Accurate modeling and parameters estimation of photovoltaic models: Analytical and artificial intelligence solutions

Abstract

The single-diode model is a widely used framework for simulating solar cell behavior. It consists of a diode, a current source, and two resistors to replicate real-world performance. This work introduces three alternative single-diode equivalent circuits with five parameters each. The current-voltage relationships of these equivalent circuits are derived using the Lambert W function, and a closed-form analytical solution is presented using Special Trans Function Theory (STFT). Additionally, the paper introduces a chaotic variant of the Gorilla Troops Optimizer (GTO), referred to as CGTO, for parameter estimation. The effectiveness of the proposed alternative single-diode models and the CGTO algorithm was validated by comparing their accuracy, using root mean square error (RMSE) metrics, against traditional models across various solar cells and conditions (different values of irradiance and temperature). The analysis confirmed that the alternative configurations could achieve comparable precision in modeling solar cell characteristics under different conditions, challenging the exclusivity of the standard single-diode model's configuration. This indicates that various single-diode solar cell models can effectively represent the current-voltage behavior of solar cells. Furthermore, the configuration of elements in the classic SDM equivalent circuit is not definitive or the most precise. This opens the door for the creation of more sophisticated models of solar cells with diverse element configurations.

Keywords

• Accurate modeling
• Equivalent circuits
• Lambert functions
• Optimization methods
• Photovoltaic