Parameter extraction of photovoltaic models using an enhanced Lévy flight bat algorithm

Abstract

In this paper, a modified version of the bat algorithm (BA), called enhanced Lévy flight bat algorithm (ELBA), is proposed for accurate and efficient parameter extraction of different photovoltaic (PV) models from experimental data. Typically, it is formulated as a multimodal nonlinear optimization problem in which the objective function is to minimize the root mean square error verified between the real data and the simulated ones by the PV model at hand, considering certain values for its parameters. In addition, the constraints are associated to the lower and upper bounds of these parameters. From the computational perspective, the main innovations of ELBA lies in the: (i) introduction of a specific mathematical expression to enhance the diversification of new solutions; (ii) adoption of a mathematical expression based on the Lévy flight to perform an effective local search; and (iii) selection of new equations for updating certain control parameters, which provide a better balance between the exploration and exploitation mechanisms of the algorithm. Simulation results comprehensively demonstrate that ELBA has a very competitive performance in terms of effectiveness, robustness, stability, convergence speed and time of simulation, in relation to other state-of-the-art metaheuristic algorithms. Therefore, the major contribution of this paper is the ELBA, a modified metaheuristic algorithm which proves to be a promising tool for parameter extraction of different PV models from experimental data.