Article(id=1228046475599479446, tenantId=1146029695717560320, journalId=1225147924628267009, issueId=1228046469559681568, articleNumber=null, orderNo=null, doi=10.16385/j.cnki.issn.1004-4523.2024.02.018, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1674662400000, receivedDateStr=2023-01-26, revisedDate=1679760000000, revisedDateStr=2023-03-26, acceptedDate=null, acceptedDateStr=null, onlineDate=1770718695782, onlineDateStr=2026-02-10, pubDate=1709049600000, pubDateStr=2024-02-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1770718695782, onlineIssueDateStr=2026-02-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1770718695782, creator=13701087609, updateTime=1770718695782, updator=13701087609, issue=Issue{id=1228046469559681568, tenantId=1146029695717560320, journalId=1225147924628267009, year='2024', volume='37', issue='2', pageStart='191', pageEnd='364', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1770718694343, creator=13701087609, updateTime=1770795432451, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1228368332575928712, tenantId=1146029695717560320, journalId=1225147924628267009, issueId=1228046469559681568, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1228368332575928713, tenantId=1146029695717560320, journalId=1225147924628267009, issueId=1228046469559681568, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=356, endPage=364, ext={EN=ArticleExt(id=1228046476916490919, articleId=1228046475599479446, tenantId=1146029695717560320, journalId=1225147924628267009, language=EN, title=A fault transfer diagnosis method for wind turbine bearings based on improved residual neural networks, columnId=null, journalTitle=Journal of Vibration Engineering, columnName=null, runingTitle=null, highlight=null, articleAbstract=

To address the low accuracy in diagnosing faults in wind turbine bearings caused by the different characteristic distribution of the source domain data and the target domain data,a fault transfer diagnosis method using improved residual neural networks is proposed. The convolution kernel and pooling kernel are set to a size suitable for the convolution operation of one-dimensional signals,allowing for direct extraction of fault features from the bearing vibration signals; Both batch normalization and case normalization are used in the one-dimensional residual network to further enhance the feature extraction ability of the model; In the model training stage,a new loss function is constructed based on the multiple kernel maximum mean discrepancy between the source domain data and the target domain data to improve the transfer learning and classification ability of the model. The effectiveness of the method is verified by conducting the experimental data of the faulty bearings. The results show that the proposed method can effectively extract the important features of bearing faults and achieve the transfer diagnosis and accurate classification of the bearing faults. This holds true even under varying speed operation conditions and when the bearing fault vibration signals are disturbed by some noise components. Therefore,this work provides a useful strategy in developing intelligent fault diagnosis technology of rotating machinery under complex working conditions.

, correspAuthors=null, 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=Lin-feng DENG, Qi WANG, Yu-qiao ZHENG), CN=ArticleExt(id=1228046484332020721, articleId=1228046475599479446, tenantId=1146029695717560320, journalId=1225147924628267009, language=CN, title=基于改进残差网络的风电轴承故障迁移诊断方法, columnId=0, journalTitle=振动工程学报, columnName=, runingTitle=null, highlight=null, articleAbstract=

针对风电轴承故障源域数据和目标域数据特征分布不同而导致的故障诊断精度偏低问题,提出一种利用改进残差神经网络进行风电轴承故障迁移诊断的方法。该方法将卷积核和池化核设定为与一维振动信号卷积运算相适应的尺寸,从振动信号直接提取轴承的故障特征;在一维残差网络中同时使用批量归一化和实例归一化,进一步增强模型的特征提取能力;在模型训练阶段,通过源域数据和目标域数据的多核最大均值差异构建新的损失函数,以提高模型在不同分布数据集上的迁移学习及分类能力。利用故障轴承实验数据对方法的有效性进行验证,结果显示,即使受到轴承变转速运行工况和故障振动信号含噪声干扰成分的双重影响,该方法仍然可提取出轴承故障的重要特征,并实现不同工况轴承故障的迁移诊断和准确分类,这对于发展复杂环境下的旋转机械智能故障诊断技术具有参考价值。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=opyNaXYU4Aw1+jtGStxvnA==, magXml=FQoOTW+Rdii8u9UORspH7A==, pdfUrl=null, pdf=7EP8cFzMsTrh8VN1QQfKAg==, pdfFileSize=2901929, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=O1BmOftqPRZUCYLlgpFbxw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=2cjIbW2Nx27Zud81A7UleA==, mapNumber=null, authorCompany=null, fund=null, authors=

邓林峰(1984—),男,博士,副教授。E-mail:

, authorsList=邓林峰, 王琦, 郑玉巧)}, authors=[Author(id=1228046487666491399, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=denglinfeng2002@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228046487754571791, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046487666491399, language=EN, stringName=Lin-feng DENG, firstName=Lin-feng, middleName=null, lastName=DENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228046487872012310, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046487666491399, language=CN, stringName=邓林峰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=兰州理工大学机电工程学院,甘肃 兰州 730050, bio={"content":"

邓林峰(1984—),男,博士,副教授。E-mail:

