锂电池荷电状态(SOC)的精确估计一直是电池管理系统的核心任务之一。电流传感器中存在非零均值的电流漂移噪声,这些噪声会造成不可避免的估计误差。为减少电流漂移噪声对估算造成的不利影响,提出了联合扩展卡尔曼滤波法,以Thevenin模型为锂电池等效电路模型,将电流漂移值作为状态变量与电池SOC进行同步预测。实验和仿真结果表明,该方法能有效抑制电流漂移噪声,提高估算精度。(The accurate estimation of the charge state (SOC) of lithium battery has always been one of the core tasks of battery management system. There are nonzero mean current drift noises in current sensors, which cause unavoidable estimation errors. In order to reduce the adverse effect of current drift noise on the estimation, a joint extended Calman filter method is proposed. The Thevenin model is used as the equivalent circuit model of lithium battery, and the current drift value is used as the state variable to predict the battery SOC synchronously. Experimental and simulation results show that the proposed method can effectively suppress current drift noise and improve estimation accuracy.)