基于参数在线辨识和SVD-UKF的锂电池SOC联合估计

Joint Estimation for SOC of Lithium Battery Based on Online Parameter Identification and SVD-UKF

锂电池的荷电状态(SOC)是电池管理系统的核心参数,准确的SOC估计对电动汽车的安全运行至关重要。针对因电池模型参数固定导致锂电池SOC估计精度不高和误差协方差非正定导致传统无迹卡尔曼滤波算法估计SOC失败的问题,提出基于参数在线辨识和SVD-UKF的锂电池SOC联合估计算法。该算法使用变遗忘因子递推最小二乘法实现电池模型参数的在线辨识,通过基于奇异值分解的无迹卡尔曼滤波算法(SVD-UKF)实现电池SOC的估计。在联邦城市运行工况下对联合估计算法进行验证,实验结果表明,联合估计算法可将SOC估计误差控制在1.53%以内,能够有效提高SOC估计的准确性和稳定性。

The state of charge(SOC) of the lithium battery is the key parameter of the battery management system, and the accurate SOC estimation is essential for the safe operation of electric vehicles. Aiming at the problems that the SOC estimation accuracy is not high due to the fixed battery model parameters and the traditional unscented Kalman filter fails to estimate the SOC because of the non-positive definite error covariance, a joint estimation algorithm for the SOC of the lithium battery based on online parameter identification and SVD-UKF is proposed. In this algorithm, the variable forgetting factor recursive least squares is used to realize the online identification of battery model parameters, and the unscented Kalman filter algorithm based on singular value decomposition(SVD-UKF) is used to realize the estimation of the battery SOC. The joint estimation algorithm is verified under the federal urban driving schedule. The experimental results show that the joint estimation algorithm can control the SOC estimation error within 1.53%, and effectively improve the accuracy and stability of SOC estimation.