基于内阻法修正的蓄电池卡尔曼滤波SOC估算

SOC estimation of battery Calman filter based on internal resistance correction

摘要:

针对蓄电池卡尔曼滤波算法荷电状态SOC（State of Charge）初始值的估计误差较大可能导致前期收敛性较差的问题，通过分析蓄电池放电实验数据，运用灰色关联模型计算电池内阻、电压和电流参数关于SOC的关联度值，将关联度最高的内阻参数作为初始SOC估计值的自变量。然后将初始SOC估计值代入由二阶等效电路模型构建的扩展卡尔曼滤波算法中，进行SOC估计。最后利用电测试平台验证SOC估计准确性，并与电压参数作为初始SOC估计值自变量的方法进行对比。实验结果表明，相对于电压法的初始估计值，内阻法的初始估计值更接近真实值，将其作为卡尔曼滤波算法的起始值，更能提高初期荷电状态估算精度。

Abstract:

In view of the high estimation error of the initial value of the charge state SOC (State of Charge) of the battery Calman filtering algorithm may lead to the poor early convergence, the correlation degree of the battery resistance, voltage and current parameters about the SOC is calculated by analyzing the experimental data of battery discharge, and the correlation degree is the most. The high internal resistance parameter is the independent variable of the initial SOC estimate. Then the initial SOC estimate is replaced by the extended Calman filtering algorithm constructed by the two order equivalent circuit model, and SOC estimation is carried out. Finally, the accuracy of SOC estimation is verified by the electrical test platform, and the method is compared with the voltage parameter as the initial SOC estimation variable. The experimental results show that the initial estimation value of the internal resistance method is closer to the real value than the initial estimation value of the voltage method. As the starting value of the Calman filtering algorithm, the estimation accuracy of the initial charge state can be improved more.