摘要
锂离子电池荷电状态(SOC)的准确估计,直接影响电动汽车的安全运行,也是电池管理系统(BMS)重要的参数之一。然而在SOC估计过程中,会受到如测量设备的精度、荷载等因素的干扰。为了提高SOC的估计精度,本工作提出了一种新的基于最小二乘支持向量机(LSSVM)机器学习的锂离子电池SOC估计模型。将测量的锂离子电池的电流、电压和温度作为模型的输入向量,SOC作为模型的输出向量,来训练LSSSVM模型。不同的是,为提高SOC的估计精度,将上一时刻估计的SOC值作为反馈向量,加入到输入向量中,用来估计当前时刻的SOC。通过实验工况采集的数据,并与传统的LSSVM、支持向量机(SVM)、神经网络(NN)比较,实验结果表明,该模型可以提高SOC的估计精度,估计误差可以控制在2%以内,验证了该模型的有效性。
To ensure the safe operation of electric vehicles, the accurate estimation of the state of charge(SOC) of the lithium-ion battery installed in the vehicle is required. SOC is also an important parameter of battery management systems. However, SOC estimation is influenced by the accuracy of the measuring equipment, applied load, and various other factors. Accordingly, to improve the accuracy of SOC estimation, this study proposes a new SOC estimation model for lithium-ion batteries based on least-squares support vector machine(LSSVM) machine learning. The LSSVM model is trained on the measured current, voltage, and temperature of the lithium ion battery, which are input in vector form,and outputs the SOC vector. To improve the accuracy of SOC estimation, the SOC value estimated at the previous time is added as a feedback vector to the input vector before estimating the SOC at the current time. In an experimental evaluation, the proposed method achieved higher SOC estimation accuracy than the LSSVM, SVM, and neural network models. Moreover, the estimation error was controlled within 1%, establishing the validity of the model.
作者
李嘉波
李忠玉
焦生杰
叶敏
徐信芯
LI Jiabo;LI Zhongyu;JIAO Shengjie;YE Min;XU Xinxin(Highway Maintenance Equipment National Engineering Laboratory,Changan University,Xi'an 710064,Shaanxi,China;Henan Gaoyuan Highway Maintenance Technology Co.Ltd.,Xinxiang 453000,Henan,China)
出处
《储能科学与技术》
CAS
CSCD
2020年第3期951-957,共7页
Energy Storage Science and Technology
基金
国家自然科学基金青年项目(51805041)
河南省交通运输厅科技计划项目(2019J3)。
作者简介
第一作者及联系人:李嘉波(1992-),男,博士研究生,研究方向为新能源,E-mail:431991454@qq.com。
