基于分层对齐迁移学习的锂离子电池容量估计

doi:10.19799/j.cnki.2095-4239.2022.0706

储能科学与技术 ›› 2023, Vol. 12 ›› Issue (4): 1223-1233.doi: 10.19799/j.cnki.2095-4239.2022.0706

基于分层对齐迁移学习的锂离子电池容量估计

翟智¹^,²(), 王福金¹^,², 邸一¹^,², 马珮羽¹^,², 赵志斌¹^,²(), 陈雪峰¹^,²

^1.西安交通大学机械工程学院，陕西西安 710049
^2.装备运行安全保障与智能监控国家地方;联合工程研究中心，陕西西安 712046

收稿日期:2022-11-30 修回日期:2022-12-30 出版日期:2023-04-05 发布日期:2023-05-08
通讯作者: 赵志斌 E-mail:zhaizhi@xjtu.edu.cn;zhaozhibin@xjtu.edu.cn
作者简介:翟智（1985—），女，博士，副研究员，主要研究方向为航天器电源系统故障诊断、航天动力系统健康管理，E-mail：zhaizhi@xjtu.edu.cn；
基金资助:
国家自然科学基金(52105116);中国博士后科学基金项目(2021TQ0263)

Hierarchical alignment transfer learning for lithium-ion battery capacity estimation

Zhi ZHAI¹^,²(), Fujin WANG¹^,², Yi DI¹^,², Peiyu MA¹^,², Zhibin ZHAO¹^,²(), Xuefeng CHEN¹^,²

^1.School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an 710049, Shaanxi, China
^2.National and Local Joint Engineering Research Center for Equipment Operation Safety Assurance and Intelligent Monitoring, Xi'an 712046, Shaanxi, China

Received:2022-11-30 Revised:2022-12-30 Online:2023-04-05 Published:2023-05-08
Contact: Zhibin ZHAO E-mail:zhaizhi@xjtu.edu.cn;zhaozhibin@xjtu.edu.cn

摘要/Abstract

摘要：

精准的容量估计对锂离子电池健康管理和预测性维护具有重要意义。近年来，数据驱动的方法被广泛应用于锂离子电池容量估计，然而现有的数据驱动方法大多假设训练和测试数据服从相同分布，当此假设不满足时，模型的预测精度快速下降。现有的基于迁移学习的锂离子电池容量估计方法旨在对齐源域和目标域的整体分布，而忽略了不同层内的特征的可迁移性。针对以上问题，研究了深度迁移学习方法不同层之间的特征可迁移属性，提出了基于分层对齐迁移学习(hierarchical alignment transfer learning， HATL)的锂离子电池容量估计方法。首先，构建了一个基于卷积神经网络的特征提取器，考虑不同层特征的可迁移性，对不同层特征施加最大均值差异约束和通道注意力一致性约束，使得特征提取器从源域和目标域提取到的特征相似且模型更加关注域不变特征；然后，特征经过一个预测器得到容量估计值。在公开的锂电池数据集上进行充分验证，并与其他方法进行对比，结果表明，本文所提的HATL方法具有更高的估计精度，明显优于其他方法。证明了迁移学习方法在跨工况容量估计任务中的有效性和优越性。

关键词: 锂离子电池, 容量估计, 分层对齐迁移学习, 最大均值差异, 通道注意力一致性

Abstract:

Accurate estimation of capacity plays an important role in the health management and predictive maintenance of lithium-ion batteries. In recent years, data-driven methods have been widely used in the capacity estimation of lithium-ion battery. However, most of these methods assume that the training and test data obey the same distribution, resulting in a rapid decline in accuracy when the test conditions change. The existing transfer learning methods for lithium-ion battery capacity estimation aim to align the global distribution of source and target domains, while ignoring the transferability of features within different layers. Thus, this study proposes a hierarchical alignment transfer learning method for lithium-ion battery capacity estimation and examines the feature transferability among different layers of the deep transfer learning. First, a feature extractor based on a convolutional neural network was designed. Considering the feature transferability within different layers, the maximum mean discrepancy constraint and channel attention consistency constraint were imposed at different layers of the feature extractor. Thus, the features extracted from the source and target domain are similar, and the feature extractor focuses more on domain-invariant features. A predictor then obtains a capacity estimation value. The experimental results were validated on a public lithium-ion battery dataset and compared with other methods. Our findings show that the proposed method has higher estimation accuracy and is significantly better than other methods. In addition, the findings demonstrated the effectiveness and superiority of the transfer learning method in cross-domain capacity estimation.

