A comparative study of benchmark function optimization for photovoltaic model parameter estimation using new intelligent optimization algorithms

This study systematically evaluates five novel intelligent optimization algorithms, namely, Artificial Lemming Algorithm (ALA), Cloud Drift Optimization (CDO), Dream Optimization Algorithm (DOA), Starfish Optimization Algorithm (SFOA), and Tornado Optimizer with Coriolis Force (TOC), to address the highly nonlinear, multimodal, and high-dimensional challenges of photovoltaic (PV) model parameter estimation. Existing methods frequently encounter difficulties in achieving a balance between accuracy, stability and computational efficiency. The present study proposes a function case-driven validation framework that integrates four benchmark function case sets and employs rigorous statistical analyses, including Friedman’s test and Wilcoxon signed rank test. The proposed the Starfish Optimization Algorithm demonstrates efficacy, exhibiting the mean root mean square error of 0.014 and an 80% reduction in standard deviation. Its rank significance is statistically confirmed (Friedman’s test, p=0.05). The proposed framework is innovative in its pioneering application of state-of-the-art algorithms for photovoltaic parameter estimation, as well as methods for robust validation in complex environments. The outcomes emphasize the prospective benefits of intelligent optimization of energy systems, offering a valuable instrument for high-precision photovoltaic modelling and enhanced operational efficiency.