Precise estimation of solar photovoltaic parameters via brown bear optimization and Differential Evolution

Abstract

Estimating the optimal parameter values for photovoltaic (PV) models is inherently challenging due to the complex and nonlinear nature of their current–voltage (I–V) characteristic curves. Precise parameter estimation is critical for ensuring the efficient operation of PV systems, as it directly influences energy output and current generation. Traditional methods for addressing this problem often suffer from convergence to local optima and require substantial computational resources, particularly concerning the count of fitness evaluations. To overcome these challenges, this paper presents an enhanced optimization method: the Brown Bear Optimization Algorithm (BBOA) hybridized with Diagonal Linear Uniform Initialization (DLUI) and the Differential Evolution (DE) algorithm, termed BBOA-DLUI-DE. This hybrid approach’s innovative design lies in integrating the DE algorithm to enhance solution diversity, ensuring better exploration and preventing premature convergence. DLUI contributes to a uniformly diverse initial population that supports rapid and robust optimization. This synergy between BBOA, DLUI, and DE addresses the limitations of existing methods by combining efficient global search capabilities with effective local refinement. The proposed BBOA-DLUI-DE method has been rigorously evaluated against state-of-the-art techniques, demonstrating superior performance in finding optimal parameter values for various PV models. Comparative statistical and practical analyses highlight that BBOA-DLUI-DE outperforms traditional methods regarding accuracy and computational efficiency. Furthermore, validation using manufacturing data sheets (MCSM55 and TFST40) confirms the practical applicability and robustness of the proposed method, making it a highly effective tool for estimating PV parameters.