Energy-efficient battery thermal management strategy for range extended electric vehicles based on model predictive control and dynamic programming

Battery thermal management system is important for improving the efficiency, lifespan, and safety of new energy vehicle batteries. An energy-efficient model predictive control algorithm based on dynamic programming solver is proposed for battery thermal management strategy. A control-oriented nonlinear battery thermal model is established for predicting temperature changes in thermal management system. The dynamic programming algorithm is utilized to solve the nonlinear optimal control problem, which includes state boundary calculation and optimal control sequence backward calculation. The effects of the weighting parameters and other hyperparameters in the proposed control strategy on the control performance are investigated to achieve the optimal trade-off between thermal equilibrium temperature control, computational efficiency, and energy savings. The simulation results show that the proposed strategy can achieve considerable energy savings in high and low environment temperatures, as well as under standard and real driving conditions. Compared to the on-off based strategy and proportional control-based strategy, the proposed strategy saves up to 8.94 % and 8.33 % of actuator energy at an environment temperature of -20 degrees C, and up to 77.83 % and 99.83 % of actuator energy at an environment temperature of 40 degrees C, utilizing the energy consumption of the proposed strategy as a baseline.