Abstract: In order to improve the estimation accuracy of state of charge (SOC) for Li-ion battery commonly used in electric vehicles, a battery model was first proposed which was equivalent to a six-parameter RC circuit. The extended Kalman filter (EKF) was then employed to estimate the SOC while dealing with the nonlinearity of the battery model. Further to cope with the disturbances caused by unknown and random noises, an adaptive extended Kalman filter (AEKF) algorithm was introduced which estimated the statistical attributes of the noises for the purpose to adaptively adjust the SOC estimation. Simulation results show that the AEKF algorithm is more robust to the external disturbances, although both EKF and AEKF algorithms have quite accurate SOC estimation.

Key words: vehicle engineering, electric vehicle, Li-ion battery, state of charge(SOC), adaptive Kalman filter

摘要： 为提高电动车锂离子动力电池荷电状态（SOC）的估算精度，提出了一种基于六参数RC等效电路的电池模型，采用扩展卡尔曼滤波开展对电池SOC的估算方法研究。在考虑未知干扰和环境噪声的影响下，进一步提出了在扩展卡尔曼滤波的基础上的自适应卡尔曼滤波算法，并开展了对电池SOC的在线估算。仿真结果表明：虽然扩展卡尔曼滤波和自适应卡尔曼滤波都对电池SOC有较好的估算精度，但在未知干扰和噪声的影响下自适应卡尔曼滤波具有更好的鲁棒性。

关键词: 车辆工程, 电动车, 锂离子电池, 荷电状态, 自适应卡尔曼滤波