Performance Optimization of Photovoltaic Water Pumping Systems: A Comparative Study of PSO-Based and ANN-Based MPPT with Real-Time Validation

This paper introduces an optimization strategy for enhancing the efficacy of Photovoltaic Water Pumping Systems (PVWPS). The proposed approach incorporates particle swarm optimization (PSO) based MPPT for the photovoltaic (PV) system and direct torque control (DTC) for the pump operating motor (IM). The PSO-based MPPT aims to improve the PV array’s energy efficiency under varying sunlight levels, while DTC improves the IM’s dynamic performance to ensure efficient pump operation. The PVWPS’s performance utilizing PSO for MPPT and DTC was evaluated and compared to a system that employs Artificial Neural Networks (ANN) for MPPT and DTC. The PSO-based approach led to significant improvements, including a 60% reduction in torque ripples, an 83.33% decrease in power oscillations, a 66.67% drop in flux ripples, a 50% improvement in system responsiveness, and a notable enhancement in water output. Real-time validation using the dSPACE control desk DS1104 aligned with the simulation outcomes, confirming the reliability and efficiency of the proposed PSO-based control strategy.