Performance Evaluation of ANN-Based MPPT in Photovoltaic Systems via Simulation: Improved Tracking Speed, Accuracy, and Power Stability

The efficiency of photovoltaic (PV) systems depends on accurately and rapidly tracking the maximum power point (MPP) under dynamic conditions. Conventional maximum power point tracking (MPPT) methods like Perturb and Observe (P&O) suffer from slow convergence, steady-state oscillations, and poor adaptability. This study proposes an artificial neural network (ANN)-based MPPT to improve tracking performance by leveraging key electrical characteristics of PV systems and DC-DC boost converters. The ANN model is trained using critical electrical parameters, including maximum current, voltage at MPP, and duty cycle variations, to predict the optimal operating point dynamically. A simulation-based analysis compares ANN-MPPT with P&O under varying irradiance. Results show ANN-MPPT achieves 7.31 W/s tracking speed, nearly 50% lower tracking errors, and reduced power oscillations (0.58% vs. 1.02%), overshoot (1.00% vs. 1.59%), and undershoot (0.69% vs. 1.06%). ANN-MPPT maintains 98.1% efficiency in steady-state and adapts better to irradiance fluctuations, confirming its robustness and suitability for real-world PV applications.