Results in Optics 12 (2023) 100445

Contents lists available at ScienceDirect

Results in Optics
journal homepage: www.sciencedirect.com/journal/results-in-optics

A novel numerical method for estimation the photovoltaic cells/modules

parameters based on dichotomy method
Abdelfattah Elhammoudy *, Mustapha Elyaqouti , Dris Ben Hmamou , El Hanafi Arjdal ,
Driss Saadaoui , Souad Lidaighbi , Imade Choulli
Laboratory of Materials, Signals, Systems and Physical Modelling, Faculty of Science, Ibn Zohr University, Agadir, Morocco

A R T I C L E I N F O A B S T R A C T

Keywords: The demand for photovoltaic (PV) energy has been growing in recent years, parallel to an increase in scientific
Photovoltaic cells research about the PV cells, such as the electrical modelling and extraction of unknown PV model parameters.
Single-diode model This paper introduces a highly efficient and straightforward numerical method to determine the unknown pa­
Parameters extraction
rameters for the single-diode model. The proposed approach utilizes Dichotomy (Bisection) to estimate the series
Dichotomy method
Numerical method
resistance Rs and an iterative process to determine the diode ideality factor a and the parameters that achieve the
minimum Root Square Error (RMSE). The proposed approach has been tested on five cells/modules from
different technologies. The results show that the proposed approach outperforms various analytical, numerical,
and meta-heuristic approaches in terms of accuracy. The RMSE produced by the proposed method is in order of
7.8514e-04 A and 2.2256e-03 A for the RTC France cell and the PWP201 module, respectively.

its electrical circuit.

The equivalent circuit made up of a current source represents the flux
1. Introduction
of incident irradiation, a diode represents the P-N junction and two re­
sistances, shunt resistance Rsh and series resistance Rs , represent the
Renewable energies have attracted much attention in the last few
contact related to the connection among the cell and its wires and the
years to reduce the consummation of fossil energies and to minimize the
semiconductor’s internal resistance (Rosa-Clot and Tina, 2018). The
greenhouse gases in the atmosphere (Gross et al., 2003). The advantage
Kirchhoff laws can be used to express the characteristic equation of the
of renewable energy sources, such as solar, hydraulic, and wind energy,
single-diode model as follows:
is that they are constantly renewed from natural processes. Most of the
( ( ) )
world focuses on solar energy and exactly in photovoltaic energy V + Rs I V + Rs I
because it directly converts sunlight into usable electricity. It allows for I = Ipv − I0 exp − 1 − (1)
aVT Rsh
various applications, such as off-grid PV systems. For example, the use of
photovoltaic energy for a water pumping system to meet the water de­ The characteristic equation has five unknown parameters: the first is
mand of the irrigation field (Sharma et al., 2020). The demand for PV the light-generated current Ipv , the second is the diode reverse saturation
energy is rapidly increasing, where the PV electricity capacity in the current I0 , the third is the series resistance Rs , the fourth is the shunt
world growth in 2021–2026 is 181.4 GW for the main case and 221.4 resistance Rsh , and finally, the diode ideality factor a. VT is the thermal
GW for the accelerated case (Solar - Fuels Technologies – IEA, 2022). voltage; it is determined based on temperature T, the Boltzmann con­
The rise in photovoltaic energy usage is accompanied by a growth in stant KB , electron charge q, and the number of cells in a series connection
scientific research, including advancements in PV technology and elec­ Ns .
trical modelling (El Chaar et al., 2010; Durgadevi et al., 2011). The The difficulty lies in determining the five unknown parameters that
electrical modelling of PV cells is an important step in controlling, define the I–V characteristics. Numerous techniques for obtaining the
optimizing, and predicting the energy produced by a PV system under parameters of PV cells and modules have been proposed using experi­
real operating conditions. There are a lot of electrical models in the mental data or information from datasheets. These methods are classi­
literature, and the single-diode model (SDM) is the most used because it fied into three types (Fahim et al., 2022), namely analytical methods
is simple and has great precision (Azzouzi et al., 2016). Fig. 1 represents (Batzelis, 2019), numerical methods (Ćalasan et al., 2022; Lidaighbi

* Corresponding author.
E-mail address: abdelfattah.elhammoudy@edu.uiz.ac.ma (A. Elhammoudy).

https://doi.org/10.1016/j.rio.2023.100445
Received 9 February 2023; Received in revised form 5 April 2023; Accepted 30 April 2023
Available online 2 May 2023
2666-9501/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Nomenclature RMSE Root Mean Square Error

SDM Single-Diode Model
STCs standard test conditions DDM Double-Diode Model
KB Boltzmann constant 1.3806485e-23 JK− 1 T temperature of the cell (K)
q electron charge 1.60217662e-19C G surface irradiance of the cell.
Vmp voltage at maximum power point (V) Isc short circuit current (A)
Imp current at maximum-power point (A) Voc open circuit voltage (V)
Pmax,e the maximum experimental power (W) I0 reverse saturation current (A)
Pmax,c the maximum power calculated (W) Ipv light-generated current (A)
Tn temperature of the cell at STCs 298.15 k Rsh shunt resistance (Ω)
Gn surface irradiance of the cell at STC 1000 Wm− 2 Rs series resistance (Ω)
KB T
VT = Ns q thermal voltage of the diode (V) a diode ideality factor
Ns number of cells connected in series
kI temperature coefficient of short-circuit current (A◦ C− 1 )
IAE Individual Absolute Error
kV temperature coefficient of open-circuit voltage (V◦ C− 1 )

Fig. 1. Single-Diode Model Equivalent Circuit.

