基于分数阶模型与IEKF的锂离子电池SOC估计研究

Research on SOC estimation of lithium ion battery based on fractional order model and IEKF

针对整数阶模型及EKF估算SOC过程中电池极化特性描述不足以及线性化误差累计的问题,提出基于分数阶微分的2阶RC等效电路模型,利用多倍率放电实验测得电池动态容量曲线,通过电压实测值和预测值的均方根误差自适应判断IEKF算法迭代步数,进行SOC估算。试验和仿真结果表明,基于分数阶模型的IEKF,在UDDS工况下SOC最大估计误差为1.24%,平均误差为0.85%,相比于整数阶EKF算法,分数阶IEKF算法在全局鲁棒性以及SOC估算精度上具有一定优势。

To address the problems of inadequate description of battery polarization characteristics and accumulated linearization errors in the process of SOC estimation by integer-order models and EKF,a second-order RC equivalent circuit model based on fractionalorder differentiation is proposed,and the dynamic capacity curve of the battery is measured by using multiplier discharge experiments,and the number of iteration steps of the IEKF algorithm is judged adaptively by the root-mean-square error of the measured and predicted voltage values for SOC estimation.The experimental and simulation results show that the IEKF,based on the fractional-order model,has a maximum SOC estimation error of 1.24%and an average error of 0.85%under UDDS conditions.Compared to the integerorder EKF algorithm,the fractional-order IEKF algorithm has certain advantages in terms of global robustness and SOC estimation accuracy.