Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI 2023)

IEEE Xplore Part Number: CFP23J32-ART; ISBN: 979-8-3503-9728-4

Particle Swarm Optimization based Maximum

Power Point Tracking
R. Vinifa S. Vetrivel E. Ragul
Department of Electrical and Department of Electrical and Department of Electrical and
Electronics Engineering, Electronics Engineering, Electronics Engineering,
Saveetha Engineering College, Saveetha Engineering College Saveetha Engineering College
Chennai, India Chennai, India Chennai, India
[email protected] vetrivelsivakumar1482001@gmail [email protected]
.com
2023 7th International Conference on Trends in Electronics and Informatics (ICOEI) | 979-8-3503-9728-4/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICOEI56765.2023.10125791

F. M ohammed Younus P. K. Pradeep

Department of Electrical and Electronics Engineering, Department of Electrical and Electronics Engineering,
Saveetha Engineering College Saveetha Engineering College
Chennai, India Chennai, India
[email protected] [email protected]

Abtract- S olar energy is revolutionizing the field of various countries for getting adapted to the
clean energy production because of its immense production of clean energy.[1]
energy yield at economical costs. Fast growing field of
renewable energy systems is facing a lot of challenges Even though the power produced from the
in power production and maintenance. This study Solar PV has less carbon footprints, the major
focuses on extraction of the maximum power from the difficulty lies in the management of generated
Photovoltaic (PV) S ystem using maximum power
power. The research work [2,3] describes the
point tracking (MPPT) controller. The need for
tracking of the maximum power point is huge, since it
various techniques of handling renewable energy
increases the efficiency even during the rapid change efficiently in order facilitate better usage profile
in the environmental conditions. Maximum power and reducing the losses.
point (MPP) is tracked by various conventional
algorithms such as Incremental Conductance Solar cells are the fundamental constituents of
algorithm and Perturb and Observe method. These PV Module, which are joined in series and parallel
algorithms are well adequate in tracking the to form a PV System. The solar irradiation and
maximum power, but it suffers a lot of power losses at temperature are the inputs to the solar cell, and the
the steady state. S o, it is necessary to device a new irradiation generates photocurrent in the cells
algorithm that is more beneficial in tracking the which produces power. [4] presents the
MPP. This work discusses about the technique called performance of the PV system at changing
Particle S warm Optimisation (PS O); a nature irradiation and temperature conditions. These
inspired algorithm like the Genetic Algorithm. The
study revolves around the various key outcomes that parameters affect the system performance at rapidly
makes the PS O differ from the conventional MPPT changing environmental conditions.
algorithms. From the analysis PS O Algorithm
efficiently tracks the MPP than the conventional
There are a variety of MPPT Algorithms
tracking techniques. proposed in the literature like Incremental
Conductance algorithm, Perturb and Observe hm,
Key words – Photo Voltaic System, Maximum Power method, these conventional algorithms are
Point Tracking, Solar Cell, Particle Swarm straightforward and can be easily put into action.
Optimization. These methods suffers a major problem of more
1.INTRODUCTION losses in the steady state due to frequent moving in
the direction of tracking. Their performance is also
Modern Power systems which are completely average in rapidly varying environmental
based on fossil fuel is in the verge of getting conditions.[5-8]
depleted. It is important to switch over to the non-
conventional energy sources in order avoid In [9], implementation of conventional
unexpected events such shortage of energy algorithms by variable step and beta method is
production. Renewable energy systems are gaining performed and analysed, clearly the variable step
a lot of advantages due to the harmful impacts of size method gives more advantages avoiding power
losses in the steady state of MPP Tracking.
the conventional energy resources on the
environment. Developing countries are much After the emergence of the Artificial
concerned about the sustainable development on Intelligence based MPPT Algorithms like ANN and
energy production. A new index called Renewable Fuzzy[10,11], the efficiency of MPP Tracking of
Energy Index (REI) establishes the motive of the the PV system has improved significantly due to

