基于双重卡尔曼滤波器电池荷电状态的估计

States of Charge Estimation of Battery Based on the Dual Kalman Filter

为了有效估计车用蓄电池的荷电状态(SOC),建立了包含迟滞因素和松弛因素的锂电池的精确模型,以自适应无迹卡尔曼滤波器算法为基础,设计了能够实现模型参数和状态同时在线估计的双重卡尔曼滤波器。通过实验和仿真结果的比较表明:该方法能够有效抑制噪声的干扰,快速修正SOC的误差,取得精确的SOC估计值,同时通过时变参数的估计为判断蓄电池的健康状态提供依据。

In order to estimate battery state-of-charge effectively, an accurate battery model was built which contained the hysteresis effect and relaxation effect. Then, on the basis of adaptive unscented Kalman filter, the dual Kalman filter was designed which could estimate the parameter and state on- line at the same time. Finally, the comparison results between test and simulation show that this method can effectively suppress noise, reduce the SOC estimate error and achieve the precise SOC es- timated value. And, the estimated value of parameter can provide evidence for the judging of state-of- health of battery.