2021 IEEE PES/IAS PowerAfrica

Robust MPPT Tracking for PV Solar Power using

Metaheuristic Algorithms
Mohammed Tebaa* Mohammed Ouassaid Youssef Ait Ali
Mohammadia School of Engineers, Mohammadia School of Engineers, Mohammadia School of Engineers,
Mohammed V University in Rabat, Mohammed V University in Rabat, Mohammed V University in Rabat,
Morocco Morocco Morocco
Engineering for Smart and Sustainable Engineering for Smart and Sustainable Engineering for Smart and Sustainable
Systems Research Center Systems Research Center Systems Research Center
Email: tebaa.mohammed1@gmail.com Email: ouassaid@emi.ac.ma Email: youssef_aitali@um5.ac.ma
perturb & observe (P&O) [7], etc. MPPT methods based on
Abstract— The emphasis of this paper is a study and classical logical sequencing have simple structures and
compare robust of MPP tracking algorithms. In order to equipment. These algorithms are ideal for MPP monitoring
achieve the maximum power, a MPPT controller is often under uniform irradiation, when there is only one physical
associated with photovoltaic panels which controls the DC MPP point that can be provided by the panel. Conversely,
voltage and current. In this work, Particle Swarm these algorithms are very weak in tracking the global peak
Optimization (PSO), Grey Wolf Optimization (GWO), Perturb under variable irradiation. To cope with this problem, global
and Observe (P&O), and Fuzzy Logic Control (FLC) are used MPPTs based on meta-heuristic optimization algorithms are
in order to achieve the optimum power conditions. The studied introduced. These algorithms include; Genetic algorithm [8],
PV system contains six 244 W PV panel implemented in series GWO [9], PSO [10], Artificial Bee Colony (ABC) [11],
feeding a resistive load through a controllable DC-DC boost
Teaching–Learning–Based Optimization (TLBO) [12],
converter. To demonstrate the effectiveness of the proposed
techniques under standard test conditions, as well as partial
Cuckoo Search (CS) [13], Golden Section Search [14] and
shading conditions, simulations are performed under the Fibonacci Search [15]. PSO and GWO are simple and
MATLAB/Simulink environment. It is revealed in simulation effective strategies. These algorithms are characterized by
results that the PSO and P&O algorithms are more efficient in the speed and simplicity of their design as well as the
locating the effective MPP. Furthermore, when working under reliability of their results.
partial shading, the traditional P&O and FLC are failing to In this context, this paper presents a PV MPPT controller
locate the real MPP. Indeed, those MPPT techniques reach the
2021 IEEE PES/IAS PowerAfrica | 978-1-6654-0311-5/21/$31.00 ©2021 IEEE | DOI: 10.1109/POWERAFRICA52236.2021.9543413

based on four MPPT optimization techniques; Particle

local power peak, while PSO and GWO reach successfully the
Swarm Optimization and Grey Wolf Optimization offer a
effective MPP with an efficiency of 99% and 95% in only 92
ms and 69 ms, respectively.
better MPP tracking efficiency and a faster convergence
time, whether working under STC conditions or partial
Keywords— PV System, MPPT, Partial Shading, PSO, GWO, shading, while, on the other side, results show that MPP
P&O, FLC. conventional tracking techniques, such as FLC and P&O fail
to reach the MPP.
I. INTRODUCTION
This paper is structured as follows: Section II describes
Energy needs have increased in recent years. In parallel, the PV system elements; the optimization techniques used is
the foreseeable end of fossil fuels has put the emphasis on presented in Section III. In section IV, a performance
the search for alternative energy sources that are renewable comparison study of PSO, GWO, P&O and FLC is
inexhaustible and non-polluting [1]. performed and analyzed in Matlab/Simulink software.
The photovoltaic (PV) system is a widely used as a Finally, section V provides some conclusions.
renewable energy source, because of its advantages. PV II. MODELING OF PV SYSTEM
systems are made up of photovoltaic modules, which are
semiconductor devices that turn sunlight into electricity [2]. A. Photovoltaic Cell
However, PV power is highly reliant on irradiation and The photovoltaic effect consists of generating an electric
temperature, which causes the production rate of current directly from sunlight by means of solar cells
photovoltaic panels to become variable over time. The operating under the photoelectric effect. This technology was
power-voltage (P-V) curve of PV panels is non-linear and introduced by Edmond Becquerel in 1839, a solar cell
varies with irradiation and temperature conditions. Thus, composed of a current source which is influenced by the

