Optimal control of a DC microgrid with busbar matrix for high power EV charging

Recent advances in the development of reconfigurable batteries pave the way for novel DC microgrid architectures that eliminate the need for DC-DC converters. The present study is focused on the control of a microgrid comprising a battery system with three reconfigurable strings to flexibly operate two electric vehicle (EV) fast chargers, a photovoltaic (PV) system, and a grid-tie inverter. The primary control tasks are to dynamically connect the individual battery strings to the other system components through a busbar matrix, and to manage the energy exchange with the AC grid. The paper formulates the control tasks as a mixed-integer linear optimization problem, virtually splitting the system into three parallel representations, each constructing the perspective of one battery string on the busbar matrix. The functionality of the proposed control is assessed through simulation scenarios using actual PV production and EV charging data of a prototype installed on the Danish island of Bornholm. To quantify the performance, the optimizer is compared with a heuristic control. Considering grid energy costs and revenues through EV charging, the optimal control increased the profit by 5.4% in the summer and 13.0% in the winter scenario, with respect to the benchmark control.