979-8-3503-9728-4/23/$31.00 ©2023 IEEE 335

Authorized licensed use limited to: ANNA UNIVERSITY. Downloaded on August 24,2023 at 06:25:34 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI 2023)
IEEE Xplore Part Number: CFP23J32-ART; ISBN: 979-8-3503-9728-4

the fast tracking and dynamic characteristics. The diode which is in parallel with a current source.
paper [12] explains the difference between an Photo current is generated when incident solar
intelligent technique of ANN and Perturb and irradiation falls on the materials. In practical cases,
Observe method. a solar cell has a no. of losses taking place due to
the material of construction, incident irradiation
Considering the wide range of performance and the temperature of the solar cell. A series
parameters is too necessary analyse the system resistance and shunt resistance are added to
when there is an immediate change in the constitute such losses, usually the effect of series
environmental conditions. Also, the PV System resistance (Rs ) is much lower than the shunt
does not always operate at equal amount of resistance (Rsh ). Shunt resistance contributes
irradiation over a large span of area, this notion is majority of losses in the form of leakage currents
termed as partial shading condition. In this which can be eliminated by improvement in
condition, there exists number of local and global fabrication of the material with desired
maximum power points. Tracking of Global characteristics. The eq. (1) gives the relationship of
Maximum power point is a tedious process because the Photo current ( ) with the Irradiation (G) and
of the problem of getting s tuck in a local Temperature (T).
maximum. This paper proposes a fuzzy logic
controller based MPPT system design which [ ( ) ]. (1)
incorporates Partial Shading, it produces a
commendable performance.[13,14]
( ) (2)
Intelligent techniques like artificial neural
networks too have certain limitations including [ ]
[ ]
accuracy of tracking the exact MPP. Hybrid MPPT
= [ ] (3)
system like Adaptive Neuro-fuzzy Inference
System discussed in [15] is good in their working,
but they are very complex to implement. So, a new = (4)
[ ]
method of tracking the MPP overcoming the
problems is of high need. ( )
( )
[ ] (5)
II.MATHEMATICAL MODELLING OF PV
SYSTEM
The shunt current ( ) is the current flow
Solar cell is Photo Voltaic cell which through the resistor ( ). The and are the
generated electric current from the incident solar panel output current and voltage respectively.
irradiation. A solar cell is nothing but a PN Other constants and parameters used in these
junction, when light is allowed to fall on the equations are mentioned as below.
junction of the p-type and n-type material, due to
the application of energy by the solar irradiation the T ABLE I. CONST ANT S AND PARAMET ERS USED IN
MODELLING
band gap of the semiconductor material becomes
narrow facilitating the flow of charges. The Photo Parameters values
current generated is directly proportional to the k 1.38e-23
incident solar irradiation. Solar cells work upon
q 1.6e-19
certain mathematical parameters and constants
which helps bring out the best characteristics and 0.0032
accurate output. 0.221
415.405
36.8 V
3.03 A
8.08 A
64

Ns - number of cells connected in series, Eg0 - band

gap energy of the semiconductor - 1.1 eV, n -
ideality factor of the diode, V - diode thermal
Fig. 1. Circuit Model of PV System voltage (V), - open circuit voltage (V), k -
Boltzmann’s constant, T n - nominal temperature.
Modelling of these parameters are necessary
to get the required system functional specification III PARTICLE SWARM OPTIMIZATION (PSO)
as of our interest. An ideal solar cell is made up of

979-8-3503-9728-4/23/$31.00 ©2023 IEEE 336

Authorized licensed use limited to: ANNA UNIVERSITY. Downloaded on August 24,2023 at 06:25:34 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI 2023)
IEEE Xplore Part Number: CFP23J32-ART; ISBN: 979-8-3503-9728-4

