Integrating Artificial Neural Networks into Hybrid Perturb and Observe Algorithm for Improved Photovoltaic System Performance

Abstract

One of the most prominent and sustainable sources of power generation is the PV solar system. However, the efficiency of energy production from the PV system is highly dependent on solar irradiance. Fluctuations in temperature can significantly impact the system's power generation, leading to suboptimal performance. To address this challenge, the implementation of a Maximum Power Point Tracking (MPPT) technique is crucial for enhancing the overall performance of the PV system. In this research paper, we propose the development of a novel MPPT algorithm based on the Perturb and Observe (P&O) technique, combined with an Artificial Neural Network (ANN). By utilizing MATLAB/Simulink, we conduct various test cases involving both uniform and non-uniform irradiance scenarios to evaluate the effectiveness of the proposed hybrid method (P&O-ANN). Furthermore, compare hybrid approach with the existing P&O-Particle Swarm Optimization (PSO) technique. The obtained results demonstrate the superiority of our proposed method, as it outperforms the P&O-PSO technique in terms of maximum power extraction and tracking speed. The P&O-ANN algorithm not only enhances the efficiency of power generation in varying environmental conditions but also ensures better accuracy and precision in tracking the maximum power point. These findings underscore the potential of our approach to significantly improve the performance of PV solar systems and contribute to the advancement of sustainable power generation.