Key words: lithium-ion battery, capacity estimation, hierarchical alignment transfer learning, maximum mean discrepancy, channel attention consistency

中图分类号:

TM 912

翟智, 王福金, 邸一, 马珮羽, 赵志斌, 陈雪峰. 基于分层对齐迁移学习的锂离子电池容量估计[J]. 储能科学与技术, 2023, 12(4): 1223-1233.

Zhi ZHAI, Fujin WANG, Yi DI, Peiyu MA, Zhibin ZHAO, Xuefeng CHEN. Hierarchical alignment transfer learning for lithium-ion battery capacity estimation[J]. Energy Storage Science and Technology, 2023, 12(4): 1223-1233.

图/表 10

图1

图2

图3

图4

图5

表1

图6

图7

表2

图8

参考文献 44

1	SCHMUCH R, WAGNER R, HÖRPEL G, et al. Performance and cost of materials for lithium-based rechargeable automotive batteries[J]. Nature Energy, 2018, 3(4): 267-278.
2	NYKVIST B, NILSSON M. Rapidly falling costs of battery packs for electric vehicles[J]. Nature Climate Change, 2015, 5(4): 329-332.
3	刘大同, 宋宇晨, 武巍, 等. 锂离子电池组健康状态估计综述[J]. 仪器仪表学报, 2020, 41(11): 1-18.
	LIU D T, SONG Y C, WU W, et al. Review of state of health estimation for lithium-ion battery pack[J]. Chinese Journal of Scientific Instrument, 2020, 41(11): 1-18.
4	SAHA B, GOEBEL K, POLL S, et al. Prognostics methods for battery health monitoring using a Bayesian framework[J]. IEEE Transactions on Instrumentation and Measurement, 2009, 58(2): 291-296.
5	SONG Y C, LIU D T, HOU Y D, et al. Satellite lithium-ion battery remaining useful life estimation with an iterative updated RVM fused with the KF algorithm[J]. Chinese Journal of Aeronautics, 2018, 31(1): 31-40.
6	王亚坤, 杨凯飞, 张婕, 等. 卫星在轨故障案例与人工智能故障诊断[J]. 中国空间科学技术, 2022, 42(1): 16-29.
	WANG Y K, YANG K F, ZHANG J, et al. Case study of in-orbit satellite failures and artificial intelligence based failure detection[J]. Chinese Space Science and Technology, 2022, 42(1): 16-29.
7	朱伟杰, 史尤杰, 雷博. 锂离子电池储能系统BMS的功能安全分析与设计[J]. 储能科学与技术, 2020, 9(1): 271-278.
	ZHU W J, SHI Y J, LEI B. Functional safety analysis and design of BMS for lithium-ion battery energy storage system[J]. Energy Storage Science and Technology, 2020, 9(1): 271-278.
8	HOSSAIN LIPU M S, HANNAN M A, HUSSAIN A, et al. A review of state of health and remaining useful life estimation methods for lithium-ion battery in electric vehicles: Challenges and recommendations[J]. Journal of Cleaner Production, 2018, 205: 115-133.
9	ZHANG Y, LI Y F. Prognostics and health management of Lithium-ion battery using deep learning methods: A review[J]. Renewable and Sustainable Energy Reviews, 2022, 161: doi: 10.1016/j.rser.2022.112282.
10	CHEN L, LÜ Z Q, LIN W L, et al. A new state-of-health estimation method for lithium-ion batteries through the intrinsic relationship between ohmic internal resistance and capacity[J]. Measurement, 2018, 116: 586-595.
11	YANG F F, SONG X B, DONG G Z, et al. A coulombic efficiency-based model for prognostics and health estimation of lithium-ion batteries[J]. Energy, 2019, 171: 1173-1182.
12	刘征宇, 朱诚诚, 尤勇, 等. 面向SOC估计的计及温度和循环次数的锂离子电池组合模型[J]. 仪器仪表学报, 2019, 40(11): 117-127.
	LIU Z Y, ZHU C C, YOU Y, et al. A Lithium-ion battery combined model considering temperature and cycle times for SOC estimation[J]. Chinese Journal of Scientific Instrument, 2019, 40(11): 117-127.
13	熊然, 王顺利, 于春梅, 等. 基于Thevenin模型和改进扩展卡尔曼的特种机器人锂离子电池SOC估算方法[J]. 储能科学与技术, 2021, 10(2): 695-704.