It is an evolutionary algorithm-based nature diversified output. and are the cognitive and
inspired Optimization and metaheuristic search- social coefficients of the swarm. For a typical
based optimization algorithm capable of solving maximization problem when, the condition eq. (3)
multi-modal maximization and minimization is satisfied it can be converted as the eq. (4).
problems. PSO is primarily inspired from the Likewise, after successive iterations the position
foraging behaviour of animals like flocking of birds continuously updated until it reaches the
and schooling of fishes. A particle may be a fish or .
a bird, where the school or a flock is called as a
swarm. A swarm contains a pool of particles each ( (3)
behaving cooperative and harmonious with the
( )
society. All the particle exchange information
among one another during successive iterations and
gradually move towards the best solution or the
prey. In order to get nearer to the solution, the A.ALGORITHM
particle has to move in a specific direction at a
specific velocity. The direction of movement of the  Initializing the random number of
particles is done by the variables, local best ( ) particles.
and the global best ( ). The particles based  Initializing the random position and
upon its fitness function mapped by an objective velocity to the particles.
function continuously varies the position and  Evaluate the Objective function of the
velocity to attain the global maximum power point particles.
illustrated in Fig. (2).  Changing the individual and of
the particles.
The position and velocity of the particles is given  Changing the velocity and position of the
by the eq. (1) and eq. (2) particles.
 Repeating steps from 2 to 5 till the
termination criteria are satisfied.
)=𝑤 )+ .( − ))
IV.SIMULATION RESULTS
+ .( − )) (1)
Simulation of Particle swarm Optimization
+1) = )+ +1) (2) algorithm was performed using
MATLAB/Simulink environment at standard
parameters maintained constant throughout the
process of simulation. It consists of the PV Array
as the primary source and the boost converter of
specifications as mentioned below in tab. (2). The
Voltage and Current from the PV Array block is
connected to the MPPT Controller which generates
PWM signal to switch the Converter. The MPPT
Controller consists of a function block which is
written with the PSO Algorithm. Externally
connected hysteresis controller is used to reduce
the errors. The PWM Generator block generates
PWM from the control signals of the PSO
Algorithm. Specifications of the MPPT System is
as mentioned in the tab. (2)

TABLE I. MPPT SYSTEM SPECIFICATIONS

PARAMETER VALUE
Fig. 2. A T ypical Particle Swarm Optimization System.
No. of Parallel strings 2

Where, ) are the velocity and ) are the No. of series strings 2
position of the particle , and represent Overall system Power 882 W
the local and global best position. 𝑤 is the inertial
weight of the particles and and are randomly Inductance 1.247 mH
chosen numbers whose values are equally Capacitance 2927 µF
distributed in the range [0, 1] for getting a more Load 6Ω

979-8-3503-9728-4/23/$31.00 ©2023 IEEE 337

Authorized licensed use limited to: ANNA UNIVERSITY. Downloaded on August 24,2023 at 06:25:34 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI 2023)
IEEE Xplore Part Number: CFP23J32-ART; ISBN: 979-8-3503-9728-4

The output power of the system is illustrated In the Fig. (4), Tracked Power output is
in the Fig. (3) where it reaches a power of 858.9 W shown with variable irradiation at temperature
at 1000 W/ 25 . At the standard conditions condition of 25 . Efficiency of the system is
PSO Algorithm performs well in tracking the MPP. calculated by the ratio between tracked power and
maximum power tabulated in Tab. (2). Steady state
oscillations produce a significant amount of power
loss in the system, reducing this steady state error
can improve the overall power output of the
system. From data, the PSO Algorithm has is
slightly narrower steady state error to that of the
other conventional Algorithms.

Fig.3. Particle Swarm Optimization Output Power

Fig. 4. Power Output at Variable Irradiation Contidions

T ABLE II. EFFICIENCY OF PSO ALGORIT HM AT VARYING IRRADIANCE AND T EMPERAT URE

Irradiance(W/m2 ) Standard Results at Temperature 25 o C Efficiency (% )

Tracked Power (W) Maximum Power (W)

800 667.2 709.22 94.07

1000 865.5 882 98.12

1200 1005 1050.89 95.63

Speed of tracking the maximum power converter is calculated by taking the difference of
point is a crucial parameter which determines the power from PV Panel and the power at the load
system performance. PSO Algorithm reaches the side. These losses are much lesser in amount in the
steady state faster than the PSO, INC and Fuzzy PSO Algorithm.
logic Algorithms.

The Switching losses do occur in the MPPT

System due to high frequency of switching, the
losses due the MPPT tracker and the boost

979-8-3503-9728-4/23/$31.00 ©2023 IEEE 338

Authorized licensed use limited to: ANNA UNIVERSITY. Downloaded on August 24,2023 at 06:25:34 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI 2023)
IEEE Xplore Part Number: CFP23J32-ART; ISBN: 979-8-3503-9728-4

