An Improved EKF Algorithm for SOC Estimation of Lithium Battery Considering Temperature Effects

Abstract

Abstract In order to improve the estimation accuracy of the state of charge (SOC) of lithium-ion power batteries during use, an estimation method that considers different temperature effects is proposed in this paper. First, the second-order resistor-capacitor (RC) circuit model is selected as the research model based on a comparative analysis of different models. Then, the parameters in the circuit model are identified online in real time under different temperatures to analyze the temperature effects on internal parameters. Lastly, the noise variance can be automatically adjusted according to the environment by using the improved extended Kalman filter (EKF) algorithm with the differential evolution algorithm, which optimizes the fixed process noise variance in the original EKF algorithm. The changing noise variance is substituted in the original EKF algorithm to estimate SOC under cycle working conditions online. The results show that the improved EKF algorithm increases the estimation accuracy of the SOC of lithium batteries. Using the improved EKF algorithm, the estimated value can quickly converge to the actual value at different temperatures, and the SOC error remains within 2% in the whole estimation process.