基于模态分解和深度学习的锂离子电池寿命预测

doi:10.19799/j.cnki.2095-4239.2024.0983

储能科学与技术 ›› 2025, Vol. 14 ›› Issue (4): 1645-1653.doi: 10.19799/j.cnki.2095-4239.2024.0983

基于模态分解和深度学习的锂离子电池寿命预测

董作林¹^,²^,³(), 宋金岩¹^,²^,³(), 孟子迪¹^,²

^1.大连海洋大学信息工程学院
^2.设施渔业教育部重点实验室
^3.辽宁省海洋信息重点实验室，辽宁大连 116023

收稿日期:2024-10-21 修回日期:2024-11-23 出版日期:2025-04-28 发布日期:2025-05-20
通讯作者: 宋金岩 E-mail:2900325200@qq.com;songjinyan@dlou.edu.cn
作者简介:董作林（2000—），男，硕士研究生，研究方向为电池寿命预测，E-mail：2900325200@qq.com；
基金资助:
辽宁省自然科学基金指导计划(20180550573);大连海洋大学校级教改项目（32），辽宁省教育厅项目(JL201918)

Lithium-ion battery life prediction based on mode decomposition and deep learning

Zuolin DONG¹^,²^,³(), Jinyan SONG¹^,²^,³(), Zidi MENG¹^,²

^1.College of Information Engineering, Dalian Ocean University
^2.Key Laboratory of Facilities Aquaculture, Ministry of Education
^3.Key Laboratory of Marine Information of Liaoning Province, Dalian 116023, Liaoning, China

Received:2024-10-21 Revised:2024-11-23 Online:2025-04-28 Published:2025-05-20
Contact: Jinyan SONG E-mail:2900325200@qq.com;songjinyan@dlou.edu.cn

摘要/Abstract

摘要：

随着新能源汽车数量的快速增长，精准预测锂离子电池的剩余使用寿命(remaining useful life, RUL)对新能源汽车产业的持续发展起到了至关重要的作用。本工作提出了一种基于集成经验模态分解(ensemble empirical mode decomposition, EEMD)和深度学习的创新方法，旨在提升锂离子电池RUL的预测精度。首先，通过EEMD对电池容量数据进行多尺度分解，得到电池容量数据的全局退化趋势和局部随机波动分量。为了减轻噪声对模型预测精确度的干扰，引入去噪自编码器(denoising autoencoder, DAE)对随机波动分量进行降噪处理。随后，分别使用长短期记忆(long short-term memory, LSTM)网络和自注意力模型(Transformer)对全局退化趋势和降噪后的随机波动分量进行建模。最后为进一步提取各模态分量中存在的信息，采用随机森林(random forest, RF)算法计算各分量的重要性权重，根据得到的权重值对预测结果加权重构。本工作在美国国家航空航天局(National Aeronautics and Space Administration, NASA)公开的电池数据集上分别使用40%和60%的历史数据进行实验，结果表明所提出的方法在精度和有效性方面均优于现有方法，验证了其在锂离子电池RUL预测中的应用潜力。

关键词: 锂离子电池, 寿命预测, 深度学习, 长短期记忆, 随机森林

Abstract:

With the rapid growth in the electric vehicle (EV) adoption, accurately predicting the remaining useful life (RUL) of lithium-ion batteries has become critical for the sustained development of the EV industry. This paper proposes an innovative approach that integrates ensemble mode decomposition (EEMD) and deep learning to improve RUL prediction accuracy for lithium-ion batteries. The proposed method begins with EEMD, which performs multiscale decomposition of battery capacity data. This process separates the global degradation trend from local random fluctuation components. To mitigate the impact of noise on model prediction accuracy, a denoising autoencoder (DAE) is introduced to remove noise from the random fluctuation components. Subsequently, long short-term memory (LSTM) networks and the transformer model are applied to model the global degradation trend and the denoised random fluctuations, respectively. To further refine predictions, a random forest (RF) algorithm calculates the importance weights of each mode component, enabling a weighted reconstruction of the prediction results. Experiments were conducted on a public battery dataset provided by the National Aeronautics and Space Administration (NASA), leveraging 40% and 60% of the historical battery data. The results demonstrate that the proposed method outperforms existing approaches in both accuracy and effectiveness, validating its potential for application in lithium-ion battery RUL prediction.

