Performance Enhancement of Solar Water Pumping Systems Using Deep Learning

Abstract

Photovoltaic energy is widely used in water pumping systems. Under partial shading, the PV power-voltage curve exhibits multiple local maxima, making it difficult for traditional MPPT algorithms to reach the global maximum power point (GMPP), resulting in energy losses. In this work, we propose the use of artificial intelligence, specifically, Deep Neural Networks (DNN) and Shallow Neural Networks (SNN) to enhance MPPT techniques and improve the overall performance of a PV pumping system. The induction machine is controlled via Direct Torque Control (DTC). Simulation results confirmed the efficiency of the proposed system in both ideal and shaded conditions, with improved tracking, reduced oscillations, and enhanced stability.