Stochastic Model Predictive Controller for Battery Thermal Management of Electric Vehicles

The performance and safety of a battery for electric vehicles get worse when the battery temperature is very low or very high. The thermal management system regulating battery temperature consumes considerable electric energy, in particular, for cooling the battery. To maximize the vehicle driving range, control of battery temperature should consider the consumed energy while regulating the battery temperature. In this paper, model predictive control is applied to the battery cooling system. An effective model predictive control (MPC) requires a good but simple system model with proper estimation of near-future disturbances. The components of the battery cooling system are modeled to represent the energy consuming and heat transferring mechanism with some assumptions for the simplicity. The future information is estimated using historical driving data in the stochastic sense. A stochastic model predictive control (SMPC) is designed to minimize the energy consumption with acceptable temperature regulation. The proposed control method shows significantly reduced energy consumption while keeping acceptable temperature regulation performance compared to a typical temperature controller, such as a thermostat-type controller. The MPC with forecasted future information enables preemptive control actions and thus prevents unnecessarily frequent cooling system operations due to instantaneous temperature changes, which reduces energy consumption.