Model Predictive Control Approach to Electric Vehicle Charging in Smart Grids

Model Predictive Control Approach to Electric Vehicle Charging in Smart Grids by Somil Bansal Master of Science in Electrical Engineering and Computer Sciences University of California, Berkeley Professor Claire Tomlin, Chair In this work, I present a method to design a predictive controller for handling Plug-andPlay (P&P) requests of Electric Vehicles (EVs) in a power distribution system. The proposed method uses a two-stage hierarchical control scheme based on the ideas of Model Predictive Control (MPC) tracking for periodic references to ensure that bus voltages track the closest possible (reachable) periodic reference to the nominal voltage while minimizing the required generation control and guaranteeing satisfaction of system constraints at all times. Next, the problem of handling real-time P&P requests is considered using a prepration-phase before the actual connection/disconnection of EVs. The only assumption made on the load is that it is time-periodic with a period of 24 hours. Under this assumption, it is proved that the proposed controller is recursively feasible, is exponentially stable and both the EVs’ State of Charge (SOC) and bus voltages converge to the desired SOC and to the optimal reference trajectory, respectively. Finally, the proposed scheme is illustrated in a set of examples.