Data-driven model predictive control of community batteries for voltage regulation in power grids subject to EV charging

In this paper, a Model Predictive Control (MPC) for community Battery Energy Storage Systems (BESS) is proposed to mitigate the Electric Vehicle (EV) charging demand while maintaining voltage regulation in residential power grids. With the increased penetration of renewable-based Distributed Energy Resources (DERs) and the EV uptake, distribution grids are highly subject to voltage fluctuations due to back power feed from DERs or significant EV charging demand. Conventionally, Distribution Network Operators (DNOs) have addressed this issue by adjusting the taps of transformers. However, managing the naturally unpredictable renewable generation and the power consumption behaviour of technology owners requires advanced real-time algorithms. In this paper, we apply a real-time control algorithm, based on Model Predictive Control (MPC), to a community BESS in order to reduce voltage fluctuations. The MPC algorithm performs in real-time based on readings from smart meters of households. It predicts the future values of voltages at households based on the current and past readings of smart meters, and calculates the BESS charging or discharging rate such that future voltage deviations are minimised. Simulations on the real-world power consumption data from Victorian residential consumers, in Australia, show that the proposed controller performs significantly better than the fixed charging/discharging BESS schedule in both voltage regulation and BESS degradation. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).