Article(id=1149776901675377370, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2309072, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1731945600000, receivedDateStr=2024-11-19, revisedDate=1729440000000, revisedDateStr=2024-10-21, acceptedDate=null, acceptedDateStr=null, onlineDate=1752057775181, onlineDateStr=2025-07-09, pubDate=1744905600000, pubDateStr=2025-04-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752057775181, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752057775180, creator=13701087609, updateTime=1752057775180, updator=13701087609, issue=Issue{id=1149776900194791454, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='11', pageStart='4397', pageEnd='4826', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752057774827, creator=13701087609, updateTime=1768456666677, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218558837930512931, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218558837930512932, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=4598, endPage=4604, ext={EN=ArticleExt(id=1149776901943812828, articleId=1149776901675377370, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Estimation Model for State of Health of Lithium-ion Battery Based on VMD and BiLSTM-ATT, columnId=1156262733675876713, journalTitle=Science Technology and Engineering, columnName=Papers·Electrical Technology, runingTitle=null, highlight=null, articleAbstract=

The estimation of the state of health (SOH) for lithium-ion batteries is considered crucial for ensuring the safe and stable operation of battery management system. However, the accurate estimation of SOH has been a challenge due to the capacity regeneration phenomenon during the discharge process of lithium-ion batteries. To improve estimation accuracy, a hybrid model based on variational mode decomposition (VMD) and bidirectional long short-term memory network with attention mechanism (BiLSTM-ATT) was proposed. First, the battery capacity was decomposed using the VMD algorithm, producing a set of stable sub-sequences. Then, permutation entropy was introduced to reconstruct the sub-sequences to reduce computational complexity. The reconstructed sequences were input into the BiLSTM-ATT model, and feature weights were assigned by the attention mechanism. The SOH values were trained and estimated by the BiLSTM model. Finally, the complete SOH estimation result was obtained by summing all estimated values. Validation was performed using the CS2_36, CS2_38, and CX2_35 datasets from the CALCE lithium battery dataset. The results show that the proposed algorithm maintains a root mean square error within 0.6% and a mean absolute error within 0.4%, which demonstrates higher accuracy and performance compared to other estimation models.

, correspAuthors=Quan-jun WU, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Ping-sheng HU, Quan-jun WU), CN=ArticleExt(id=1149776924467225582, articleId=1149776901675377370, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于变分模态分解和BiLSTM-ATT的锂电池健康状态估计模型, columnId=1156262734506353627, journalTitle=科学技术与工程, columnName=论文·电工技术, runingTitle=null, highlight=null, articleAbstract=

锂离子电池健康状态(state of health,SOH)估计对于保证锂离子电池管理系统的安全稳定运行至关重要。然而,由于锂离子电池在放电过程中存在容量再生现象,SOH的准确估计一直是一个挑战。为了提高估计精度,提出了一种基于变分模态分解(variational mode decomposition,VMD)和双向长短期记忆网络-注意力机制(bidirectional long short term memory-attention,BiLSTM-ATT)的混合模型估计方法。首先,采用VMD分解算法对锂电池容量进行分解,得到一组相对稳定的子序列。为了降低后续的计算规模,通过引入了排列熵的方法对各个子序列进行重构。然后,将重构后的序列输入到BiLSTM-ATT模型中,利用注意力机制来分配隐藏层的特征权重,并通过双向长短期记忆网络(bidirectional long short term memory, BiLSTM)模型对SOH值进行训练和估计。最后,将所有估计值进行相加得到完整的SOH估计结果。通过在CALCE锂电池数据集上的CS2_36、CS2_38和CX2_35进行验证,实验结果表明所提出算法的均方根误差始终保持在0.6%以内,平均绝对误差始终保持在0.4%以内,相比其他估计模型表现出更高的精度和性能。

, correspAuthors=吴泉军, authorNote=null, correspAuthorsNote=
* 吴泉军(1982—),男,汉族,山东汶上人,博士,副教授。研究方向:锂电池健康监测。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=eChuhVXTykwJYb/Nfip8NQ==, magXml=5/wyohSGo0kdCByjxjlYtA==, pdfUrl=null, pdf=ITNf1CQxaUPr+NuDAFQLQQ==, pdfFileSize=5985202, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=ElvbxXHaOV3j7zuGtjNDww==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=PB3SaEi1dFOiZfJ/HZKLSA==, mapNumber=null, authorCompany=null, fund=null, authors=

胡平生(1996—),男,汉族,安徽合肥人,硕士研究生。研究方向:深度学习、新能源电池的管理系统。E-mail:

, authorsList=胡平生, 吴泉军)}, authors=[Author(id=1218843912366637349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=hu942658636@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1218843912475689268, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, authorId=1218843912366637349, language=EN, stringName=Ping-sheng HU, firstName=Ping-sheng, middleName=null, lastName=HU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843912584741184, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, authorId=1218843912366637349, language=CN, stringName=胡平生, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=上海电力大学,数理学院智慧能源数学研究中心, 上海 201306, bio={"content":"

胡平生(1996—),男,汉族,安徽合肥人,硕士研究生。研究方向:深度学习、新能源电池的管理系统。E-mail:

