中文摘要：

在充电式混合动力电动汽车（plug-in hybrid electricvehicle，PHEV）和电动汽车（electric vehicle，EV）中，对电池进行精确、可靠的荷电状态估计（state of charge，SOC）非常重要。传统估计方法存在计算量大、估计不精确等缺点，提出一种平方根无迹卡尔曼滤波（square root unscented Kalman filter，SRUKF）算法对SOC进行实时估计及更新。利用无迹变换（unscented transformation，UT）精确估计系统方程的均值和协方差，使估算值达到二阶精度。利用平方根算法保证状态协方差的半正定性，提高数字计算的稳定性。通过实验对比，验证了该算法的有效性。结果表明，该方法可使状态估计值具有较小的误差和快速跟随性，满足了SOC估计的实际需求。

英文摘要：

To estimate the state of charge （SOC） of batteries accurately and reliably is of great importance for plug-in hybrid electric vehicles （PHEV） and electric vehicles （EVs）. However, the traditional methods have drawbacks of heavy computation and inaccurate estimation. In this study, a square root unscented Kalman filter （SRUKF） algorithm was proposed for estimating the SOC of Lithium-Ion batteries. An optimization algorithm was used to update the model＇s state vector during a charge/discharge period. When the estimation of the means and covariance of the state vector was used, the unscented transformation （UT） takes advantages of deterministic sampling. The square root algorithm of the filter improves the numerical stability by ensuring the state covariance, which is always semi-positive definite. The proposed method has been validated experimentally and the results are compared with the unscented Kalman filter. Experimental results have shown that the proposed method has better performance in terms of lower error and shorter convergence time, and can meet the actual requirements.