et al., 2022), and meta-heuristic methods (Rezk et al., 2021). Analytical 2014) uses several iterations between Rsh ; Rs and Ipv , resulting the minor
methods offer the benefit of simplicity and can determine unknown possible error, for the starting values of Rsh , Rs and Ipv , respectively
parameters speedily because it is based on mathematical equations, such 0.15Ω, 10Ω and Isc . Another method given by Wang et al. (Wang et al.,
as; Sera’s method presented in (Sera et al., 2008) considered that the 2017) uses the derivative of the power at the maximum power point and
series resistance is the most important parameter and ignores the shunt gives the first series resistance Rs,n1 the value of zero then determines the
resistance. Several analytical methods, including De Blas’s (De Blas first derivative of power dP1 . After, calculate the second series resistance
et al., 2002), Firoz Khan’s (Khan et al., 2013), Phang’s (Phang et al., Rs,n2 then determines the second derivative of power dP2 and verified
1984), Celik’s (Celik and Acikgoz, 2007), Senturk’s (Senturk and Eke, that (dP1dP− 1dP2 ) ≤ 0. Ishaque et al. (Ishaque et al., 2011) presented a simple
2017), Singh’s (Singh et al., 2013) and Ruschel’s (Ruschel et al., 2016)
method using the increase of Rs from the initial value is zero until the
methods, use the inverse of the slope at the short-circuit and open-circuit
maximum power calculated Pmax,c is nearly equivalent to the maximum
points to determine the parameters. Cubas (Cubas et al., 2014) proposed
experimental power Pmax,e and the initial value of Rsh is determined by
two analytical methods; the first is a direct method based on the inverse
using the electrical characteristics. All the previous methods considered
of the slope at the short circuit point, and the second is an explicit
that the diode’s factor ideality a is a constant dependent only on the
method using the lambert W-function. It is also used by Batzelis (Batzelis
solar cell material. On the contrary, Ishibashi (Ishibashi et al., 2008)
and Papathanassiou, 2016) to determine an auxiliary parameter used for
proposed a method uses an iterative process for the series resistance Rs
calculating the unknown parameters.
and the diode’s factor ideality a until the condition |δ|≪1 is validated.
In order to address the weaknesses of analytical methods, numerical
Easwarakhanthan’s method (Easwarakhanthan et al., 1986) uses the
methods are utilized, considering the complete set of measured I–V data
Newton model modified with the Levenberg parameter and a least-
rather than individual data points. The numerical methods are an iter­
squares optimization algorithm to determine the unknown parameters.
ative approach and an algorithmic process for solving systems of equa­
The meta-heuristics approach uses optimization algorithms to
tions. It starts from an initial point considered as a first draft solution and
determine the unknown PV parameters; they are generally inspired by
proceeds by iterations, during which it determines a solution that fulfils
biology, physics, sociology, and mathematics (Yang et al., 2020). Among
the condition. The most celebrated method is given by Villalva et al.
the popular meta-heuristic algorithms used are Genetic Algorithm (GA)
(Villalva et al., 2009); it is centred on altering the series resistance Rs
that used and developed by several researchers to determine the PV
until the maximum power calculated Pmax,c is nearly equivalent to the
parameters. Such as, Saadaoui et al. (Saadaoui et al., 2021), Bastidas
maximum experimental power Pmax,e and the first value of Rs is zero.
et al. (Bastidas-Rodriguez et al., 2017) and Zhang (Zhang et al., 2016).
Moballegh et al. (Moballegh and Jiang, 2011) based on the same thing as
Another algorithm is Differential Evolution (DE) is an evolutionary
Villalva et al. however, the starting value of Rs is represented by the
optimization algorithm that principally uses the mutation operation to
inverse of the slopes at the open-circuit point. Vika (Håvard Breisnes,
find the probable regions in the search space. Liang et al. (Liang et al.,

2
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

2. Dichotomy method

The Dichotomy method, also known as the Bisection method (Pota­

pov, 2011), is a simple and an exact numerical technique for solving
nonlinear equations with a single unknown. It is one of several numer­
ical methods, including the Secant method (Amat and Busquier, 2003),
Newton-Raphson method (Abbasbandy, 2003), and fixed-point method
(Saqib et al., 2015) used to solve nonlinear equations. Despite its
simplicity, the Dichotomy method can provide accurate solutions to
various nonlinear problems. It helps find an approximate solution to an
equation f(x) = 0, in which f is continuous on the interval [a, b] with f(a)
≤ 0 and f(b) ≥ 0. At least one real x0 exists in the interval [a, b] with
f(x0 ) = 0.
The dichotomy method consists of dividing the interval into halves
by calculating c1 = a+b
2 , there is tow possibilities (Potapov, 2011):

• f(c1 ) ≤ 0 then x0 ∈ [c1 , b]

• f(c1 ) ≥ 0 then x0 ∈ [a, c1 ]
Fig. 2. The concept of the dichotomy method.
After dividing the new intervals into halves by calculating c1 , c2 ⋯cn
2020) estimate the PV parameters via a variant of (DE) named self- until that |cn − cn− 1 |〈ε which ε is an absolute error and sufficiently small
adaptive ensemble-based differential evolution (SEDE). Gao et al. (Gao as shown in Fig. 2.
et al., 2021) also propose a new variant of (DE) named directional The algorithm
permutation differential evolution algorithm (DPDE), and Kharchouf
(Kharchouf et al., 2022) proposes a hybrid method between the differ­
ential evolution (DE) and Lambert W-function. Hachana et al. (Hachana
et al., 2013) have proposed a combination between the Artificial Bee
Colony algorithm and Differential Evolution (ABC-DE).
There are other methods use the teaching–learning-based optimiza­
tion algorithm to determine the PV parameters. Such as, the one given
by Li et al. (Li et al., 2021), namely dynamic self-adaptive and mutual-
comparison teaching–learning-based optimization (DMTLBO), they use
two differentiated and personalized teaching strategies to improve the
teacher and the learner phases. Yu et al. (Yu et al., 2017) propose a
variant of (TLBO) named self-adaptive teaching–learning-based opti­
mization (SATLBO); the learners can self-adaptively choose other
learning phases using their knowledge.. However, other algorithms
based on artificial intelligence can also be used, such as particle swarm
optimization (PSO) inspired by the social behaviour of a herd of birds or
swarms of fish. This algorithm is used to determine the PV parameters by
(Ben Hmamou et al., 2022b), W. Ahmed et al. (Ahmed et al., 2022), and 3. The proposed approach
S. Ebrahimi et al. (Ebrahimi et al., 2019).
Another type of method combines numerical and analytical ap­ Most studies in the field utilize three significant points from the I–V
proaches. Such as, Lidaighbi et al.’s method (Lidaighbi et al., 2022) uses characteristic as presented in Fig. 3, namely the short-circuit point, the
an increment of 0.01VT for the diode ideality factor to achieve the open-circuit point, and the maximum power point, to construct four
minimum value of RMSE for single-diode model SDM. Also, (Ben hma­ equations.
mou et al., 2022a) mixed analytical and numerical methods; they At short-circuit point (SC: V = 0; I = Isc ):
determined three parameters I0 ; Ipv , and a analytically using the elec­ ( ( ) )
trical characteristics under STCs, for Rs and Rsh are obtained with an Rs Isc Rs Isc
Isc = Ipv − I0 exp − 1 − (2)
iterative method similar to Villalva’s method (Villalva et al., 2009). aVT Rsh
This work presents a simple approach to extracting the PV parame­ At open-circuit point (OC: V = Voc ; I = 0):
ters for a single-diode model. The approach’s originality lies in its ( ( ) )
straightforward method of solving a nonlinear equation for Rs . The other 0 = Ipv − I0 exp
Voc
− 1 −
Voc
(3)
parameters Rsh , I0 and Ipv , are calculated for several values of the diode aVT Rsh
ideality factor; the parameters will be selected based on the lowest At maximum power point (MP: V = Vmp ; I = Imp ):
RMSE. The proposed method works on STCs and real conditions and [ ( ) ]
outperforms analytical, numerical, and meta-heuristic algorithms. The Imp = Ipv − I0 exp
Vmp + Rs Imp
− 1 −
Vmp + Rs Imp
(4)
rest of this paper is made up of three sections. An introduction to the aVT Rsh
dichotomy method is in the second section. The third section presents ⃒
the proposed approach. The fourth section is a validation of the pro­ dP ⃒⃒
=0 (5)
posed approach on five PV cells/modules and discusses its performance dV ⃒ V = Vmpn
I = Impn
compared with other methods from the literature, followed by a
conclusion of this work in the last section. The proposed approach is built in two steps; the first step is calcu­
lating the series resistance Rs , using the dichotomy method and calcu­
lating the other parameters, Rsh , I0 , and Ipv for 1001 value of the diode

3
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Fig. 3. I–V characteristic curve of the PV cell.

