文章摘要
卫志农,原康康,成乐祥,等.基于多新息最小二乘算法的锂电池参数辨识[J].电力系统自动化. DOI: 10.7500/AEPS20180814005.
WEI Zhinong,YUAN Kangkang,CHENG Lexiang, et al.Multi-innovation Least Squares Algorithm Based Parameter Identification of Lithium-ion Battery[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180814005.
基于多新息最小二乘算法的锂电池参数辨识
Multi-innovation Least Squares Algorithm Based Parameter Identification of Lithium-ion Battery
DOI:10.7500/AEPS20180814005
关键词: 锂离子电池  参数辨识;最小二乘  多新息
KeyWords: Lithium-ion battery  Parameter identification  Least square  Multi-innovation
上网日期:2019-06-11
基金项目:国家自然科学基金
作者单位E-mail
卫志农 河海大学能源与电气学院 wzn_nj@263.net 
原康康 河海大学能源与电气学院 503478066@qq.com 
成乐祥 国网江苏省电力有限公司南京供电分公司 clxxt123@126.com 
王春宁 国网江苏省电力有限公司南京供电分公司 wangchunning@sina.com 
许洪华 国网江苏省电力有限公司南京供电分公司 827054516@qq.com 
孙国强 河海大学能源与电气学院 hhusunguoqiang@163.com 
臧海祥 河海大学能源与电气学院 zanghaixiang@hhu.edu.cn 
摘要:
      为了保证电池管理系统的安全可靠运行,需要对锂离子电池模型参数进行准确地辨识。本文以磷酸铁锂电池为研究对象,建立RC等效电路模型,并基于该模型进行锂离子电池模型参数辨识。锂离子电池模型参数受外部因素影响较大并且参数辨识结果受在线信息采集限制的影响,本文采用多新息最小二乘辨识算法进行锂离子电池模型参数在线辨识。本文通过进行三种不同的充放电实验来采集数据,并根据实验数据在不同初值下进行参数辨识,通过比较由辨识结果估计出的端口电压值与实际值的误差来描述辨识结果的准确度。实验结果表明,多新息最小二乘辨识算法具有快速收敛性与高精确性。
Abstract:
      In order to ensure the safe and reliable operation of the battery management system, it is necessary to accurately identify the parameters of the lithium-ion battery model. In this paper, a RC equivalent circuit model for lithium iron phosphate battery is established, and the parameters of lithium ion battery model are identified based on this model. The parameters of lithium-ion battery model are greatly influenced by external factors and the results of parameter identification are limited by online information acquisition. In this paper, multi-innovation least squares identification algorithm is used to identify the parameters of lithium-ion battery model online. In this paper, three different charging and discharging experiments are carried out to collect data, and the parameters are identified according to the experimental data under different initial values. The accuracy of the identification results is described by comparing the errors between the estimated port voltage and the actual value. The experimental results show that the multi-innovation least squares identification algorithm has fast convergence and high accuracy.
查看全文   查看/发表评论  下载PDF阅读器