电工技术学报  2022, Vol. 37 Issue (23): 6065-6073    DOI: 10.19595/j.cnki.1000-6753.tces.211366
新型储能系统应用关键技术专题(特约主编:李建林 教授 梅生伟 教授 李军徽 教授) |
基于灰色关联度分析-长短期记忆神经网络的锂离子电池健康状态估计
周才杰, 汪玉洁, 李凯铨, 陈宗海
中国科学技术大学自动化系 合肥 230027
State of Health Estimation for Lithium-Ion Battery Based on Gray Correlation Analysis and Long Short-Term Memory Neural Network
Zhou Caijie, Wang Yujie, Li Kaiquan, Chen Zonghai
Department of Automation University of Science and Technology of China Hefei 230027 China
全文: PDF (1784 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 电池的健康状态是电池健康管理的核心,准确的锂离子电池健康状态估计对保证电池安全、可靠、长寿命运行具有重要意义。为此,该文提出了一种基于增量容量曲线和灰色关联度分析(GRA)以及长短期记忆(LSTM)神经网络的锂离子电池健康状态估计方法。该方法通过分析电池在老化过程中的充电增量容量曲线变化模式,提取电池老化特征。为了降低计算复杂度,引入灰色关联度分析法进行特征分析与筛选,并将其作为长短时间记忆神经网络的输入,进行网络预训练进而估计电池的健康状态。最后,利用三种不同工况的电池加速老化测试数据集对所提出的健康状态估计方法进行了验证。实验结果表明,所提出的方法表现出优秀的电池健康状态估计性能,并在不同工况以及不同训练循环周期数条件下表现出良好的鲁棒性。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
周才杰
汪玉洁
李凯铨
陈宗海
关键词 锂离子电池健康状态估计增量容量曲线灰色关联度分析长短期记忆神经网络    
Abstract:The state of health is the core of battery health management. An accurate estimation of the state of health is of great significance to ensure the safe, reliable, and long-life operation of lithium-ion batteries. To this end, this paper proposes a method for estimating the state of health of lithium-ion batteries based on incremental capacity curves and gray correlation analysis long short-term memory neural networks. This method extracts several different aging characteristics by analyzing the attenuation mode of battery charging incremental capacity curves during the aging process. In order to reduce the computational complexity, the gray correlation analysis method is introduced for features analysis and screening. Then, the extracted aging features are used as the input to train the long short-term memory network and estimate the battery health status. Finally, accelerated battery aging tests based on three different working conditions are conducted to verify the proposed method. The experimental results show that the proposed method exhibits excellent performance in estimating the state of health of the battery, and it shows good robustness under different working conditions and different number of training cycles.
Key wordsLithium-ion battery    state of health estimation    incremental capacity curve    gray correlation analysis (GRA)    long short-term memory (LSTM) neural network   
收稿日期: 2021-09-03     
PACS: TM911  
基金资助:国家自然科学基金项目(61803359)、安徽省高校协同创新项目(GXXT-2019-002)和中国科学技术大学“统筹推进世界一流大学和一流学科建设专项资金”(YD2350002002)资助
通讯作者: 汪玉洁 男,1990年生,博士,副研究员,研究方向为节能与新能源汽车技术,复杂系统建模、仿真与控制,燃料电池系统管理与优化控制,人工智能方法在能源系统中的应用等。E-mail: wangyujie@ustc.edu.cn   
作者简介: 周才杰 男,1997年生,硕士研究生,研究方向为锂电池建模、状态估计与充电优化控制。E-mail:cjzhou19@mail.ustc.edu.cn
引用本文:   
周才杰, 汪玉洁, 李凯铨, 陈宗海. 基于灰色关联度分析-长短期记忆神经网络的锂离子电池健康状态估计[J]. 电工技术学报, 2022, 37(23): 6065-6073. Zhou Caijie, Wang Yujie, Li Kaiquan, Chen Zonghai. State of Health Estimation for Lithium-Ion Battery Based on Gray Correlation Analysis and Long Short-Term Memory Neural Network. Transactions of China Electrotechnical Society, 2022, 37(23): 6065-6073.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.211366          https://dgjsxb.ces-transaction.com/CN/Y2022/V37/I23/6065