"}, bioImg=null, bioContent=

胡平生(1996—),男,汉族,安徽合肥人,硕士研究生。研究方向:深度学习、新能源电池的管理系统。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843912190476560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, xref=null, ext=[AuthorCompanyExt(id=1218843912207253780, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China), AuthorCompanyExt(id=1218843912224030998, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海电力大学,数理学院智慧能源数学研究中心, 上海 201306)])]), Author(id=1218843912697987406, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wuquanjun@shhiep.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1218843912828010839, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, authorId=1218843912697987406, language=EN, stringName=Quan-jun WU, firstName=Quan-jun, middleName=null, lastName=WU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843912932868448, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, authorId=1218843912697987406, language=CN, stringName=吴泉军, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=上海电力大学,数理学院智慧能源数学研究中心, 上海 201306, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843912190476560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, xref=null, ext=[AuthorCompanyExt(id=1218843912207253780, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China), AuthorCompanyExt(id=1218843912224030998, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海电力大学,数理学院智慧能源数学研究中心, 上海 201306)])])], keywords=[Keyword(id=1218843913130000754, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=1, keyword=lithium-ion battery), Keyword(id=1218843913239052665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=2, keyword=state of health), Keyword(id=1218843913478128004, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=3, keyword=variational mode decomposition), Keyword(id=1218843913591374217, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=4, keyword=permutation entropy), Keyword(id=1218843913687843218, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=5, keyword=attention mechanism), Keyword(id=1218843913763340698, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, orderNo=6, keyword=bidirectional long short-term memory), Keyword(id=1218843913901752735, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=1, keyword=锂离子电池), Keyword(id=1218843914006610346, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=2, keyword=健康状态), Keyword(id=1218843914170188216, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=3, keyword=变分模态分解), Keyword(id=1218843914300211658, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=4, keyword=排列熵), Keyword(id=1218843914438623697, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=5, keyword=注意力机制), Keyword(id=1218843914593812955, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, orderNo=6, keyword=双向长短期记忆网络)], refs=[Reference(id=1218843918108639975, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2017, volume=357, issue=null, pageStart=209, pageEnd=219, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Richardson R R, Osborne M A, Howey D A, journalName=Journal of Power Sources, refType=null, unstructuredReference=Richardson R R, Osborne M A, Howey D A. Gaussian process regression for forecasting battery state of health[J]. Journal of Power Sources, 2017, 357: 209-219., articleTitle=Gaussian process regression for forecasting battery state of health, refAbstract=null), Reference(id=1218843918234469102, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=15, pageStart=6279, pageEnd=6289, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=王志福, 杨忠义, 罗崴, journalName=科学技术与工程, refType=null, unstructuredReference=王志福, 杨忠义, 罗崴, 等. 基于数据驱动的锂离子动力电池剩余使用寿命预测方法综述[J]. 科学技术与工程, 2023, 23(15): 6279-6289., articleTitle=基于数据驱动的锂离子动力电池剩余使用寿命预测方法综述, refAbstract=null), Reference(id=1218843918343521014, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=15, pageStart=6279, pageEnd=6289, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Wang Zhifu, Yang Zhongyi, Luo Wei, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Zhifu, Yang Zhongyi, Luo Wei, et al. Review of remaining useful life prediction for lithium-ion battery based on data-driven[J]. Science Technology and Engineering, 2023, 23(15): 6279-6289., articleTitle=Review of remaining useful life prediction for lithium-ion battery based on data-driven, refAbstract=null), Reference(id=1218843918477738752, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=5, pageStart=1928, pageEnd=1933, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=林娜, 朱武, 邓安全, journalName=科学技术与工程, refType=null, unstructuredReference=林娜, 朱武, 邓安全. 基于融合方法预测锂离子电池剩余寿命[J]. 科学技术与工程, 2020, 20(5): 1928-1933., articleTitle=基于融合方法预测锂离子电池剩余寿命, refAbstract=null), Reference(id=1218843918746174221, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=5, pageStart=1928, pageEnd=1933, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Lin Na, Zhu Wu, Deng Anquan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Lin Na, Zhu Wu, Deng Anquan. Remaining useful life prediction of the lithium-ion battery based on fusion method[J]. Science Technology and Engineering, 2020, 20(5): 1928-1933., articleTitle=Remaining useful life prediction of the lithium-ion battery based on fusion method, refAbstract=null), Reference(id=1218843918884586262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=50, issue=10, pageStart=66, pageEnd=72, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=李凯, 胡丽, 宋婷婷, journalName=山东电力技术, refType=null, unstructuredReference=李凯, 胡丽, 宋婷婷. 基于CNN-Bi-LSTM的锂离子电池健康状态估算[J]. 山东电力技术, 2023, 50(10): 66-72., articleTitle=基于CNN-Bi-LSTM的锂离子电池健康状态估算, refAbstract=null), Reference(id=1218843919006221084, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=50, issue=10, pageStart=66, pageEnd=72, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Li Kai, Hu Li, Song Tingting, journalName=Shandong Electric Power, refType=null, unstructuredReference=Li Kai, Hu Li, Song Tingting. Health state estimation of lithium-ion batteries based on CNN-Bi-LSTM[J]. Shandong Electric Power, 2023, 50(10): 66-72., articleTitle=Health state estimation of lithium-ion batteries based on CNN-Bi-LSTM, refAbstract=null), Reference(id=1218843919144633121, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2020, volume=32, issue=null, pageStart=101741, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=Fan Y, Xiao F, Li C, journalName=Journal of Energy Storage, refType=null, unstructuredReference=Fan Y, Xiao F, Li C, et al. A novel deep learning framework for state of health estimation of lithium-ion battery[J]. Journal of Energy Storage, 2020, 32: 101741., articleTitle=A novel deep learning framework for state of health estimation of lithium-ion battery, refAbstract=null), Reference(id=1218843919245296422, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2020, volume=459, issue=null, pageStart=228069, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=Li P, Zhang Z, Xiong Q, journalName=Journal of Power Sources, refType=null, unstructuredReference=Li P, Zhang Z, Xiong Q, et al. State-of-health estimation and remaining useful life prediction for the lithium-ion battery based on a variant long short term memory neural network[J]. Journal of Power Sources, 2020, 459: 228069., articleTitle=State-of-health estimation and remaining useful life prediction for the lithium-ion battery based on a variant long short term memory neural network, refAbstract=null), Reference(id=1218843919366931242, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2018, volume=384, issue=null, pageStart=387, pageEnd=395, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=Yang D, Zhang X, Pan R, journalName=Journal of Power Sources, refType=null, unstructuredReference=Yang D, Zhang X, Pan R, et al. A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve[J]. Journal of Power Sources, 2018, 384: 387-395., articleTitle=A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve, refAbstract=null), Reference(id=1218843919492760370, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2019, volume=176, issue=null, pageStart=91, pageEnd=102, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=Deng Y, Ying H, E J, journalName=Energy, refType=null, unstructuredReference=Deng Y, Ying H, E J, et al. Feature parameter extraction and intelligent estimation of the state-of-health of lithium-ion batteries[J]. Energy, 2019, 176: 91-102., articleTitle=Feature parameter extraction and intelligent estimation of the state-of-health of lithium-ion batteries, refAbstract=null), Reference(id=1218843919614395191, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=29, pageStart=12900, pageEnd=12908, url=null, language=null, rfNumber=[9], rfOrder=11, authorNames=郝可青, 吕志刚, 邸若海, journalName=科学技术与工程, refType=null, unstructuredReference=郝可青, 吕志刚, 邸若海, 等. 基于鲸鱼算法优化长短时记忆神经网络的锂电池剩余寿命预[J]. 科学技术与工程, 2022, 22(29): 12900-12908., articleTitle=基于鲸鱼算法优化长短时记忆神经网络的锂电池剩余寿命预, refAbstract=null), Reference(id=1218843919727641404, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=29, pageStart=12900, pageEnd=12908, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=Hao Keqing, Lü Zhigang, Di Ruohai, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Hao Keqing, Zhigang, Di Ruohai, et al. Remaining useful life prediction of lithium battery based on long short-term memory optimized by whale optimization algorithm[J]. Science Technology and Engineering, 2022, 22(29): 12900-12908., articleTitle=Remaining useful life prediction of lithium battery based on long short-term memory optimized by whale optimization algorithm, refAbstract=null), Reference(id=1218843919849276226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=7, pageStart=902, pageEnd=906, url=null, language=null, rfNumber=[10], rfOrder=13, authorNames=张岸, 杨春德, journalName=电源技术, refType=null, unstructuredReference=张岸, 杨春德. 