基于BASOA-IEKF融合算法的电池SOC估计

Battery SOC estimation based on BASOA-IEKF fusion algorithm

标准扩展卡尔曼滤波是常用的电池荷电状态(state of charge,SOC)估计算法,针对其系统线性化误差和噪声矩阵依赖问题会导致电池SOC估计的准确性下降的情况,提出了一种基于边界自适应人群搜索(boundary adaptive seeker optimization algorithm,BASOA)-迭代扩展卡尔曼滤波(iterated extended Kalman filter,IEKF)的融合滤波算法,通过状态估计值多次迭代和系统噪声矩阵智能寻优来提升SOC估计效果.结果表明:在静态工况下BASOA-IEKF算法的SOC最大估算绝对误差为3.74%,混合功率脉冲特性(hybrid pulse power characterization,HPPC)工况下SOC估算绝对误差小于3.00%,城市道路循环(urban dynamometer driving schedule,UDDS)工况下估算绝对误差小于2.50%,相较于单一IEKF算法,BASOA-IEKF算法的SOC估计精度更高,SOC误差曲线在收敛后波动更小,表现更稳定,全局鲁棒性更优.

Standard extended Kalman filter(EKF)is currently and commonly used algorithm for estimating the state of charge(SOC)of batteries.However,the accuracy of battery SOC estimation is compromised due to the linearization errors and the dependence on noise matrices in the system.To address the issue,a fusion filtering algorithm based on boundary adaptive seeker optimization algorithm and iterated extended Kalman filter(BASOA-IEKF)was proposed to improve the SOC estimation accuracy by iteratively updating the state estimates and intelligently optimizing the system noise matrix.The static and dynamic simulation results show that under static conditions,the maximum estimation error of SOC is 3.74%.Under hybrid pulse power characterization(HPPC)conditions,the estimation error is less than 3.00%,while under urban dynamometer driving schedule(UDDS)conditions,it is less than 2.50%.Compared to the single IEKF algorithm,the BASOA-IEKF algorithm achieves higher accuracy in SOC estimation with smaller fluctuations in the SOC error curve after convergence and demonstrates better stability and global robustness.