基于动态遗忘因子最小二乘与EKF的电池SOC估计

电池荷电状态SOC（State Of Charge）作为电池管理系统中尤为重要的一部分，其准确估计成为锂离子电池研究的重点。为了提高动态工况下的SOC估计精度，对锂离子电池等效模型进行分析，基于AIC（赤池信息）准则确定二阶RC电路为等效电路模型，使用递推最小二乘算法对模型参数进行在线辨识，为提高辨识精度，提出了改进带动态遗忘因子递推最小二乘算法，对算法加入遗忘因子，通过电压结果误差实时动态调整算法遗忘因子取值。将递推最小二乘算法和含动态遗忘因子最小二乘算法分别与扩展卡尔曼滤波（EKF）算法进行SOC联合估计，并对比其预测效果，结果表明含有动态遗忘因子最小二乘与EKF联合估计模型具有更高的精度和鲁棒性。

Battery SOC Estimation based on Dynamic Forgetting Factor Least Squares and EKF

As a particularly important part of the battery management system, the accurate estimation of the battery SOC (State Of Charge) has become the focus of lithium-ion battery research. In order to improve the SOC estimation accuracy under dynamic conditions, the equivalent model of lithium-ion batteries is analyzed, the second-order RC circuit is determined as the equivalent circuit model based on the AIC (Akaike Information) criterion, and the recursive least squares algorithm was used to identify the model parameters online, and in order to improve the identification accuracy, an improved least squares algorithm with dynamic forgetting factor was proposed, the forgetting factor is added to the recursive least squares algorithm, and the forgetting factor of the algorithm is dynamically adjusted in real time through the voltage result error. The recursive Least Squares algorithm and the Least Squares algorithm with dynamic forgetting factor are combined with the extended Kalman filtering (EKF) algorithm for SOC joint estimation respectively, compared the prediction results, the results showed that the joint estimation model containing the least squares with dynamic forgetting factor and EKF has higher accuracy and robustness.