摘要
针对目前梯次利用锂离子电池受老化的影响,导致在采用传统带遗忘因子的最小二乘法进行等效模型在线参数辨识后,会出现模型参数过饱和、无法跟踪时变参数的问题。本工作提出一种可变遗忘因子的最小二乘法,通过不断地对遗忘因子进行矫正更新,从而更好地跟踪梯次利用电池老化特性。以锂电池一阶RC等效电路模型为对象,搭建测试平台进行充放电试验,并与传统带遗忘因子的最小二乘法的参数辨识结果进行对比。实验结果表明,该方法能够快速收敛及动态跟踪,同时模型端电压参数平均误差减小到25 mV以内,所提出的方法在DST和储能系统典型工况下运行时,其对应的参数辨识精度提高了38.33%,证明该方法具有较高的准确性。
The current challenges of modeling aging lithium-ion batteries include oversaturated model parameters and time-varying parameters,which cannot be evaluated with an online parameter identification of the model using the traditional least squares method with a fixed forgetting factor.This paper proposes a least squares method with a variable forgetting factor,which continuously updates the forgetting factor to better track the run-time utilization of battery aging characteristics.Using the first-order RC equivalent circuit model of the lithium battery as a model,a test platform was established for charge and discharge experiments,and the results were compared with the traditional least squares method with a fixed forgetting factor.The experimental results indicated that the proposed method can quickly converge and dynamically track battery aging.The average absolute error of the voltage parameters at the model terminal was found to be less than 25 mV.When the proposed method was run under a dynamic stress test with the typical working conditions of an energy storage system,the corresponding parameter identification accuracy was improved by 38.33%,indicating that the proposed method is highly accurate.
作者
杜帮华
张宇
吴铁洲
何衍林
李子龙
DU Banghua;ZHANG Yu;WU Tiezhou;HE Yanlin;LI Zilong(Hubei University of Technology,Hubei Provincial Key Laboratory of Solar Energy Efficient Utilization and Energy Storage Operation Control,Wuhan 430068,Hubei,China;State Grid New Source Hubei Bailianhe Pumped Storage Co.Ltd.,Huanggang 438600,Hubei,China)
出处
《储能科学与技术》
CAS
CSCD
北大核心
2021年第1期342-348,共7页
Energy Storage Science and Technology
基金
湖北省科技厅重大专项项目(2018AAA056)。
关键词
梯次利用
锂离子电池
退役电池
等效电路模型
参数辨识
遗忘因子
最小二乘法
SOC估算
echelon use
Li-ion battery
decommissioned battery
equivalent circuit model
parameter identification
forgetting factor
least square method
SOC estimation