series resistor R and a shunt resistor R that represents the

extracting the maximum power is an important issue. effect of light, a diode representing a P-N junction cell, a
In a PV system, an individual solar cell generates a very
small amount of voltage. To overcome this problem, a series internal resistance of the current flow in the PV cell [16].
of cells are connected together to form a solar module with a The PV generator has a non-linear voltage-current
sufficiently high energy yield. Furthermore, the PV module characteristic which is highly dependent on the insulation
maximum output power is affected by changing atmospheric [17]. The current generated by a solar cell is expressed as

I = I −I −I
conditions. Therefore, the point of the P-V characteristic follows:

where I is the current produced by incident solar light, Id is

corresponding to the real power consumption profile must be (1)
adjusted in static converters based on power electronics. The

resistor R is denoted by I . The current and voltage of the

strategy for ensuring this operation is well known as the diode current, the current flowing through the parallel
"Maximum Power Point Tracking" [3].
PV cell are related by the following expression:
I =N I −N I e −1 −N
Until these days, several common Algorithms for MPPT
have been proposed, including hill climbing [4], incremental ! "
# ! $
(2)
resistance [5], incremental conductance (INC) [6] and

978-1-6654-0311-5/21/$31.00 ©2021 IEEE

Authorized licensed use limited to: Yildirim Beyazit Univ. Downloaded on September 12,2023 at 13:26:24 UTC from IEEE Xplore. Restrictions apply.
 2021 IEEE PES/IAS PowerAfrica

where Irs represents the reverse saturation current of the cell,

V presents the PV cell output voltage (in Volt), the ideality B. Particle Swarm optimization
constant of the diode is denoted by A, the electron charge is PSO is a metaheuristic optimization algorithm based on
denoted by q (1.6021764610-19 C), the operating temperature the concept of swarm intelligence. It has the ability to solve

T and K, respectively. The photocurrent produced I is

and Boltzmann constant (1.380650310-23 J/K) are denoted by the mathematics problems applied in engineering. It is simple
in design which has a fast convergence rate and it presents a
satisfactory result in solving global optimization problems.

I = )I * + K - T − T /
proportional to the solar light which is expressed in eq. (3).
' Indeed, The PSO technique consists of a number of particles,
(
where I * represents the short-circuit current at the reference
(3) each one offers an individual solution to the problem at hand.

temperature and irradiation conditions, G (W/m2) and T are

During the specific search process, the collected data is
exchanged between the particles in order to achieve the best

respectively. K - is the temperature coefficient when the

the solar irradiation and the PV cell reference temperature, solution. The movement of these particles in the search space

current value equals I * .

is expressed in a simple mathematical equation. The position
and velocity of the particles are updated on the basis of the
eq. (6) and eq. (7). In this work, the position of the particle
B. DC Boost Converter swarm and the fitness are considered, respectively, as the

d- is updated by the following relations:

duty cycle and the output PV power. The particle at position
Boost converter is a DC-to-DC power converter that
d8- ( = d8- + V-8 (
increases voltage while decreasing current from its input to
(6)

V-8 = ωV-8 + c( r( )P<= > − d8- ( / + c? r? G<= > − d8- (7)

its output. It is a type of switched-mode power converter that
contains at least one uncontrolled and one controlled switch. (
Filters are attached to the converter output in order to

are denoted by c( and c? , respectively. r( and r? are random

minimize ripple. [18]. The input and output voltage are where, the individual and the collective learning coefficients

V = V values in [0,1]. ω represent the inertia moment.

