基于FFRLS算法的电池模型

Battery model based on FFRLS algorithm

建立二阶Thevenin等效电路电池模型。以最小二乘法(recursive least squares,RLS)为基础,对电池模型进行参数辨识,模型迭代过程中电路端电压的估计误差随数据的微小变化而陡然增大。引入遗忘因子,采用遗忘因子最小二乘法(forgetting factor recursive least squares,FFRLS)进行参数辨识,以削弱迭代中旧数据对参数的影响,增强新数据对参数的影响,结果使收敛速度提高、误差波动减小。经验证,运用该方法进行参数辨识的电池模型具有良好的精度。

The battery model of a second-order Thevenin equivalent circuit is established in this paper.Based on the RLS(recursive least squares)method,the battery model is identified by parameters.However,the error of the estimated terminal voltage during the iteration of the model increases sharply with small changes in the data.The forgetting factor is introduced,and the FFRLS(forgetting factor recursive least squares)is used to identify the parameters,so as to weaken the influence of the old data on the parameters in the iteration and enhance the influence of the new data on the parameters.The result is that the convergence speed is increased and the error fluctuation is reduced.It has been proved that the battery model using this method to identify parameters has good precision.