AI-Enhanced Maximum Power Point Tracking for PV–Battery Systems: Modeling, Simulation, and Comparative Analysis

Abstract

The integration of photovoltaic (PV) systems with energy storage is essential to ensure reliable power delivery under variable environmental conditions. A critical component in such systems is the maximum power point tracking (MPPT) controller, which directly influences overall efficiency. Conventional methods, such as Perturb and Observe (P&O), are widely used due to their simplicity but suffer from steady-state oscillations and slow response under rapidly changing irradiance. This paper presents a comparative study between a conventional P&O–based MPPT with multistage battery charging and an artificial intelligence (AI)–enhanced fuzzy logic MPPT integrated with the same charging strategy. System models of the PV array, DC-DC buck converter, and lithium-ion battery are developed to provide an accurate simulation framework. Results demonstrate that the fuzzy MPPT effectively mitigates oscillations, improves tracking efficiency, and ensures smoother battery charging profiles compared to the conventional approach, albeit with increased computational complexity and simulation time.