	XIONG R, WANG S L, YU C M, et al. An estimation method for lithium-ion battery SOC of special robots based on Thevenin model and improved extended Kalman[J]. Energy Storage Science and Technology, 2021, 10(2): 695-704.
14	戴海峰, 姜波, 魏学哲, 等. 基于充电曲线特征的锂离子电池容量估计[J]. 机械工程学报, 2019, 55(20): 52-59.
	DAI H F, JIANG B, WEI X Z, et al. Capacity estimation of lithium-ion batteries based on charging curve features[J]. Journal of Mechanical Engineering, 2019, 55(20): 52-59.
15	韩云飞, 谢佳, 蔡涛, 等. 结合高斯过程回归与特征选择的锂离子电池容量估计方法[J]. 储能科学与技术, 2021, 10(4): 1432-1438.
	HAN Y F, XIE J, CAI T, et al. Capacity estimation of lithium-ion batteries based on Gaussian process regression and feature selection[J]. Energy Storage Science and Technology, 2021, 10(4): 1432-1438.
16	周子游, 刘永刚, 杨阳, 等. 考虑混杂充电数据的锂离子电池容量估计[J]. 机械工程学报, 2021, 57(14): 1-9.
	ZHOU Z Y, LIU Y G, YANG Y, et al. Capacity estimation of lithium ion battery considering hybrid charging data[J]. Journal of Mechanical Engineering, 2021, 57(14): 1-9.
17	舒星, 刘永刚, 申江卫, 等. 基于改进最小二乘支持向量机与Box-Cox变换的锂离子电池容量预测[J]. 机械工程学报, 2021, 57(14): 118-128.
	SHU X, LIU Y G, SHEN J W, et al. Capacity prediction for lithium-ion batteries based on improved least squares support vector machine and box-cox transformation[J]. Journal of Mechanical Engineering, 2021, 57(14): 118-128.
18	HU X S, XU L, LIN X K, et al. Battery lifetime prognostics[J]. Joule, 2020, 4(2): 310-346.
19	RAUF H, KHALID M, ARSHAD N. Machine learning in state of health and remaining useful life estimation: Theoretical and technological development in battery degradation modelling[J]. Renewable and Sustainable Energy Reviews, 2022, 156: doi: 10.1016/j.rser.2021.111903.
20	LI P H, ZHANG Z J, GROSU R, et al. An end-to-end neural network framework for state-of-health estimation and remaining useful life prediction of electric vehicle lithium batteries[J]. Renewable and Sustainable Energy Reviews, 2022, 156: doi: 10.1016/j.rser.2021.111843.
21	DU Z K, ZUO L, LI J J, et al. Data-driven estimation of remaining useful lifetime and state of charge for lithium-ion battery[J]. IEEE Transactions on Transportation Electrification, 2022, 8(1): 356-367.
22	YANG Y X. A machine-learning prediction method of lithium-ion battery life based on charge process for different applications[J]. Applied Energy, 2021, 292: doi: 10.1016/j.apenergy.2021.116897.
23	ZHANG Q S, YANG L, GUO W C, et al. A deep learning method for lithium-ion battery remaining useful life prediction based on sparse segment data via cloud computing system[J]. Energy, 2022, 241: doi: 10.1016/j.energy.2021.122716.
24	ZHUANG F Z, QI Z Y, DUAN K Y, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76.
25	LI Y Y, SHENG H M, CHENG Y H, et al. State-of-health estimation of lithium-ion batteries based on semi-supervised transfer component analysis[J]. Applied Energy, 2020, 277: doi: 10.1016/j.apenergy.2020.115504.
26	LIU Y, SHU X, YU H, et al. State of charge prediction framework for lithium-ion batteries incorporating long short-term memory network and transfer learning[J]. Journal of Energy Storage, 2021, 37: doi: 10.1016/j.est.2021.102494.
27	LI Y H, LI K, LIU X, et al. Lithium-ion battery capacity estimation—a pruned convolutional neural network approach assisted with transfer learning[J]. Applied Energy, 2021, 285: doi: 10.1016/j.apenergy.2020.116410.