"}, bioImg=null, bioContent=

邓林峰(1984—),男,博士,副教授。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228046485850358783, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, xref=null, ext=[AuthorCompanyExt(id=1228046485862941696, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China), AuthorCompanyExt(id=1228046485871329280, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=兰州理工大学机电工程学院,甘肃 兰州 730050)])]), Author(id=1228046487955898401, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228046488039784490, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046487955898401, language=EN, stringName=Qi WANG, firstName=Qi, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228046488127864880, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046487955898401, language=CN, stringName=王琦, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=兰州理工大学机电工程学院,甘肃 兰州 730050, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228046485850358783, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, xref=null, ext=[AuthorCompanyExt(id=1228046485862941696, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China), AuthorCompanyExt(id=1228046485871329280, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=兰州理工大学机电工程学院,甘肃 兰州 730050)])]), Author(id=1228046488211750969, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228046488291442752, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046488211750969, language=EN, stringName=Yu-qiao ZHENG, firstName=Yu-qiao, middleName=null, lastName=ZHENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228046488383717446, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, authorId=1228046488211750969, language=CN, stringName=郑玉巧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=兰州理工大学机电工程学院,甘肃 兰州 730050, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228046485850358783, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, xref=null, ext=[AuthorCompanyExt(id=1228046485862941696, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China), AuthorCompanyExt(id=1228046485871329280, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=兰州理工大学机电工程学院,甘肃 兰州 730050)])])], keywords=[Keyword(id=1228046488534712400, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, orderNo=1, keyword=fault diagnosis), Keyword(id=1228046488631181397, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, orderNo=2, keyword=wind turbine bearing), Keyword(id=1228046488752816224, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, orderNo=3, keyword=vibration signal), Keyword(id=1228046488840896615, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, orderNo=4, keyword=convolutional neural network), Keyword(id=1228046490216628331, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, orderNo=5, keyword=residual neural network), Keyword(id=1228046490304708722, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, orderNo=1, keyword=故障诊断), Keyword(id=1228046490405372023, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, orderNo=2, keyword=风电轴承), Keyword(id=1228046490552172671, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, orderNo=3, keyword=振动信号), Keyword(id=1228046490661224579, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, orderNo=4, keyword=卷积神经网络), Keyword(id=1228046490774470795, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, orderNo=5, keyword=残差网络)], refs=[Reference(id=1228046494914249032, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2020, volume=145, issue=null, pageStart=642, pageEnd=650, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Wang Jinjiang, Liang Yuanyuan, Zheng Yinghao, journalName=Renewable Energy, refType=null, unstructuredReference=Wang JinjiangLiang YuanyuanZheng Yinghao, et al. An integrated fault diagnosis and prognosis approach for predictive maintenance of wind turbine bearing with limited samples[J]. Renewable Energy2020145: 642-650., articleTitle=An integrated fault diagnosis and prognosis approach for predictive maintenance of wind turbine bearing with limited samples, refAbstract=null), Reference(id=1228046495006523727, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2020, volume=149, issue=null, pageStart=107002, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=Liu Zepeng, Zhang Long, journalName=Measurement, refType=null, unstructuredReference=Liu ZepengZhang Long. A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings[J]. Measurement2020149: 107002., articleTitle=A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings, refAbstract=null), Reference(id=1228046495098798422, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2023, volume=205, issue=null, pageStart=108867, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=Ma Zhipeng, Zhao Ming, Luo Mourui, journalName=Signal Processing, refType=null, unstructuredReference=Ma ZhipengZhao MingLuo Mourui, et al. An integrated monitoring scheme for wind turbine main bearing using acoustic emission[J]. Signal Processing2023205: 108867., articleTitle=An integrated monitoring scheme for wind turbine main bearing using acoustic emission, refAbstract=null), Reference(id=1228046495178490203, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2019, volume=133, issue=null, pageStart=620, pageEnd=635, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=Stetco A, Dinmohammadi F, Zhao X, journalName=Renewable Energy, refType=null, unstructuredReference=Stetco ADinmohammadi FZhao X, et al. Machine learning methods for wind turbine condition monitoring: a review[J]. Renewable Energy2019133: 620-635., articleTitle=Machine learning methods for wind turbine condition monitoring: a review, refAbstract=null), Reference(id=1228046495295930720, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=4, pageStart=949, pageEnd=957, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=姚齐水, 别帅帅, 余江鸿, journalName=振动工程学报, refType=null, unstructuredReference=姚齐水, 别帅帅, 余江鸿, 等. 一种结合改进Inception V2模块和CBAM的轴承故障诊断方法[J]. 振动工程学报202235(4): 949-957., articleTitle=一种结合改进Inception V2模块和CBAM的轴承故障诊断方法, refAbstract=null), Reference(id=1228046495409176933, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=4, pageStart=949, pageEnd=957, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=Yao Qishui, Bie Shuaishuai, Yu Jianghong, journalName=Journal of Vibration Engineering, refType=null, unstructuredReference=Yao QishuiBie ShuaishuaiYu Jianghong, et al. A bearing fault diagnosis method combining improved inception V2 module and CBAM[J]. Journal of Vibration Engineering202235(4): 949-957., articleTitle=A bearing fault diagnosis method combining improved inception V2 module and CBAM, refAbstract=null), Reference(id=1228046495501451630, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=182, issue=null, pageStart=615, pageEnd=626, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=Xu Zifei, Mei Xuan, Wang Xinyu, journalName=Renewable Energy, refType=null, unstructuredReference=Xu ZifeiMei XuanWang Xinyu, et al. Fault diagnosis of wind turbine bearing using a multi-scale convolutional neural network with bidirectional long short term memory and weighted majority voting for multi-sensors[J]. Renewable Energy2022182: 615-626., articleTitle=Fault diagnosis of wind turbine bearing using a multi-scale convolutional neural network with bidirectional long short term memory and weighted majority voting for multi-sensors, refAbstract=null), Reference(id=1228046495585337715, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2023, volume=36, issue=2, pageStart=565, pageEnd=573, url=null, language=null, rfNumber=[7], rfOrder=7, authorNames=韩特, 刘超, 沈长青, journalName=振动工程学报, refType=null, unstructuredReference=韩特, 刘超, 沈长青, 等. 深度嵌入度量学习的机械跨工况故障识别方法[J].振动工程学报202336(2):565-573., articleTitle=深度嵌入度量学习的机械跨工况故障识别方法, refAbstract=null), Reference(id=1228046495698583932, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2023, volume=36, issue=2, pageStart=565, pageEnd=573, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=Han Te, Liu Chao, Shen Changqing, journalName=Journal of Vibration Engineering, refType=null, unstructuredReference=Han TeLiu ChaoShen Changqing, et al. Deep embedding metric learning for machinery fault identification across different working conditions[J]. Journal of Vibration Engineering202336(2):565-573., articleTitle=Deep embedding metric learning for machinery fault identification across different working conditions, refAbstract=null), Reference(id=1228046495774081408, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2020, volume=138, issue=null, pageStart=106587, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=Lei Yaguo, Yang Bin, Jiang Xinwei, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=Lei YaguoYang BinJiang Xinwei, et al. Applications of machine learning to machine fault diagnosis: a review and roadmap[J]. Mechanical Systems and Signal Processing2020138: 106587., articleTitle=Applications of machine learning to machine fault diagnosis: a review and roadmap, refAbstract=null), Reference(id=1228046495857967494, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2019, volume=136, issue=null, pageStart=27, pageEnd=51, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=Gomez J L, Khelf I, Bourdon A, journalName=Mechanism and Machine Theory, refType=null, unstructuredReference=Gomez J LKhelf IBourdon A, et al. Angular modeling of a rotating machine in non-stationary conditions: application to monitoring bearing defects of wind turbines with instantaneous angular speed[J]. Mechanism and Machine Theory2019136: 27-51., articleTitle=Angular modeling of a rotating machine in non-stationary conditions: application to monitoring bearing defects of wind turbines with instantaneous angular speed, refAbstract=null), Reference(id=1228046495962825102, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2021, volume=70, issue=null, pageStart=3525828, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=Zhao Zhibin, Zhang Qiyang, Yu Xiaolei, journalName=IEEE Transactions on Instrumentation and Measurement, refType=null, unstructuredReference=Zhao ZhibinZhang QiyangYu Xiaolei, et al. Applications of unsupervised deep transfer learning to intelligent fault diagnosis: a survey and comparative study[J]. IEEE Transactions on Instrumentation and Measurement202170: 3525828., articleTitle=Applications of unsupervised deep transfer learning to intelligent fault diagnosis: a survey and comparative study, refAbstract=null), Reference(id=1228046496046711183, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=242, issue=null, pageStart=108381, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=Su Hao, Yang Xin, Xiang Ling, journalName=Knowledge-Based Systems, refType=null, unstructuredReference=Su HaoYang XinXiang Ling, et al. A novel method based on deep transfer unsupervised learning network for bearing fault diagnosis under variable working condition of unequal quantity[J]. Knowledge-Based Systems2022242: 108381., articleTitle=A novel method based on deep transfer unsupervised learning network for bearing fault diagnosis under variable working condition of unequal quantity, refAbstract=null), Reference(id=1228046496126402962, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2018, volume=29, issue=2, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=Guo Jun, Lu Siliang, Zhai Chao, journalName=Measurement Science and Technology, refType=null, unstructuredReference=Guo JunLu SiliangZhai Chao, et al. Automatic bearing fault diagnosis of permanent magnet synchronous generators in wind turbines subjected to noise interference[J]. Measurement Science and Technology201829(2): 025002., articleTitle=Automatic bearing fault diagnosis of permanent magnet synchronous generators in wind turbines subjected to noise interference, refAbstract=null), Reference(id=1228046496222871960, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=1, pageStart=148, pageEnd=156, url=null, language=null, rfNumber=[13], rfOrder=14, authorNames=董绍江, 裴雪武, 吴文亮, journalName=机械工程学报, refType=null, unstructuredReference=董绍江, 裴雪武, 吴文亮, 等. 基于多层降噪技术及改进卷积神经网络的滚动轴承故障诊断方法[J]. 