Enhanced Efficiency and Robustness in PV Water Pumping Systems via a Novel Weighting Factor-Free MPC with Online Loss Minimization

Battery-free photovoltaic water pumping systems (PV-WPSs) represent a practical solution for utilizing solar energy in applications such as irrigation and potable water supply, particularly in remote regions. This article proposes a standalone PV-WPS featuring a two-stage converter configuration designed to enhance system efficiency and reliability. The first stage employs a hybrid Maximum Power Point Tracking (MPPT) strategy, combining an Artificial Neural Network with Sliding Mode Control (ANN-SM). This technique is selected for its rapid and robust performance under highly dynamic atmospheric conditions, ensuring maximum power extraction from the PV panels. The primary contribution of this work lies in the second stage, where a novel Model Predictive Control (MPC) algorithm is developed for the three-phase inverter driving the induction motor (IM). The proposed MPC is designed for real-time minimization of power losses, thereby maximizing overall system efficiency. A key innovation of this approach is the elimination of weighting factors, which are typically difficult to tune and can limit the implementation and reliability of traditional MPC systems under varying operating conditions. The entire PV-WPS is modeled and simulated in MATLAB/Simulink under fast-changing environmental profiles. A comparative analysis demonstrates that the proposed system significantly improves both tracking performance and overall efficiency. In particular, the ANN-SM MPPT enhances tracking efficiency by approximately 3.92% and reduces power ripple by 95.4% compared to conventional methods. Furthermore, the proposed weighting-factor-free MPC with online loss minimization reduces motor power losses by 16.67% and improves dynamic response, achieving a 43.75% faster settling time. These improvements result in faster response, reduced torque ripples, and increased water pumping capability. Ultimately, by removing the need for weighting factors, the proposed approach simplifies controller design and improves robustness, while the integration of online loss minimization maximizes solar energy utilization, representing a significant advancement in PV-WPS technology.