Improvement teaching-learning-based optimization algorithm for solar cell parameter extraction in photovoltaic systems

Introduction. This study investigates parameter extraction methods for solar cell analytical models, which are crucial for accurate photovoltaic (PV) system design and performance. Problem. Traditional single-diode models, while widely used, often lack precision, leading to inefficiencies in parameter extraction essential for reliable PV systems. Goal. The work aims to improve the Teaching-Learning-Based Optimization (TLBO) algorithm to enhance the accuracy of parameter extraction in PV models. Methodology. We adopt an enhanced single-diode model, integrating modifications into the TLBO algorithm, including dynamic teaching factor adjustment, refined partner selection, and targeted local searches with the fmincon function. Comparative analysis with experimental data from four PV systems validates the model’s accuracy. Results. The enhanced TLBO algorithm achieves superior convergence and reliability in parameter extraction, as evidenced by 500 independent runs. Originality. Key contributions include methodological improvements such as dynamic adjustment of the teaching factor and a new approach to partner selection, which significantly optimizes the algorithm’s performance. Practical value. This research provides a robust framework for solar cell parameter extraction, offering practical benefits for PV system designers and researchers in improving model accuracy and efficiency. References 35, table 1, figures 15.