V.CONCLUSION [10] R. Vinifa and A. Kavitha, "Fuzzy based maximum power

point tracking on a PV system", Proc. Int. Conf. Energy
The Overall Efficiency of the system is Efficient T echnol. Sustain. (ICEET S), pp. 329 -333, 2013.
determined by the ratio between output power and [11] Vinifa R, Kavitha A, Immanuel Selwynraj A (2017)
maximum power produced under standard Maximum power pointtracking of boost converter on a PV
conditions. Efficiency of the Particle Swarm system using fuzzy logic. International Journal of Mechanical
Optimisation Algorithm is higher than that of the Engineering and T echnology (IJMET ) 8(12): 584–593.
Perturb and Observe, INC, Fuzzy, etc at almost all [12] Khanam, J. and Foo, S.Y., 2018, April. Neural Networks
varying Irradiation conditions. T echnique for Maximum Power Point T racking of Photovoltaic
Array. In SoutheastCon 2018 (pp. 1-4). IEEE.
The power extracted from the PSO MPPT
system is higher than that of the other MPPT [13] Mohamed, A.E., Marei, M.I. and Elkhattam, W., 2018,
December. A maximum power point tracking technique for PV
technique. An important aspect of this study is to under partial shading condition. In 2018 8th IEEE India
determine how an intelligent technique outstands International Conference on Power Electronics (IICPE) (pp. 1-
the conventional MPPT tracking algorithms and it 6). IEEE.
established well the differences between the [14] Gupta, A., Kumar, P., Pachauri, R.K. and Chauhan, Y.K.,
conventional MPPT Algorithms and Particle 2014, December. Performance analysis of neural network and
Swarm Optimization Algorithm performance on fuzzy logic based MPPT techniques for solar PV systems. In
various parameters of interest with accurate and 2014 6th IEEE power India international conference (PIICON)
(pp. 1-6). IEEE.
satiable outputs.
[15] Mohammed, S.S., Devaraj, D. and Ahamed, T .I., 2016,
REFERENCES January. Maximum power point tracking system for stand alone
solar PV power system using Adaptive Neuro-Fuzzy Inference
[1] Hashim, I.J., 2021, February. A new renewable energy System. In 2016 biennial international conference on power and
index. In 2021 6th International Conference on Renewable energy systems: towards sustainable energy (PEST SE) (pp. 1-
Energy: Generation and Applications (ICREGA) (pp. 229-232). 4). IEEE.
IEEE.

[2] Sharma, A. and Sharma, M., 2017, November. Power &

energy optimization in solar photovoltaic and concentrated solar
power systems. In 2017 IEEE PES Asia-Pacific Power and
Energy Engineering Conference (APPEEC) (pp. 1 -6). IEEE.

[3] Deepa Kumar Chy., Md. Khal luzzaman, “Exper mental

Assessment of PV Arrays Connected to Buck-Boost Converter
using MPPT and Non-MPPT T echnique by Implementing in
Real T me Hardware”, 3rd Internat onal Conference on
Advances in Electrical Engineering, pg.no. 306-309.

[4] Kumar, R., Sinha, S.K. and Pandey, K., 2016, July. Effect of
temperature, irradiation, humidity and wind on ideal/double
diode PV system performance. In 2016 IEEE 1st International
Conference on Power Electronics, Intelligent Control and
Energy Systems (ICPEICES) (pp. 1-5). IEEE.

[5] Shi, S., Wang, Y. and Jin, P., 2013, November. Study of
maximum power point tracking methods for photovoltaic power
generation system. In 2013 Chinese Automation Congress (pp.
835-840). IEEE.

[6] Y zhe J , B n Wu “A MZ Modulator B as Control System

Based on Var able Step P&O Algor thm”, IEEE Photonics
T echnology Letters, Volume: 32, Issue: 23.

[7] Dezso Sera, Laszlo Mathe, et al., ”On the Perturb & Observe
and Incremental Conductance MPPT Methods for PV Systems”,
IEEE Journal of Photovoltaics, VOL. 3, pg.no. 1070 -78.

[8] Amjad Al , Khal d Almuta r , et al., “Invest gat on of MPPT

T echniques Under Uniform and Non-Uniform Solar Irradiation
Condition - A Retrospect on”, IEEE Access.

[9] Li, X. and Wen, H., 2016, December. Evaluation of different

Maximum power point tracking techniques by using EN 50530
dynamic test standard. In 2016 IEEE International Conference
on Power Electronics, Drives and Energy Systems (PEDES) (pp.
1-6). IEEE.

979-8-3503-9728-4/23/$31.00 ©2023 IEEE 339

Authorized licensed use limited to: ANNA UNIVERSITY. Downloaded on August 24,2023 at 06:25:34 UTC from IEEE Xplore. Restrictions apply.