Available online at www.sciencedirect.

com
ScienceDirect

Energy Reports 8 (2022) 1179–1188

www.elsevier.com/locate/egyr

2021 8th International Conference on Power and Energy Systems Engineering (CPESE 2021),
10–12 September 2021, Fukuoka, Japan

Optimal parameter identification of triple diode model for solar

photovoltaic panel and cells
Hegazy Rezka,b ,∗, Mohammad Ali Abdelkareemc,d,e
a
College of Engineering at Wadi Addawaser, Prince Sattam Bin Abdulaziz University, Saudi Arabia
b Electrical Engineering Department, Faculty of Engineering, Minia University, Egypt
c Center for Advanced Materials Research, University of Sharjah, PO Box 27272, Sharjah, United Arab Emirates
d Chemical Engineering Department, Minia University, Elminia, 61516, Egypt
e Department of Sustainable and Renewable Energy Engineering, University of Sharjah, PO Box 27272, Sharjah, United Arab Emirates

Received 5 November 2021; accepted 8 November 2021

Available online 27 November 2021

Abstract
The correct parameter determination of the photovoltaic module and the solar cell is considered an important phase to
deliver a reliable simulation for the PV system characteristics. The triple diode model (TDM) has been examined to model
the PVM 752 GaAs thin-film PV solar cell (SC), STM6 PV module, and RTC SC. A set of the measured I-V data at various
levels of the weather conditions has been considered for the studied PV solar cell/modules. A new marine predators algorithm
(MPA) is used to identify the TDM parameters based on measured datasets. Throughout the optimization procedure, the nine
undetermined parameters of TDM are employed as decision variables but the fitness function requires to be least is the RMSE
between the experimental and calculated datasets. The achieved results by MPA are compared with different algorithms. The
comparison confirmed the superiority of MPA.
© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems Engineering,
CPESE, 2021.

Keywords: Modelling; Marine predators algorithm; Photovoltaic; Solar cells

1. Introduction
A wide usage of fossil fuels and conventional energy resources for power generation results in greenhouse
gases and raises the global warming effect and adversely disturbs the human life and the environment [1]. Thereby
renewable energy resources (RERs) have been penetrated as alternatives to fossil fuels to minimize the greenhouse
and environmental effects. RERs such solar energy, wind energy, geothermal energy, biomass energy, . . . etc.,
demonstrated a promising result in several applications [2,3].
∗ Corresponding author at: College of Engineering at Wadi Addawaser, Prince Sattam Bin Abdulaziz University, Saudi Arabia.
E-mail address: hr.hussien@psau.edu.sa (H. Rezk).

https://doi.org/10.1016/j.egyr.2021.11.179
2352-4847/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http:
//creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 2021 8th International Conference on Power and Energy Systems
Engineering, CPESE, 2021.
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