机械工程学报202157(1): 148-156., articleTitle=基于多层降噪技术及改进卷积神经网络的滚动轴承故障诊断方法, refAbstract=null), Reference(id=1228046496302563742, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2021, volume=57, issue=1, pageStart=148, pageEnd=156, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=Dong Shaojiang, Pei Xuewu, Wu Wenliang, journalName=Journal of Mechanical Engineering, refType=null, unstructuredReference=Dong ShaojiangPei XuewuWu Wenliang, et al. Rolling bearing fault diagnosis method based on multilayer noise reduction technology and improved convolutional neural network[J]. Journal of Mechanical Engineering202157(1): 148-156., articleTitle=Rolling bearing fault diagnosis method based on multilayer noise reduction technology and improved convolutional neural network, refAbstract=null), Reference(id=1228046496369672612, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2019, volume=148, issue=null, pageStart=106906, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=16, authorNames=Ren He, Liu Wenyi, Shan Mengchen, journalName=Measurement, refType=null, unstructuredReference=Ren HeLiu WenyiShan Mengchen, et al. A new wind turbine health condition monitoring method based on VMD-MPE and feature-based transfer learning[J]. Measurement2019148: 106906., articleTitle=A new wind turbine health condition monitoring method based on VMD-MPE and feature-based transfer learning, refAbstract=null), Reference(id=1228046496445170089, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=15, pageStart=8394, pageEnd=8402, url=null, language=null, rfNumber=[15], rfOrder=17, authorNames=Zhu Jun, Chen Nan, Shen Changqing, journalName=IEEE Sensors Journal, refType=null, unstructuredReference=Zhu JunChen NanShen Changqing. A new deep transfer learning method for bearing fault diagnosis under different working conditions[J]. IEEE Sensors Journal202020(15): 8394-8402., articleTitle=A new deep transfer learning method for bearing fault diagnosis under different working conditions, refAbstract=null), Reference(id=1228046496524861870, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=194, issue=null, pageStart=249, pageEnd=258, url=null, language=null, rfNumber=[16], rfOrder=18, authorNames=Zhang Yan, Liu Wenyi, Wang Xin, journalName=Renewable Energy, refType=null, unstructuredReference=Zhang YanLiu WenyiWang Xin, et al. A novel wind turbine fault diagnosis method based on compressed sensing and DTL-CNN[J]. Renewable Energy2022194: 249-258., articleTitle=A novel wind turbine fault diagnosis method based on compressed sensing and DTL-CNN, refAbstract=null), Reference(id=1228046496612942260, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=33, issue=9, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=19, authorNames=Zhang Yan, Liu Wenyi, Gu Heng, journalName=Measurement Science and Technology, refType=null, unstructuredReference=Zhang YanLiu WenyiGu Heng, et al. A novel wind turbine fault diagnosis based on deep transfer learning of improved residual network and multi-target data[J]. Measurement Science and Technology202233(9): 095007., articleTitle=A novel wind turbine fault diagnosis based on deep transfer learning of improved residual network and multi-target data, refAbstract=null), Reference(id=1228046496688439736, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2023, volume=118, issue=null, pageStart=105656, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=20, authorNames=Liang Pengfei, Wang Bin, Jiang Guoqian, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Liang PengfeiWang BinJiang Guoqian, et al. Unsupervised fault diagnosis of wind turbine bearing via a deep residual deformable convolution network based on subdomain adaptation under time-varying speeds[J]. Engineering Applications of Artificial Intelligence2023118: 105656., articleTitle=Unsupervised fault diagnosis of wind turbine bearing via a deep residual deformable convolution network based on subdomain adaptation under time-varying speeds, refAbstract=null), Reference(id=1228046496751354303, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=[19], rfOrder=21, authorNames=He Kaiming, Zhang Xiangyu, Ren Shaoqing, journalName=null, refType=null, unstructuredReference=He KaimingZhang XiangyuRen Shaoqing, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, N.V., USA, 2016: 770-778., articleTitle=Deep residual learning for image recognition, refAbstract=null), Reference(id=1228046496856211908, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2022, volume=593, issue=null, pageStart=126907, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=22, authorNames=Gong Lihua, Xiang Lingzhi, Liu Sihang, journalName=Physica A: Statistical Mechanics and its Applications, refType=null, unstructuredReference=Gong LihuaXiang LingzhiLiu Sihang, et al. Born machine model based on matrix product state quantum circuit[J]. Physica A: Statistical Mechanics and its Applications2022593: 126907., articleTitle=Born machine model based on matrix product state quantum circuit, refAbstract=null), Reference(id=1228046496940097991, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2020, volume=143, issue=null, pageStart=106427, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=23, authorNames=Che Changchang, Wang Huawei, Ni Xiaomei, journalName=Computers & Industrial Engineering, refType=null, unstructuredReference=Che ChangchangWang HuaweiNi Xiaomei, et al. Domain adaptive deep belief network for rolling bearing fault diagnosis[J]. Computers & Industrial Engineering2020143: 106427., articleTitle=Domain adaptive deep belief network for rolling bearing fault diagnosis, refAbstract=null), Reference(id=1228046497049149902, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2018, volume=100, issue=null, pageStart=439, pageEnd=453, url=null, language=null, rfNumber=[22], rfOrder=24, authorNames=Zhang Wei, Li Chuanhao, Peng Gaoliang, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=Zhang WeiLi ChuanhaoPeng Gaoliang, et al. A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load[J]. Mechanical Systems and Signal Processing2018100: 439-453., articleTitle=A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load, refAbstract=null), Reference(id=1228046497158201811, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2019, volume=66, issue=7, pageStart=5525, pageEnd=5534, url=null, language=null, rfNumber=[23], rfOrder=25, authorNames=Li Xiang, Zhang Wei, Ding Qian, journalName=IEEE Transactions on Industrial Electronics, refType=null, unstructuredReference=Li XiangZhang WeiDing Qian. Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics201966(7): 5525-5534., articleTitle=Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks, refAbstract=null), Reference(id=1228046500412981723, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, doi=null, pmid=null, pmcid=null, year=2018, volume=41, issue=12, pageStart=3071, pageEnd=3085, url=null, language=null, rfNumber=[24], rfOrder=26, authorNames=Long Mingsheng, Cao Yue, Cao Zhangjie, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=Long MingshengCao YueCao Zhangjie, et al. Transferable representation learning with deep adaptation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence201841(12): 3071-3085., articleTitle=Transferable representation learning with deep adaptation networks, refAbstract=null)], funds=[Fund(id=1228046494679367994, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, awardId=62241308, language=CN, fundingSource=国家自然科学基金资助项目(62241308), fundOrder=null, country=null), Fund(id=1228046494771642691, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, awardId=22CX8GA130, language=CN, fundingSource=甘肃省技术创新引导计划-科技专员专项资助项目(22CX8GA130), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1228046485850358783, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, xref=null, ext=[AuthorCompanyExt(id=1228046485862941696, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China), AuthorCompanyExt(id=1228046485871329280, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, companyId=1228046485850358783, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=兰州理工大学机电工程学院,甘肃 兰州 730050)])], figs=[ArticleFig(id=1228046490925465748, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.1, caption=Structure of residual block, figureFileSmall=iJeRLEP49MwUUhoLICYLmg==, figureFileBig=O1BmOftqPRZUCYLlgpFbxw==, tableContent=null), ArticleFig(id=1228046491017740442, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图1, caption=残差块结构, figureFileSmall=iJeRLEP49MwUUhoLICYLmg==, figureFileBig=O1BmOftqPRZUCYLlgpFbxw==, tableContent=null), ArticleFig(id=1228046491290370214, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.2, caption=Improved residual block, figureFileSmall=2lh+3HWJBgPf0w4g5t9odw==, figureFileBig=0BKRh+KyseAEs73/hyn8zA==, tableContent=null), ArticleFig(id=1228046491386839211, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图2, caption=改进后的残差块, figureFileSmall=2lh+3HWJBgPf0w4g5t9odw==, figureFileBig=0BKRh+KyseAEs73/hyn8zA==, tableContent=null), ArticleFig(id=1228046491487502514, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.3, caption=Structure of one-dimensional residual neural network model, figureFileSmall=+ySR5O5Xn82meCps1LoBjQ==, figureFileBig=spqXuUMPqkTECs57Feh6UQ==, tableContent=null), ArticleFig(id=1228046491575582904, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图3, caption=一维残差神经网络模型结构, figureFileSmall=+ySR5O5Xn82meCps1LoBjQ==, figureFileBig=spqXuUMPqkTECs57Feh6UQ==, tableContent=null), ArticleFig(id=1228046491667857596, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.4, caption=Training process of the model, figureFileSmall=EyokEqnPyb8Ngep6ARRFew==, figureFileBig=YajWELXccQMgI3+nC2qgkQ==, tableContent=null), ArticleFig(id=1228046491764326593, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图4, caption=模型的训练过程, figureFileSmall=EyokEqnPyb8Ngep6ARRFew==, figureFileBig=YajWELXccQMgI3+nC2qgkQ==, tableContent=null), ArticleFig(id=1228046491864989895, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.5, caption=Flowchart of the bearing fault transfer diagnosis, figureFileSmall=aMXEgckqBbPWMYjSvk1A+Q==, figureFileBig=YTyA+181tPVaiwElbOYw4Q==, tableContent=null), ArticleFig(id=1228046491948875983, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图5, caption=轴承故障迁移诊断流程图, figureFileSmall=aMXEgckqBbPWMYjSvk1A+Q==, figureFileBig=YTyA+181tPVaiwElbOYw4Q==, tableContent=null), ArticleFig(id=1228046492049539287, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.6, caption=Fault diagnosis accuracies of the three different methods, figureFileSmall=0KhFl6aIodTulhltOSy1zA==, figureFileBig=KJUVgCvPeyvhIa4aiB6C/Q==, tableContent=null), ArticleFig(id=1228046492129231067, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图6, caption=3种方法的故障诊断准确率, figureFileSmall=0KhFl6aIodTulhltOSy1zA==, figureFileBig=KJUVgCvPeyvhIa4aiB6C/Q==, tableContent=null), ArticleFig(id=1228046492221505763, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.7, caption=HZXT-DS-001 experiment rig, figureFileSmall=zi9I65ucQynvJWjr0ws0Hg==, figureFileBig=OkQHhtdw7MvFn04vrZoYTw==, tableContent=null), ArticleFig(id=1228046492330557672, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图7, caption=HZXT-DS-001型实验台, figureFileSmall=zi9I65ucQynvJWjr0ws0Hg==, figureFileBig=OkQHhtdw7MvFn04vrZoYTw==, tableContent=null), ArticleFig(id=1228046492418638059, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.8, caption=Four different fault forms of the bearing, figureFileSmall=DJx5Su1DVHKZaiWNQrV5CQ==, figureFileBig=rD6ylETIn1wJ1omeiSXjUQ==, tableContent=null), ArticleFig(id=1228046492502524146, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图 8, caption=轴承故障的4种形式, figureFileSmall=DJx5Su1DVHKZaiWNQrV5CQ==, figureFileBig=rD6ylETIn1wJ1omeiSXjUQ==, tableContent=null), ArticleFig(id=1228046492586410234, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Fig.9, caption=Performance comparison of the three different methods, figureFileSmall=V+zSh2swguPWJj6nAmbGdA==, figureFileBig=0Lt+105RUEvlYIsOc9udSg==, tableContent=null), ArticleFig(id=1228046492653519103, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=图9, caption=3种方法的性能对比, figureFileSmall=V+zSh2swguPWJj6nAmbGdA==, figureFileBig=0Lt+105RUEvlYIsOc9udSg==, tableContent=null), ArticleFig(id=1228046492754182408, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Tab.1, caption=

