Optimization of 1st-Order Sliding Mode Control Using ANNs for MPPT in Photovoltaic Systems

This study explores the optimization of photovoltaic (PV) systems, aiming to enhance energy efficiency at the maximum power point (MPPT) in the face of non-linearity and fluctuating environmental conditions. The approach combines first-order sliding mode control (1st-order SMC) with artificial neural networks (ANN) to achieve robust and adaptive control. While 1st order SMC ensures effective MPPT by maintaining rapid convergence and low sensitivity to environmental variations, it is prone to chattering, which can destabilize the system. To overcome this limitation, the integration of ANN facilitates dynamic adaptation of control parameters, reducing chattering and improving tracking accuracy. Simulation results demonstrate that this hybrid ANN-SMC approach enhances the performance of the PV system, ensuring stable energy extraction, minimizing external disturbances, and optimizing the management of uncertainties, thus improving energy efficiency and maintaining optimal performance under varying conditions.