Key words: lithium-ion battery, life prediction, deep learning, long short-term memory, random forest

中图分类号:

TM 912

董作林, 宋金岩, 孟子迪. 基于模态分解和深度学习的锂离子电池寿命预测[J]. 储能科学与技术, 2025, 14(4): 1645-1653.

Zuolin DONG, Jinyan SONG, Zidi MENG. Lithium-ion battery life prediction based on mode decomposition and deep learning[J]. Energy Storage Science and Technology, 2025, 14(4): 1645-1653.

图/表 10

图1

图2

图3

图4

表1

表2

图5

表3

表4

图6

参考文献 14

1	王海波, 林虹, 宋文龙, 等. 2024年上半年中国电池行业运行情况[J]. 电池, 2024, 54(4): 445-449. DOI: 10.19535/j.1001-1579. 2024.04.002.
	WANG H B, LIN H, SONG W L, et al. Operation of China's battery industry in the first half of 2024[J]. Battery Bimonthly, 2024, 54(4): 445-449. DOI: 10.19535/j.1001-1579.2024.04.002.
2	陈素华, 白莹. 锂离子动力电池热失控机理及热管理技术研究进展[J]. 中国科学基金, 2023, 37(2): 187-198. DOI: 10.16262/j.cnki. 1000-8217.2023.02.005.
	CHEN S H, BAI Y. Thermal runway mechanism and research progress on thermal management of lithium-ion power batteries[J]. Bulletin of National Natural Science Foundation of China, 2023, 37(2): 187-198. DOI: 10.16262/j.cnki.1000-8217.2023. 02.005.
3	NUHIC A, TERZIMEHIC T, SOCZKA-GUTH T, et al. Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methods[J]. Journal of Power Sources, 2013, 239: 680-688. DOI: 10.1016/j.jpowsour.2012. 11.146.
4	YOU K S, QIU G Q, GU Y K. Remaining useful life prediction of lithium-ion batteries using EM-PF-SSA-SVR with gamma stochastic process[J]. Measurement Science and Technology, 2024, 35(1): 015015. DOI: 10.1088/1361-6501/acfbef.
5	CHEN L, ZHANG Y, ZHENG Y, et al. Remaining useful life prediction of lithium-ion battery with optimal input sequence selection and error compensation[J]. Neurocomputing, 2020, 414: 245-254. DOI: 10.1016/j.neucom.2020.07.081.
6	LYU Z Q, WANG G, GAO R J. Synchronous state of health estimation and remaining useful lifetime prediction of Li-Ion battery through optimized relevance vector machine framework[J]. Energy, 2022, 251: 123852. DOI: 10.1016/j.energy.2022. 123852.
7	GUO W, HE M. An optimal relevance vector machine with a modified degradation model for remaining useful lifetime prediction of lithium-ion batteries[J]. Applied Soft Computing, 2022, 124: 108967. DOI: 10.1016/j.asoc.2022.108967.
8	VENUGOPAL P, SHANKAR S S, JEBAKUMAR C P, et al. Analysis of optimal machine learning approach for battery life estimation of Li-ion cell[J]. IEEE Access, 2021, 9: 159616-159626. DOI: 10.1109/ACCESS.2021.3130994.
9	PARK K, CHOI Y, CHOI W J, et al. LSTM-based battery remaining useful life prediction with multi-channel charging profiles[J]. IEEE Access, 2020, 8: 20786-20798. DOI: 10.1109/ACCESS.2020.2968939.
10	XIE Q L, LIU R C, HUANG J H, et al. Residual life prediction of lithium-ion batteries based on data preprocessing and a priori knowledge-assisted CNN-LSTM[J]. Energy, 2023, 281: 128232. DOI: 10.1016/j.energy.2023.128232.
11	HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society of London Series A: Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903-995. DOI: 10.1098/rspa.1998.0193.
12	WU Z H, HUANG N E. Ensemble empirical mode decomposition: A noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1): 1-41. DOI: 10.1142/s1793536909000047.
13	BREIMAN L. Random forests[J]. Machine learning, 2001, 45: 5-32.
14	YU Y, SI X S, HU C H, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235-1270. DOI: 10.1162/neco_a_01199.