(
related by the following equation:: .
(12
(4)
The output voltage Vo is bigger than that of the input Vs, This technique is based on the flowchart depicted in Fig.

converter R connected to the panel can be presented

as can be shown in Eq (4). The output resistance of the boost 3. Each particle, defined by its position is compared in each

according to R 3 as shown in eq. (5). recorded performance is attributed to P<= > , a comparison
iteration with its last position and therefore the maximum

R =
!45
determine G<= > .
between the best positions of each particle allows to
(12 6
(5)
The main component of the PV system is the MPPT C. Grey Wolf optimization
controller. Its role is to achieve the maximum possible PV
The GWO technique follows the hunting strategy carried out
power. The MPPT controller block is programmed by MPPT
by grey wolves groups in the wild. Grey wolves are
optimization algorithm. The PV voltage and current
considered one of the species that live in groups. They are
represent the two inputs of the MPPT controller. It provides a
arranged in a four-tiered hierarchy representing the
duty cycle as an output, which is directly applied to the
leadership hierarchy; alpha (α), beta (β), delta (δ) and omega
IGBT of the converter. They are maintained at the point
(ω).In order to model the hierarchy of wolves in GWO
where the MPP is recorded. The duty cycle value is changed
design, α is taken as the most suitable solution. Therefore, β
in accordance with the MPP monitoring. The output voltage
and δ represent the second and third best solutions. ω is
can step up by using boost converter depicted in Fig. 1 [19,
assumed for the remaining candidate solutions. The main
20].
steps of GWO, including the hunting stage, prey tracking and
III. MPPT OPTIMIZATION TECHNIQUES tracing, encircling and attacking prey are used to design the
A. Perturb and Observe Algorithm GWO algorithm [24]. Alpha is the group leader which is in
P&O algorithm is based on changing the voltage (ΔV), charge of making decisions about hunting, sleeping areas,
the voltage can be decremented or incremented at each and waking up time. Encircling behaviors is modeled as
iteration of calculation according to the value of ΔV [21], follows:
[22].
The change in power is directly recorded after changing
voltage. At the MPP, the rate of change in power is close to
zero. The voltage variation is in the same direction when
ΔV>0 leads to ΔP>0. Otherwise, it occurs in the opposite
direction [23]. Fig 2 illustrates the whole process of the P&O
algorithm.

Fig .1: DC-DC boost converter. Fig .2: P&O algorithm Flowchart.

Authorized licensed use limited to: Yildirim Beyazit Univ. Downloaded on September 12,2023 at 13:26:24 UTC from IEEE Xplore. Restrictions apply.
 2021 IEEE PES/IAS PowerAfrica

Fig .3: PSO algorithm flowchart.

BD
B⃗ = DCB⃗. BBBB⃗
X − BX⃗ t D
BX⃗ t + 1 = DX B⃗ . BBB⃗
BBBB⃗ t − A DD
(8)

BBBB⃗ ⃗
where t represents the current iteration, JK and J represent
(9) Fig .4. GWO algorithm flowchart.

So FLC does not require a precise mathematical model.

L⃗ and M⃗ are coefficient vectors which are calculated

the prey position and the grey wolf position, respectively. The FLC is usually composed of two inputs and one output
and is based on three parts: fuzzification and inference

