An Improved Fast-Charging Strategy for Batteries Integrating Lithium Plating Detection and Online Optimization

An online fast-charging strategy based on double optimizations (DOBs) is proposed for lithium-ion batteries (LIBs). A dual RC electrical model is developed to calculate the anode potential, which is constrained to be larger than 0 mV to avoid lithium plating. The thermal model and degradation model are coupled with the electrical model. They are applied to calculate the battery temperature and capacity loss during charging. To reach a high safety, the proposed fast-charging strategy limits the battery temperature and terminal voltage and lets them never surpass $45~^{\circ }$ C and 4.3 V. To extend the battery lifespan, the algorithm constrains the maximum current rate. Moreover, the Pareto curve is used to balance the charging time and capacity loss, and the DOBs based on the model predictive control (MPC) and enumeration method are employed to plan the current profile for a short charging time and small capacity fall. Then, the proposed charging strategy is verified and compared to the standard charging strategy (SCS) at different ambient temperatures. According to the results, the charging time of the proposed strategy is reduced by about 38%, and its capacity loss only increases by 0.50% at the cycle number of 495. Under the test conditions, the anode potential achieved by the proposed strategy is always above 0 mV, the terminal voltage is never beyond 4.3 V, and the maximum temperature is always below $40~^{\circ }$ C. In order to further verify the performance of the proposed strategy, it is compared with the charging strategy based on pulse (PCS). The result demonstrates again that the proposed strategy achieves a better performance.