摘要
为提高锂离子电池荷电状态(state of charge,SOC)的估计精度并准确估计健康状态(state of health,SOH),以二阶RC等效电路模型为研究对象,基于Sage-Husa自适应滤波的思想,对传统的平方根无迹卡尔曼滤波(square-root unscented Kalman filter,SRUKF)进行改进,提出一种自适应SRUKF(adaptive square-root unscented Kalman filter,ASRUKF)算法,该算法通过对状态方差阵和噪声方差阵平方根的递推估算,确保了状态和噪声方差阵的对称性和非负定性。验证结果显示,相比于SRUKF算法,ASRUKF算法能够得到精度更高的SOC估计值,并在FUDS工况下将最大SOC估计误差降低4%。针对电池欧姆内阻和容量参数随着电池的老化而变化的现象,对内阻和容量进行实时在线估计,在此基础上完成对SOH参数的预测。验证结果表明,联合估计算法对电池的欧姆电阻和容量有一个较好的估计,进一步提升了电池状态的估计精度。
In order to improve the accuracy of the state of charge(SOC)estimation of the Lithium-ion battery and predict the state of health(SOH)precisely, a second-order RC equivalent circuit model was introduced. On the basis of Sage-Husa adaptive filter, the traditional square-root unscented Kalman filter(SRUKF) was improved and an adaptive SRUKF(ASRUKF) method was proposed. The method guaranteed the symmetry and non-negativity of the variance matrices of state as well as noise by estimating the square root of state variance matrix and noise variance matrix recursively. The validation results demonstrate that, compared with SRUKF, ASRUKF method can obtain more accurate SOC estimation value and reduce the maximum SOC error of FUDS test by 4%. Considering the changes of internal resistance and capacity through battery aging process, real-time online estimation of internal resistance and capacity was realized simultaneously when estimating SOC, on basis of which SOH prediction was completed. The simulation results illustrate that the joint algorithm can estimate the internal resistance and capacity of battery accurately, so the accuracy of state estimation is further improved.
作者
程泽
杨磊
孙幸勉
CHENG Ze;YANG Lei;SUN Xingmian(School of Electrical and Information Engineering, Tianjin University, Nankai District, Tianjin 300072, Chin)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2018年第8期2384-2393,共10页
Proceedings of the CSEE
基金
国家自然科学基金项目(61374122)~~