B⃗ = 2. Ba⃗. BBB⃗
A r( − aB⃗
according to eq. (10) and eq. (11). engine then defuzzification [25]. Four fuzzy subgroups are
used to assign membership functions to linguistic variables:
B⃗ = 2. BBB⃗
C r?
(10)
LV (Low voltage), MV1 (medium voltage1) MV2 (medium

where a is a vector containing elements that linearly reduced

(11) voltage 2), HV (High voltage). The fuzzy subsets partition
and the membership functions’ shape are adapted to the
system as shown in Fig. 5. The determination of membership
with the following expression; a = 2 – 2 Q t/MaxIt, r1 and
from 2 to 0 during the algorithm iterations in accordance
functions of voltage and current is started by applying
manual duty cycle values (D) to determinate exactly the
r2 represent arbitrary vectors in [0 1]. GWO is used for direct value of D corresponding to the MPP as depicted in Fig. 6.
control of the duty cycle D. At the MPP, D keeps the same The membership functions of current and voltage is divided
value. As a result, the steady-state fluctuations discovered in into four domains based on their powers as mentioned in
traditional MPPT techniques are limited, and the power loss fig.5 and Fig. 6. The output of the fuzzy control system (the
due to oscillations is decreased which increase the efficiency duty cycle) and the input (voltage and current) are related to
of the system. In this case, D represents the position of the four rules (table in Fig. 5).

D- k + 1 = D- k − A. D
gray wolves. Therefore, eq. (9) will be written as follows:
(12) Voltage can reach values between 158V and 171 V
which belong to the area between high and medium 2
Equation. (12) can be written in function of the hunt voltages. The same goes for the current which its value is

BBBBB⃗
DW = DC BBBB⃗( BBBB⃗
X W t − BX⃗D ; BBBB⃗ XW − BBBB⃗
X( = BBBB⃗ A( . BBBBB⃗
DW
entities (α), (β), (δ), as expressed in eq. (13) and eq. (14). between 8.37 A and 8.56A. We obtain High domain duty

VBBBB⃗ BBBB⃗ BBBB⃗ B ⃗ BBBB⃗ BBBB⃗ BBBB⃗ BBBB⃗

cycle values in these ranges of current and voltage.
DY = DC? XY t − XD ; X ? = X Y − A ? DY
.
BBBB⃗ BBBB⃗[ BBBB⃗
XZ t − BX⃗D ; BBBB⃗ XZ − BBBB⃗
X[ = BBBB⃗ A[ . BBBB⃗
DZ
(13)
DZ = DC

BX⃗ = \]
BBBBB⃗ \
BBBBB⃗
6 BBBBB⃗
\^
[
(14)
The fitness function of GWO is defined as:
P)d8- / _ P)d81(
- /
where measured power is denoted by P, i and k represent the
(15)

current grey wolf number and the current iteration,

respectively. Fig. 4 depicts the GWO algorithm flowchart.
D. Fuzzy Logic Control
Fuzzy logic is a very powerful control method. For problem
solving, this technique uses multiple rules and can solve Fig .5: Proposed MPPT Fuzzy logic control.
multi-variable problems. Fuzzy logic controllers (FLC) have
the ability to work with non-linearities and imprecise input,

Authorized licensed use limited to: Yildirim Beyazit Univ. Downloaded on September 12,2023 at 13:26:24 UTC from IEEE Xplore. Restrictions apply.
 2021 IEEE PES/IAS PowerAfrica

Optimum values:
D=0.26-0.32
P=1399-1430W
0 V=172-184V
I=7.7-8.37A

HD : 1388-1430 W
MD2 : 1127-1311 W
MD1 : 1124-1318 W
LD : 0-787 W

Fig .6: Duty cycle values corresponding of the optimum MPPT.

Fig .7: P-V curve under STC and P.Sh.
IV. RESULTS AND DISSCUSIONS
A. Simulation environment
The PV array is powering a feeding load through a
controlled DC-DC boost. The MPPT optimization methods
proposed in this work are implemented in Matlab/Simulink
environment. The main components of the converter used in
the simulation are chosen as follows: inductance L = 10 mH,
capacity C = 500 µF, fs = 50 kHz as a switching frequency
of the PWM generator and the resistive Load R=45Ω. As
mentioned, in this paper four MPPT optimization techniques
are used PSO, GWO, P&O and fuzzy logic control. The PV
Fig .8: I-V curve under STC and P.Sh.
array contains six panels. Each panel generates a maximum
power of 240W. In order to decrease the current and deliver
a higher voltage value, the panels are connected in series.