( ) ( )
We have proposed that exp Rs Isc
aVT
Voc
≪exp aV T

Isc (Rs + Rsh ) − Voc

I0 = ( ) (9)
Voc
Rsh exp aV T

For the shunt resistance Rsh , it is necessary to calculate the derivative

of the power P at MPP, and for determining its derivative, it should
replace Ipv and I0 expressions in the Eq. (1).

I = f (I, V)
( ) ( )
Isc Rs − Rs I − V Isc Rs − Voc V + Rs I − Voc
= Isc + − Isc + exp (10)
Rsh Rsh aVT

dI ∂f (I, V) dI ∂f (I, V) dV
= + (11)
dV ∂I dV ∂V dV
∂f (I,V)
dI
= ∂V
∂f (I,V)
(12)
dV 1 − ∂I

From Eqs. (10) and (12), the derivative of the current I = f(I, V) by
the voltage V is given by the following expression:
( ) ( )
dI − Isc RshaV
− Voc − Isc Rs
exp V+RaVs I−T Voc − R1sh
(13)
T Rsh
= ( ) ( )
dV 1 + Rs Isc Rsh − Voc − Isc Rs exp V+Rs I− Voc + Rs
aVT Rsh aVT Rsh

The derivative of the power P = IV by the voltage V can be expressed

by the expression below:
dP d(IV) dI
= = I +V (14)
dV dV dV
Fig. 4. The flow chart of the proposed method.
To determine the shunt resistance is necessary to use the expressions
(5), (13), and (14) in order to obtain the following expression:
ideality factor a. Generally, its value is included in the interval [1,2], it is
⎛ ⎞
staring with an initial value is 1 and changes it with an increment of i = ( ) (
Isc Rsh − Voc − Isc Rs
)
0.001. Rsh , I0 , and Ipv are calculated using mathematical equations ⎜ − exp V+RaVs I−T Voc − R1sh ⎟
(15)
aVT Rsh
Imp + Vmp ⎜ ( ) ( ) ⎟=0
extracted from the three remarkable points. The light-generated current ⎝
1 + Rs Isc RshaV− Voc − Isc Rs
exp V+RaVs I−T Voc + RRshs
⎠
T Rsh
Ipv is derived from Eq. (3).
[ ( ) ]
Voc Voc From the expression (15), the shunt resistance is determined by the
Ipv = I0 exp − 1 + (6)
aVT Rsh following equation:
( ( ) )( ) ( )
The reverse saturation current I0 is expressed by replacing the Eq. (6) Vmp + Rs Isc − Imp Imp Rs − Vmp − Imp Rs − Vmp aVT
Rsh = ( )( ) (16)
in (2). Iscn − Imp Imp Rs − Vmp + Imp aVT
[ ( ) ] ( ( ) )
Voc Voc Rs Isc Rs Isc To apply the dichotomy method is necessary to create a non-linear
Isc = I0 exp − 1 + − I0 exp − 1 − (7)
aVT Rsh aVT Rsh equation of the series resistance f(Rs ). To determine this function, we
should replace the expressions of Ipv and I0 in the Eq. (4) given by the
maximum power point.
Voc − Rs Isc
Isc −
(8)
Rsh
I0 = ( ) ( )
Voc Rs Isc
exp aVT
− exp aVT

4
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

[ ( ) ] [ ( ) ]
Table 1 Voc Voc Vmp + Rs Imp Vmp + Rs Imp
The characteristics of five cell / modules for G = 1000 W/m2 .
Impn = I0 exp − 1 + − I0n exp − 1 −
aVT Rsh aVT Rsh
The RTC PWP201 Shell Shell Shell (17)
characteristic France S115 SM55 ST20
( )
Isc [A] 0.7603 1.0317 4.7 3.4 1.54 Isc (Rs + Rsh ) − Voc Vmp + Rs Imp − Voc Vmp + Rs Imp − Isc (Rs + Rsh )
Impn + exp +
Voc [V] 0.5728 16.7785 32.8 21.7 22.9 Rsh aVT Rsh
Imp [A] 0.6894 0.9120 4.29 3.16 1.28
=0
Vmp [V] 0.4507 12.6490 26.8 17.4 15.6
Pmax,e [W] 0.3107 11.5359 115 55 20 (18)
Ns 1 36 54 36 42 ( )
ki [A/◦ C] − 0.115 1.4e-03 0.2e-03 Isc (Rs + Rsh ) − Voc Vmp + Rs Imp − Voc
kv [V/◦ C] 0.002 − 0.076 − 0.1
f (Rs ) = Impn + exp
Rsh aVT
T[◦ C] 33 45 25 25 25 (19)
Vmp + Rs Imp − Isc (Rs + Rsh )
+
Rsh
Two initial values of the series resistance must be entered to start the
calculation Rs 1 and Rs 2 ; they represent the extremities of the interval of
Rs , where Rs 1 <Rs <Rs 2 .
Rs 1 =0 (20)

Table 2
The calculated parameters and RMSE for RTC France cell.
Parameters a Rs [Ω] Rsh [Ω] Ipv [A] I0 [A] RMSE[A] RMSE%

The proposed method 1.4760 0.0367 53.1394 0.7608 3.0667e-07 7.8514e-04 0.1423
Numerical Bouzidi et al. (Bouzidi et al., 2007) 1.4816 0.0364 60.2409 0.7607 3.267e-07 1.4654e-03 0.2657
Easwarakhanthan et al. (Easwarakhanthan et al., 1986) 1.4837 0.0364 53.7634 0.7608 3.2230e-07 5.8556e-03 1.0616
Ishibashi (Ishibashi et al., 2008) 1.4000 0.0370 32.0000 0.7700 2.0000e-07 8.0757e-02 14.6405
Ishaque et al. (Ishaque et al., 2011) 1.600 0.0313 64.1025 0.7617 9.9800e-07 9.8170e-03 1.7797
Villalva et all. (Villalva et al., 2009) 1.3000 0.0400 18.2173 0.7620 4.0699e-07 1.4000e-03 0.2538
Wang et al. (Wang et al., 2017) 1.5000 0.0343 58.2790 0.7607 3.9335e-07 3.0959e-03 0.5613

