagh Gupta; P. Arvind
Algeria” 2015 3rd International Renewable and
“P&O algorithm based MPPT technique for
solar PV system under different weather Sustainable Energy Conference (IRSEC).
conditions” 2017 International Conference on [16] Homare Hiroi; Noriyuki Sakai; Yasuaki Iwata;
Circuit, Power and Computing Technologies Takuya Kato; Hiroki Sugimoto “Impact of
(ICCPCT). buffer layer on kesterite solar cells” 2015 IEEE
[7] S. Uma Ramani; Sathish Kumar Kollimalla; B. 42nd Photovoltaic Specialist Conference
Arundhati “Comparitive study of P&O and (PVSC).
incremental conductance method for PV [17] Taohid Latif; Syed R. Hussain “Design of a
system” 2017 International Conference on charge controller based on SEPIC and buck
Circuit, Power and Computing Technologies topology using modified Incremental
Conductance MPPT” 8th International
(ICCPCT).
Conference on Electrical and Computer
[8] Marco Morales-Caporal; Jose Rangel-
Engineering (Year: 2014).
Magdaleno; R. Morales-Caporal “Digital

@ IJTSRD | Unique Paper ID – IJTSRD52182 | Volume – 6 | Issue – 6 | September-October 2022 Page 1875