Figure. 7 and 8 show the P-V and I-V curves,

respectively, under different irradiation conditions. The blue
curve presents the PV power under STC conditions. The PV
array can supply a maximal power of 1458 W. Red curve
shows the PV array power under partial shading (P.Sh)
where two power picks appear due to the irradiation change.
This could be explained by the fact that three panels are
under an irradiation of 1000W/m2, while the three other
panels are under 750W/m2. The global pick presents the
Fig .9: Output power using the four MPPT techniques under STC.
maximal power of the six panels with a PV power array of
1132W.
B. Simulation results and analysis
1) Results under STC conditions
The performance of the MPPT optimization techniques,
is assessed under the standard testing conditions (irradiation
= 1000 W/m2 and temperature = 25 °C). Results in Fig. 9 and
10 describe the evolution of power, voltage and current
curves using the four MPPT techniques.
Figure. 9 shows that the proposed fuzzy logic control
technique has the fastest response time compared to PSO, Fig .10: Output Current and voltage using all the four MPPT techniques.
GWO and P&O with a value of 8.4 ms. PSO and P&O on the
other side have a better MPP Tracking efficiency with 98% It also has a stable signal when the power reaches
compared to 96 % provided by the FLC. 1428W. Even with its good response time, GWO is
Results show that PSO has proven its high efficiency in considered as the less efficient MPPT technique among the
tracking the maximum power point with an acceptable four techniques because of its low efficiency. Indeed, the
response time. It shows, also, its stability during the maximum power that can be achieved by this method could
simulation when reaching the power of 1429W, a power not exceed 1185 W.
variation of ΔPPSO=1W is recorded with the PSO algorithm. 2) Results under partial shading
FLC has the fastest time of response compared to the other In order to analyze the effectiveness of the suggested
three MPPT techniques which is two times faster than PSO. MPPT methods under partial shading, six PV panels are
In addition, it presents an acceptable MPP tracking connected in series, three among them are under 750 W/m2
efficiency, it is evaluated at 96% with a good stability. irradiation power, when the three other panels are under
Furthermore, P&O has also presented a high MPP tracking STC. Fig. 11 and 12 show the results, which describe the
efficiency with a response time less than 0. 244s. evolution of the power,

Authorized licensed use limited to: Yildirim Beyazit Univ. Downloaded on September 12,2023 at 13:26:24 UTC from IEEE Xplore. Restrictions apply.
 2021 IEEE PES/IAS PowerAfrica

