High efficiency electric vehicle charging strategy based on model predictive control

The accuracy of lithium-ion battery charging control has a profound effect on the cycle life and safety of battery system in electric vehicles. Therefore, we propose a highly efficient charging method based on model predictive control. A high-precision SOC estimation approach can be carried out by unscented Kalman filter based on Thevenin model, and parameters are identified by simulated annealing algorithm. Further, the model predict control is presented and the implementation is realized. Finally, precision of SOC estimation is validated by experiment. Simulation results show that compared with the CCCV method, the proposed battery charging strategy has advantages of both high accuracy and high efficiency, hence delivering its practicability to be extended to battery systems applying complex operating conditions.