Analytical Celik (Celik and Acikgoz, 2007) 1.3562 0.0506 46.3143 0.7611 5.4253e-07 6.5000e-03 1.1784
Pindado et al. (Pindado and Cubas, 2017) 1.4800 0.0362 52.0000 0.7610 3.2000e-07 9.5802e-04 0.1737
Senturk (Senturk and Eke, 2017) 1.0000 0.0522 28.0522 0.7617 2.7586e-08 1.5000e-02 2.7194
Peng et al. (Peng et al., 2014) 1.4841 0.0364 54.0540 0.7609 3.2200e-07 6.9472e-03 1.2595
Zhang et al. (Zhang et al., 2011) 1.4561 0.0373 42.0000 0.7611 2.4220e-07 6.8858e-03 1.2483
Louzazni (Louzazni et al., 2019) 1.4837 0.0364 53.7634 0.7608 3.2230e-07 5.8556e-03 1.0616

Meta-heuristic DE (SD) (Kharchouf et al., 2022) 1.4709 0.0363 54.1134 0.7607 3.2090e-07 7.7692e-04 0.1408
ABC-DE (Hachana et al., 2013) 1.47986 0.03637 53.7185 0.76077 3.2302e-06 9.8602e-04 0.1788
DMTLBO (Li et al., 2021) 1.4812 0.0363 53.7183 0.7608 3.2302e-06 9.8602e-04 0.1788
DPDE (Gao et al., 2021) 1.4811 0.0363 53.7185 0.7607 3.2302e-07 9.8602e-04 0.1788
SEDE (Liang et al., 2020) 1.4811 0.0363 53.7185 0.7607 3.6377e-07 9.8602e-04 0.1788
SATLBO (Yu et al., 2017) 1.4812 0.0363 53.7256 0.7608 3.2320e-07 9.8602e-04 0.1788

Table 3
The calculated parameters and RMSE for PWP201 module.
Parameters a Rs [Ω] Rsh [Ω] Ipv [A] I0 [A] RMSE[A] RMSE%

The proposed method 1.3250 1.2375 7.1039e + 02 1.0335 2.7054e-06 2.2256e-03 0.0139
Numerical Bouzidi et al. (Bouzidi et al., 2007) 1.3380 1.2024 1.2779e + 02 1.0339 3.0760e-06 4.7611e-03 0.0297
Easwarakhanthan et al. (Easwarakhanthan et al., 1986) 1.3458 1.2024 1.2779e + 02 1.0318 3.2876e-06 5.4811e-03 0.0342
Ishibashi (Ishibashi et al., 2008) 1.3055 1.2996 8.2999e + 02 1.030 2.3000e-06 4.7611e-03 0.0297
Ishaque et al. (Ishaque et al., 2011) 1.3413 1.2024 7.1428e + 02 1.0313 3.1756e-06 9.8367e-03 0.0613
Villalva et al. (Villalva et al., 2009) 1.3000 0.9360 2.8770e + 02 1.0351 2.0444e-06 2.0197e-02 0.1259
Wang et al. (Wang et al., 2017) 1.4000 1.0872 9.6256e + 02 1.0329 5.4957e-06 6.9321e-03 0.0432

Analytical Celik (Celik and Acikgoz, 2007) 1.0000 1.9980 4.3918e + 02 1.0364 2.7347e-08 7.8704e-02 0.4905
Pindado et al. (Pindado and Cubas, 2017) 1.2500 1.5588 3.5499e + 03 1.0300 1.2800e-06 1.4481e-02 0.0902
Senturk (Senturk and Eke, 2017) 1.0000 1.1592 4.6977e + 02 1.0343 4.1344e-08 4.1798e-02 0.2605
Peng et al. (Peng et al., 2014) 1.3422 1.2132 6.2499e + 02 1.0313 3.2212e-06 6.6709e-03 0.0416
Zhang et al. (Zhang et al., 2011) 1.2739 1.3104 6.0229e + 02 1.0332 1.5970e-06 3.8455e-03 0.0240
Louzazni (Louzazni et al., 2019) 1.3458 1.2024 5.4944e + 02 1.0318 3.2876e-06 5.4811e-03 0.0342

Meta- heuristic DE (SD) (Kharchouf et al., 2022) 1.3139 1.2348 8.2165e + 02 1.0314 2.6380e-06 2.0529e-03 0.0128
ABC-DE (Hachana et al., 2013) 1.3443 1.2060 8.4524e + 02 1.0318 3.2774e-06 2.6513e-03 0.0165
DMTLBO (Li et al., 2021) 1.3511 1.1988 9.8197e + 02 1.0305 3.4823e-06 2.4251e-03 0.0151
DPDE (Gao et al., 2021) 1.3511 1.1988 9.8197e + 02 1.0305 3.4822e-06 2.4250e-03 0.0151
SEDE (Liang et al., 2020) 1.3511 1.1988 9.8197e + 02 1.0305 3.4822e-06 2.4250e-03 0.0151
SATLBO (Yu et al., 2017) 1.351 1.1988 9.8240e + 02 1.0305 3.4827e-06 2.4250e-03 0.0151

5
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Table 4
Experimental and predicted data of current and voltage for RTC France and PWP201.
RTC France cell PWP201 module

Experimental data Predicted data Experimental data Predicted data

V[V] Iexp [A] I[A] IAE[A] V[V] Iexp [A] I[A] IAE[A]

1 − 0.2057 0.7640 0.7641 1.6828e-04 0.1248 1.0315 1.0315 1.6743e-05

2 − 0.1291 0.7620 0.7627 7.2776e-04 1.8093 1.0300 1.0291 8.7097e-04
3 − 0.0588 0.7605 0.7614 9.0562e-04 3.3511 1.0260 1.0269 8.9844e-04
4 0.0057 0.7605 0.7601 3.0791e-04 4.7622 1.0220 1.0247 2.7363e-03
5 0.0646 0.7600 0.7590 9.1812e-04 6.0538 1.0180 1.0225 4.4651e-03
6 0.1185 0.7590 0.7580 9.4156e-04 7.2364 1.0155 1.0197 4.2376e-03
7 0.1678 0.7570 0.7570 9.7943e-05 8.3189 1.0140 1.0159 1.9207e-03
8 0.2132 0.7570 0.7561 8.5840e-04 9.3097 1.0100 1.0099 5.6598e-05
9 0.2545 0.7555 0.7550 1.6112e-04 10.2163 1.0035 1.0002 3.3469e-03
10 0.2924 0.7540 0.7536 3.3558e-04 11.0449 0.9880 0.9843 3.7315e-03
11 0.3269 0.7505 0.7514 9.0105e-04 11.8018 0.9630 0.9595 3.4908e-03
12 0.3585 0.7465 0.7473 8.8729e-04 12.4929 0.9255 0.923 2.4808e-03
13 0.3873 0.7385 0.7401 1.6753e-03 13.1231 0.8725 0.8726 8.9610e-05
14 0.4137 0.7280 0.7275 4.7593e-04 13.6983 0.8075 0.8072 3.2423e-04
15 0.4373 0.7065 0.7071 6.2526e-04 14.2221 0.7265 0.7276 1.0543e-03
16 0.459 0.6755 0.6754 8.5655e-06 14.6995 0.6345 0.6357 1.2478e-03
17 0.4784 0.6320 0.6310 9.2976e-04 15.1346 0.5345 0.5347 2.1817e-04
18 0.496 0.5730 0.5722 7.7921e-04 15.5311 0.4275 0.4277 2.1330e-04
19 0.5119 0.4990 0.4995 5.6110e-04 15.8929 0.3185 0.3176 8.8126e-04
20 0.5265 0.4130 0.4134 4.9583e-04 16.2229 0.2085 0.2070 1.4519e-03
21 0.5398 0.3165 0.3171 6.8395e-04 16.5241 0.1010 0.0980 3.0393e-03
22 0.5521 0.2120 0.2120 7.9396e-05 16.7987 − 0.0080 − 0.0080 2.6917e-07
23 0.5633 0.1035 0.1027 7.3069e-04 17.0499 − 0.1110 − 0.1101 8.8265e-04
24 0.5736 − 0.010 − 0.0090 9.3206e-04 17.2793 − 0.2090 − 0.2075 15013e-03
25 0.5833 − 0.1230 − 0.1240 1.0060e-03 17.4885 − 0.3030 − 0.2996 3.4212e-03
26 0.5900 − 0.2100 − 0.2086 1.3594e-03

