DSM Approach for Water Heater Control Strategy Utilizing Elman Neural Network

This paper describes an artificial neural network based demand-side management (DSM) strategy to shift the peaks of the average residential electrical water heater power demand profile from periods of high demand to off peak periods. The DSM strategy is achieved by dividing the water heaters connected to certain distribution feeder into blocks and controlling each block by a different individual neural network controller. The proposed control schemes will consider an adequate representation of the customers´ specifications and preferences. Simulation results are presented to show the effectiveness of the proposed DSM strategy to shift the average electrical water heater peak demand to off peak periods and to level the utility distribution demand profile.