Experiment datasets

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集训练样本数测试样本数转速/(r·min-1)
A10×80010×2001797
B10×80010×2001772
C10×80010×2001750
D10×80010×2001730
), ArticleFig(id=1228046492859040015, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=表1, caption=

实验数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集训练样本数测试样本数转速/(r·min-1)
A10×80010×2001797
B10×80010×2001772
C10×80010×2001750
D10×80010×2001730
), ArticleFig(id=1228046492951314709, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Tab.2, caption=

Parameters of the model

, figureFileSmall=null, figureFileBig=null, tableContent=
网络层具体参数输出维度其他参数
输入层-2048×1批尺寸:16

优化器:Adadelta
卷积层16×64×8256×16
残差块116×1×1
32×3×1
256×32
残差块232×1×1
64×3×1
256×64
残差块364×1×1
128×3×1
256×128
池化层-128×1
输出层1010×1
), ArticleFig(id=1228046493051978014, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=表2, caption=

模型参数

, figureFileSmall=null, figureFileBig=null, tableContent=
网络层具体参数输出维度其他参数
输入层-2048×1批尺寸:16

优化器:Adadelta
卷积层16×64×8256×16
残差块116×1×1
32×3×1
256×32
残差块232×1×1
64×3×1
256×64
残差块364×1×1
128×3×1
256×128
池化层-128×1
输出层1010×1
), ArticleFig(id=1228046493169418537, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=EN, label=Tab.3, caption=

Experiment datasets

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集训练样本数测试样本数转速/(r·min-1)
A5×6005×2002600
B5×6005×2002800
C5×6005×2003000
D5×6005×2003200
), ArticleFig(id=1228046493282664752, tenantId=1146029695717560320, journalId=1225147924628267009, articleId=1228046475599479446, language=CN, label=表3, caption=

