Model Predictive Control Based Battery Energy Management For EV Ancillary Service Provision*

This study presents an optimal energy management method for the ancillary service provision of an electric vehicle (EV) battery pack while the EV is parked. An electro-thermal model of cylindrical Li-ion battery cells is employed, in which the Unnewehr universal model is used to approximate the relation between the open-circuit voltage and the state-of-charge (SoC) by a linear function. Therefore, the physical limits on the cell’s SoC and terminal voltage can be represented by linear constraints of the charging-discharging current. On the other hand, the physical limit on battery cells’ temperatures is in form of a second-order cone constraint. As a result, the considered energy management problem of the EV battery pack is formulated as a second-order cone program (SOCP). However, the temperatures of battery cells are dependent on the ambient temperature which can only be predicted and not precisely known in advance. To cope with this issue, the model predictive control (MPC) framework is utilized to solve the formulated SOCP problem. Simulation results then depicts an interesting relation between the charging-discharging current and the length of the ancillary service period, where the higher the former is, the shorter the latter becomes. Furthermore, the temperatures of battery cells is significantly increased as the charging-discharging current is larger, which affects to the cell’s lifespan.