基于GAN-CNN-LSTM的锂电池SOH估计[J]. 电源技术, 2021, 45(7): 902-906., articleTitle=基于GAN-CNN-LSTM的锂电池SOH估计, refAbstract=null), Reference(id=1218843919941550918, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=7, pageStart=902, pageEnd=906, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=Zhang An, Yang Chunde, journalName=Chinese Journal of Power Sources, refType=null, unstructuredReference=Zhang An, Yang Chunde. SOH estimation of lithium batteries based on GAN-CNN-LSTM[J]. Chinese Journal of Power Sources, 2021, 45(7): 902-906., articleTitle=SOH estimation of lithium batteries based on GAN-CNN-LSTM, refAbstract=null), Reference(id=1218843920079962958, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=43, issue=7, pageStart=985, pageEnd=994, url=null, language=null, rfNumber=[11], rfOrder=15, authorNames=史永胜, 施梦琢, 丁恩松, journalName=工程科学学报, refType=null, unstructuredReference=史永胜, 施梦琢, 丁恩松, 等. 基于CEEMDAN-LSTM组合的锂离子电池寿命预测方法[J]. 工程科学学报, 2021, 43(7): 985-994., articleTitle=基于CEEMDAN-LSTM组合的锂离子电池寿命预测方法, refAbstract=null), Reference(id=1218843920197403477, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=43, issue=7, pageStart=985, pageEnd=994, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=Shi Yongsheng, Shi Mengzhuo, Ding Ensong, journalName=Chinese Journal of Engineering, refType=null, unstructuredReference=Shi Yongsheng, Shi Mengzhuo, Ding Ensong, et al. Combined prediction method of lithium-ion battery life based on CEEMDAN-LSTM[J]. Chinese Journal of Engineering, 2021, 43(7): 985-994., articleTitle=Combined prediction method of lithium-ion battery life based on CEEMDAN-LSTM, refAbstract=null), Reference(id=1218843920302261077, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2022, volume=10, issue=null, pageStart=89402, pageEnd=89413, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=Ding G, Wang W, Zhu T, journalName=IEEE Access, refType=null, unstructuredReference=Ding G, Wang W, Zhu T. Remaining useful life prediction for lithium-ion batteries based on CS-VMD and GRU[J]. IEEE Access, 2022, 10: 89402-89413., articleTitle=Remaining useful life prediction for lithium-ion batteries based on CS-VMD and GRU, refAbstract=null), Reference(id=1218843920419701596, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2019, volume=84, issue=null, pageStart=105739, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=Zhu Q, Zhang F, Liu S, journalName=Applied Soft Computing, refType=null, unstructuredReference=Zhu Q, Zhang F, Liu S, et al. A hybrid VMD-BiGRU model for rubber futures time series forecasting[J]. Applied Soft Computing, 2019, 84: 105739., articleTitle=A hybrid VMD-BiGRU model for rubber futures time series forecasting, refAbstract=null), Reference(id=1218843920579085162, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=1, pageStart=99, pageEnd=106, url=null, language=null, rfNumber=[14], rfOrder=19, authorNames=伍济钢, 文港, journalName=航天器环境工程, refType=null, unstructuredReference=伍济钢, 文港. 基于优化多尺度排列熵和卷积神经网络的滚动轴承故障诊断方法[J]. 航天器环境工程, 2023, 40(1): 99-106., articleTitle=基于优化多尺度排列熵和卷积神经网络的滚动轴承故障诊断方法, refAbstract=null), Reference(id=1218843920688137070, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=1, pageStart=99, pageEnd=106, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=Wu Jigang, Wen Gang, journalName=Spacecraft Environment Engineering, refType=null, unstructuredReference=Wu Jigang, Wen Gang. Rolling bearing fault diagnosis method based on optimized multi-scale permutation entropy and convolutional neural network[J]. Spacecraft Environment Engineering, 2023, 40(1): 99-106., articleTitle=Rolling bearing fault diagnosis method based on optimized multi-scale permutation entropy and convolutional neural network, refAbstract=null), Reference(id=1218843920801383282, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2019, volume=7, issue=null, pageStart=87178, pageEnd=87191, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=Qu J, Liu F, Ma Y, journalName=IEEE Access, refType=null, unstructuredReference=Qu J, Liu F, Ma Y, et al. A neural-network-based method for RUL prediction and SOH monitoring of lithium-ion battery[J]. IEEE Access, 2019, 7: 87178-87191., articleTitle=A neural-network-based method for RUL prediction and SOH monitoring of lithium-ion battery, refAbstract=null), Reference(id=1218843920889463674, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2022, volume=520, issue=null, pageStart=230823, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=22, authorNames=Han T, Wang Z, Meng H, journalName=Journal of Power Sources, refType=null, unstructuredReference=Han T, Wang Z, Meng H. 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: 230823., articleTitle=End-to-end capacity estimation of lithium-ion batteries with an enhanced long short-term memory network considering domain adaptation, refAbstract=null), Reference(id=1218843920981738368, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=8, pageStart=3387, pageEnd=3395, url=null, language=null, rfNumber=[17], rfOrder=23, authorNames=吴昉, 王伟, 刘卫朋, journalName=科学技术与工程, refType=null, unstructuredReference=吴昉, 王伟, 刘卫朋. 结合注意力机制和卷积神经网络的X射线焊缝缺陷检测[J]. 科学技术与工程, 2023, 23(8): 3387-3395., articleTitle=结合注意力机制和卷积神经网络的X射线焊缝缺陷检测, refAbstract=null), Reference(id=1218843921082401670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=8, pageStart=3387, pageEnd=3395, url=null, language=null, rfNumber=[17], rfOrder=24, authorNames=Wu Fang, Wang Wei, Liu Weipeng, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wu Fang, Wang Wei, Liu Weipeng. X-ray weld defect detection combined with attention mechanism and convolutional neural network[J]. Science Technology and Engineering, 2023, 23(8): 3387-3395., articleTitle=X-ray weld defect detection combined with attention mechanism and convolutional neural network, refAbstract=null), Reference(id=1218843921178870668, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=4, pageStart=95, pageEnd=100, url=null, language=null, rfNumber=[18], rfOrder=25, authorNames=郭海龙, 杨康, 吉龙军, journalName=三峡大学学报(自然科学版), refType=null, unstructuredReference=郭海龙, 杨康, 吉龙军, 等. 基于注意力机制的改进LSTM锂电池健康状态估计方法[J]. 三峡大学学报(自然科学版), 2023, 45(4): 95-100., articleTitle=基于注意力机制的改进LSTM锂电池健康状态估计方法, refAbstract=null), Reference(id=1218843921292116881, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=4, pageStart=95, pageEnd=100, url=null, language=null, rfNumber=[18], rfOrder=26, authorNames=Guo Hailong, Yang Kang, Ji Longjun, journalName=Journal of China Three Gorges University(Natural Sciences), refType=null, unstructuredReference=Guo Hailong, Yang Kang, Ji Longjun, et al. Method of improved LSTM lithium battery health estimation based on attention mechanism[J]. Journal of China Three Gorges University(Natural Sciences), 2023, 45(4): 95-100., articleTitle=Method of improved LSTM lithium battery health estimation based on attention mechanism, refAbstract=null), Reference(id=1218843921396974481, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2022, volume=308, issue=null, pageStart=118348, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=Khaleghi S, Hosen M S, Karimi D, journalName=Applied Energy, refType=null, unstructuredReference=Khaleghi S, Hosen M S, Karimi D, et al. Developing an online data-driven approach for prognostics and health management of lithium-ion batteries[J]. Applied Energy, 2022, 308: 118348., articleTitle=Developing an online data-driven approach for prognostics and health management of lithium-ion batteries, refAbstract=null), Reference(id=1218843921518609303, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=229, issue=null, pageStart=120682, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=Zang H, Xu R, Cheng L, journalName=Energy, refType=null, unstructuredReference=Zang H, Xu R, Cheng L, et al. Residential load forecasting based on LSTM fusing self-attention mechanism with pooling[J]. Energy, 2021, 229: 120682., articleTitle=Residential load forecasting based on LSTM fusing self-attention mechanism with pooling, refAbstract=null), Reference(id=1218843921644438430, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, doi=null, pmid=null, pmcid=null, year=2021, volume=184, issue=null, pageStart=109935, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=29, authorNames=Li X, Ma Y, Zhu J, journalName=Measurement, refType=null, unstructuredReference=Li X, Ma Y, Zhu J. An online dual filters RUL prediction method of lithium-ion battery based on unscented particle filter and least squares support vector machine[J]. Measurement, 2021, 184: 109935., articleTitle=An online dual filters RUL prediction method of lithium-ion battery based on unscented particle filter and least squares support vector machine, refAbstract=null)], funds=[Fund(id=1218843917777289936, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, awardId=61903244, language=CN, fundingSource=国家自然科学基金(61903244), fundOrder=null, country=null), Fund(id=1218843917856981719, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, awardId=H2019-269, language=CN, fundingSource=上海辰仕科技发展有限公司一般企事业单位资助项目(H2019-269), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1218843912190476560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, xref=null, ext=[AuthorCompanyExt(id=1218843912207253780, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China), AuthorCompanyExt(id=1218843912224030998, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, companyId=1218843912190476560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海电力大学,数理学院智慧能源数学研究中心, 上海 201306)])], figs=[ArticleFig(id=1218843914891608562, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.1, caption=Structure unit of LSTM, figureFileSmall=ciovd1LX0Ss0NLbOuaS+Pg==, figureFileBig=tGNnAW/29So5utqQVaHFSw==, tableContent=null), ArticleFig(id=1218843915009049083, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图1, caption=LSTM的结构单元

