An adaptive power management approach for hybrid PV-wind desalination plant using recurrent neural networks

In hybrid Photovoltaic (PV)-wind desalination plants, power management is necessary because renewable energy sources such as PV and wind are highly variable. An optimal power management system also helps extend the equipment's life by preventing overloading or underloading of the system. This paper proposes a recurrent neural network (RNN) based power management for a desalination plant. In the proposed work, renewable energy sources like solar and wind are utilized to power the reverse osmosis (RO) desalination unit. The developed RNN model optimizes renewable energy sources while maintaining a stable function of the desalination process. The RNN power management module utilizes historical data to predict the power generation of the PV and wind sources and then adjusts the system's output to meet the power demand of the desalination plant while considering the limitations of the RO unit and the water profile requirements. The developed model was implemented in MATLAB tool, and the results are estimated. Simulation outcomes demonstrate that the developed model improves the hybrid system's performance in terms of resource and battery storage utilization and minimizes energy loss.