Maximum power extraction of partially shaded solar photovoltaic array using reconfiguration method

ABSTRACT

This article presents a reconfiguration strategy, named the Cheetah Optimization Algorithm (COA), for maximizing power generation in partially shaded solar photovoltaic (PSPV) arrays. The proposed method’s originality aims to mitigate the negative impacts of partial shading, improve the output power of the PV system, and prevent thermal failure of shaded solar panels. The strategy combines efficient maximum power point tracking techniques with suitable photovoltaic system design topologies. The COA is done on the MATLAB platform and is analyzed with the existing methods such as the Wild Horse Optimizer (WHO), Salp Swarm Algorithm and Heap-based Optimizer. The outcome shows the superior performance of the COA in achieving the extraction of maximum power compared to the other methods. Additionally, by comparing the two patterns of partial shading, it is shown that the COA extracts more power than the existing methods and gets to the global power value faster. The COA is compared with existing methods, including WHO, SSA, and HBO. In terms of accuracy, the proposed system achieves the highest value of 0.93, surpassing the accuracy of HBO, WHO, and SSA, which stand at 0.8, 0.9, and 0.9, respectively. The specificity, precision and recall are also reflect the superiority of the proposed system with the values of 0.87, 0.9, and 0.86, respectively, outperforming the existing algorithms in most categories.

GRAPHICAL ABSTRACT

This article proposes a reconfiguration strategy for a solar PV array that is partially shaded in order to generate the maximum power. The proposed approach is Cheetah Optimization Algorithm (COA). Commonly it is named as COA. The primary goal of the suggested strategy is to cause the thermal failure of shaded solar energy panels and enhance the output power produced by the photovoltaic (PV) system.

KEYWORDS:

• PV array
• maximum power
• global power
• resistors
• Cheetah optimization algorithm (COA)
• and irradiance

Nomenclature

Symbols and Abbreviations	=
$\mathit{Iph}$	=	Current produced by occurrence particle emission
$\mathit{Ish}$	=	Passage of electricity through the shunt resistor
K	=	Boltzmann constant
$V_m$	=	Module voltage
q	=	Electron Charge
$I_{ph}^{STC}$	=	image current under the STC
G	=	Emission of the Photovoltaic module
$\mathit{R}$	=	Random generation
Xt + 1i,j	=	Cheetah i’s upcoming and existing places in organization j
${\mathrm{Y}}_{\mathrm{i},\mathrm{j}}^{\mathrm{t}+1}$and $Y_{i,j}^t$	=	Cheetah i’s most recent and present locations in arrangement j
$r_{\mathit{ij}}\mathit{and}{\beta }_{i,j}^t$	=	modification reason and boundary reason associated to the cheetah i
PV	=	Photovoltaic
COA	=	Cheetah Optimization Algorithm
SSA	=	Salp Swarm Algorithm
HBO	=	Heap-Based Optimizer
TCT	=	Total Cross-Tied
WO	=	Whale Optimization
SSD	=	Static Shade Dispersion
GMPP	=	Global Maximum Power Point
ID	=	Diode current
a	=	Idealist factor
T	=	Temperature
$R_s$ and $R_p$	=	parallel resistors
$I_m$	=	module current
$k_i$	=	Thermal coefficient for short-circuit current
$\mathit{V}$t	=	Predicted thermal voltage
$X_{i,j}^t$	=	modern situation of cheetah
${{\hat{r}}_{\mathit{i},\mathit{j}}}^{-1}$ and ${{\mathit{\alpha }}_{\mathit{i},\mathit{j}}}^{\mathit{t}}$	=	step duration and randomization parameter
$Y_{\beta ,j}^t$	=	present location of the prey in preparation j
MPPT	=	Maximum power point tracking
PS	=	Partial shading
PSPV	=	Partly Shaded Photovoltaic
WHO	=	Wild Horse Optimizer
PSCs	=	Partial Shadow Circumstances
GP	=	Global Peak
MRL	=	Mimetic Reinforcement Learning
SEPICs	=	Single-Ended Primary-Inductor Converters
LMPP	=	Local Maximum Power Point

Disclosure statement

No potential conflict of interest was reported by the author(s).