mean(IAE) 6.8494e-04 1.7031e-03

∑
IAE 0.0178 0.0426

0.8 0.3 Proposed method

Experimental data
0.6 0.2

0.4 0.1
P [W]
I [A]

0.2 0

0 -0.1
Proposed method
-0.2 Experimental data -0.2

-0.2 0 0.2 0.4 0.6 -0.2 0 0.2 0.4 0.6

V [V] V [V]
Fig. 5. I–V Proposed method’s curve and experimental data for RTC France Fig. 6. P–V Proposed method’s curve and experimental data for RTC France
solar cell. solar cell.

Voc − Vmp average difference between the current values predicted by the model Ii
Rs 2 = (21) and the actual current values from the experimental data Iexp,i , divided
Imp
by the number of measurements n. Fig. 4 depicts the flowchart of the
The shunt resistance’s initial value is calculated using the electrical proposed method.
characteristics. √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
√∑
√n
Vmp √ (Ii − Iexp,i )2
Rsh 0 =m (22) √
Isc − Imp RMSE = i=1 (23)
n
Where m is a constant takes the values of 2.
The second step involves selecting the five parameters that resulted Under reals conditions, it is necessary to update the values of short-
in the lowest root mean squared error (RMSE), which represents the circuit current, open-circuit voltage, and the maximum power point

6
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

1.2

0.8

0.6
I [A]

0.4

0.2

-0.2 Proposed method

Experimental data
-0.4
0 5 10 15
V [V] Fig. 9. I–V Proposed method’s curve and experimental data for
Shell S115panel.
Fig. 7. I–V Proposed method’s curve and experimental data for PWP201 panel.

Proposed method
10 Experimental data

5
P [W]

-5
0 5 10 15 Fig. 10. I–V Proposed method’s curve and experimental data for
V [V] Shell S115panel.

Fig. 8. P–V Proposed method’s curve and experimental data for PWP201 panel.
4. Results and discussion

current and voltage using the formulas below. To validate the suggested method, two parts are required; the first is
G a comparison with other methods from the literature for the RTC France
Isc = Iscn + kI (T − Tn ) (24) cell and the PWP201 module. Their experimental data are taken from (Li
Gn
et al., 2021). The second part is testing the proposed method’s accuracy
G for different module technologies, the Shell S115 (poly-crystalline), the
Voc = Vocn + kV (T − Tn ) (25)
Gn Shell SM55 (mono-crystalline) and the Shell ST20 (thin-film) under
different meteorology conditions. Their experimental data are obtained
Imp = Impn + kI (T − Tn )
G
(26) from the manufacturer’s datasheet. Table 1 present the electrical char­
Gn acteristics of the five cell/modules used.
Most researchers use the RTC France cell and the PWP201 module to
G
Vmp = Vmpn + kV (T − Tn ) (27) validate their methods and compare the results with other methods in
Gn
the literature. Following in their footsteps, the performance of our
Iscn , Vocn , Impn , and Vmpn are the short circuit current, the open circuit method will be tested on the cell and the module that has already been
voltage, the current and the voltage at the maximum power point under mentioned and compare its results with eighteen methods classified into
real conditions. The diode ideality factor has the same value for STCs three categories; six iterative methods are Bouzidi et al. (Bouzidi et al.,
and real conditions because it is constant and depends only on the PV 2007), Easwarakhanthan et al. (Easwarakhanthan et al., 1986), Ishiba­
cell’s technology. shi (Ishibashi et al., 2008), Ishaque et al. (Ishaque et al., 2011), Villalva
et al. (Villalva et al., 2009) and Wang et al. (Wang et al., 2017), six
analytical methods are Celik (Celik and Acikgoz, 2007), Pindado et al.

7
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Fig. 11. I–V Proposed method’s curve and experimental data for Fig. 13. I–V Proposed method’s curve and experimental data for Shell
Shell SM55panel. ST20 panel.

Fig. 14. I–V Proposed method’s curve and experimental data for Shell
Fig. 12. I–V Proposed method’s curve and experimental data for ST20 panel.
Shell SM55panel.

(Pindado and Cubas, 2017), Senturk (Senturk and Eke, 2017), Peng et al. Table 5
The calculated parameters and RMSE for Shell S115, Shell SM55 and Shell ST20.
(Peng et al., 2014), Zhang et al. (Zhang et al., 2011) and Louzazni
(Louzazni et al., 2019), and six meta-heuristic methods are DE (SD) Poly-crystalline Mono-crystalline Thin-film
Shell S115 Shell SM55 Shell ST20
(Kharchouf et al., 2022), ABC-DE (Hachana et al., 2013), DMTLBO (Li
et al., 2021), DPDE (Gao et al., 2021), SEDE (Liang et al., 2020) and a 1.0000 1.0560 1.5950
SATLBO (Yu et al., 2017). Rs [Ω] 0.5469 0.4975 2.8509
Rsh [Ω] 282.3313 521.3824 313.2428
Tables 2 and 3 present the values of the five parameters for the RTC Ipv [A] 4.7091 3.4032 1.5540
France cell and the PWP201 module, they also present the RMSE I0 [A] 1.6490e-10 7.5480e-10 2.4672e-06
calculated by the proposed method, and its percentage RMSE% is RMSE[A] 0.0037 0.0041 0.0031
calculated by dividing the RMSE by the average value of the experi­ max(IAE)[A] 0.0128 0.0111 0.0083
2.2685e-05 5.3755e-05 5.4789e-05
mental data compared with other methods. For all methods, the RMSE% min(IAE)[A]
Time complexity [s] 0.345437 0.29562 0.39266
is less than 10%, which means that the accuracy of all methods is
excellent. The values given by the proposed method for RMSE are
comparable with the other methods, especially meta-heuristics methods. straightforward and, with the lowest time complexity, took only 0.2106
We can observe that for RTC France and PWP201, the values of RMSE s to extract the five parameters for the RTC France cell and 0.3244 s for
given by the proposed method are reasonable and lower than that pro­ the PWP201 module.
vided by the analytical and numerical methods. It is also lower than the Table 4 presents the experimental and predicted voltage and current
meta-heuristic methods excluding DE (SD) (Kharchouf et al., 2022); the data for the RTC France cell and PWP201 module. It also presents their
last one based on a complicated algorithm is the differential evolution absolute error (IAE), the difference between the experimental and the
with the highest time complexity. However, the proposed method is estimated current. As provided in the table, the values predicted

