Solar photovoltaic parameter estimation using an improved equilibrium optimizer

• Published in: Solar energy Vol. 209; pp. 694 - 708
• Main Authors: Abdel-Basset, Mohamed, Mohamed, Reda, Mirjalili, Seyedali, Chakrabortty, Ripon K., Ryan, Michael J.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.10.2020; Pergamon Press Inc
• Subjects: Algorithms; Boundaries; Double diode model; Entrapment; Equilibrium Optimizer; Minima; Optimization; Parameter estimation; Photovoltaic cells; Photovoltaic solar cells; Photovoltaics; Single diode model; Solar cells; Solar energy; Photovoltaic solar cells; Parameter estimation; Equilibrium Optimizer; Single diode model; Double diode model
• ISSN: 0038-092X; 1471-1257; 1471-1257
• DOI: 10.1016/j.solener.2020.09.032

•A Novel algorithm is proposed as the new method for identifying the parameters of solar PV.•The accuracy and convergence time of the proposed method is verified.•The parameters of single diode and double diode models of solar PV is estimated.•A set of feasible optimal cell parameters is achieved from several runs.•The significance of experimental data on cell I-V characteristic is validated.•Comparison with recent methods in the literature show the superiority of our proposed algorithm. In this paper, a recent optimization algorithm called Equilibrium Optimizer (EO) is first improved using a linear reduction diversity technique (LRD) and local minima elimination method (MEM). The improved EO (IEO) reduces the diversity of the population until enabling them to get better solutions. This method is centered around improving the particles with the worst fitness values within the population by moving them toward the best-so-far solution as an attempt to increase the convergence toward the near-optimal solution. As a side effect, LRD increases the probability of entrapment into local minima if it could not find a better solution. Therefore, another method known as local minima elimination method (MEM) is used to take the current solution either within the boundaries of two particles selected randomly or within the search boundaries of the problem itself. The extensive comparative experiments demonstrate that the proposed IEO is competitive and often superior compared to recent algorithms. We applied the proposed IEO algorithm to R.T.C France commercial solar cells using a single diode model (SDM), the double diode model (DDM), and three photovoltaic (PV) modules in addition to two commercial ones.