A hybrid observer for SOC estimation of lithium-ion battery based on a coupled electrochemical-thermal model

An accurate prediction of the state of charge (SOC) of lithium-ion battery provides crucial information for the Battery Management System (BMS). In this paper, a simplified uneven electrochemical model coupled to a bulk thermal model of Lithium-ion battery is built up. The parameters of battery model identified in previous research can provide a more accurate description of the battery dynamic characteristics. This implementation of the coupled model is then applied to amalgamation with the extend Kalman filter combined with smoothing variable structure filter (EK-SVSF) to obtain an accurate and robust SOC result. MATLAB/Simulink and experiments, including 1C pulse discharge and NEDC cycle tests are adopted to evaluate the performance of the proposed hybrid algorithm. The simulation results indicate that the SOC error is less than 2%, therefore the algorithm is suitable for the new energy vehicle power BMS.