8
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Fig. 15. IAE values for S115, SM55 and ST20 obtained by the proposed method at STCs.

m2), (T = 25 ◦ C, G = 600 W/m2),and (T = 25 ◦ C, G = 200 W/m2). Under

Table 6 different temperatures and irradiation levels, the (I–V) curves obtained
Comparison of RMSE at different values of irradiation and temperature for Shell based on the recovered optimal parameters and the measured curves
SM55. show good agreement.
In addition, Table 5 presents the values of the five parameters and
( ( ) )
G W/m2 , T(C◦ ) (1000, (600,25) (200,25) (1000, (1000, 50)
25) 30) the RMSE calculated by the proposed method under STCs for the three
The proposed 4.1016e- 4.0913e- 5.9486e- 7.6799e- 5.5872e- panels. The values of RMSE are in order of 10− 3 A, and it takes around
method 03 03 03 03 03 half a minute to determine the five parameters. They are good values to
Femia et al. (Femia 0.02731 0.16770 0.15347 0.11304 0.13723 say that the proposed method evaluates the five parameters with high
et al., 2013)
Villalva et al. ( 3.8441e- 2.5189e- 2.8283e- 7.4161e- 8.1868e-
accuracy and performance. Table 5 also summarizes the maximum and
Villalva et al., 02 02 02 02 02 the minimum values of the IAE for the three modules under Standard
2009) Test Conditions (STCs). The highest value is 0.0128A, and the lowest is
SEDE (Liang et al., 1.8432e- 1.9359e- 2.0233e- 2.0444e- 2.0684e- 2.2685e-05A; they are recorded for the Shell S115 module. Further­
2020) 02 02 02 02 02
more, Fig. 15 shows the variation of the IAE for the three modules at
STCs.
Table 6 summarizes the RMSE values for the Shell SM55 module
Table 7 calculated at different values of irradiation and temperature compared
A comparison of the dichotomy and false position methods. with three different methods. The first is analytical, given by Femia et al.
RTC France PWP201
(Femia et al., 2013), the second is numerical, proposed by Villalva et al.
(Villalva et al., 2009) and a meta-heuristic method is SEDE (Liang et al.,
The method Dichotomy False Dichotomy False
2020) based on differential evolution algorithm (DE). After the table, for
Position Position
RMSE[A] 7.851435e- 7.524027e- 2.225699e- 1.055337e- different irradiation and temperature values, the proposed approach
04 03 03 02 yields lower RMSE than other methods, demonstrating its efficiency
mean(IAE)[A] 6.849378e- 6.421039e- 1.703135e- 8.336715e- under different environmental conditions.
04 03 03 03
The proposed approach used the dichotomy method to solve the Eq.
Time complexity 0.2568 13.2286 0.4061 16.2803
[s]
(19). However, we will employ another numerical method namely the
False Position. The RTC France cell and PWP201 module will be sub­
jected to evaluate their performance based on metrics such as RMSE, the
currents are almost identical to the experimental, in which the mean of average of IAE, and the time complexity, as presented in Table 7. The
absolute error is 6.8494e-04 A and 1.7031e-03 A for RTC France and dichotomy method is superior to the False Position as it yields smaller
PWP, respectively. It is very clear in Figs. 5, 6, 7 and 8, they illustrate the errors and has a lower time complexity for the RTC France cell and the
I–V and P–V curves of the RTC France cell and the PWP201 module. The PWP module. This implies that the False Position method has both a
obtained curves show that the data calculated through the estimated slow rate of convergence and, furthermore, failed to attain the optimal
optimal parameters are nearly identical to the experimental data. solution of Eq. (19). It is confirmed that the most effective way to solve
Accordingly, the proposed method estimates I–V characteristics the equation for Rs is by using the dichotomy method. This is because it
perfectly for the RTC France cell and PWP201 module. always leads to the optimal solution and does not involve any complex
⃒ ⃒ calculations. Furthermore, it has not been previously employed for
IAE = ⃒Iexp − I ⃒ (28)
identifying the PV parameter.
This subsection demonstrates the performance of the suggested
approach and applies it to three panels from various technologies, such 5. Conclusion
as polycrystalline, mono-crystalline, and thin film. The Figs. 9–14 pre­
sent the I–V curves obtained by the suggested method compared with This paper introduces a novel numerical method that employs the
the experimental data of three panels, namely Shell S115, Shell SM55, dichotomy technique and the Root Mean Square Error (RMSE) to iden­
and Shell ST20, for five different meteorology conditions:(T = 25 ◦ C, G tify the five unknown parameters for the single-diode model. This paper
= 1000 W/m2), (T = 30 ◦ C, G = 1000 W/m2), (T = 50 ◦ C, G = 1000 W/ is divided into four sections: the introduction, the electrical modelling of

