摘要
基于UTS分容柜所测得的实验数据,建立了18650锂电池的三阶Thevenin模型。将扩展卡尔曼滤波算法(Extened Kalman Filter,EKF)作为粒子滤波算法(Particle Filter,PF)的重要密度函数形成了扩展卡尔曼粒子滤波算法(Extened Kalman Particle Filter,EKPF)。对于EKPF算法在重采样过程中存在的样本退化、多样性丧失的问题,提出了一种通过权值排序的优胜劣汰粒子选择算法。采用通过该方法改进的EKPF算法对所建立的三阶Thevenin模型进行电池荷电状态(State of Charge,SOC)估计,实验结果表明,改进EKPF算法的SOC估计精度优于EKF算法和PF算法的SOC估计精度。
The three order Thevenin model of 18650 Lithium-Ion battery is established based on the experimental data of UTS divided capacity tester. The extended kalman filtering(EKF) algorithm is adopted as the important density function of particle filter(PF) algorithm, and the extended Kalman particle filter(EKPF) algorithm is formed. The sample degradation and lack of diversity in the re-sampling stage of EKPF algorithm is optimized by an improved re-sampling algorithm which based on a weight sorting and survival of the fittest particles. The improved EKPF algorithm is applied to estimate the State of Charge(SOC) of the three order Thevenin model of batteries. The experimental results show that the SOC estimation accuracy of the improved EKPF algorithm is better than that of the EKF algorithm and the PF algorithm.
作者
夏飞
王志成
郝硕涛
彭道刚
余贝丽
黄毅敏
Xia Fei;Wang Zhicheng;Hao Shuotao;Peng Daogang;Yu Beili;Huang Yimin(College of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Zhejiang Electric Transmission and Transformer Co.,Ltd.,Hangzhou 310020,China;Fangshan Power Supply Company,Beijing Power Company,Beijing 102401,China;State Power Investment Group,Wuhu Power Generation Co.,Ltd,Anhui,Wuhu 241009,China)
出处
《系统仿真学报》
CAS
CSCD
北大核心
2020年第1期44-53,共10页
Journal of System Simulation