28	SHU X, SHEN J W, LI G, et al. A flexible state-of-health prediction scheme for lithium-ion battery packs with long short-term memory network and transfer learning[J]. IEEE Transactions on Transportation Electrification, 2021, 7(4): 2238-2248.
29	KIM S, CHOI Y Y, KIM K J, et al. Forecasting state-of-health of lithium-ion batteries using variational long short-term memory with transfer learning[J]. Journal of Energy Storage, 2021, 41: doi: 10.1016/j.est.2021.102893.
30	WANG F J, ZHAO Z B, REN J X, et al. A transferable lithium-ion battery remaining useful life prediction method from cycle-consistency of degradation trend[J]. Journal of Power Sources, 2022, 521: doi: 10.1016/j.jpowsour.2022.230975.
31	WANG F J, ZHAO Z B, ZHAI Z, et al. Remaining useful life prediction of lithium-ion battery based on cycle-consistency learning[C]//2021 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD). October 21-23, 2021, Nanjing, China. IEEE, 2022: 1-6.
32	陈峥, 赵广达, 沈世全, 等. 基于迁移模型的老化锂离子电池SOC估计[J]. 储能科学与技术, 2021, 10(1): 326-334.
	CHEN Z, ZHAO G D, SHEN S Q, et al. SOC estimation of aging lithium-ion battery based on a migration model[J]. Energy Storage Science and Technology, 2021, 10(1): 326-334.
33	HAN T, WANG Z, MENG H X. End-to-end capacity estimation of Lithium-ion batteries with an enhanced long short-term memory network considering domain adaptation[J]. Journal of Power Sources, 2022, 520: doi: 10.1016/j.jpowsour.2021.230823.
34	YE Z, YU J B. State-of-health estimation for lithium-ion batteries using domain adversarial transfer learning[J]. IEEE Transactions on Power Electronics, 2022, 37(3): 3528-3543.
35	YOSINSKI J, CLUNE J, BENGIO Y, et al. How transferable are features in deep neural networks?[A]. Proceedings of the 27th International Conference on Neural Information Processing Systems-Volume 2[C]. Cambridge, MA, USA: MIT Press, 2014: 3320-3328.
36	NEYSHABUR B, SEDGHI H, ZHANG C Y. What is being transferred in transfer learning? [EB/OL]. 2020. [2021-12-01]. https://arxiv.org/abs/2008.11687.
37	BOUSMALIS K, TRIGEORGIS G, SILBERMAN N, et al. Domain separation networks[EB/OL]. 2016. [2021-12-01]. https://arxiv.org/abs/1608.06019.
38	OYEWOLE I, CHEHADE A, KIM Y. A controllable deep transfer learning network with multiple domain adaptation for battery state-of-charge estimation[J]. Applied Energy, 2022, 312: doi: 10.1016/j.apenergy.2022.118726.
39	MAATEN L V D, HINTON G. Visualizing data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9: 2579-2625.
40	GRETTON A, BORGWARDT K M, RASCH M, et al. A kernel two-sample test[J]. Journal of Machine Learning Research, MIT Press, 2012, 13: 723-773.
41	DENG Z Y, ZHOU K Y, YANG Y X, et al. Domain attention consistency for multi-source domain adaptation[EB/OL]. 2021. [2021-12-01]. https://arxiv.org/abs/2111.03911.
42	ATTIA P M, GROVER A, JIN N, et al. Closed-loop optimization of fast-charging protocols for batteries with machine learning[J]. Nature, 2020, 578(7795): 397-402.
43	SUN B C, SAENKO K. Deep coral: Correlation alignment for deep domain adaptation[C]//European Conference on Computer Vision. Cham: Springer, 2016: 443-450.
44	GANIN Y, LEMPITSKY V. Unsupervised domain adaptation by backpropagation[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning-Volume 37. New York: ACM, 2015: 1180-1189.

