摘要
电池荷电状态(State of Charge,SOC)作为电池状态重要的评价指标,很难通过直接测量获得。传统方法通过物理指标侧面估算SOC的大小,存在一定的局限性。因此,研究拓展卡尔曼滤波法(Extended Kalman Filter,EKF)、BP神经网络(BP Neural Network)以及模糊控制方法等估计SOC值的方法,分析各种方法的实现过程及优缺点,提出了一种BP和EKF相结合的方法用于SOC值的估计。该方法可提高EKF的收敛性,并增加了SOC值估计的准确度。
The State of Charge(SOC)of the battery is an important evaluation indicator of the battery state,which is difficult to obtain through direct measurement.Traditional methods estimate the size of SOC through physical indicators.Based on this,the methods of estimating SOC value such as EKF extended Kalman filter method,BP neural network and fuzzy control method are studied,the realization process and advantages and disadvantages of various methods are analyzed,and a BP and EKF combination is proposed.The method is used to estimate the SOC value.This method can improve the convergence of EKF and increase the accuracy of SOC value estimation.
作者
朱立宗
黄煜
ZHU Lizong;HUANG Yu(School of Automobile and Information Engineering,Guangxi Vocational and Technical College of Ecological Engineering,Liuzhou 545004)
出处
《现代制造技术与装备》
2021年第2期166-167,共2页
Modern Manufacturing Technology and Equipment
基金
2019年广西高校中青年教师基础能力提升项目“电动汽车锂离子电池SOC估算研究”(2019KY1426)。
关键词
电池荷电状态
模糊控制
神经网络
battery state of charge
fuzzy control
neural networks