实验数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集训练样本数测试样本数转速/(r·min-1)
A5×6005×2002600
B5×6005×2002800
C5×6005×2003000
D5×6005×2003200
)], attaches=null, journal=Journal(id=1225147830491308032, delFlag=0, nameCn=振动工程学报, nameEn=Journal of Vibration Engineering, nameHistory1=null, nameHistory2=null, issn=1004-4523, eissn=null, cn=32-1349/TB, coden=null, periodic=0, language=CN, oaType=null, 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=null, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Vibration Engineering, journalRemark=null, publicationField=null, createdTime=1770027604939, updatedTime=1770169610881, createdBy=18614031015, updatedBy=18614031015, firstLetterCn=J, firstLetterEn=J, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=null, picEn=null, jcr=null, cjcr=null, exts=[JournalExt(id=1225743346702925905, language=CN, name=振动工程学报, nameHistory1=null, nameHistory2=null, managedBy=中国科学技术协会, sponsoredBy=中国振动工程学会, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1770169587064, updatedTime=1770169587064, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.manuscripts.com.cn/zdgcxb, submissionEditorUrl=https://www.manuscripts.com.cn/zdgcxb, submissionReviewUrl=https://www.manuscripts.com.cn/zdgcxb, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1225743346765840466, language=EN, name=Journal of Vibration Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1770169587079, updatedTime=1770169587079, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.manuscripts.com.cn/zdgcxb, submissionEditorUrl=https://www.manuscripts.com.cn/zdgcxb, submissionReviewUrl=https://www.manuscripts.com.cn/zdgcxb, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1225147924628267009, websiteList=[Website(id=1225150618881404985, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1225147924628267009, 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/zdgcxb/CN, language=CN, createTime=1770028269739, createBy=18614031015, updateTime=1770028293069, updateBy=18614031015, name=振动工程学报-中文, tplId=1146099689490845704, title=振动工程学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1225151164178673750, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=articleTextType, value=kx, createTime=1770028399748, updateTime=1770028399748, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164157702227, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=banner, value=null, createTime=1770028399743, updateTime=1770028399743, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164203839577, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=grayFlag, value=0, createTime=1770028399754, updateTime=1770028399754, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164145119314, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=logo, value=https://castjournals.cast.org.cn/joweb/zdgcxb/EN/file/pic?fileId=L7mSU8YPwm66NWFMoTG4aQ==, createTime=1770028399740, updateTime=1770028399740, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164212228187, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=minRunFlag, value=0, createTime=1770028399756, updateTime=1770028399756, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164170285141, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zdgcxb/CN/file/pic, createTime=1770028399746, updateTime=1770028399746, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164208033882, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=silenceFlag, value=0, createTime=1770028399755, updateTime=1770028399755, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164166090836, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1770028399745, updateTime=1770028399745, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164187062359, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=themeColor, value=null, createTime=1770028399750, updateTime=1770028399750, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151164195450968, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150618881404985, code=themeStyle, value=null, createTime=1770028399752, updateTime=1770028399752, creator=18614031015, updator=18614031015)]), Website(id=1225150619003039804, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1225147924628267009, 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/zdgcxb/EN, language=EN, createTime=1770028269768, createBy=18614031015, updateTime=1770028309190, updateBy=18614031015, name=振动工程学报-英文, tplId=1146101810881728533, title=Journal of Vibration Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1225151193366835296, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=articleTextType, value=kx, createTime=1770028406707, updateTime=1770028406707, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193350058077, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=banner, value=null, createTime=1770028406703, updateTime=1770028406703, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193387806819, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=grayFlag, value=0, createTime=1770028406712, updateTime=1770028406712, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193341669468, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=logo, value=https://castjournals.cast.org.cn/joweb/zdgcxb/EN/file/pic?fileId=L7mSU8YPwm66NWFMoTG4aQ==, createTime=1770028406701, updateTime=1770028406701, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193400389733, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=minRunFlag, value=0, createTime=1770028406715, updateTime=1770028406715, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193362640991, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/zdgcxb/EN/file/pic, createTime=1770028406706, updateTime=1770028406706, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193392001124, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=silenceFlag, value=0, createTime=1770028406713, updateTime=1770028406713, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193354252382, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1770028406704, updateTime=1770028406704, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193371029601, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=themeColor, value=null, createTime=1770028406708, updateTime=1770028406708, creator=18614031015, updator=18614031015), WebsiteProps(id=1225151193379418210, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1225150619003039804, code=themeStyle, value=null, createTime=1770028406710, updateTime=1770028406710, creator=18614031015, updator=18614031015)])], journalTitle=振动工程学报, weixinUrl=null, journalUrl=http://zdgcxb.csve.org.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Vibration Engineering, journalPhotoCn=null, journalPhotoEn=null, journalFirstLetter=J, 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=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/zdgcxb/CN/10.16385/j.cnki.issn.1004-4523.2024.02.018, detailUrlEn=https://castjournals.cast.org.cn/joweb/zdgcxb/EN/10.16385/j.cnki.issn.1004-4523.2024.02.018, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zdgcxb/CN/PDF/10.16385/j.cnki.issn.1004-4523.2024.02.018, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zdgcxb/EN/PDF/10.16385/j.cnki.issn.1004-4523.2024.02.018, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于改进残差网络的风电轴承故障迁移诊断方法
收藏切换
PDF下载
邓林峰 , 王琦 , 郑玉巧
振动工程学报 | 2024,37(2): 356-364
收起
收藏切换
振动工程学报 | 2024, 37(2): 356-364
基于改进残差网络的风电轴承故障迁移诊断方法
全屏
邓林峰 , 王琦, 郑玉巧
作者信息
  • 兰州理工大学机电工程学院,甘肃 兰州 730050
  • 邓林峰(1984—),男,博士,副教授。E-mail:

A fault transfer diagnosis method for wind turbine bearings based on improved residual neural networks
Lin-feng DENG , Qi WANG, Yu-qiao ZHENG
Affiliations
  • School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,China
出版时间: 2024-02-28 doi: 10.16385/j.cnki.issn.1004-4523.2024.02.018
文章导航
收藏切换

针对风电轴承故障源域数据和目标域数据特征分布不同而导致的故障诊断精度偏低问题,提出一种利用改进残差神经网络进行风电轴承故障迁移诊断的方法。该方法将卷积核和池化核设定为与一维振动信号卷积运算相适应的尺寸,从振动信号直接提取轴承的故障特征;在一维残差网络中同时使用批量归一化和实例归一化,进一步增强模型的特征提取能力;在模型训练阶段,通过源域数据和目标域数据的多核最大均值差异构建新的损失函数,以提高模型在不同分布数据集上的迁移学习及分类能力。利用故障轴承实验数据对方法的有效性进行验证,结果显示,即使受到轴承变转速运行工况和故障振动信号含噪声干扰成分的双重影响,该方法仍然可提取出轴承故障的重要特征,并实现不同工况轴承故障的迁移诊断和准确分类,这对于发展复杂环境下的旋转机械智能故障诊断技术具有参考价值。

故障诊断  /  风电轴承  /  振动信号  /  卷积神经网络  /  残差网络

To address the low accuracy in diagnosing faults in wind turbine bearings caused by the different characteristic distribution of the source domain data and the target domain data,a fault transfer diagnosis method using improved residual neural networks is proposed. The convolution kernel and pooling kernel are set to a size suitable for the convolution operation of one-dimensional signals,allowing for direct extraction of fault features from the bearing vibration signals; Both batch normalization and case normalization are used in the one-dimensional residual network to further enhance the feature extraction ability of the model; In the model training stage,a new loss function is constructed based on the multiple kernel maximum mean discrepancy between the source domain data and the target domain data to improve the transfer learning and classification ability of the model. The effectiveness of the method is verified by conducting the experimental data of the faulty bearings. The results show that the proposed method can effectively extract the important features of bearing faults and achieve the transfer diagnosis and accurate classification of the bearing faults. This holds true even under varying speed operation conditions and when the bearing fault vibration signals are disturbed by some noise components. Therefore,this work provides a useful strategy in developing intelligent fault diagnosis technology of rotating machinery under complex working conditions.

