Domestic peak-load management including vehicle-to-grid and battery storage unit using an artificial neural network

Controlled charging and discharge of electric vehicles (EVs) in distributed power systems can reduce the peak demand on the grid. This paper presents an energy management strategy (EMS) using an artificial neural network to shave the domestic peak grid load by the coordinated response of distributed energy resource (DER) units including photovoltaic (PV) systems, V2G (vehicle to grid)-capable EVs, and battery energy storage systems (BESS). The developed EMS is implemented on a test system including 15 houses and the realistic load pattern of California, U.S. From the simulation results for this network it is found that an artificial-neural-network controller can effectively coordinate the system to reshape the load curve and shave the domestic peak loading by up to 77%.