基于最大相关熵扩展卡尔曼滤波算法的锂离子电池荷电状态估计

The traditional extended Kalman filter(EKF)algorithm has low accuracy in estimating the state of charge(SOC)of lithium-ion battery under the non-Gaussian noise interference. Therefore, a new extended Kalman filter (MCC-EKF) algorithm based on maximum correlation-entropy criterion was proposed. Firstly, the Thevenin equivalent circuit of the lithium-ion battery was model and its parameters was identified. Secondly, the proposed algorithm MCC-EKF and EKF algorithm were used to estimate the SOC under different noise interference. The experimental results show that, compared with the EKF algorithm, the running time of the new algorithm increases by 0.282s and the estimation accuracy increases by 19% under Gaussian noise interference; under non-Gaussian noise interference, the running time of the new algorithm increases by 0.418s and the estimation accuracy increases by 51%. In addition, given the wrong initial SOC value, the new algorithm can converge to the true value within 10s after the battery starts working, indicating that the new algorithm has better robustness. The proposed algorithm has high estimation accuracy and good robustness while the increase of running time is small, and it is an effective SOC estimation method.