ft为当前时刻遗忘门的状态;it为当前时刻输入门状态;Ott时刻输出门的信息选择程度; C ~ t为记忆细胞的历史选择信息;Ctt时刻的记忆细胞;Ct-1t-1时刻的记忆细胞;ht-1为前一个输出的隐藏状态;xt当前的输出;Wb分别为相关单元状态的权重和偏置

, figureFileSmall=ciovd1LX0Ss0NLbOuaS+Pg==, figureFileBig=tGNnAW/29So5utqQVaHFSw==, tableContent=null), ArticleFig(id=1218843915168432650, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.2, caption=BiLSTM network structure, figureFileSmall=j+Zz6VTyp73GznYAnpD0nw==, figureFileBig=e3KSr+j3ZslsFn7gc2Ut7A==, tableContent=null), ArticleFig(id=1218843915323621907, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图2, caption=BiLSTM网络结构, figureFileSmall=j+Zz6VTyp73GznYAnpD0nw==, figureFileBig=e3KSr+j3ZslsFn7gc2Ut7A==, tableContent=null), ArticleFig(id=1218843915441062429, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.3, caption=Attention mechanism structure diagram, figureFileSmall=bTL4/WFBeHeqj1lCcAcS+w==, figureFileBig=wIiYbuAaqKKxkSFbaHRNZw==, tableContent=null), ArticleFig(id=1218843915524948514, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图3, caption=注意力机制结构图, figureFileSmall=bTL4/WFBeHeqj1lCcAcS+w==, figureFileBig=wIiYbuAaqKKxkSFbaHRNZw==, tableContent=null), ArticleFig(id=1218843915625611820, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.4, caption=Battery capacity degradation curve, figureFileSmall=eUayGIsza6tvzYLrNBf/Iw==, figureFileBig=OebRsgJhPKIGbstbS6M2Gg==, tableContent=null), ArticleFig(id=1218843915755635258, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图4, caption=电池容量退化曲线, figureFileSmall=eUayGIsza6tvzYLrNBf/Iw==, figureFileBig=OebRsgJhPKIGbstbS6M2Gg==, tableContent=null), ArticleFig(id=1218843915881464387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.5, caption=Modal components obtained from VMD decomposition of CS2_36 battery, figureFileSmall=1qHE8PP9US6SqleyBxNHTQ==, figureFileBig=kcDd/oY7tBUZtbjc/gDSeQ==, tableContent=null), ArticleFig(id=1218843916024070732, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图5, caption=CS2_36号电池VMD分解得到的模态分量, figureFileSmall=1qHE8PP9US6SqleyBxNHTQ==, figureFileBig=kcDd/oY7tBUZtbjc/gDSeQ==, tableContent=null), ArticleFig(id=1218843916166677077, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.6, caption=Estimation framework of the VMD-BiLSTM-ATT method, figureFileSmall=gOhuHduayo1LWceuJuNguw==, figureFileBig=bknZCEl092gysF3iapot8Q==, tableContent=null), ArticleFig(id=1218843916284117604, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图6, caption=VMD-BiLSTM-ATT方法估计框架, figureFileSmall=gOhuHduayo1LWceuJuNguw==, figureFileBig=bknZCEl092gysF3iapot8Q==, tableContent=null), ArticleFig(id=1218843916435112558, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.7, caption=Battery state of health estimation results, figureFileSmall=8M0pi+n2m4oya+zstXJDiw==, figureFileBig=ezbuktS4kV6yz77MNa/brA==, tableContent=null), ArticleFig(id=1218843916535775865, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图7, caption=电池SOH估计结果, figureFileSmall=8M0pi+n2m4oya+zstXJDiw==, figureFileBig=ezbuktS4kV6yz77MNa/brA==, tableContent=null), ArticleFig(id=1218843916653216384, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Fig.8, caption=Battery state of health estimation error, figureFileSmall=gxFVw0wHPSXBlv4IgkxN2Q==, figureFileBig=ESvvzPIorAUlbxFF1TwSQA==, tableContent=null), ArticleFig(id=1218843916825182861, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=图8, caption=电池SOH估计误差, figureFileSmall=gxFVw0wHPSXBlv4IgkxN2Q==, figureFileBig=ESvvzPIorAUlbxFF1TwSQA==, tableContent=null), ArticleFig(id=1218843916925846164, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Table 1, caption=

Normalized permutation entropy values for components of CS2_38 battery

, figureFileSmall=null, figureFileBig=null, tableContent=
分量 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6
PE值 0.435 0.755 0.916 0.982 0.989 0.905
), ArticleFig(id=1218843917060063904, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=表1, caption=

CS2_38号电池各分量归一化后的排列熵值

, figureFileSmall=null, figureFileBig=null, tableContent=
分量 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6
PE值 0.435 0.755 0.916 0.982 0.989 0.905
), ArticleFig(id=1218843917181698729, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Table 2, caption=

Reconstruction results for components of CS2_38 battery

, figureFileSmall=null, figureFileBig=null, tableContent=
序列 L1 L2 L3 L4
分量 IMF1 IMF2 IMF3 IMF4+IMF5+IMF6
), ArticleFig(id=1218843917282362035, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=表2, caption=

