Backstepping-Based Artificial Intelligence Algorithm for Photovoltaic-Quadratic Boost Converter System

Abstract

With energy demands rising, fossil fuel resources depleting, and global warming caused by carbon emissions, there’s an urgent need for efficient, renewable, and clean energy. Solar energy, especially photovoltaic systems, is gaining popularity due to its simple structure, low energy production costs, and eco-friendliness. However, these systems suffer from insufficient efficiency due to environmental factors. To address this issue, a Maximum Power Point Tracking interface is required to extract the full power from the photovoltaic panel and increase its efficiency. This study proposes a Backstepping Maximum Power Point Tracking algorithm for photovoltaic systems based on an artificial intelligence algorithm that involves a Quadratic Boost Converter as a power electronics adaptation stage. The control scheme comprises two steps, where an Artificial Neuro Fuzzy Interference System generates photovoltaic reference voltages during the first loop, which serves as an input for the backstepping controller in the second loop. To validate the effectiveness of the suggested algorithm, simulations are performed using MATLAB/Simulink tools, and the results are compared with those of the conventional Perturb and Observe and Perturb and Observe-based Backstepping algorithms. As a result, the suggested Artificial Neuro Fuzzy Interference System-based Backstepping controller exhibits excellent dynamic behavior, indicating that the system is stable, fast, and has accurate tracking.