中文摘要：

为解决锂离子电池欧姆内阻随温度变化特性对电池状态估计精度的影响，提出了一种考虑内阻时变特性的两步无迹H∞滤波锂电池状态估计方法；该方法首先分析了锂离子电池常用物理模型，并在此基础上对锂离子电池和内阻抗进行分开建模，在电池Thevenin模型的基础上构建内阻抗预测模型，实时修正模型参量；接着，在扩展H∞滤波框架内实现电池状态的滤波估计，并采用无迹变换取代泰勒级数近似截断处理，有效降低了截断近似误差和测量噪声对估计精度的扰动；最后，在实验室环境下对电池进行充放电实验，分别针对降温和升温情况下的内阻值及电池端电压的估计进行了详细的实验分析；实验结果表明，同传统的估计方法相比，该算法明显提升了不同环境下电池荷电状态的估计精度，具有较好的应用价值。

英文摘要：

In order to solve the effect of the lithium ion battery ohm resistance changes with temperature to the battery state estimation precision, this paper proposes a two step unscented h--infinity filtering lithium battery state estimation method with considering internal re sistance time--varying characteristics. First of all, the paper models the lithium battery and internal impedance separately, and correct model parameters real--time based on the battery internal resistance prediction model in Thevenin model. Then, the battery state estimation is achieved within the framework of the extended h--infinity filtering filter. The disturbance of measurement noise on estimation accuracy and Taylor series approximation truncation errors were reduced by the extension of unscented transform into h--infinity filtering framework. Fi- nally, the battery charge and discharge experiments in the laboratory environment was conducted in detail. The estimation of internal resist ance and voltage of the battery were analyzed in cooling and heating conditions respectively. Experimental results show that, compared with the traditional estimation method, the algorithm significantly improved the battery charged state estimation precision under different environment, has good application value.