CS2_38号电池各分量重构结果

, figureFileSmall=null, figureFileBig=null, tableContent=
序列 L1 L2 L3 L4
分量 IMF1 IMF2 IMF3 IMF4+IMF5+IMF6
), ArticleFig(id=1218843917408191161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=EN, label=Table 3, caption=

Comparative results of evaluation metrics for different methods

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 CS2_36 CS2_38 CX2_35
RMSE MAE RMSE MAE RMSE MAE
LSTM 0.026 0 0.016 6 0.029 3 0.020 0 0.013 3 0.008 6
BiLSTM 0.012 9 0.010 3 0.018 8 0.014 6 0.010 3 0.007 4
BiLSTM-
ATT
0.009 3 0.006 4 0.011 9 0.009 1 0.008 8 0.006 8
本文方法 0.005 3 0.003 8 0.004 7 0.002 7 0.003 6 0.002 8
), ArticleFig(id=1218843917542408897, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776901675377370, language=CN, label=表3, caption=

不同方法评估指标的对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 CS2_36 CS2_38 CX2_35
RMSE MAE RMSE MAE RMSE MAE
LSTM 0.026 0 0.016 6 0.029 3 0.020 0 0.013 3 0.008 6
BiLSTM 0.012 9 0.010 3 0.018 8 0.014 6 0.010 3 0.007 4
BiLSTM-
ATT
0.009 3 0.006 4 0.011 9 0.009 1 0.008 8 0.006 8
本文方法 0.005 3 0.003 8 0.004 7 0.002 7 0.003 6 0.002 8
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2309072, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2309072, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2309072, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2309072, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于变分模态分解和BiLSTM-ATT的锂电池健康状态估计模型
收藏切换
PDF下载
胡平生 , 吴泉军 *
科学技术与工程 | 论文·电工技术 2025,25(11): 4598-4604
收起
收藏切换
科学技术与工程 | 论文·电工技术 2025, 25(11): 4598-4604
基于变分模态分解和BiLSTM-ATT的锂电池健康状态估计模型
全屏
胡平生 , 吴泉军*
作者信息
  • 上海电力大学,数理学院智慧能源数学研究中心, 上海 201306
  • 胡平生(1996—),男,汉族,安徽合肥人,硕士研究生。研究方向:深度学习、新能源电池的管理系统。E-mail:

通讯作者:

* 吴泉军(1982—),男,汉族,山东汶上人,博士,副教授。研究方向:锂电池健康监测。E-mail:
Estimation Model for State of Health of Lithium-ion Battery Based on VMD and BiLSTM-ATT
Ping-sheng HU , Quan-jun WU*
Affiliations
  • Smart Energy Mathematics Research Center of College of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 201306, China
出版时间: 2025-04-18 doi: 10.12404/j.issn.1671-1815.2309072
文章导航
收藏切换

锂离子电池健康状态(state of health,SOH)估计对于保证锂离子电池管理系统的安全稳定运行至关重要。然而,由于锂离子电池在放电过程中存在容量再生现象,SOH的准确估计一直是一个挑战。为了提高估计精度,提出了一种基于变分模态分解(variational mode decomposition,VMD)和双向长短期记忆网络-注意力机制(bidirectional long short term memory-attention,BiLSTM-ATT)的混合模型估计方法。首先,采用VMD分解算法对锂电池容量进行分解,得到一组相对稳定的子序列。为了降低后续的计算规模,通过引入了排列熵的方法对各个子序列进行重构。然后,将重构后的序列输入到BiLSTM-ATT模型中,利用注意力机制来分配隐藏层的特征权重,并通过双向长短期记忆网络(bidirectional long short term memory, BiLSTM)模型对SOH值进行训练和估计。最后,将所有估计值进行相加得到完整的SOH估计结果。通过在CALCE锂电池数据集上的CS2_36、CS2_38和CX2_35进行验证,实验结果表明所提出算法的均方根误差始终保持在0.6%以内,平均绝对误差始终保持在0.4%以内,相比其他估计模型表现出更高的精度和性能。

锂离子电池  /  健康状态  /  变分模态分解  /  排列熵  /  注意力机制  /  双向长短期记忆网络

The estimation of the state of health (SOH) for lithium-ion batteries is considered crucial for ensuring the safe and stable operation of battery management system. However, the accurate estimation of SOH has been a challenge due to the capacity regeneration phenomenon during the discharge process of lithium-ion batteries. To improve estimation accuracy, a hybrid model based on variational mode decomposition (VMD) and bidirectional long short-term memory network with attention mechanism (BiLSTM-ATT) was proposed. First, the battery capacity was decomposed using the VMD algorithm, producing a set of stable sub-sequences. Then, permutation entropy was introduced to reconstruct the sub-sequences to reduce computational complexity. The reconstructed sequences were input into the BiLSTM-ATT model, and feature weights were assigned by the attention mechanism. The SOH values were trained and estimated by the BiLSTM model. Finally, the complete SOH estimation result was obtained by summing all estimated values. Validation was performed using the CS2_36, CS2_38, and CX2_35 datasets from the CALCE lithium battery dataset. The results show that the proposed algorithm maintains a root mean square error within 0.6% and a mean absolute error within 0.4%, which demonstrates higher accuracy and performance compared to other estimation models.