fault diagnosis  /  wind turbine bearing  /  vibration signal  /  convolutional neural network  /  residual neural network
邓林峰, 王琦, 郑玉巧. 基于改进残差网络的风电轴承故障迁移诊断方法. 振动工程学报, 2024 , 37 (2) : 356 -364 . DOI: 10.16385/j.cnki.issn.1004-4523.2024.02.018
Lin-feng DENG, Qi WANG, Yu-qiao ZHENG. A fault transfer diagnosis method for wind turbine bearings based on improved residual neural networks[J]. Journal of Vibration Engineering, 2024 , 37 (2) : 356 -364 . DOI: 10.16385/j.cnki.issn.1004-4523.2024.02.018
轴承是风电机组传动系统的关键部件,承受着风机转动部件的重量,并支承这些部件进行旋转将风能转化为电能,其健康状况对风电机组的性能、稳定性和使用寿命会产生巨大影响1-3。因此,对风电轴承故障进行监测与诊断,是保障风电机组安全高效运行的重要措施。轴承振动信号中包含着轴承故障信息,通过对其进行分析处理可识别轴承的健康状况。然而随着风电设备的监测数据量与日俱增,传统故障诊断方法已无法适应大数据故障诊断分析的发展需求4
近年来,基于大数据深度学习的模式识别方法在机械故障诊断领域快速发展5。其中,卷积神经网络(Convolutional Neural Network,CNN)具有局部计算、权值共享和端对端等特点,表现出优异的故障诊断性能6-7。但是,这些方法有效的前提是基于两个基本假设8:1)有大量的标记数据可用;2)训练集和测试集独立且服从同一分布。然而,在工程现场,风电机组运行环境复杂,存在多种引起风电机组运行状态发生变化的因素,使得风电轴承出现故障时采集到的振动信号往往具有明显的非平稳特性9。在这种状况下,从部分振动信号提取的故障特征值将发生偏离,导致故障训练样本和测试样本的分布特性存在一定差异10-11,从而影响深度学习模型在测试集上的泛化性能,致使故障诊断结果的准确性较低,难以满足轴承故障监测的实际需求。
针对上述问题,Guo等12利用重采样技术和自适应随机共振滤波技术,实现强噪声干扰下风力发电机轴承故障的高效自动诊断。董绍江等13联合奇异值分解和经验模态分解对故障信号进行处理,并在CNN池化层引入注意力机制增强模型的特征提取能力,实现了强噪声环境下轴承故障的准确诊断。Ren等14将变分模态分解、多尺度排列熵和特征迁移学习方法相结合,对变工况下的风机齿轮箱轴承故障具有较高的分类准确率。Zhu等15提出了一种迁移学习的CNN模型,将服从不同分布的两组数据的多核最大均值差异作为训练CNN的约束条件,可有效识别不同负载下轴承的故障类别。Zhang等16-17针对风电机组轴承、齿轮的故障信号和诊断模型的特点,采用数模联动策略进行数据形式的变换与模型结构的改进,建立了可靠有效的风电轴承、齿轮的故障迁移诊断方法。Liang等18提出基于子域自适应深度可变形残差网络的无监督风机轴承故障诊断方法,实现了变转速轴承故障的有效诊断。虽然上述方法针对噪声或负载变化工况下的轴承复杂故障诊断有一定效果,但模型的特征提取能力和迁移学习能力还有进一步提升的空间。
综上所述,本文提出一种利用残差神经网络的风电轴承故障迁移诊断方法。通过对残差网络的卷积运算进行一维化处理并加入批量归一化和实例归一化,以增强残差网络的特征提取能力;在模型训练阶段,利用不同分布特性的数据建立基于多核最大均值差异的新损失函数,用于提高模型在不同分布数据集上的迁移学习及数据分类能力。在此基础上,利用滚动轴承故障实验数据对该方法的性能进行验证和对比分析,证实其对风电轴承复杂故障进行迁移诊断的有效性。
为解决深度神经网络在训练过程中出现的梯度消失和梯度爆炸问题,He等19提出了残差神经网络(Residual Network,ResNet)。ResNet本质上是由多个残差块堆叠而成的,残差块在普通CNN的基础上使用跳跃连接,达到使网络学习残差的目的。残差块结构如图1所示,图1(a)和(b)分别表示输入与残差映射输出特征维度相同和不同时的残差块。
图1可见,残差块输出y为输入x和残差映射输出Fx)的叠加值,其中Fx)是对x进行卷积和激活运算所得,其具体计算过程如下式所示:
式中  x表示残差块的输入;c表示通道数量;w表示卷积核的权重矩阵;b表示偏置项;*表示进行卷积运算;xl表示卷积运算的结果。
式中  xl为式(1)卷积运算得到的输出特征图;ylxl经过ReLU激活处理后的输出值。
需要注意的是,当输入x和残差映射输出Fx)的通道数量不同时,则需要在图1(a)的跳跃连接的基础上,通过使用1×1卷积运算保证它们的特征维度相同,如图1(b)所示。
批量归一化(Batch Normalization,BN)操作类似于一种标准化操作,在CNN中添加BN层的目的是减少内部协变量转移,从而提高网络性能。其计算过程为:
式中  表示样本均值;表示样本方差;表示样本批尺寸;表示卷积核的原始输出特征图xi经标准化处理后的数据;yi表示的批量归一化处理结果;ε表示常值向量;γβ分别表示尺度因子和移动因子的两个可学习参数。
实例归一化(Instance Normalization,IN)的运算过程与BN相似,不同之处在于:BN是通过计算每批次样本的均值和方差实现归一化,而IN则是通过计算每个通道的均值μI和方差实现归一化;对于一维信号而言,则是通过计算单个样本信号所有数据点的均值μI和方差进行实例归一化,因此IN的结果不受批尺寸的影响。
在机器学习方法中,度量准则必不可少。最大均值差异(Maximum Mean Discrepancy,MMD)20是迁移学习方法尤其是域适应中常用的度量准则之一,其主要用来度量两组不同但相关数据在再生核希尔伯特空间(Reproducing Kernel Hilbert Space,RKHS)中的分布距离,是一种核学习方法。假设存在两组服从不同分布的数据X={x1,x2,…,xn}和Y={y1,y2,…,ym},则XYMMD计算式为:
式中 ||·||H表示RKHS中的分布距离;φ(·)为映射函数,用于把原变量映射到RKHS。
将式(4)的平方展开后,RKHS中的内积就可以转换为核函数,即
式中  n为源域样本的数量;m为目标域样本的数量;k(·,·)为核函数。
高斯核函数可以将原始特征映射到无穷维的高维特征空间20,因此,本研究使用高斯核函数计算MMD,即式(5)中的核函数为:
式中  h为核参数,可控制函数的径向作用范围。
由式(6)可知,核参数h越大,高斯核函数的值就越分散,其函数图像就会越平坦,两个样本之间的欧式距离对高斯核函数的值影响较小,使得分类模型的学习能力有限,数据的分类训练准确率较低;反之,函数值就越集中,函数图像就会越尖锐,高斯核函数对两点之间的欧式距离就会很敏感,使分类模型对源域数据过分拟合,而对目标域数据的预测效果较差。显而易见,参数h的选择很关键,会对MMD的计算结果产生一定影响。为解决该问题,Gretton等提出多核最大均值差异(Multiple Kernel MMD,MK-MMD)21,其在计算MMD时,通过同时使用多个核函数来避免核参数选择不当所带来的计算误差。因此,MK-MMD将传统核参数的选择问题转换为多个核函数的凸优化问题21,如下式所示:
式中  K表示Nk个高斯核函数集的凸组合;ku表示第u个核;{βu}表示不同核函数的权重参数集。
He等19提出残差网络的初衷是用于解决复杂图像的识别问题,因此其卷积核及其卷积运算都是针对二维图像设计和计算的。为使残差网络适合处理轴承一维振动信号,需对其进行改进,具体改进措施如下:
1) 采集的轴承一维振动信号为一维时间序列,为适应其结构特性,所构建的残差网络模型的各层卷积核和池化核均适用于一维信号的一维卷积运算。
2) 模型的第一层采用大尺寸卷积核,目的是为了增大感受野,一次获取更多的数据点并为网络后续层提供有用信息。同时,对于含噪振动信号而言,大卷积核可以更好地抑制高频噪声22
3) 为了增强模型的特征提取能力,在图1所示的残差块中同时使用BN和IN两种归一化方法,具体操作如图2所示。
图2中可见,改进残差块是在图1所示残差块的基础上先对残差映射部分的第一个卷积层输出依次进行BN和激活操作;然后,对第二个卷积层的输出进行BN操作;最后,对残差块的输出y依次进行IN和激活操作。此外,需注意的是,当残差块的输入x和残差映射的输出Fx)具有相同特征维度时,使用图2(a)所示的残差块,反之使用图2(b)所示的残差块。综上所述,最终所构建的一维残差神经网络模型结构如图3所示。
图3中可见,改进后的一维残差神经网络模型主要由1个卷积层、3个残差块以及1个全局平均池化层构成。其中,前2个残差块为图2所示改进后的完整残差块,而第3个残差块仅有BN层无IN层。
在构建好一维残差神经网络模型后,还需要对其进行训练才可用于实际一维信号的模式分类。首先以原始一维振动信号作为输入,前向传播并逐层完成信号特征的自动提取;然后利用全局平均池化操作对所提取的特征进行全局空间整合,并通过Softmax函数进行分类;最后以交叉熵度量Softmax函数的输出概率分布和信号类别概率分布间的差异,并将其作为损失函数。交叉熵的计算公式为:
式中  px)表示样本x的分类的真实概率;qx)表示观测到的样本x实际分类结果的概率。
为增强模型在不同分布特性数据集上的迁移学习和推广泛化能力,将数据集的MK-MMD引入到交叉熵损失函数上,即
