Comparative Performance Analysis of Metaheuristic Optimization Algorithms for Parameter Identification of Photovoltaic Cell/Module

Abstract

PV (photovoltaic) cells are key elements in the conversion process of the sunlight into electrical energy. Therefore, the accurate modeling of PV cells or modules is of amount importance to not only predict their performance in terms of energy production but also to analyse their degradation process and detect their faults, which may be of different types. In this regard, various metaheuristic algorithms have been proposed over the past years to identify the parameters of the PV cells/modules model with the aim to accurately predict their behaviour under different operation conditions. This paper intends to compare three parameters extraction algorithms: Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Grey Wolf Optimizer (GWO). From random weather conditions, the measured I-V curves are first translated into Standard Test Conditions (STC) of 1000 W/m<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and 25°C. Thereafter, the five standard parameters of the one-diode model are extracted using the aforementioned algorithms. A comparative study is carried out based on experimental data of an Isofoton 106W-12V monocrystalline PV module. It reveals the superiority of the GA in terms of accuracy. It shows also that the PSO outperforms the remaining algorithms from the point of view computational complexity. Moreover, the worst performances in terms of accuracy and computational complexity are obtained with the GWO.