As per the assessed reports of the International Energy Agency (IEA), the installation size of photovoltaic systems
(PVS) is increasing to a great level. Thereby, it is important to estimate the output power accurately under different
conditions. To achieve this, an accurate model of PVS components is highly required. To achieve this efficient
model, the manufacturer cannot provide required parameters. Additionally, because of the photovoltaic non-linearity
characteristics, an exact PV model becomes a difficult task.
There are three familiar PV models: single diode model (SDM), double diode model (DDM), and triple diode
model (TDM) [4]. The TDM is recognized to give an appropriate model for solar PV cell/module characteristics
under various conditions [5]. Determining the appropriate and accurate parameters of the TDM is the crucial task to
provide a consistent model. With this motivation, numerous researchers concentrating on designing various methods
for efficient PV modelling.
Several approaches are used until now to build a precise PV model. These approaches include analytical,
deterministic, and meta-heuristic. The drawback of analytical approaches is assumptions should be made to decrease
parameters number. The inappropriate assumptions of these values deteriorate the overall performance. The iterative
procedures like Newton–Raphson with hood estimator [6] and Gauss–Seidel [7] have being employed to remove the
limitations of the analytical procedures. The accuracy of such methods is low as the primary solutions sometimes
being far from the optimal solutions [8].
The limitations of the analytical and deterministic procedures are removed thanks to using metaheuristic
algorithms. They are used in various fields of applications as well. Up to date, several optimizers are used to deal the
PV parameter identification process. Such optimizers include imperialist competitive algorithm [9], enhanced moth
search algorithm [10], artificial ecosystem-based optimization [11], stochastic fractal search optimization algorithm
[12], chaotic whale optimization algorithm [13] and Water Cycle Algorithm [14].
Nevertheless, the stated optimizers presented high ability in convergence speed and global search, these
optimizers have some limits for instance most of these algorithms diverged their accuracy between SDM and
DDM. Yet, it is challenging to accomplish consistent accuracy. Consequently, the precision and consistency in
PV parameter determination yet to be done. As Well, few research works were concentrated on TDM because it
has nine parameters that increase the difficulty of the optimization process. Therefore, in the current research work,
the triple TDM has been examined to model the PVM 752 GaAs thin-film PV solar cell, STM6 PV module, and
RTC solar cell using a recent marine predators algorithm (MPA). Throughout the optimization procedure, the nine
unidentified parameters of TDM are employed as decision variables but the fitness function that needs to be least
is the RMSE between the measured data and calculated data.
The primary contributions of the current research work are summarized as follows:
• A new application of MPA optimizer to identify the parameters of the TDM for PVM 752 GaAs thin-film PV
solar cell, STM6 PV module, and RTC solar cell.
• The obtained results by MPA are compared with other algorithms.
• The accuracy and superiority of MPA in detraining parameters of the TDM are proved.
The rest of the work is arranged as following. Section 2 offers the mathematical representation of the TDM and
problem formulation. A brief overview of the MAP optimizer is presented in Section 3. The obtained results are
described and discussed in Section 4. Lastly, Section 5 outlines the conclusions of the current research work.

2. Triple diode model and problem formulation

A mathematical model of the triple-diode model of solar PV cell and module is described in this section. The
equivalent circuit of TDM of PV solar cell (SC) is explained in the Fig. 1.
The solar cell output current considering TDM can be estimated as following.
[ ( ) ] [ ( ) ]
V + I Rs V + I Rs
I = I pv − I01 E x p − 1 − I02 E x p −1
n 1 Vt n 2 Vt
[ ( ) ] ( )
V + I Rs V + I Rs
− I03 E x p −1 − (1)
n 3 Vt Rp
where;
I P V denotes the photo-generated current.
Vt is the thermal voltage.
Rs and R p are the series and parallel resistance respectively.
1180
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

Fig. 1. TDM circuit of SC.

n1 , n2 and n3 denote the ideality factor of D1 , D2 and D3 respectively.

To increase the rating of the voltage and current, several solar cells are connected. Therefore, considering no. of
parallel strings (N p ) and no. series SCc (Ns ), the PV output current I can be defined by the following relation:
⎡ ⎛( )⎞ ⎤ ⎡ ⎛( )⎞ ⎤
V I Rs V I Rs
Ns
+ Np N
+ Np
I = I pv N P − I01 N P ⎣ E x p ⎝ ⎠ − 1⎦ − I01 N P ⎣ E x p ⎝ s ⎠ − 1⎦
n 1 Vt n 1 Vt
⎡ ⎛( )⎞ ⎤ ⎡(VN )⎤
V I Rs p
Ns
+ Np Ns
+ I R s
− I01 N P ⎣ E x p ⎝ ⎠ − 1⎦ − ⎣ ⎦ (2)
nn 1 Vt Rp

Considering Eq. (5), the TDM model contains nine unknown parameters (I pv , I01 , I01 , I01 , n 1 , n 1 , n 1 , Rs , R p ).
Accurate values unknown parameters must be estimated to get the exact characteristics of the solar PV module. The
nine unknown parameters of TDM require to be identified accurately. Throughout the optimization procedure, these
parameters are selected to be decision variables while the RMSE between the measured and estimated datasets are
used as a cost function that requires to be minimum [15].