域	充电策略	电池编号(通道)
源域	4.4-5.6-5.2-4.252 C^①	8, 15, 18, 32, 48
目标域	7-4.8-4.8-3.652 C	3, 25, 26, 28, 44

方法	误差	电池编号
方法	误差	3	25	26	28	44
HATL	MSE/%	0.064	0.080	0.091	0.125	0.076
HATL	MAE/%	1.786	2.278	2.340	2.325	2.016
ResNet	MSE/%	0.611	1.161	1.306	0.399	0.435
ResNet	MAE/%	6.546	10.075	10.723	5.186	5.119
MMD	MSE/%	0.082	0.174	0.166	0.206	0.162
MMD	MAE/%	2.504	2.704	3.600	2.659	2.911
CORAL	MSE/%	0.701	0.983	0.557	0.270	0.152
CORAL	MAE/%	8.093	9.337	7.148	3.892	1.811
DANN	MSE/%	0.562	0.516	1.170	0.334	0.259
DANN	MAE/%	6.798	5.221	10.524	6.798	4.263

[1]	宋缙华, 张兴浩, 丰震河, 程广玉, 顾洪汇, 顾海涛, 王可. 基于原位参比的氧化亚硅-石墨复合负极循环衰减机制[J]. 储能科学与技术, 2023, 12(4): 1059-1065.
[2]	胡力月, 姚行艳. 基于正交试验的锂离子电池热失控仿真[J]. 储能科学与技术, 2023, 12(4): 1268-1277.
[3]	刘峰, 陈海忠. 基于CEEMDAN和ISOA-ELM的锂电池荷电状态预测[J]. 储能科学与技术, 2023, 12(4): 1244-1256.
[4]	王朋凯, 张新燕, 张光昊. 基于ResNet-Bi-LSTM-Attention的锂离子电池剩余使用寿命预测[J]. 储能科学与技术, 2023, 12(4): 1215-1222.
[5]	赵立禹, 孙桓五, 刘世闯, 闫志远. 重卡辅助动力电池加热系统能耗对比及优化[J]. 储能科学与技术, 2023, 12(4): 1139-1147.
[6]	杨妮, 苏岳锋, 王联, 李宁, 马亮, 朱晨. 聚焦离子束显微镜技术在锂离子电池领域的研究进展[J]. 储能科学与技术, 2023, 12(4): 1283-1294.
[7]	成雪莉, 张维福, 罗城城, 袁小亚. 一步水热法制备三维石墨烯/Fe₃O₄ 复合材料及其储锂性能[J]. 储能科学与技术, 2023, 12(4): 1066-1074.
[8]	李奇松, 陈荣, 李慧芳. 阻抗分析法在锂离子电池析锂阈值检测中的应用[J]. 储能科学与技术, 2023, 12(4): 1278-1282.
[9]	董渊昌, 庞晓琼, 贾建芳, 史元浩, 温杰, 李笑, 张鑫. 基于SVD-SAE-GPR的锂离子电池RUL预测[J]. 储能科学与技术, 2023, 12(4): 1257-1267.
[10]	管敏渊, 沈建良, 徐国华, 汤舜, 张炜鑫, 曹元成. 锂离子电池储能系统靶向消防装备设计与性能[J]. 储能科学与技术, 2023, 12(4): 1131-1138.
[11]	常修亮, 李希超, 贾隆舟, 韦守李, 王敬豪, 戴作强, 郑莉莉. 过充循环老化电池产热特性[J]. 储能科学与技术, 2023, 12(3): 685-697.
[12]	刘树港, 蒙波, 李政隆, 杨亚雄, 陈建. Li _x （Mg， Ni， Zn， Cu， Co） ₁_-_x O高熵氧化物负极材料电化学储锂特性[J]. 储能科学与技术, 2023, 12(3): 743-753.
[13]	姚逸鸣, 栾伟玲, 陈莹, 孙敏. 基于光学显微镜的锂离子电池材料老化衰减原位研究进展[J]. 储能科学与技术, 2023, 12(3): 777-791.
[14]	李奎杰, 楼平, 管敏渊, 莫金龙, 张炜鑫, 曹元成, 程时杰. 锂离子电池热失控多维信号演化及耦合机制研究综述[J]. 储能科学与技术, 2023, 12(3): 899-912.
[15]	封迈, 陈楠, 陈人杰. 锂离子电池低温电解液的研究进展[J]. 储能科学与技术, 2023, 12(3): 792-807.

基于分层对齐迁移学习的锂离子电池容量估计

Hierarchical alignment transfer learning for lithium-ion battery capacity estimation

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 44

相关文章 15

编辑推荐

Metrics

本文评价