式中  l表示训练一维残差网络模型的损失函数;lc表示交叉熵损失函数;DsDt分别表示存在分布差异的源域和目标域故障数据;MK1DsDt)和MK2DsDt)分别表示DsDt图3所示一维残差网络模型的第3个残差块和全局平均池化层的多核最大均值差异;λ1λ2分别为用于控制MK1DsDt)和MK2DsDt)的权重。
为提高模型的分类性能,本研究通过联合使用5个高斯核函数计算DsDt的MK-MMD值,其中5个高斯核函数的核参数h分别设置为1,2,4,8和1623
完成损失函数的计算后,再通过反向传播分别计算损失函数关于模型各层权值和偏置的导数,最后使用优化算法更新模型的权值和偏置。迭代循环上述过程,直到损失函数值达到最小。综上,改进的一维残差神经网络模型的训练过程如图4所示。
图4中可见,结合轴承复杂故障信号训练一维残差网络的过程为:将服从不同分布的源域数据Ds和目标域数据Dt共同输入模型进行前向传播;然后,分别计算MK1DsDt)和MK2DsDt)的值,通过式(9)计算模型的损失函数l的值,并将其进行反向传播,以更新模型参数使损失函数l达到最小,从而完成模型训练;最后,分别输出源域和目标域数据的分类结果。
针对风电轴承故障振动特征分布不同的特点,基于改进的一维残差神经网络模型及其训练过程,设计复杂工况下的风电轴承故障迁移诊断方法,该方法主要分为三个环节:1) 数据集准备;2) 模型训练;3) 模型测试。其具体实施步骤如下:
步骤1 将采集到的轴承故障振动信号分为源域和目标域,并对源域数据进行故障状态标记,目标域数据不标记;其中,目标域数据进一步划分为训练数据子集和测试数据子集,将训练数据子集与源域数据合并构成完整的训练数据集。
步骤2 建立适应轴承故障振动信号的改进一维残差神经网络,并对网络参数进行初始化。
步骤3 将训练数据集输入网络模型,并逐层进行前向传播,获得不同层上输出的故障特征。
步骤4 利用一维残差神经网络第3个残差块和平均池化层的输出特征图计算相应的MK-MMD值以及输出层的交叉熵,进而得到模型的损失函数值,并将该值进行反向传播,以更新模型每层的权重和偏置,使模型的下一轮次训练损失值减小。
步骤5 若满足训练终止条件,训练过程结束;若不满足,则返回步骤3继续循环迭代。
步骤6 将轴承信号的测试数据子集输入训练好的网络模型,对其进行故障分类与识别,实现复杂工况下的风电轴承故障迁移诊断。
与上述步骤对应的流程如图5所示。
实验数据选用美国凯斯西储大学轴承数据中心的滚动轴承故障数据集5。轴承型号为SKF6205深沟球轴承,其运行状态包括正常(N)、内圈故障(IF)、外圈故障(OF)和滚动体故障(BF)共4种类型,且每类故障都包含了0.1778,0.3556和0.5334 mm三种故障直径,因此每一转速下可形成10种运行状态。
本研究选用驱动端轴承数据,因原始数据样本有限,采用重叠采样22的方式处理原始数据,使每个样本数据点为2048。由于具有4种转速,按不同转速将原始数据划分为4个数据集,具体信息如表1所示。
残差神经网络模型用于具体的分类任务时,选择恰当的参数可有效发挥模型的性能。因此,通过预训练方式,得到第2节所建立一维残差神经网络模型的参数,具体参数如表2所示。
为说明本文方法的优势,将其与ResNet和DAN24方法进行对比分析,其中ResNet与本研究构建的网络模型相比,其前两个残差块只使用BN操作。3种方法对轴承故障的分类准确率如图6所示。
图6(a),(b),(c)和(d)分别表示在测试数据中加入信噪比为-4,-2,2和4 dB高斯白噪声后的故障诊断结果。从图6中可见,在同时受到噪声污染和轴承转速变化影响的情况下,ResNet的故障识别准确率最低;DAN的故障识别率略高于ResNet,因为DAN方法同时进行故障特征提取和减小源域与目标域数据分布差异操作,但该方法却未能消除噪声干扰所带来的影响。相比之下,本文方法在同时受到噪声和转速变化影响下可以达到很高的故障识别率,当信噪比为-4 dB时,平均准确率相较于ResNet和DAN方法分别高出了26.7%和16.4%;信噪比为4 dB时其平均准确率高达99.2%。由此说明,本研究方法可实现复杂工况下的轴承故障迁移诊断,并具有较高的准确率。
为进一步验证本研究方法的实用性,将该方法用于图7所示的HZXT-DS-001型双跨转子轴承系统综合故障模拟实验台的轴承故障识别。
实验台的轴承型号为NSK6038深沟球轴承,在轴承上设置的故障形式如图8所示。
轴承设置了4种转速,信号采样频率为8 kHz。为使用与3.2节表2中相同的模型参数,使每个样本的采样点数为2048。同样,按轴承转速不同,本实验共准备了4个数据集,其具体信息如表3所示。
本实验中轴承故障不区分严重程度,故只有5种运行状态,即每个数据集均包含5种状态,因此需将模型输出向量长度设置为5。仍然与ResNet和DAN方法进行诊断性能的对比分析,结果如图9所示。
图9中可见,在原始故障数据中加入不同信噪比的高斯白噪声的情况下,ResNet和DAN方法的准确率明显低于本研究方法的准确率,且信噪比越低,其他两种方法的故障诊断准确率与本研究方法的故障诊断准确率的差距越大;对于任何一种信噪比,本研究方法的故障识别准确率都超过了90%。由此说明,在轴承故障信号受到严重噪声污染的情况下,本研究方法仍可有效解决变转速下的轴承故障诊断问题,其故障识别准确率明显高于其他两种方法,表现出了较好的故障迁移诊断能力和抗噪性能。因此,本研究方法在复杂工况下轴承故障诊断应用中的有效性和广泛适用性得到证实。
风电轴承故障诊断技术可为风电装备的安全运行提供保障。针对风电轴承故障数据不同分布特性而导致的故障诊断精度偏低问题,提出一种利用改进残差神经网络的风电轴承故障迁移诊断方法。通过处理轴承故障实验数据以及与其他典型故障诊断方法的比较分析,对该方法的性能进行了验证。获得的主要结论如下:
(1) 本研究提出的一维残差神经网络模型由于对卷积运算进行了一维化处理,可直接将轴承故障原始一维振动信号作为网络的输入数据,无需对一维振动信号进行人工特征提取;通过在模型的前两层残差块中同时使用批量归一化和实例归一化,能够有效降低噪声对模型性能的影响;利用源域特征和目标域特征之间的最大均值差异建立模型的损失函数,可提高模型在目标域数据上的分类性能。
(2) 与其他两种故障诊断方法进行实验和对比分析,结果表明在噪声污染和轴承转速变化同时存在的情况下,本研究方法依然可以有效实现滚动轴承复杂故障的迁移诊断,其故障诊断准确率较ResNet和DAN方法更高,且准确率达到了90%以上,说明本研究方法不仅可以实现复杂工况下轴承故障的迁移诊断,而且具有较好的抗噪性能。
(3) 在实验验证阶段,本研究考虑的情况是轴承在不同转速之间的故障迁移诊断任务,且所建模型在固定参数下表现出良好的故障迁移诊断能力,而对于模型在不同参数下的诊断效果并未考虑。下一步将尝试对不同参数下模型的鲁棒性进行研究,揭示模型参数对模型识别结果的影响机制,并建立适合于处理风电轴承振动信号的一维深度卷积网络模型,发展复杂工况下风电装备的智能故障诊断方法与技术。
  • 国家自然科学基金资助项目(62241308)
  • 甘肃省技术创新引导计划-科技专员专项资助项目(22CX8GA130)
参考文献 引证文献
排序方式:
[1]
Wang JinjiangLiang YuanyuanZheng Yinghao, et al. An integrated fault diagnosis and prognosis approach for predictive maintenance of wind turbine bearing with limited samples[J]. Renewable Energy2020145: 642-650.
[2]
Liu ZepengZhang Long. A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings[J]. Measurement2020149: 107002.
[3]
Ma ZhipengZhao MingLuo Mourui, et al. An integrated monitoring scheme for wind turbine main bearing using acoustic emission[J]. Signal Processing2023205: 108867.
[4]
Stetco ADinmohammadi FZhao X, et al. Machine learning methods for wind turbine condition monitoring: a review[J]. Renewable Energy2019133: 620-635.
[5]
姚齐水, 别帅帅, 余江鸿, 等. 一种结合改进Inception V2模块和CBAM的轴承故障诊断方法[J]. 振动工程学报202235(4): 949-957.
Yao QishuiBie ShuaishuaiYu Jianghong, et al. A bearing fault diagnosis method combining improved inception V2 module and CBAM[J]. Journal of Vibration Engineering202235(4): 949-957.
[6]
Xu ZifeiMei XuanWang Xinyu, et al. Fault diagnosis of wind turbine bearing using a multi-scale convolutional neural network with bidirectional long short term memory and weighted majority voting for multi-sensors[J]. Renewable Energy2022182: 615-626.