3. Marine predators algorithm

The idea of the MPA has been originated from the biologic interface between the predator and the prey. The
model of MPA optimizer is clearly illustrated in [16]. The optimization process of MPA involves three periods
based on the speed fraction between prey as well as predator. High speed fraction is the dominant in the first period
while the unity and low speed are assigned second and third periods, respectively.
Period 1 (High-speed fraction): In this period, throughout the initial third of iterations, the prey is quicker than
the predator. The following equation can be used to adjust the position of prey [16].
Si = R B × (Elitei − R B × Z i ) , i = 1, 2, . . . , n (3)
Z i = Z i + P.R × Si (4)
where, R ϵ [0, 1] is a vector of random numbers, P = 0.5 and R B is Brownian motion vector.
Period 2 (Unity speed fraction): Throughout the middle of the optimization procedure, the speed of prey equal
to predator speed. The predator and prey move based on the Brownian and Lévy flight strategies, respectively. The
following relations can be used to adjust the positions [16].
Si = R L × (Elitei − R L × Z i ) , i = 1, 2, . . . , n/2 (5)
Z i = Z i + P.R × Si (6)
where R L is a variable created arbitrarily by Lévy distribution.
Si = R B × (R B × Elitei − Z i ) , i = 1, 2, . . . , n/2 (7)
) t
t 2 tmax
(
Z i = Elitei + P.C F × Si , C F = 1 − (8)
tmax
1181
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

where t and tmax are present and upper limit of iterations.

Period 3 (low-speed fraction): during the last third of iterations, in contrast with the first period, the predator is
quicker than prey. The position is adjusted considering the following relation:
Si = R L × (R L × Elitei − Z i ) , i = 1, 2, . . . , n (9)
) t
t 2 tmax
(
Z i = Elitei + P.C F × Si , C F = 1 − (10)
tmax

4. Results and discussion

To demonstrate the superiority of the suggested MPA optimizer, it is employed to define the nine unknown
parameters of TDM. Three different scenarios are considered for the evaluation. A solar PV module STM6-40/36 is
considered for the first scenario. It comprises 36 SCs in series of mono-crystalline silicon cells. The experimental
data was measured at 51 ◦ C. In the second scenario, the common data of a R.T.C. France silicon SC working at
a 33 ◦ C and 1000 W/m2 respectively for the temperature and radiation. For the third scenario, a dataset of PVM
752 GaAs thin-film PV solar cell at a temperature of 44 ◦ C is considered. Table 1 presents the specifications of the
studied PV modules and SCs.
Table 1. General data related to PVM 752 GaAs thin-film PV SC, STM6 PV module and RTC solar cell.
Parameter Type of solar cell and panel
PVM 752 GaAs STM6-40/36 module R.T.C solar cell
Number of samples 44 18 26
Temperature, C 25 51 33
Irradiance, W/m2 1000 1000 1000
Short circuit current, A 0.1001 1.663 0.7605
Open circuit voltage, V 0.9926 19.08 0.5727
Current @ MPP, A 0.0937 1.5 0.6755
Voltage @ MPP, V 0.8053 16.98 0.4590
Number of cells 1 36 1

4.1. Results of first scenario

Based on the data of STM6-40/36 PV module, the suggested approach of MPA is applied to estimate the optimum
parameters of TDM. The obtained parameters are given in Table 2. Table 2 also illustrates a comparison between
the MPA approach and other methods. The minimum RMSE of 1.778E−03 is attained by the MPA. Additionally,
the coefficient of determination for TDM applying MPA is 0.9998. This validates the superiority of the proposed
MPA in determining the parameters of the TDM STM6-40/36 PV module. As demonstrated in Fig. 2, there is a
complete agreement between the calculated data by MPA and measured data.
The absolute error versus the measured PV module voltage applying various approaches is presented in Fig. 3.
The maximum rates of absolute error values were 0.006013, 0.007, 0.00515, 0.00518, and 0.00598 respectively
for CWOA, ELPSO, MPA (DDM), MPA (TDM), and STLBO. The mean absolute error using MPA is 1.22E−03.
This demonstrates the superiority of MPA in comparison with methods. Changing the fitness function throughout
identification of STM6-40/36 PV module applying MPA optimizer for both DDM and TDM is shown in Fig. 4.
Both curves convergence to optimal solution 700 iterations. The best solutions are 1.781E−03 and 1.778E−03
respectively for DDM and TDM.

