State of charge estimation based on improved LiFePO4 battery model and Kalman filtering

Abstract

The Lithium iron phosphate (LiFePO4) batteries are widely used as power source for electric vehicles. By reason of safety issues, it is significant to estimate the state of charge (SOC) of a battery precisely. On the basis of the characteristics of LiFePO4 battery, the improved Thevenin equivalent circuit model is proposed, and the model parameters are identified by experimental testing. Additionally, a novel algorithm, which combines the Open-circuit voltage method, Ampere-hour integration and Kalman filtering, for SOC online estimation is proposed. The experimental verification of the improved battery model and SOC estimation algorithm is presented. The improved battery model shows better estimation accuracy than traditional Thevenin equivalent circuit model. The proposed SOC estimation algorithm is able to obtain LiFePO4 SOC precisely even with inaccurate initial values and distinguish the performances of different batteries.