电池SOC的自适应平方根无极卡尔曼滤波估计算法  

Adaptive square root unscented Kalman filter for SOC estimation of battery

作　　者：胡志坤[1] 刘斌[1] 林勇 王文祥[1] 彭小奇[1] 

机构地区：[1]中南大学物理与电子学院,湖南长沙410083 [2]湖南科力远高技术股份有限公司,湖南长沙410083

出　　处：《电机与控制学报》2014年第4期111-116,共6页Electric Machines and Control

基　　金：国家自然科学基金(61273159;61105080);湖南省自然科学基金委员会与株洲市政府自然科学联合基金资助(13JJ9038)

摘　　要：针对利用平方根无极卡尔曼算法估算电池SOC时,因噪声协方差为常量带来的误差,在平方根无极卡尔曼滤波(SR-UKF)算法的基础,改进了算法,把每次测量的输出值残差的协方差作为噪声的协方差,得到自适应平方根无极卡尔曼滤波算法,使得噪声协方差随时间的更新而更新,解决了噪声协方差为常量带来的误差。实验表明,利用自适应平方根无极卡尔曼滤波算法对在常温下电池放电过程的SOC估计,精确度在总体上得到了提高,在电池工作区间0.2≤YSOC≤0.9内估计误差在1.5%以内。自适应平方根无极卡尔曼滤波算法对电池常温放电过程的SOC估计能满足电动汽车电池SOC估计的实际要求。As the noise covariance is a constant that can produce error when using the square root un-scented Kalman Filter（SR-UKF） algorithm for SOC estimation, the study improves the SR-UKF algo-rithm. Each measurement error covariance of output was put as the covariance of the noise, updating with time. As a result, the adaptive SR-UKF algorithm was obtained and the disadvantages of the noise covari-ance unchanged in SR-UKF algorithm was overcome. The experimental results show that the adaptive SR-UKF algorithm makes SOC estimation precision improved in general and error less than 1. 5% in the bat-tery work interval 0. 2≤YSOC≤0. 9. Using the adaptive SR-UKF algorithm to estimate state of charge of battery can meet the actual requirements.

关 键 词：电池 荷电状态估计 最小二乘法 

分 类 号：TN0[电子电信—物理电子学]