4.2. Results of second scenario

To prove the accuracy of MPA to estimate the parameters of the SC, a dataset of GaAs solar cell has been used
in the second scenario. The suggested approach of MPA is used to determine the optimum parameters of TDM.
The obtained parameters are given in Table 3. Table 3 also shows a comparison between the MAP and some other
methods. The minimum RMSE of 5.67E−05 is accomplished by the MPA. Besides, the coefficient of determination
for TDM using MPA is 0.999998. This proves the superiority of MPA in determining the parameters of TDM for
1182
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

Table 2. Optimum values of SDM, DDM and TDM parameters of STM6-40/36 PV module.
Ref Method Model Iph Io1 Io2 Io3 n1 n2 n3 Rs RP RMSE
[17] ELPSO SDM 1.66626 4.60E−07 – – 50.45864 – – 0.5 497.747 2.18E−03
[17] CPSO SDM 1.66626 4.60E−07 – – 50.45865 – – 0.5 497.748 2.18E−03
[17] BSA SDM 1.65940 6.34E−07 – – 51.76275 – – 0.52734 723.391 3.63E−03
[17] ABS SDM 1.66718 4.66E−07 – – 50.47517 – – 0.5 495.520 2.40E−03
[17] ELPSO DDM 1.66484 1.67E−08 6.21E−06 – 41.99348 67.34 – 0.5 606.888 1.83E−03
[17] CPSO DDM 1.66474 2.53E−08 8.78E−06 – 42.81694 70.73 – 0.5 611.747 1.83E−03
[17] BSA DDM 1.66111 1.20E−06 8.92E−07 – 100 52.87 – 0.5 924.813 4.03E−03
[17] ABS DDM 1.66347 8.94E−06 1.10E−12 – 71.46498 27.79 – 1.23643 938.209 2.05E−03
[18] CWOA SDM 1.6646 1.43E−06 – – 1.4994 – – 5(m) 14.9371 1.90E−03
[13] BMO SDM 1.7 1.41E−06 – – 1.5 – – 5(m) 15.4 1.90E−03
[13] STLBO SDM 1.7 1.63E−06 – – 1.5 – – 5(m) 15.4 1.80E−03
Proposed MPA DDM 1.66403 2.046E−06 1.11E−06 – 1.53335 2.947 – 0.07248 546.321 1.781E−03
Proposed MPA SDM 1.66412 1.24E−06 5.56E−06 2.20E−06 1.48521 2.265 4.7889 0.13682 555.087 1.778E−03

Fig. 2. The experimental data vs. the calculated data for STM6 PV module by MPA for TDM.

Fig. 3. Absolute errors vs. measured voltage for SDM, DDM and TDM of STM6 by different methods.

GaAs solar cell. As demonstrated in Fig. 5, there is a great agreement between the calculated data and measured
dataset.
The absolute error versus the measured PV module voltage applying various approaches is presented in Fig. 6.
The maximum rates of absolute errors were 0.0006, 0.0042, 0.057, and 0.000137 respectively for ACT, ELPSO
(TDM), ELPSO (SDM), and MPA (TDM). The mean absolute error using MPA is 4.68E−05. This proves the
superiority of the suggested MPA in comparison with other approaches. The change in the objective function value
throughout the optimization process estimation of GaAs solar cell applying MPA approach for TDM is illustrated
in Fig. 7. The proposed strategy succeeds to reach the minimum RMSE of 5.67E−05.
1183
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

Fig. 4. Objective function variation throughout optimization process of STM6-40/36 using MPA optimizer.

Fig. 5. The experimental data vs. the calculated for GaAs solar cell using MPA algorithm for TDM.