电池编号	拟合优度	IMF1权重	IMF2权重	IMF3权重
B0005	0.85540	0.45478	0.30492	0.24029
B0006	0.91451	0.52218	0.27672	0.20109
B0007	0.86730	0.43125	0.29544	0.27330
B0018	0.86863	0.48244	0.29831	0.21925

参数	Transformer	LSTM
λ	0.001	0.001
s	64	256
ι	2	4
η	8	None
θ	Adan	Adan
б	32	6

电池	模型	RE	MAE	RMSE	MAPE
B0005	EEMD-LSTM	0.00878	0.00923	0.01223	0.00673
	DaTransformer	0.00824	0.00889	0.01322	0.00645
	EEMD-LSTM-DaTransformer	0.00719	0.00838	0.01074	0.00608
	Our model	0.00719	0.00714	0.00913	0.00519

B0006	EEMD-LSTM	0	0.01063	0.01468	0.00815
	DaTransformer	0.01639	0.01284	0.01485	0.00996
	EEMD-LSTM-DaTransformer	0	0.00878	0.01116	0.00670
	Our model	0.01869	0.00817	0.00997	0.00624

B0007	EEMD-LSTM	—	0.00846	0.01103	0.00512
	DaTransformer	—	0.00663	0.00939	0.00449
	EEMD-LSTM-DaTransformer	—	0.00731	0.00911	0.00493
	Our model	—	0.00617	0.00740	0.00416

B0018	EEMD-LSTM	0.05263	0.01194	0.01525	0.00852
	DaTransformer	1	0.01545	0.02520	0.01088
	EEMD-LSTM-DaTransformer	0.01053	0.01003	0.01460	0.00715
	Our model	0.01053	0.00754	0.01239	0.00583

电池	模型	RE	MAE	RMSE	MAPE
B0005	EEMD-LSTM	0.00813	0.00975	0.01442	0.00680
	DaTransformer	0	0.00996	0.01559	0.00685
	EEMD-LSTM-DaTransformer	0.00813	0.00786	0.01084	0.00548
	Our model	0.00813	0.00739	0.00990	0.00514

B0006	EEMD-LSTM	0.00935	0.01514	0.02213	0.01107
	DaTransformer	0.01802	0.01222	0.01998	0.00886
	EEMD-LSTM-DaTransformer	0.00935	0.01276	0.01605	0.00931
	Our model	0.00935	0.01090	0.01353	0.00795

B0007	EEMD-LSTM	—	0.00836	0.01358	0.00548
	DaTransformer	—	0.00770	0.01215	0.00510
	EEMD-LSTM-DaTransformer	—	0.00771	0.01098	0.00504
	Our model	—	0.00626	0.00897	0.00409

B0018	EEMD-LSTM	0.04211	0.01117	0.01611	0.00812
	DaTransformer	1	0.01788	0.02581	0.01214
	EEMD-LSTM-DaTransformer	0.01053	0.01153	0.01524	0.00795
	Our model	0.01053	0.01049	0.01436	0.00719

