摘要
锂离子电池是电力系统中不可或缺的重要储能元件,脉冲大倍率工况下运行的锂离子电池具有单次放电时间短、放电循环多、状态变化频繁、非线性极化现象明显等特点。该文以脉冲大倍率工况下锂离子电池模型为研究对象,针对电化学模型和等效电路模型对模型依赖度高、模型参数难以获取以及脉冲大倍率工况下非线性极化现象导致拟合精度不足等问题,提出基于长短期记忆循环神经网络(long short term memory recurrent neural network,LSTM-RNN)以实现准确的锂离子电池建模。该方法利用LSTM-RNN的动态逼近和长时记忆能力,以获取脉冲大倍率工况下锂离子电池性能参数和电池端电压、荷电状态、电流、温度之间的非线性关系。在6种脉冲大倍率放电工况下对磷酸铁锂电池进行建模,实验结果表明,所提出的基于长短期记忆循环神经网络的锂离子电池模型均能够准确表征磷酸铁锂电池工作特性。
Lithium-ion battery is the key energy storage element in power system.This paper studied the problem of the lithium-ion battery simulation modeling under pulsed high rate condition.Under the pulsed high rate condition,the lithium-ion battery has many characteristics,such as short discharging time,more discharge cycles,frequent state change and obvious nonlinear polarization.To deal with these problems,this paper presented a new method to perform accurate modeling of the lithium-ion battery using long short term memory recurrent neural network(LSTM-RNN).The proposed modeling method utilizes dynamic approximation and long-term memory ability of LSTM-RNN to establish nonlinear relationship between the performance parameters,terminal voltage,the state of charge,current and temperature under pulsed high rate condition.The experiment result of LiFePO4 batteries modelling under 6 kinds of pulsed high rate discharge conditions show that the proposed modeling method could accurately characterize the lithium-ion batteries.
作者
李超然
肖飞
樊亚翔
张振宇
杨国润
LI Chaoran;XIAO Fei;FAN Yaxiang;ZHANG Zhengyu;YANG Guorun(National Key Laboratory of Science and Technology on Vessel Integrated Power System(Naval University of Engineering),Wuhan 430033,Hubei Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2020年第9期3031-3041,共11页
Proceedings of the CSEE
基金
国防科技创新特区(18-163-00-TS-004-072-01)。
关键词
锂离子电池
电池模型
脉冲大倍率工况
长短期记忆循环神经网络
lithium-ion battery
battery model
pulsed high rate condition
long short term memory recurrent neural network