9
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

PV cells, the definition of the dichotomy technique, the presentation of Ćalasan, M., Al-Dhaifallah, M., Ali, Z.M., Abdel Aleem, S.H.E., 2022. Comparative
analysis of different iterative methods for solving current-voltage characteristics of
the proposed approach, and its validation on five cells/modules from
double and triple diode models of solar cells. Mathematics 10 (17), pp. https://doi.
various technologies: RTC France, PWP201, Shell S115, Shell SM55, and org/10.3390/math10173082.
Shell ST20. The results show that the suggested approach outperforms Celik, A.N., Acikgoz, N., 2007. Modelling and experimental verification of the operating
analytical and numerical methods in terms of accuracy; it is comparable current of mono-crystalline photovoltaic modules using four- and five-parameter
models. Appl. Energy 84 (1), 1–15. https://doi.org/10.1016/j.
with meta-heuristic methods; the RMSE values are 7.8514e-04 A and apenergy.2006.04.007.
2.2256e-03 A, with a lowest time complexity around 0.2106 s and Cubas, J., Pindado, S., Victoria, M., 2014. On the analytical approach for modeling
0.3244 s for, respectively RTC France cell and PWP201 module. The photovoltaic systems behavior. J. Power Sour. 247, 467–474.
De Blas, M.A., Torres, J.L., Prieto, E., García, A., 2002. Selecting a suitable model for
proposed method, such as all the methods, has a limitation; it is neces­ characterizing photovoltaic devices. Renew. Energy 25 (3), 371–380. https://doi.
sary to use the experimental data because it is based on the RMSE to org/10.1016/S0960-1481(01)00056-8.
determine the diode ideality factor. Apart from this, with the excellent A. Durgadevi, S. Arulselvi, and S. P. Natarajan, “Photovoltaic modeling and its
characteristics,” 2011 Int. Conf. Emerg. Trends Electr. Comput. Technol. ICETECT 2011,
matching of I–V and P–V curves with experimental data, we can use the pp. 469–475, 2011, doi: 10.1109/ICETECT.2011.5760162.
proposed approach as an efficiency tool for extracting the PV single- Easwarakhanthan, T., Bottin, J., Bouhouch, I., Boutrit, C., 1986. Nonlinear minimization
diode model parameters for different applications such as the MPPT. algorithm for determining the solar cell parameters with microcomputers. Int. J. Sol.
Energy 4 (1), 1–12. https://doi.org/10.1080/01425918608909835.
Ebrahimi, S.M., Salahshour, E., Malekzadeh, M., Gordillo, F., 2019. Parameters
CRediT authorship contribution statement identification of PV solar cells and modules using flexible particle swarm
optimization algorithm. Energy 179, 358–372. https://doi.org/10.1016/j.
energy.2019.04.218.
Abdelfattah Elhammoudy: Conceptualization, Validation, Writing El Chaar, L., Lamont, L.A., Elzein, N., 2010. PV technology – Industry update. IEEE PES
– original draft, Writing – review & editing. Mustapha Elyaqouti: Gen. Meet. PES 2010, 1–6. https://doi.org/10.1109/PES.2010.5589568.
Conceptualization, Validation, Writing – original draft, Writing – review Fahim, S.R., Hasanien, H.M., Turky, R.A., Aleem, S.H.E.A., Ćalasan, M., 2022.
A comprehensive review of photovoltaic modules models and algorithms used in
& editing. Dris Ben Hmamou: Conceptualization, Validation, Writing –
parameter extraction. Energies 15 (23), 1–56. https://doi.org/10.3390/
original draft, Writing – review & editing. El hanafi Arjdal: Concep­ en15238941.
tualization, Methodology, Writing – review & editing, Software, Formal N. Femia, G. Petrone, G. Spagnuolo, and M. Vitelli, MPPT Efficiency: Noise Sources and
Methods for Reducing Their Effects. 2013. doi: 10.1201/b14303-3.
analysis, Funding acquisition. Driss Saadaoui: Conceptualization,
Gao, S., Wang, K., Tao, S., Jin, T., Dai, H., Cheng, J., 2021. A state-of-the-art differential
Methodology, Writing – review & editing, Software, Formal analysis, evolution algorithm for parameter estimation of solar photovoltaic models. Energy
Funding acquisition. Souad Lidaighbi: Conceptualization, Methodol­ Convers. Manage. 230, 113784 https://doi.org/10.1016/j.enconman.2020.113784.
ogy, Writing – review & editing, Software, Formal analysis, Funding Gross, R., Leach, M., Bauen, A., 2003. Progress in renewable energy. Environ. Int. 29 (1),
105–122. https://doi.org/10.1016/S0160-4120(02)00130-7.
acquisition. Imade Choulli: Conceptualization, Methodology, Writing – Hachana, O., Hemsas, K.E., Tina, G.M., Ventura, C., 2013. Comparison of different
review & editing, Software, Formal analysis, Funding acquisition. metaheuristic algorithms for parameter identification of photovoltaic cell/module.
J. Renew. Sustain. Energy 5 (5), 053122.
V. Håvard Breisnes, “Modelling of Photovoltaic Modules with Battery Energy Storage in
Declaration of Competing Interest Simulink / Matlab Håvard Breisnes Vika,” no. June, p. 136, 2014.
Ishaque, K., Salam, Z., Taheri, H., 2011. Simple, fast and accurate two-diode model for
photovoltaic modules. Sol. Energy Mater. Sol. Cells 95 (2), 586–594. https://doi.
The authors declare that they have no known competing financial org/10.1016/j.solmat.2010.09.023.
interests or personal relationships that could have appeared to influence Ishibashi, K.I., Kimura, Y., Niwano, M., 2008. An extensively valid and stable method for
the work reported in this paper. derivation of all parameters of a solar cell from a single current-voltage
characteristic. J. Appl. Phys. 103 (9), 1–7. https://doi.org/10.1063/1.2895396.
Khan, F., Baek, S.H., Park, Y., Kim, J.H., 2013. Extraction of diode parameters of silicon
Data availability solar cells under high illumination conditions. Energy Convers. Manage. 76,
421–429. https://doi.org/10.1016/j.enconman.2013.07.054.
Y. Kharchouf R. Herbazi A. Chahboun Parameter’s extraction of solar photovoltaic
No data was used for the research described in the article. models using an improved differential evolution algorithm Energy Convers. Manag.
251 November 2022 2021, p. 114972 10.1016/j.enconman.2021.114972.
References Li, L., Xiong, G., Yuan, X., Zhang, J., Chen, J., 2021. Parameter extraction of photovoltaic
models using a dynamic self-adaptive and mutual- comparison teaching-learning-
based optimization. IEEE Access 9, 52425–52441. https://doi.org/10.1109/
Abbasbandy, S., 2003. Improving Newton-Raphson method for nonlinear equations by
ACCESS.2021.3069748.
modified Adomian decomposition method. Appl. Math. Comput. 145 (2–3),
Liang, J., Qiao, K., Yu, K., Ge, S., Qu, B., Xu, R., Li, K.e., 2020. Parameters estimation of
887–893. https://doi.org/10.1016/S0096-3003(03)00282-0.
solar photovoltaic models via a self-adaptive ensemble-based differential evolution.
Ahmed, W.A.E.M., Mageed, H.M.A., Mohamed, S.A.E., Saleh, A.A., 2022. Fractional
Sol. Energy 207, 336–346.
order Darwinian particle swarm optimization for parameters identification of solar
Lidaighbi, S., Elyaqouti, M., Ben Hmamou, D., Saadaoui, D., Assalaou, K., Arjdal, E.,
PV cells and modules. Alexand. Eng. J. 61 (2), 1249–1263. https://doi.org/10.1016/
2022. A new hybrid method to estimate the single-diode model parameters of solar
j.aej.2021.06.019.
photovoltaic panel. Energy Convers. Manage. X 15, 100234.
Amat, S., Busquier, S., 2003. On a higher order secant method. Appl. Math. Comput. 141
Lidaighbi, S., Elyaqouti, M., Assalaou, K., Ben Hmamou, D., Saadaoui, D., H’roura, J.,
(2–3), 321–329. https://doi.org/10.1016/S0096-3003(02)00257-6.
2022. Parameter estimation of photovoltaic modules using analytical and
Azzouzi, M., Popescu, D., Bouchahdane, M., 2016. Modeling of electrical characteristics
numerical/iterative approaches: a comparative study. Mater. Today Proc. 52, 1–6.
of photovoltaic cell considering single-diode model. J. Clean Energy Technol. 4 (6),
https://doi.org/10.1016/j.matpr.2021.10.021.
414–420. https://doi.org/10.18178/jocet.2016.4.6.323.
Louzazni, M., Khouya, A., Al-Dahidi, S., Mussetta, M., Amechnoue, K., 2019. Analytical
Bastidas-Rodriguez, J.D., Petrone, G., Ramos-Paja, C.A., Spagnuolo, G., 2017. A genetic
optimization of photovoltaic output with Lagrange Multiplier Method. Optik (Stuttg)
algorithm for identifying the single diode model parameters of a photovoltaic panel.
199, 163379. https://doi.org/10.1016/j.ijleo.2019.163379.
Math. Comput. Simul. 131, 38–54. https://doi.org/10.1016/j.matcom.2015.10.008.
Moballegh, S., Jiang, J., 2011. Partial shading modeling of photovoltaic system with
Batzelis, E., 2019. Non-iterative methods for the extraction of the single-diode model
experimental validations. IEEE Power Energy Soc. Gen. Meet. https://doi.org/
parameters of photovoltaic modules: A review and comparative assessment. Energies
10.1109/PES.2011.6039281.
12 (3), 14–28. https://doi.org/10.3390/en12030358.
Peng, L., Sun, Y., Meng, Z., 2014. An improved model and parameters extraction for
Batzelis, E.I., Papathanassiou, S.A., 2016. A Method for the analytical extraction of the
photovoltaic cells using only three state points at standard test condition. J. Power
single-diode PV model parameters. IEEE Trans. Sustain. Energy 7 (2), 504–512.
Sources 248 (51907117), 621–631. https://doi.org/10.1016/j.
https://doi.org/10.1109/TSTE.2015.2503435.
jpowsour.2013.07.058.
Ben hmamou, D., Elyaqouti, M., Arjdal, E.H., Saadaoui, D., Lidaighbi, S., Chaoufi, J.,
Phang, J.C.H., Chan, D.S.H., Phillips, J.R., 1984. Accurate analytical method for the
Ibrahim, A., Aqel, R., El fatmi, D., Obukhov, S., 2022a. A novel hybrid numerical
extraction of solar cell model parameters. Electron. Lett. 20 (10), 406–408. https://
with analytical approach for parameter extraction of photovoltaic modules. Energy
doi.org/10.1049/el:19840281.
Convers. Manage. X 14, 100219.
Pindado, S., Cubas, J., 2017. Simple mathematical approach to solar cell/panel behavior
Ben Hmamou, D., Elyaqouti, M., Arjdal, E., Chaoufi, J., Saadaoui, D., Lidaighbi, S.,
based on datasheet information. Renew. Energy 103, 729–738. https://doi.org/
Aqel, R., 2022b. Particle swarm optimization approach to determine all parameters
10.1016/j.renene.2016.11.007.
of the photovoltaic cell. Mater. Today Proc. 52, 7–12.
M. M. Potapov, “Dichotomy method,” Encycl. Math., no. 2, pp. 1–2, 2011, [Online].
Bouzidi, K., Chegaar, M., Bouhemadou, A., 2007. Solar cells parameters evaluation
Available: http://www.encyclopediaofmath.org/index.php?title=Dichotomy_meth
considering the series and shunt resistance. Sol. Energy Mater. Sol. Cells 91 (18),
od&oldid=12591.
1647–1651. https://doi.org/10.1016/j.solmat.2007.05.019.