[1] P. A. Owusu and S. Asumadu-Sarkodie, “A review of renewable energy sources,

sustainability issues and climate change mitigation,” Cogent Eng., vol. 3, no. 1, p.
1167990, Dec. 2016.
[2] G. K. Singh, “Solar power generation by PV (photovoltaic) technology: A
review,” Energy, vol. 53, pp. 1–13, May 2013.
[3] B. Bendib, H. Belmili, and F. Krim, “A survey of the most used MPPT methods:
Conventional and advanced algorithms applied for photovoltaic systems,” Renew.
Sustain. Energy Rev., vol. 45, pp. 637–648, May 2015.
[4] P. Motsoeneng, J. Bamukunde, and S. Chowdhury, “Comparison of Perturb &
Observe and Hill Climbing MPPT Schemes for PV Plant Under Cloud Cover and
Varying Load,” in 2019 10th International Renewable Energy Congress (IREC),
Sousse, Tunisia, Mar. 2019, pp. 1–6.
[5] E. M. Ahmed, M. Orabi, and M. Shoyama, “High efficient variable step size
incremental resistance maximum power point tracker for PV battery charging
applications,” in 2013 IEEE Energy Conversion Congress and Exposition, Denver,
CO, USA, Sep. 2013, pp. 2435–2439.
[6] M. A. Elgendy, B. Zahawi, and D. J. Atkinson, “Assessment of the Incremental
Conductance Maximum Power Point Tracking Algorithm,” IEEE Trans. Sustain.
Energy, vol. 4, no. 1, pp. 108–117, Jan. 2013.
Fig .11: Output power using all the four MPPT techniques under P.Sh. [7] B. Subudhi and R. Pradhan, “A Comparative Study on Maximum Power Point
Tracking Techniques for Photovoltaic Power Systems,” IEEE Trans. Sustain. Energy,
vol. 4, no. 1, pp. 89–98, Jan. 2013.
[8] P. Megantoro, Y. D. Nugroho, F. Anggara, Suhono, and E. Y. Rusadi, “Simulation
and Characterization of Genetic Algorithm Implemented on MPPT for PV System
under Partial Shading Condition,” in 2018 3rd International Conference on
Information Technology, Information System and Electrical Engineering (ICITISEE),
Yogyakarta, Indonesia, Nov. 2018, pp. 74–78.
[9] S. K. Cherukuri and S. R. Rayapudi, “Enhanced Grey Wolf Optimizer based
MPPT Algorithm of PV system under Partial Shaded Condition,” Int. J. Renew.
Energy Dev., vol. 6, no. 3, p. 203, Nov. 2017.
[10] K. Ishaque, Z. Salam, M. Amjad, and S. Mekhilef, “An Improved Particle Swarm
Optimization (PSO)–Based MPPT for PV With Reduced Steady-State Oscillation,”
IEEE Trans. Power Electron., vol. 27, no. 8, pp. 3627–3638, Aug. 2012.
[11] S. Goyal, S. Mishra, and A. Bhatia, “Optimization of size of PV/wind/biodiesel
by using Artificial Bee Colony (ABC) algorithm,” in 2017 Recent Developments in
Fig .12: Output Current and voltage using all the four MPPT techniques. Control, Automation & Power Engineering (RDCAPE), Noida, Oct. 2017, pp. 220–
223.
voltage and current curves using the proposed algorithms. [12] Y. Kumar and R. Khare, “TLBO based cost analysis of Renewable mix in island
mode accounting employment creation and human development index,” in 2016 IEEE
Fig. 11 shows that the response time of the GWO is the Uttar Pradesh Section International Conference on Electrical, Computer and
Electronics Engineering (UPCON), Varanasi, India, 2016, pp. 189–194.
fastest, it is evaluated at 69 ms. PSO has recorded a response [13] D. A. Nugraha, K. L. Lian, and Suwarno, “A Novel MPPT Method Based on
time of 92 ms. In addition, only the PSO and GWO have Cuckoo Search Algorithm and Golden Section Search Algorithm for Partially Shaded
succeeded in locating the global MPP with efficiencies PV System,” Can. J. Electr. Comput. Eng., vol. 42, no. 3, pp. 173–182, 2019.
[14] B. Chen, T. Chen, and C. Xiao, “An Improved Maximum Power Point Tracking
evaluated at 99% and 95%, respectively, while the MPPT Algorithm with Golden Section Search for Photovoltaic System,” in 2018 5th
efficiency of P&O technique is so mediocre. This proves its International Conference on Systems and Informatics (ICSAI), Nanjing, Nov. 2018,
failure in the partial shading case with an efficiency that pp. 187–191.
[15] R. Ramaprabha, B. Mathur, A. Ravi, and S. Aventhika, “Modified Fibonacci
could not overcome 78%. Search Based MPPT Scheme for SPVA Under Partial Shaded Conditions,” in 2010
