An Accurate Co-Estimation of Core Temperature and State of Charge for Lithium-Ion Batteries With Electrothermal Model

Lithium-ion batteries (LIBs) have become the main power source for electric vehicles and energy storage systems. But, the hazardous overcharging and overheating can be easily occurred during improper operations or accidents, which are urgent problems to be improved. Therefore, excellent core temperature and state of charge (SOC) estimations are indispensable prerequisites for safe and efficient applications of LIBs. However, the core temperature and SOC play important roles in each other due to coupled electrochemical reactions inside LIBs, so that their accurate co-estimation is difficult to achieve. Aiming to tackle this problem, this article proposes an electrothermal model consisting of a temperature-dependent dual-polarization equivalent circuit model (DPM) and a two-state thermal model (TSM). Also, the dual time windows method and particle swarm optimization (PSO) algorithm are used to identify model parameters. Then, dual unscented Kalman filters (DUKFs) are integrated with the electrothermal model to achieve accurate and adaptive co-estimation of core temperature and SOC. The experiments are finished under different ambient temperatures and dynamic conditions. The results quantitatively verify the accuracy and robustness of the proposed co-estimation algorithm. Furthermore, the comparisons with other methods demonstrate the superiority of states estimation performance.