Table 3. Optimum values of SDM, DDM and TDM parameters of GaAs SC.
Ref Method Model Iph Io1 Io2 Io3 n1 n2 n3 Rs RP RMSE
[19] ACT SDM 0.099985 1.94E−11 – – 1.73411 – – 0.61657 684.519 2.10E−04
[20] ELPSO SDM 0.115016 0.0 – – 1.77E+00 – – 0.15905 14.42951 2.54E−02
[20] CPSO SDM 0.11653 0.00 – – 1.62E+00 – – 0.34658 14.24198 2.54E−02
[20] BSA SDM 0.103903 8.49E−11 – – 1.858574 – – 0.5 100 2.15E−03
[20] ABS SDM 0.103312 3.20E−11 – – 1.774159 – – 0.5 100 2.04E−03
[20] ELPSO DDM 0.103192 1.76E−10 1.00E−12 – 2 1.5711 – 0.5 100 2.08E−03
[20] CPSO DDM 0.102688 1.00E−12 1.00E−12 – 1.572718 1.5727 – 0.5 100 2.30E−03
[20] BSA DDM 0.102497 1.00E−12 8.06E−11 – 1.63559 1.8743 – 0.5 100 2.18E−03
[20] ABS DDM 0.103252 4.00E−11 1.00E−12 – 1.792987 2.00 – 0.5 100 2.04E−03
[21] BA SDM 0.0999 1.94E−11 – – 1.7341 – – 0.6165 684.516 2.10E−04
Prop. MPA TDM 0.0999 1.08E−15 6.32E−09 1.99E−07 1.2159 2.5576 4.9353 0.72603 1030.383 5.67E−05

4.3. Results of the third scenario

Using data of R.T.C SC, the suggested approach of MPA is applied to estimate the optimum parameters of TDM.
The obtained parameters are given in Table 4. Table 4 also displays a comparison between the MPA approach
and the other algorithms. The minimum RMSE of 6.85E−04 is achieved by the proposed strategy. Moreover, the
coefficient of determination for TDM using MPA is 0.999995. This proves the superiority of the MPA in determining
the parameters of TDM for R.T.C SC. As explained in Fig. 8, there is a good agreement between the calculated
data and measured datasets.
The absolute error versus the measured PV module voltage applying various approaches is presented in
Fig. 9. The maximum rates of absolute errors were 2.51E−03, 2.51E−03, 1.34E−03, 2.50E−03, and 4.53E−03
respectively for BBOM, IMFO, MPA, OBWOA, and SA. The mean absolute error using MPA is 5.73E−04. This
1184
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

Fig. 6. Absolute errors vs. measured voltage for SDM, DDM and TDM of GaAs solar cell employing all algorithms.

Fig. 7. Objective function variation through parameter estimation of GaAs solar cell using MPA optimizer.

Fig. 8. Experimental data vs. calculated for RTC solar cell using MPA algorithm for TDM.

proves the superiority of the suggested MPA in comparison with other approaches. The change in the objective
function value throughout the optimization process estimation of R.T.C SC applying the MPA approach for TDM
is illustrated in Fig. 10. The proposed strategy succeeds to reach the minimum RMSE of 6.85E−04.

5. Conclusion
In this work, the authors proposed and validated a recent marine predators algorithm (MPA) for effective
modelling of triple diode model (TDM). Nine unidentified parameters are effectively calculated. The efficacy of
the suggested MPA has been confirmed via I-V and P-V curves of considered PV solar cell/module using MPA.
1185
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