3rd International Conference on Emerging Trends in Engineering and Technology,
Figure. 12 shows, respectively, the output PV current and Goa, Nov. 2010, pp. 379–384.
voltage under partial shading, using the four proposed MPPT [16] Z. Ben Mahmoud, M. Hamouda, and A. Khedher, “Effects of series and shunt
techniques. In terms of stability, all the techniques prove the resistances on the performance of PV panel under temperature variations,” in 2016
International Conference on Electrical Sciences and Technologies in Maghreb
stability of the output signal after reaching their maximum (CISTEM), Marrakech & Bengrir, Morocco, Oct. 2016, pp. 1–7.
power points. [17] S. Messalti, A. G. Harrag, and A. E. Loukriz, “A new neural networks MPPT
controller for PV systems,” in IREC2015 The Sixth International Renewable Energy
V. CONCLUSION Congress, Sousse, Tunisia, Mar. 2015, pp. 1–6.
This paper has proposed a comparison study of MPP [18] R. Patil and H. Anantwar, “Comparitive analysis of fuzzy based MPPT for buck
and boost converter topologies for PV application - IEEE Conference Publication,”
tracking techniques applied to a PV array with the aim to 2017.
evaluate their ability to track successfully the effective MPP [19] S. Singh, S. Manna, M. I. Hasan Mansoori, and A. K. Akella, “Implementation
of Perturb & Observe MPPT Technique using Boost converter in PV System,” in 2020
under STC and partial shading conditions. Results have International Conference on Computational Intelligence for Smart Power System and
shown that, under standard test conditions, PSO algorithm Sustainable Energy (CISPSSE), Keonjhar, Odisha, India, Jul. 2020, pp. 1–4.
[20] D. K. Chy and Md. Khaliluzzaman, “Experimental assessment of PV arrays
has proven its high efficiency to track the maximum power. connected to Buck-Boost converter using MPPT and Non-MPPT technique by
FLC algorithms is considered as the fastest technique in implementing in real time hardware,” in 2015 International Conference on Advances
achieving the MPP, its tracking efficiency is acceptable. The in Electrical Engineering (ICAEE), Dhaka, Dec. 2015, pp. 306–309.
[21] J. J. Nedumgatt, K. B. Jayakrishnan, S. Umashankar, D. Vijayakumar, and D. P.
classical P&O has also shown a high-power tracking Kothari, “Perturb and observe MPPT algorithm for solar PV systems-modeling and
efficiency. Finally, GWO technique present the lower simulation,” in 2011 Annual IEEE India Conference, Hyderabad, India, Dec. 2011,
pp. 1–6.
performance among the proposed techniques. [22] M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs, “Theoretical and experimental
Under partial shading conditions, PSO algorithm has analyses of photovoltaic systems with voltageand current-based maximum power-
point tracking,” IEEE Trans. Energy Convers., vol. 17, no. 4, pp. 514–522, Dec. 2002.
proven its high-power efficiency that could overcome 99%. [23] S. U. Ramani, S. K. Kollimalla, and B. Arundhati, “Comparitive study of P&O
Also, it presents a good response time of 92 ms. and incremental conductance method for PV system,” in 2017 International
Furthermore, GWO algorithm shows good results in terms Conference on Circuit, Power and Computing Technologies (ICCPCT), Kollam,
India, Apr. 2017, pp. 1–7.
of power efficiency that can reach 95% in only 69 ms, [24] S. Mohanty, B. Subudhi, and P. K. Ray, “A New MPPT Design Using Grey Wolf
which is the fastest response time between the four MPPT Optimization Technique for Photovoltaic System Under Partial Shading Conditions,”
IEEE Trans. Sustain. Energy, vol. 7, no. 1, pp. 181–188, Jan. 2016.
techniques. The classical FLC and P&O have proved their [25] N. Khaehintung, P. Sirisuk, and A. Kunakorn, “Grid-connected Photovoltaic
failure in locating the global maximum power point. System with Maximum Power Point Tracking using Self-Organizing Fuzzy Logic
Controller,”p.4.
REFERENCES

Authorized licensed use limited to: Yildirim Beyazit Univ. Downloaded on September 12,2023 at 13:26:24 UTC from IEEE Xplore. Restrictions apply.