动力锂电池荷电状态的联合估计与实验研究

摘要

针对动力锂电池常用的荷电状态（SOC）估计算法存在的扩展卡尔曼滤波法精度低、无迹卡尔曼滤波法收敛速度慢等问题，在动力锂电池的 Randles 等效模型的基础上，通过脉冲放电实验对模型参数进行辨识；并设计了一种基于迭代扩展卡尔曼滤波（IEKF）与无迹卡尔曼滤波（UKF）联合估计的 SOC 估计法。在电池实验平台上设计模拟工况实验，实验分析表明：该算法的 SOC 初值修正速度快于 EKF 和 UKF，计算量比 UKF 小，且稳态误差不超过1.5％，相对扩展卡尔曼滤波（EKF）提高了40％，是一个收敛快、计算量少、静差小的迭代估计算法。%In view of the current estimation algorithm of state of charge (SOC) for power lithium battery, the extended Kalman filter method has low precision and slow convergence speed , and so on.With these problems in hand, based on Randles equivalent model of power Li -ion battery, parameter identification is finished through pulse discharge experiment.Moreover, an joint estimation algorithm of SOC, which is on basis of iterative extended Kal-man filter and unscented Kalman filter, is presented in this paper.The working condition simulation experiment is designed on battery experimental platform , and experimental analysis shows that the speed of revising initial SOC is faster than that of EKF and UKF, the calculation is smaller than UKF, the static error is no more than 1.5%, which is better than that of EKF algorithm by 40%.It’s evidently that it is an estimation algorithm with satisfactory convergence, static error, and lower computation.