[1]	彭磊, 倪照鹏, 于越, 孙福鹏, 夏修龙, 张鹏, 孙思博. 过充导致三元锂电池电动汽车火灾的试验研究[J]. 储能科学与技术, 2025, 14(4): 1484-1495.
[2]	申江卫, 折亦鑫, 舒星, 刘永刚, 魏福星, 夏雪磊, 陈峥. 基于短时随机充电数据和优化卷积神经网络的锂电池健康状态估计[J]. 储能科学与技术, 2025, 14(4): 1585-1595.
[3]	刘瑞昊, 马小乐, 张宇萱, 朱曰莹, 刘仕强, 白广利. 基于绝热量热仪的锂离子电池热物性参数测试影响因素研究[J]. 储能科学与技术, 2025, 14(4): 1596-1602.
[4]	徐桂培, 刘浩, 赖洁文, 卢毅锋, 黄辉, 邸会芳, 王振兵. 干法电极技术在超级电容器和锂离子电池中的研究进展[J]. 储能科学与技术, 2025, 14(4): 1445-1460.
[5]	岳金明, 刘媛丽, 陈一霞, 禹习谦, 李泓. GC-MS检测锂离子电池电解液分离条件的研究[J]. 储能科学与技术, 2025, 14(4): 1564-1573.
[6]	廖兴群, 杨睿, 于立娟, 胡大林, 肖峰, 胡菁, 卢周广. 多功能电解液添加剂2，6-吡啶二甲腈稳定高电压钴酸锂[J]. 储能科学与技术, 2025, 14(4): 1331-1339.
[7]	王鹏, 周俊, 伍星, 刘韬. 改进Sine混沌映射CO-ELM锂离子电池RUL预测[J]. 储能科学与技术, 2025, 14(4): 1603-1616.
[8]	蔡志端, 张吴哲, 吴成傲, 童嘉阳. 强干扰下基于VMD三次分解的锂电池健康状态估计方法[J]. 储能科学与技术, 2025, 14(4): 1631-1644.
[9]	武小兰, 马彭杰, 白志峰, 刘成龙, 郭桂芳, 张锦华. 一种锂离子电池组智能PID双层主动均衡控制方法[J]. 储能科学与技术, 2025, 14(3): 1150-1159.
[10]	曾帅波, 李涌仪, 彭静, 何梓星, 梁倬健, 徐伟, 蓝凌霄, 梁兴华. 基于三元锂离子电池的导电剂优化设计[J]. 储能科学与技术, 2025, 14(3): 1187-1197.
[11]	张朝龙, 陈阳, 刘梦玲, 张俣峰, 华国庆, 阴盼昐. 一种基于ICA-T特征和CNN-LA-BiLSTM的锂离子电池健康状态估计方法[J]. 储能科学与技术, 2025, 14(3): 1258-1269.
[12]	李南, 马静, 黄挺秀, 沈毅星, 沈旻, 江依义, 洪涛, 马国强, 马紫峰. 腈类化合物在高电压电解液中的研究进展[J]. 储能科学与技术, 2025, 14(3): 997-1009.
[13]	许陈程, 王湛, 李爽, 蒋江民, 鞠治成. 锂离子电池预锂化技术研究进展及工程化应用展望[J]. 储能科学与技术, 2025, 14(3): 930-946.
[14]	陈会明, 蔡艺嘉, 尹文骥, 陈美芬, 黄有国, 胡思江, 王红强, 李庆余. 铬钼双掺杂调控富锂锰基正极材料结构和电化学性能[J]. 储能科学与技术, 2025, 14(3): 1123-1132.
[15]	周丽萍, 周德清, 郑锋华, 潘齐常, 胡思江, 蒋永杰, 王红强, 李庆余. 锂离子电池Si@Void@C复合负极材料的制备及其应用[J]. 储能科学与技术, 2025, 14(3): 1115-1122.

基于模态分解和深度学习的锂离子电池寿命预测

Lithium-ion battery life prediction based on mode decomposition and deep learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 14

相关文章 15

编辑推荐

Metrics

本文评价