基于无迹Kalman滤波算法的电池内部温度估计

Estimation of battery internal temperature based on unscented Kalman filter algorithm

为实时监测车用锂离子动力电池内部温度,提高电池性能,提出了一种非线性无迹Kalman滤波(UKF)估计算法。对某一2.6 Ah三元单体锂离子电池,建立等效可变参数热模型;用状态方程分析法,建立电池内部外部温度的关联并离散化;用递推最小二乘法(RLS)辨识热模型中时间、表面温度、环境温度、输入电流4种热参数,实时更新系统状态与观测方程的参数矩阵,结合UKF算法,实现电池内部温度估计。通过Matlab搭建仿真模型,用混合动力脉冲能力特性(HPPC)、动态应力测试(DST)以及恒流3种工况,来验证算法精度。结果表明:对于这3种工况,该UKF算法均可在1℃内估计电池内部温度。

A nonlinear unscented Kalman filter(UKF)estimation algorithm was proposed for real-time monitoring lithium-ion power battery’s internal temperature to improve the vehicle power battery’s performances.An equivalent variable parameter thermal model was established for a 2.6 Ah ternary lithium ion battery;The correlation between the internal and external temperature of the battery is established and discretized by the state equation analysis method;The recursive least square(RLS)method is used to identify the four thermal parameters of time,surface temperature,ambient temperature and input current in the thermal model,update the parameter matrix of system state and observation equation in real time,and realize the estimation of battery internal temperature combined with UKF algorithm.The simulation model is built by MATLAB,and the accuracy of the algorithm is verified by three working conditions:hybrid impulse capability characteristic(HPPC),dynamic stress test(DST)and constant current.The results show that the UKF algorithm can estimate the internal temperature of the battery within 1℃for these three working conditions.