Table 4. Optimum values of SDM, DDM and TDM parameters of RTC SC.
Ref Method Model Iph Io1 Io2 Io3 n1 n2 n3 Rs RP RMSE
[18] EPSO TDM 0.7679 1.35E−07 1.59E−07 6.44E−07 1.4117 1.9035 1.8329 0.037287 56.533 7.54E−04
[18] FC-EEPSO3 TDM 0.7607 1.68E−07 9.20E−07 2.59E−07 1.4257 1.9949 2 0.037215 55.742 7.44E−04
[22] OBWOA TDM 0.7607 2.35E−07 2.22E−05 4.57E−07 1.4543 2 2 0.03668 55.4448 9.85E−04
[22] ABC TDM 0.7607 2.00E−07 5.00E−07 2.10E−07 1.4414 1.9 2.05 0.03687 55.8344 9.85E−04
[22] STLBO TDM 0.7608 2.35E−07 2.30E−07 4.44E−07 1.4541 2 2 0.0367 55.2641 9.83E−04
[22] OBWOA DDM 0.7607 2.23E−07 6.20E−07 na 1.4945 2 na 0.03671 55.399 9.83E−04
[14] ER-WCA SDM 0.7607 3.23E−07 na na 1.4811 na na 0.036381 53.691 9.86E−04
[14] ER-WCA DDM 0.7607 2.26E−07 7.51E−07 na 1.4509 2 na 0.036742 55.4858 9.82E−04
[23] BBO-M DDM 0.7608 5.91E−07 2.452E−07 na 2 1.45798 na 0.03664 55.0494 9.83E−04
[23] BBO-M SDM 0.7607 3.19E−07 na na 1.4798 na na 0.03642 53.36227 9.86E−04
[24] IMFO SDM 0.7607 3.23E−07 na na 1.4812 na na 0.03638 53.71456 9.86E−04
[24] MFO SDM 0.7609 3.01E−07 na na 1.4694 na na 0.03596 51.81957 9.95E−04
[24] WCMFO SDM 0.7607 3.23E−07 na na 1.4812 na na 0.03638 53.69502 9.86E−04
[25] SCA SDM 0.7650 6.79E−07 na na 1.5609 na na 0.03544 50.14796 5.81E−03
[25] IMFO DDM 0.7608 2.34E−07 6.84E−07 na 1.4537 2 na 0.03671 55.2997 9.83E−04
[25] MFO DDM 0.7609 1.60E−07 3.07E−07 na 1.4655 1.82735 na 0.03699 53.27466 9.87E−04
[24] WCMFO DDM 0.7608 2.40E−07 4.43E−07 na 1.4568 1.90457 na 0.03661 55.11475 9.84E−04
[24] CSO SDM 0.7608 3.23E−07 na na 1.4812 na na 0.03638 53.7185 9.86E−04
[26] SA SDM 0.762 4.80E−07 na na 1.5172 na na 0.0345 43.103 1.70E−03
[26] NM-MPSO SDM 0.7608 3.23E−07 na na 1.4812 na na 0.03638 53.7222 9.86E−04
[26] NM-MPSO DDM 0.7608 2.25E−07 7.55E−07 na 1.4505 1.99998 na 0.03675 55.5296 9.83E−04
Prop. MPA TDM 0.7610 5.946E−09 1.835E−07 2.2017E−04 1.4308 1.43074 5.00 0.037754 79.3102 6.85E−04

Fig. 9. Absolute errors vs. measured voltage for SDM, DDM and TDM of RTC solar cell using different strategies.

Fig. 10. Changing the objective function throughout parameter estimation of RTC solar cell using MPA optimizer.

1186
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

The PV solar cell/modules characteristics generated using the proposed method precisely fit with experimental
datasets with lower errors. The performance metrics such as RMSE, MAE, and coefficient of determination have
been calculated and compared with other techniques for the considered solar PV cell/modules. From the presented
performance metrics, it is noteworthy to mention that the proposed MPA attains less RMSE than other counterparts.
Accordingly, it can be decided that the suggested MPA is recommended as an effective algorithm to determine the
unknown parameters of TDM.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could
have appeared to influence the work reported in this paper.

