Predictive Battery Preconditioning Strategy Considering Charging Time, Battery Degradation, and Energy Consumption

During fast charging, battery temperature plays a critical role in determining charging time, battery degradation, and auxiliary energy consumption. If the battery is not within the optimal temperature range at the start of charging, performance can deteriorate significantly. To enable optimal performance, the battery must be thermally preconditioned before arriving at the charging station. This work presents an optimal battery preconditioning strategy that considers charging time, battery degradation, and energy consumption. The approach utilizes route-based velocity prediction and optimizes the battery thermal management strategy on the way to the charging station using nonlinear programming. The optimization is based on precomputed maps that represent charging time, degradation, and energy consumption as functions of battery temperature and state of charge at the start of charging. These maps are generated through fast charging simulations using a validated high-fidelity simulation model and a 1D electrochemical battery model. The strategy then determines the optimal battery temperature trajectory to achieve the desired state upon arrival. This trajectory is enforced using a nonlinear model predictive controller (NMPC), which leverages the information of the velocity prediction to control the battery thermal system, specifically the coolant pump, PTC heater and compressor. Depending on user-defined priorities, the strategy can minimize charging time, battery degradation, energy consumption, or achieve a trade-off among these objectives. Both the velocity prediction and the control-oriented models are developed and validated using comprehensive measurement data from a state-of-the-art battery electric vehicle (BEV). Finally, the complete strategy is deployed on a validated high-fidelity simulation model. Results demonstrate that the proposed method effectively leverages predictive information to minimize charging time, battery degradation or energy consumption and outperforms rule-based preconditioning strategies.