lithium-ion battery  /  state of health  /  variational mode decomposition  /  permutation entropy  /  attention mechanism  /  bidirectional long short-term memory
胡平生, 吴泉军. 基于变分模态分解和BiLSTM-ATT的锂电池健康状态估计模型. 科学技术与工程, 2025 , 25 (11) : 4598 -4604 . DOI: 10.12404/j.issn.1671-1815.2309072
Ping-sheng HU, Quan-jun WU. Estimation Model for State of Health of Lithium-ion Battery Based on VMD and BiLSTM-ATT[J]. Science Technology and Engineering, 2025 , 25 (11) : 4598 -4604 . DOI: 10.12404/j.issn.1671-1815.2309072
为了准确稳定地估计电池的电池健康状态(state of health,SOH),锂电池研究领域提出了多种方法。目前,锂离子电池SOH估计算法主要分为两大类:基于模型的方法和基于数据驱动的方法[1-2]。基于模型的方法主要包括等效电路模型、电化学模型和经验降解模型。然而,基于模型的方法原理复杂,易受外界因素影响,难以建立稳定的模型[3]。值得注意的是,基于数据驱动的方法无需进行复杂的建模和内在机理分析,而是直接根据历史数据进行建模,由于其灵活性和易操作性,数据驱动的方法在该领域内受到了广泛关注[4]
目前数据驱动的方法主要包括高斯过程回归(gaussian process regression,GPR)、支持向量回归(support vector regression,SVR)和深度学习方法等[5-6]。Yang等[7]在输入变量的相似性度量和协方差函数设计等方面改进了GPR模型,提高了SOH估计精度。Deng等[8]提出用最小二乘法支持向量机结合灰色关联分析来估计SOH。与传统的机器学习方法相比,深度学习技术采用多层次的方法从原始输入中逐步提取更高层次的特征。郝可青等[9]采用鲸鱼优化算法对长短期记忆网络(long short term memory,LSTM)进行参数优化,提高了模型的精确度。张岸等[10]采用生成对抗网络(generative adversarial network,GNN)对原始数据进行数据增强,利用卷积神经网络(convolutional neural network,CNN)和LSTM对数据进行特征提取并预测。
与上述方法有所不同的是,在锂离子电池健康状态估计的实际建模中,本文中考虑了锂电池容量退化序列的非平稳性,即容量再生现象的存在;提出了一种基于变分模态分解(variational mode decomposition,VMD)算法和双向长短期记忆网络-注意力机制(bidirectional long short term memory-attention, BiLSTM-ATT)的电池SOH估计方法。该方法采用VMD分解算法结合排列熵对预测的电池SOH进行模态分解,以降低了容量再生现象带来的噪声影响。此外,通过结合双向长短期记忆网络(bidirectional long short term memory,BiLSTM)和注意力机制,对BiLSTM隐藏层的权重结构进行注意力分数计算,以增强对不同权重特征的学习能力,进一步提高方法的估计性能。
锂离子电池的寿命衰减是多种机理相互作用的过程,如电化学反应、材料膨胀、温度变化等,从而导致电池的容量变化的非线性和非平稳变化,传统的预测算法往往难以准确估计预测这种复杂情况[11]。经验模态分解(empirical mode decomposition,EMD)、集合经验模态分解(ensemble empirical mode decomposition,EEMD)和 完全自适应噪声集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)等算法能够自适应地将复杂的电池容量序列数据划分为一系列的本征模态函数(intrinsic mode function,IMF)。然而,它们缺乏坚实的理论基础,导致在处理高度复杂序列时存在欠分解和过分解的问题[12]。为了解决这个问题,后来提出了变分模态分解算法。VMD算法是基于变分原理的信号分解方法,通过迭代地提取模态函数和包络函数,将信号分解成一组相对平稳的子信号。相比之下,VMD算法具有良好的理论基础和噪声鲁棒性。它在提取复杂序列波动模式和提高预测模型精度方面具有显著的优势。VMD算法理论参考文献[13]。
由于锂离子电池容量变化呈现非平稳性特征, VMD分解后会得到一系列的IMF。然而,过多的IMF分量明显增加了模型的训练难度,同时降低了预测精度。为了解决这个问题,提出一种基于排列熵(permutation entropy,PE)的IMF重构策略[14],排列熵是用于量化IMF分量的复杂程度,其计算步骤如下:
(1) 相空间重构。假设VMD分解后的分量为x(i)(i=1,2,…,N),对其相空间进行重构处理,得到矩阵Y,表达式为
Y= x ( 1 ) x ( 1 + t ) x [ 1 + ( m - 1 ) t ] x ( 2 ) x ( 2 + t ) x [ 2 + ( m - 1 ) t ] x ( K ) x ( K + t ) x [ K + ( m - 1 ) t ]
式(1)中:m为嵌入维数;t为延迟时间;其中K=N-(m-1)t,表示重构矩阵的行数。
(2)提取符号序列。矩阵Y中的每一行向量定义为重构向量S(g),共有K个重构分量。将每一个重构分量按照其元素值从小到大排列。
S(g)={i1,i2,…,im}
式(2)中:g=1,2…,K,Km,i1,i2,…,imS(g)每个元素的列序号。
(3)求符号序列出现概率。计算每一种符号序列出现的概率,即{P1,P2,…,PK},得到 g = 1 KPg=1。
(4) 根据Shannon熵的计算公式,可将x(i)的排列熵定义为
Hp=-( g = 1 Kpglnpg)
(5) 归一化处理。为了方便实际应用,通常将排列熵进行归一化处理,以便更好地表示,可表示为
H=Hp/ln(m!)
式(4)中:H取值范围[0,1],H的值表示时间序列的随机程度,H越小,代表序列越简单,反之,则表示序列越复杂。
当循环神经网络(recursive neural network,RNN)处理时间序列时,神经元的部分输出可以作为输入再次传递给神经单元,这样可以有效地利用先前的信息。然而,RNN的记忆和存储能力是有限的,随着序列长度的增加,RNN逐渐失去了对过去信息的学习能力,并且容易受到梯度消失和梯度爆炸问题的影响[15]。为了解决这一问题,LSTM引入了细胞状态和门控机制[16],增强了记忆学习能力,并解决了RNN递归神经网络中梯度消失和梯度爆炸的问题。LSTM的结构单元如图1所示。
LSTM模型由一系列相同的时序模块组成。每个模块包含三个门:遗忘门、输入门和输出门,三个控制门的计算均与前一个输出ht-1和当前的输出xt有关。具体计算方式如下。
遗忘门是LSTM模型中的关键组成部分,负责选择性遗忘前一时刻的细胞状态信息,以提高模型对时间序列中的长期依赖关系的捕捉能力,表达式为
ft=σ(Wf[ht-1,xt]+bf)
式(5)中:ft为当前时刻遗忘门的状态;Wfbf分别为遗忘门单元状态的权重和偏置;σ为sigmoid激活函数。
输入门负责更新细胞状态,将新的信息融入到当前时刻的状态中。计算公式为
it=σ(Wi[ht-1,xt]+bi)
C ~ t=tanh(Wc[ht-1,xt]+bc)
Ct=ftCt-1+it C ~ t
式中: i t为当前时刻输入门状态;Wibi分别为输入门的权重和偏置;Wcbc分别为输入单元状态的权重和输入单元状态的偏置; C ~ t为记忆细胞的历史选择信息;Ctt时刻的记忆细胞;Ct-1t-1时刻的记忆细胞;tanh为双曲正切激活函数;☉为哈达玛积运算。
输出门负责控制当前时刻的细胞状态Ct输出到LSTM单元中输出值ht的比例。计算公式为
Ot=σ(Wo[ht-1,xt]+bo)
ht=Ot☉tanh(Ct)
式中:Wobo分别为输出门单元状态的偏置和权重;Ott时刻输出门的信息选择程度。
考虑到时间序列数据中前后关联性的重要性,本文中引入了双向长短期记忆网络BiLSTM来进一步改进模型。相比于传统的单向LSTM,BiLSTM可以更全面地学习锂电池容量退化数据的时序关联信息。通过前向和后向的输入序列传递,BiLSTM能够更有效地捕捉序列中的前后依赖关系,从而提高估计性能,其网络结构如图2所示。
该结构采用式(11)和式(12)独立计算前向传播和反向传播的隐藏层状态,最后通过式(13)将两个方向的输出拼接起来作为该网络的最终输出。
h t=LSTM(Xt, h t - 1)
h t=LSTM(Xt, h t + 1)
yt=Wt h t+Ut h t+bt
式中: h t h t分别为前向LSTM和反向LSTM输出的隐藏层状态; WtUt分别为对应的输出权重;yt为BiLSTM网络结构的输出状态;bt为偏置向量。
