Intelligent Artificial power sharing of water desalination plant fed by renewable source

This paper presents an energy management system based on artificial neural network (ANN) strategy integrated with a hybrid renewable production system (photovoltaic-wind) coupled to a brackish water reverse osmosis desalination unit in order to share the power flow during the system operation. A dynamic simulator of the proposed system which includes the renewable generators, three motor-pumps, reverse osmosis desalination unit, three water tanks, annual consumption data, annual meteorological data, and the energy management system is developed. The ANN models has been developed in the objective of sharing the generated power into the three motor pumps system (pumping motor-pump, desalination motor-pump, and storage motor-pump) taking into account not only the available hybrid power but also the three water tanks levels (brackish water tank, permeate water tank and fresh water tank). For this reason, the obtained results by an optimized fuzzy logic- genetic algorithm are used as training data. After training and testing, it is concluded that the ANN models show good performance. The results of yearly hybrid power test showed that the used ANN architecture can estimate the power sharing with a coefficient of correlation between the actual and estimated values equal to 99.8%.