[7]
韩特, 刘超, 沈长青, 等. 深度嵌入度量学习的机械跨工况故障识别方法[J].振动工程学报202336(2):565-573.
Han TeLiu ChaoShen Changqing, et al. Deep embedding metric learning for machinery fault identification across different working conditions[J]. Journal of Vibration Engineering202336(2):565-573.
[8]
Lei YaguoYang BinJiang Xinwei, et al. Applications of machine learning to machine fault diagnosis: a review and roadmap[J]. Mechanical Systems and Signal Processing2020138: 106587.
[9]
Gomez J LKhelf IBourdon A, et al. Angular modeling of a rotating machine in non-stationary conditions: application to monitoring bearing defects of wind turbines with instantaneous angular speed[J]. Mechanism and Machine Theory2019136: 27-51.
[10]
Zhao ZhibinZhang QiyangYu Xiaolei, et al. Applications of unsupervised deep transfer learning to intelligent fault diagnosis: a survey and comparative study[J]. IEEE Transactions on Instrumentation and Measurement202170: 3525828.
[11]
Su HaoYang XinXiang Ling, et al. A novel method based on deep transfer unsupervised learning network for bearing fault diagnosis under variable working condition of unequal quantity[J]. Knowledge-Based Systems2022242: 108381.
[12]
Guo JunLu SiliangZhai Chao, et al. Automatic bearing fault diagnosis of permanent magnet synchronous generators in wind turbines subjected to noise interference[J]. Measurement Science and Technology201829(2): 025002.
[13]
董绍江, 裴雪武, 吴文亮, 等. 基于多层降噪技术及改进卷积神经网络的滚动轴承故障诊断方法[J]. 机械工程学报202157(1): 148-156.
Dong ShaojiangPei XuewuWu Wenliang, et al. Rolling bearing fault diagnosis method based on multilayer noise reduction technology and improved convolutional neural network[J]. Journal of Mechanical Engineering202157(1): 148-156.
[14]
Ren HeLiu WenyiShan Mengchen, et al. A new wind turbine health condition monitoring method based on VMD-MPE and feature-based transfer learning[J]. Measurement2019148: 106906.
[15]
Zhu JunChen NanShen Changqing. A new deep transfer learning method for bearing fault diagnosis under different working conditions[J]. IEEE Sensors Journal202020(15): 8394-8402.
[16]
Zhang YanLiu WenyiWang Xin, et al. A novel wind turbine fault diagnosis method based on compressed sensing and DTL-CNN[J]. Renewable Energy2022194: 249-258.
[17]
Zhang YanLiu WenyiGu Heng, et al. A novel wind turbine fault diagnosis based on deep transfer learning of improved residual network and multi-target data[J]. Measurement Science and Technology202233(9): 095007.
[18]
Liang PengfeiWang BinJiang Guoqian, et al. Unsupervised fault diagnosis of wind turbine bearing via a deep residual deformable convolution network based on subdomain adaptation under time-varying speeds[J]. Engineering Applications of Artificial Intelligence2023118: 105656.
[19]
He KaimingZhang XiangyuRen Shaoqing, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, N.V., USA, 2016: 770-778.
[20]
Gong LihuaXiang LingzhiLiu Sihang, et al. Born machine model based on matrix product state quantum circuit[J]. Physica A: Statistical Mechanics and its Applications2022593: 126907.
[21]
Che ChangchangWang HuaweiNi Xiaomei, et al. Domain adaptive deep belief network for rolling bearing fault diagnosis[J]. Computers & Industrial Engineering2020143: 106427.
[22]
Zhang WeiLi ChuanhaoPeng Gaoliang, et al. A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load[J]. Mechanical Systems and Signal Processing2018100: 439-453.
[23]
Li XiangZhang WeiDing Qian. Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics201966(7): 5525-5534.
[24]
Long MingshengCao YueCao Zhangjie, et al. Transferable representation learning with deep adaptation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence201841(12): 3071-3085.
2024年第37卷第2期
PDF下载
75
36
引用本文
BibTeX
文章信息
doi: 10.16385/j.cnki.issn.1004-4523.2024.02.018
  • 接收时间:2023-01-26
  • 首发时间:2026-02-10
  • 出版时间:2024-02-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2023-01-26
  • 修回日期:2023-03-26
基金
国家自然科学基金资助项目(62241308)
甘肃省技术创新引导计划-科技专员专项资助项目(22CX8GA130)
作者信息
    兰州理工大学机电工程学院,甘肃 兰州 730050
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zdgcxb/CN/10.16385/j.cnki.issn.1004-4523.2024.02.018
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