为了提高BiLSTM模型的估计性能,本文中引入了注意力机制来优化BiLSTM。注意力机制旨在通过注意力权重来捕捉最相关的信息,而非全局信息。在预测输出序列时,注意力机制将对BiLSTM模型的输入序列中的关键数据给予更多的关注,使网络能够有选择的专注于在特定的信息上[17-18],注意力机制的结构如图3所示。鉴于注意力机制与各种LSTM的集成在许多预测任务中以被证明能提高性能[19-20],所以将BiLSTM-ATT方法应用于锂离子电池SOH的估计中,以期进一步改进预测性能。
图3可以看出,输入序列是y1,y2,…,yt,对应的隐藏层状态值为h1,h2,…,ht,计算公式为
Sti=tanh(Wht+Uhi+b), i=1,2,…,t-1
ati= e x p ( S t i ) k = 1 t e x p ( S t k ), i=1,2,…,t-1
F= i = 1 tati×hi, i=1,2,…,t-1
h't=f(F,ht,yt)
式中: Sti为每个隐藏层对模型最终输出的注意力分数,通过Softmax函数将注意力分数转化为注意力权重,第i个元素的注意力权重为ati;F为注意力权重与对应的输入隐藏状态值hi的加权和;h't为最终的特征向量。WUb为模型的学习参数,这些参数随着模型训练过程不断更新。
本实验采用马里兰大学CALCE中心提供的锂离子电池老化数据集[21],其中包括CS系列和CX系列数据,选取标号为CS2_36、CS2_38和CX2_35的锂电池作为研究对象。如图4所示为本次实验所选的CALCE数据集的容量退化数据。
SOH是电池管理系统的一个关键参数,代表这电池老化的程度,SOH通常是用容量或阻抗表示。在本文中。容量表示为SOH,并且SOH的定义式为
SOH= C i C 0×100%
式(18)中:Ci为第i次循环测得的电池放电容量;C0为电池的额定容量。
原始电池容量退化序列中含有大量的噪声,给模型的预测估计带来了挑战。因此,本文中采用VMD分解算法对电池容量退化序列进行分解,以CS2_36号电池为例,通过手工调节参数,分解后的模态分量如图5所示。可以观察到各个IMF的分量的频率相对稳定,没有出现模态混叠的现象。其中IMF1代表锂电池容量衰减的分量,具有全局衰减趋势,而剩余分量可以看作局部容量再生现象的波动分量。
通过观察图5,可以发现VMD分解后的信号中,某些分量具有相似的波动频率。因此,为了更有效地预测电池的SOH,采用归一化后的PE值量化各个IMF分量的复杂程度,并将熵值相近的IMF分量进行重组,如表1所示,展示了各分量对应的归一化PE值,分析表1可以得出IMF4、IMF5和IMF6分量的PE值非常接近,因此将它们直接合并成一组,得到4组重构序列。这样的处理方式不仅减少了方法的运算量,还降低了运算误差。后面模型的预测次数也从6次减少到4次,有效提高了计算效率。这4组重组序列分别记为L1、L2、L3和L4,合并结果如表2所示。
现提出一种基于VMD-BiLSTM-ATT模型的锂离子电池SOH估计方法,其过程框架(图6)如下。
(1) 对电池容量退化数据进行预处理,并将其划分成训练集和测试集。利用VMD算法将数据分解为多个模态分量。
(2) 应用排列熵算法计算分解后模态的排列熵值,以评估各子序列的复杂程度,并根据复杂程度进行合并和重构。
(3) 将重构分量输入到多层的BiLSTM-ATT神经网络中进行训练和估计。
(4) 将每个重构分量的估计值相加求和,得到最终的SOH估计结果。
为了评估不同算法的估计性能,本次实验采用均方根误差(root mean square error, RMSE)和平均绝对误差(mean absolute error, MAE)作为评估标准,其计算公式分别为
RMSE= 1 n i = 1 n ( Y i - Y ' i ) 2
MAE= 1 n i = 1 n Y i - Y ' i
式中:YiY'i分别为第i个样本的真实值与预测值;n为样本的数量;RMSE、MAE的值越小,则表示模型的估计性能越好。
在锂离子电池的SOH估计阶段,本文中采用CALCE数据集CS系列(CS2_36、CS2_38)和CX系列(CX2_35)对不同估计方法进行训练和测试。为了体现本文所提出来的方法(VMD-BiLSTM-ATT)在估计锂离子电池SOH方面的性能,分别与LSTM模型、BiLSTM模型和BiLSTM-ATT模型进行对比实验。
选取50%的数据作为训练集,剩下50%的数据作为测试集。如图7所示为本文所选锂电池数据集上各种估计方法的估计结果。如图8所示为这些方法在这些数据集上的估计误差。分析图7图8可知,VMD-BiLSTM-ATT方法提供了最准确且误差最小的估计。单一的LSTM模型在不同电池上的泛化能力较差,并且误差较大。而BiLSTM考虑了更多的电池容量退化数据的前后关联信息,提高了估计精度,但效果仍不理想。BiLSTM-ATT相对于LSTM和BiLSTM的单一方法,具有相对有效的估计精度,但未考虑电池容量退化序列的波动性,估计的SOH误差波动较大。而VMD-BiLSTM-ATT方法表现出稳定的估计误差波动,能够持续准确地估计电池SOH,使估计误差控制在0.4%以内。
表3所示为不同方法下对CS2_36、CS2_38和CX2_35电池SOH估计结果的评价指标。RMSE和MAE越小,方法的估计精确越高。结果显示,本文所提出的方法在相同条件下的估计性能明显优于其他实验方法,具有能好的拟合能力。RMSE和MSE均有不同程度的降低,估计效果显著提升。此外,在三组电池的SOH数据上,BiLSTM的估计效果均优于LSTM方法,即BiLSTM方法可以综合考虑历史和未来信息,提高方法估计精度。而通过将BiLSTM-ATT与LSTM和BiLSTM单个模型方法对比,可以看出引入注意力机制可以将BiLSTM模型转化为一个注意力模型,通过对输入序列中每个位置的注意力进行学习,使模型更加关注与当前预测相关的特征,从而提高模型的性能。最后,将本文算法与BiLSTM-ATT方法进行对比可知,在三个数据集上,本文提出的方法的最大均方根误差在0.6%以内,最大的MAE在0.4%以内,表明本文方法具有最高的估计精度。相较于BiLSTM-ATT方法,三个数据集的RMSE和MAE平均值分别下降了0.005 5和0.004 3。这是因为本文中引入VMD分解算法对电池容量退化序列进行分解,降低了噪声,并利用排列熵算法计算排列熵值,对分解后的序列进行重构,避免了多次预测造成的误差累积。综上所述,本文提出的方法在锂电池健康状态估计中展现了良好的估计效果,并具有一定的优越性。
锂离子电池的容量衰退是实际应用中至关重要的课题。为了提升锂离子电池SOH估计的准确性,本文提出了一种基于VMD和BiLSTM-ATT的SOH估计方法。通过理论分析和实验验证,得出了以下结论。
(1)针对锂离子电池在放电过程中出现的容量再生现象,采用VMD算法对容量衰退曲线进行分解,有效减少了数据的不平稳性,从而提升了模型输入数据的质量。
(2)引入排列熵算法后,能够对具有相似排列熵值的子序列进行重构。这种处理不仅减少了模型的计算量,提升了计算效率,还有效降低了计算误差。
(3)将重构后的序列输入到BiLSTM-ATT组合模型中,增强了模型对关键特征的关注,弥补了单一LSTM等神经网络在捕捉数据潜在有效特征方面的不足。
(4)在不同锂电池数据集上的实验结果表明,所提出的方法在SOH估计精度方面表现出色,均方根误差不超过0.6%,平均绝对误差不超过0.4%,具有较高的实际应用价值。
  • 国家自然科学基金(61903244)
  • 上海辰仕科技发展有限公司一般企事业单位资助项目(H2019-269)
参考文献 引证文献
排序方式:
[1]
Richardson R R, Osborne M A, Howey D A. Gaussian process regression for forecasting battery state of health[J]. Journal of Power Sources, 2017, 357: 209-219.
[2]
王志福, 杨忠义, 罗崴, 等. 基于数据驱动的锂离子动力电池剩余使用寿命预测方法综述[J]. 科学技术与工程, 2023, 23(15): 6279-6289.
Wang Zhifu, Yang Zhongyi, Luo Wei, et al. Review of remaining useful life prediction for lithium-ion battery based on data-driven[J]. Science Technology and Engineering, 2023, 23(15): 6279-6289.
[3]
林娜, 朱武, 邓安全. 基于融合方法预测锂离子电池剩余寿命[J]. 科学技术与工程, 2020, 20(5): 1928-1933.
Lin Na, Zhu Wu, Deng Anquan. Remaining useful life prediction of the lithium-ion battery based on fusion method[J]. Science Technology and Engineering, 2020, 20(5): 1928-1933.
[4]
李凯, 胡丽, 宋婷婷. 基于CNN-Bi-LSTM的锂离子电池健康状态估算[J]. 山东电力技术, 2023, 50(10): 66-72.
