摘要 针对平方根容积卡尔曼滤波(SRCKF)估算SOC时需要准确获得系统状态及测量噪声协方差这一缺陷,将基于电池模型输出电压残差序列的协方差匹配思想引入平方根容积卡尔曼滤波,提出了自适应平方根容积卡尔曼滤波算法(ASRCKF)。以18650型锂电池为实验对象,建立了戴维南等效电路模型,采用递推最小二乘法辨识电池模型参数,最后,利用UDDS电池实验数据对ASRCFK算法进行了仿真。实验结果表明,传统的SRCKF算法估算SOC产生的均方根误差为3.41%;而提出的ASRCKF算法估算SOC产生的均方根误差仅为0.97%,与传统算法相比具有更高的精度,对噪声的适应能力更强。 Aiming at the disadvantage of accurate system state and measure noise covariance needed by the square root cubature Kalman filter(SRCKF),the covariance matching idea based on the battery model output voltage residual sequence is introduced into the SRCKF,the adaptive square root cubature Kalman filter algorithm is proposed.Taking the 18650 lithium battery as the experimental object,the Thevenin equivalent circuit model is established.The battery model parameters are identified by recursive least squares method.Finally,the ASRCKF algorithm is simulated by UDDS battery experimental data.The results show that the root mean square error(rmse)of traditional SRCKF is 3.41%,the rmse of ASRCKF is only 0.97%,which has higher precision and more adaptable than traditional algorithm.
机构地区 湖南大学机械与运载工程学院
出处 《机械设计与制造》 北大核心 2021年第4期53-55,61,共4页 Machinery Design & Manufacture
关键词 平方根容积卡尔曼滤波 噪声协方差 噪声自适应估计 Square Root Cubature Kalman Filter Noise Covariance Noise Adaptive Estimator
