Development of an Enhanced MPPT Controller in Solar PV Systems Using a Hybrid ANN and Improved Grey Wolf Optimizer

Maximum Power Point Tracking (MPPT) is critical for maximizing energy extraction from photovoltaic (PV) systems under variable irradiance and temperature. This study proposes a hybrid Artificial Neural Network-Improved Grey Wolf Optimizer (ANN-IGWO) MPPT controller, integrating ANN-based prediction with IGWO-driven optimization to improve tracking accuracy, dynamic response, and computational efficiency. A detailed PV system was modeled in MATLAB/Simulink to capture nonlinear behavior under varied irradiance and temperature. The hybrid controller was evaluated against standalone ANN and conventional Perturb and Observe (P&O) methods. The average tracking efficiency of ANN-IGWO, ANN, and P&O across all tested conditions are 95.6%, 90.5%, and 84.8%, respectively. Error metrics show Root Mean Square Errors (RMSE) of 0.2603, 0.7159, and 0.9762, and Mean Absolute Errors (MAE) of 0.125, 0.450, and 0.625 for ANN-IGWO, ANN, and P&O, respectively. Computational time is also reduced for ANN-IGWO at 17.55 s, compared to 37.44 s for ANN and 52.65 s for P&O. These results demonstrate that the hybrid ANN-IGWO MPPT controller offers an effective and computationally efficient solution for real-time PV applications.