Li Kai, Hu Li, Song Tingting. Health state estimation of lithium-ion batteries based on CNN-Bi-LSTM[J]. Shandong Electric Power, 2023, 50(10): 66-72.
[5]
Fan Y, Xiao F, Li C, et al. A novel deep learning framework for state of health estimation of lithium-ion battery[J]. Journal of Energy Storage, 2020, 32: 101741.
[6]
Li P, Zhang Z, Xiong Q, et al. State-of-health estimation and remaining useful life prediction for the lithium-ion battery based on a variant long short term memory neural network[J]. Journal of Power Sources, 2020, 459: 228069.
[7]
Yang D, Zhang X, Pan R, et al. A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve[J]. Journal of Power Sources, 2018, 384: 387-395.
[8]
Deng Y, Ying H, E J, et al. Feature parameter extraction and intelligent estimation of the state-of-health of lithium-ion batteries[J]. Energy, 2019, 176: 91-102.
[9]
郝可青, 吕志刚, 邸若海, 等. 基于鲸鱼算法优化长短时记忆神经网络的锂电池剩余寿命预[J]. 科学技术与工程, 2022, 22(29): 12900-12908.
Hao Keqing, Zhigang, Di Ruohai, et al. Remaining useful life prediction of lithium battery based on long short-term memory optimized by whale optimization algorithm[J]. Science Technology and Engineering, 2022, 22(29): 12900-12908.
[10]
张岸, 杨春德. 基于GAN-CNN-LSTM的锂电池SOH估计[J]. 电源技术, 2021, 45(7): 902-906.
Zhang An, Yang Chunde. SOH estimation of lithium batteries based on GAN-CNN-LSTM[J]. Chinese Journal of Power Sources, 2021, 45(7): 902-906.
[11]
史永胜, 施梦琢, 丁恩松, 等. 基于CEEMDAN-LSTM组合的锂离子电池寿命预测方法[J]. 工程科学学报, 2021, 43(7): 985-994.
Shi Yongsheng, Shi Mengzhuo, Ding Ensong, et al. Combined prediction method of lithium-ion battery life based on CEEMDAN-LSTM[J]. Chinese Journal of Engineering, 2021, 43(7): 985-994.
[12]
Ding G, Wang W, Zhu T. Remaining useful life prediction for lithium-ion batteries based on CS-VMD and GRU[J]. IEEE Access, 2022, 10: 89402-89413.
[13]
Zhu Q, Zhang F, Liu S, et al. A hybrid VMD-BiGRU model for rubber futures time series forecasting[J]. Applied Soft Computing, 2019, 84: 105739.
[14]
伍济钢, 文港. 基于优化多尺度排列熵和卷积神经网络的滚动轴承故障诊断方法[J]. 航天器环境工程, 2023, 40(1): 99-106.
Wu Jigang, Wen Gang. Rolling bearing fault diagnosis method based on optimized multi-scale permutation entropy and convolutional neural network[J]. Spacecraft Environment Engineering, 2023, 40(1): 99-106.
[15]
Qu J, Liu F, Ma Y, et al. A neural-network-based method for RUL prediction and SOH monitoring of lithium-ion battery[J]. IEEE Access, 2019, 7: 87178-87191.
[16]
Han T, Wang Z, Meng H. 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: 230823.
[17]
吴昉, 王伟, 刘卫朋. 结合注意力机制和卷积神经网络的X射线焊缝缺陷检测[J]. 科学技术与工程, 2023, 23(8): 3387-3395.
Wu Fang, Wang Wei, Liu Weipeng. X-ray weld defect detection combined with attention mechanism and convolutional neural network[J]. Science Technology and Engineering, 2023, 23(8): 3387-3395.
[18]
郭海龙, 杨康, 吉龙军, 等. 基于注意力机制的改进LSTM锂电池健康状态估计方法[J]. 三峡大学学报(自然科学版), 2023, 45(4): 95-100.
Guo Hailong, Yang Kang, Ji Longjun, et al. Method of improved LSTM lithium battery health estimation based on attention mechanism[J]. Journal of China Three Gorges University(Natural Sciences), 2023, 45(4): 95-100.
[19]
Khaleghi S, Hosen M S, Karimi D, et al. Developing an online data-driven approach for prognostics and health management of lithium-ion batteries[J]. Applied Energy, 2022, 308: 118348.
[20]
Zang H, Xu R, Cheng L, et al. Residential load forecasting based on LSTM fusing self-attention mechanism with pooling[J]. Energy, 2021, 229: 120682.
[21]
Li X, Ma Y, Zhu J. An online dual filters RUL prediction method of lithium-ion battery based on unscented particle filter and least squares support vector machine[J]. Measurement, 2021, 184: 109935.
2025年第25卷第11期
PDF下载
377
155
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2309072
  • 接收时间:2024-11-19
  • 首发时间:2025-07-09
  • 出版时间:2025-04-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-11-19
  • 修回日期:2024-10-21
基金
国家自然科学基金(61903244)
上海辰仕科技发展有限公司一般企事业单位资助项目(H2019-269)
作者信息
    上海电力大学,数理学院智慧能源数学研究中心, 上海 201306

通讯作者:

* 吴泉军(1982—),男,汉族,山东汶上人,博士,副教授。研究方向:锂电池健康监测。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2309072
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