Performance Optimization and Reliability Enhancement of PV Desalination via ANN-Based Control

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

Abstract The increasing global demand for renewable energy and potable water has accelerated the adoption of photovoltaic (PV)-powered desalination systems as a sustainable alternative. This paper presents an artificial neural network (ANN)-based control framework for performance optimization and reliability enhancement of a 100 kW PV desalination system connected to the electrical grid. Multiple ANN-based trained algorithms, including Bayesian Regularization (BR), Scaled Conjugate Gradient (SCG), and Levenberg–Marquardt (LM), are investigated to address challenges of forecasting accuracy, adaptability to dynamic environmental variations, and operational robustness. The proposed ANN controllers optimize key operational parameters such as solar energy utilization and desalination throughput, ensuring reliable performance under both PV-dominant and grid-support modes. The methodology is validated through MATLAB/Simulink simulations and real-time hardware-in-the-loop (HIL) implementation on an OPAL-RT platform. Experimental results demonstrate strong consistency with simulation outcomes, confirming the controllers’ robustness, real-time feasibility, and superior reliability. Comparative analysis reveals substantial improvements in output power stability, transient response, and disturbance rejection across fluctuating temperature and irradiation profiles. The findings highlight the potential of ANN-based control as a practical, scalable solution to enhance efficiency, reliability, and resilience in PV-driven desalination systems.

Links and sources

Comments