10
A. Elhammoudy et al. Results in Optics 12 (2023) 100445

Rezk, H., Babu, T.S., Al-Dhaifallah, M., Ziedan, H.A., 2021. A robust parameter Singh, N.S., Jain, A., Kapoor, A., 2013. A new method to determine the optimum load of
estimation approach based on stochastic fractal search optimization algorithm a real solar cell using special trans function theory (STFT). Int. J. Renew. Energy Res.
applied to solar PV parameters. Energy Rep. 7, 620–640. https://doi.org/10.1016/j. 3 (2), 378–382.
egyr.2021.01.024. “Solar - Fuels & Technologies - IEA.” https://www.iea.org/fuels-and-technologies/solar
Rosa-Clot, M., Tina, G.M., 2018. Introduct. PV Plants. https://doi.org/10.1016/b978-0- (accessed Jun. 27, 2022).
12-812149-8.00003-x. Villalva, M.G., Gazoli, J.R., Filho, E.R., 2009. Comprehensive approach to modeling and
Ruschel, C.S., Gasparin, F.P., Costa, E.R., Krenzinger, A., 2016. Assessment of PV simulation of photovoltaic arrays. IEEE Trans. Power Electron. 24 (5), 1198–1208.
modules shunt resistance dependence on solar irradiance. Sol. Energy 133, 35–43. https://doi.org/10.1109/TPEL.2009.2013862.
https://doi.org/10.1016/j.solener.2016.03.047. Wang, G., Zhao, K.e., Shi, J., Chen, W., Zhang, H., Yang, X., Zhao, Y., 2017. An iterative
Saadaoui, D., Elyaqouti, M., Assalaou, K., Ben hmamou, D., Lidaighbi, S., 2021. approach for modeling photovoltaic modules without implicit equations. Appl.
Parameters optimization of solar PV cell/module using genetic algorithm based on Energy 202, 189–198.
non-uniform mutation. Energy Convers. Manage. X 12, 100129. Yang, B.o., Wang, J., Zhang, X., Yu, T., Yao, W., Shu, H., Zeng, F., Sun, L., 2020.
Saqib, M., Iqbal, M., Ahmed, S., Ali, S., Ismaeel, T., 2015. New modification of fixed Comprehensive overview of meta-heuristic algorithm applications on PV cell
point iterative method for solving nonlinear equations. Appl. Math. 06 (11), parameter identification. Energy Convers. Manage. 208, 112595.
1857–1863. https://doi.org/10.4236/am.2015.611163. Yu, K., Chen, X., Wang, X., Wang, Z., 2017. Parameters identification of photovoltaic
Senturk, A., Eke, R., 2017. A new method to simulate photovoltaic performance of models using self-adaptive teaching-learning-based optimization. Energy Convers.
crystalline silicon photovoltaic modules based on datasheet values. Renew. Energy Manage. 145, 233–246. https://doi.org/10.1016/j.enconman.2017.04.054.
103, 58–69. https://doi.org/10.1016/j.renene.2016.11.025. Zhang, Y., Lyden, S., De La Barra, B.A.L., Haque, M.E., 2016. A genetic algorithm
D. Sera, R. Teodorescu, and P. Rodriguez, “Photovoltaic module diagnostics by series approach to parameter estimation for PV modules. IEEE Power Energy Soc. Gen.
resistance monitoring and temperature and rated power estimation,” IECON Proc. Meet. 2016, 4–8. https://doi.org/10.1109/PESGM.2016.7741781.
(Industrial Electron. Conf., pp. 2195–2199, 2008, doi: 10.1109/ Zhang, C., Zhang, J., Hao, Y., Lin, Z., Zhu, C., 2011. A simple and efficient solar cell
IECON.2008.4758297. parameter extraction method from a single current-voltage curve. J. Appl. Phys. 110
Sharma, R., Sharma, S., Tiwari, S., 2020. Design optimization of solar PV water pumping (6), 064504.
system. Mater. Today Proc. 21, 1673–1679. https://doi.org/10.1016/j.
matpr.2019.11.322.

11