References
[1] Olabi AG, Wilberforce T, Sayed ET, Elsaid K, Rezk H, et al. Recent progress of graphene based nanomaterials in bioelectrochemical
systems. Sci Total Environ 2020;749:141225.
[2] Abdelkareem MA, Sayed ET, Mohamed HO, Obaid M, Rezk H, et al. Nonprecious anodic catalysts for low-molecular-hydrocarbon
fuel cells: Theoretical consideration and current progress. Prog Energy Combust Sci 2020;77:100805.
[3] Mustafa RJ, Gomaa MR, Al-Dhaifallah M, Rezk H. Environmental impacts on the performance of solar photovoltaic systems.
Sustainability 2020;12:608.
[4] Khanna V, Das BK, Bisht D, Singh PK. A three diode model for industrial solar cells and estimation of solar cell parameters using
PSO algorithm. Renew Energy 2015;78:105–13.
[5] Ibrahim IA, Hossain MJ, Duck BC, Nadarajah M. An improved wind driven optimization algorithm for parameters identification of a
triple-diode photovoltaic cell model. Energy Convers Manage 2020;213:112872.
[6] Ayang A, Wamkeue R, Ouhrouche M, Djongyang N, Salomé NE, et al. Maximum likelihood parameters estimation of single-diode
model of photovoltaic generator. Renew Energy 2019;130:111–21.
[7] Chatterjee A, Keyhani A, Kapoor D. Identification of photovoltaic source models. IEEE Trans Energy Convers 2011;26:883–9.
[8] Blaifi SA, Moulahoum S, Taghezouit B, Saim A. An enhanced dynamic modeling of PV module using Levenberg–Marquardt algorithm.
Renew Energy 2019;135:745–60.
[9] Fathy A, Rezk H. Parameter estimation of photovoltaic system using imperialist competitive algorithm. Renew Energy 2017;111:307–20.
[10] Fathy A, Elaziz MA, Sayed ET, Olabi AG, Rezk H. Optimal parameter identification of triple-junction photovoltaic panel based on
enhanced moth search algorithm. Energy 2019;188:116025.
[11] Yousri D, Rezk H, Fathy A. Identifying the parameters of different configurations of photovoltaic models based on recent artificial
ecosystem-based optimization approach. Int J Energy Res 2020;44:11302–22.
[12] Rezk H, Babu TS, Dhaifallah MA, Ziedan HA. A robust parameter estimation approach based on stochastic fractal search optimization
algorithm applied to solar PV parameters. Energy Rep 2021;7:620–40.
[13] Oliva D, Aziz MAE, Hassanien AE. Parameter estimation of photovoltaic cells using an improved chaotic whale optimization algorithm.
Appl Energy 2017;200:141–54.
[14] Kler D, Sharma P, Banerjee A, Rana KPS, Kumar V. PV cell and module efficient parameters estimation using Evaporation Rate
based Water Cycle Algorithm. Swarm Evol Comput 2017;35:93–110.
[15] Ibrahim MN, Rezk H, Dhaifallah MA, Sergeant P. Solar array fed synchronous reluctance motor driven water pump: An improved
performance under partial shading conditions. IEEE Access 2019;7:77100–15.
[16] Faramarzi A, Heidarinejad M, Mirjalili S, Gandomi AH. Marine Predators Algorithm: A nature-inspired metaheuristic. Expert Syst
Appl 2020;152:113377.
[17] Jordehi AR. Enhanced leader particle swarm optimisation (ELPSO): An efficient algorithm for parameter estimation of photovoltaic
(PV) cells and modules. Sol Energy 2018;159:78–87.
[18] Yousri D, Thanikanti SB, Allam D, Ramachandaramurthy VK, Eteiba MB. Fractional chaotic ensemble particle swarm optimizer for
identifying the single, double, and three diode photovoltaic models’ parameters. Energy 2020;195:116979.
[19] Muhammad FF, Sangawi K, Hashim AW, Ghoshal S, Abdullah SK, et al. Simple and efficient estimation of photovoltaic cells and
modules parameters using approximation and correction technique. PLoS One 2019;14:0216201.
[20] Jordehi AR. Enhanced leader particle swarm optimisation (ELPSO): An efficient algorithm for parameter estimation of photovoltaic
(PV) cells and modules. Sol Energy 2018;159:78–87.
[21] Muhammadsharif FF, Hashim S, Hameed SS, Ghoshal SK, Abdullah IK, et al. Brent’s algorithm based new computational approach
for accurate determination of single-diode model parameters to simulate solar cells and modules. Sol Energy 2019;193:782–98.
[22] Elaziz MA, Oliva D. Parameter estimation of solar cells diode models by an improved opposition-based whale optimization algorithm.
Energy Convers Manage 2018;171:1843–59.
[23] Niu Q, Zhang L, Li K. A biogeography-based optimization algorithm with mutation strategies for model parameter estimation of solar
and fuel cells. Energy Convers Manage 2014;86:1173–85.
[24] Zhang H, Heidari AA, Wang M, Zhang L, Chen H, et al. Orthogonal Nelder–Mead moth flame method for parameters identification
of photovoltaic modules. Energy Convers Manage 2020;211:112764.
1187
H. Rezk and M.A. Abdelkareem Energy Reports 8 (2022) 1179–1188

[25] Sheng H, Li C, Wang H, Yan Z, Xiong Y, et al. Parameters extraction of photovoltaic models using an improved moth-flame
optimization. Energies 2019;12:3527.
[26] Hamid NFA, Rahim NA, Selvaraj J. Solar cell parameters identification using hybrid Nelder–Mead and modified particle swarm
optimization. J Renew Sustain Energy 2016;8:015502.

1188