Article(id=1149780467467116891, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403576, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1715702400000, receivedDateStr=2024-05-15, revisedDate=1736438400000, revisedDateStr=2025-01-10, acceptedDate=null, acceptedDateStr=null, onlineDate=1752058625332, onlineDateStr=2025-07-09, pubDate=1744041600000, pubDateStr=2025-04-08, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752058625332, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752058625332, creator=13701087609, updateTime=1752058625332, updator=13701087609, issue=Issue{id=1149780466032669506, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='10', pageStart='3969', pageEnd='4395', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752058624990, creator=13701087609, updateTime=1768456644259, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218558743898411553, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218558743898411554, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=4199, endPage=4205, ext={EN=ArticleExt(id=1149780467710386525, articleId=1149780467467116891, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Evolution Indicators of Heavy-Duty Railway Rail Wave Wear Damage Based on t-SNE, columnId=1156262729162810294, journalTitle=Science Technology and Engineering, columnName=Papers·Automation and Computational Technology, runingTitle=null, highlight=null, articleAbstract=

Understanding the evolution law of rail service performance is of great significance for reducing the operation and maintenance costs of heavy-duty railway rails. Due to the complex and variable operating environment of rail tracks, which makes it difficult to construct scientifically effective damage evolution indicators to reflect objective development patterns, a method based on t-SNE(t-distributed stochastic neighbor embedding) was proposed for constructing the evolution law of corrugation damage. Firstly, the time-domain, frequency-domain, statistical, and entropy features were extracted from the original rail corrugation vibration signal. The random forest algorithm was then used to rank the features by importance, and the top-ranked features were selected to construct the feature vector. Dimensionality reduction was performed using t-SNE and other methods, and it is found that t-SNE demonstrates superior performance. The final temporal damage degradation index is obtained through Euclidean distance metric and median filtering for smoothing. The results indicate that this method provides good discrimination, anti-interference capability, and practical applicability for damage stages classification.

, correspAuthors=Zhong-mei WANG, 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=Zhong-mei WANG, Wei DENG, Jian-hua LIU, Peng-xuan NIE, Hai-bo WU, Wen-kun WANG), CN=ArticleExt(id=1149780484881867298, articleId=1149780467467116891, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于t-SNE的重载铁路钢轨波磨伤损演化指标, columnId=1156262729783567290, journalTitle=科学技术与工程, columnName=论文·自动化技术、计算机技术, runingTitle=null, highlight=null, articleAbstract=

认识钢轨服役性能演化规律对降低重载铁路钢轨运维成本具有重要意义。针对钢轨运行环境复杂多变,难以构建科学有效的伤损演化指标以反映客观发展规律的问题,提出了基于t-分布随机近邻分布(t-distributed stochastic neighbor embedding,t-SNE)的波磨伤损演化规律构建方法。首先,对钢轨原始波磨振动信号提取时域、频域、统计学、熵的特征指标;然后,使用随机森林算法对特征进行特征重要性排名,选取排名靠前的特征构建特征矢量;接着使用t-SNE等方式降维,验证了t-SNE更具优势,使用欧氏距离度量和中值滤波法进行平滑处理得到最终的时序性伤损退化指标。结果表明本文方法对于伤损阶段的划分具有较好的区分度、抗干扰能力和工程实用性。

, correspAuthors=王忠美, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=Q9ZXDfsgzE8O3Jqn2cyMpQ==, magXml=pz3q59oDWG4+D9crKlwfcQ==, pdfUrl=null, pdf=TsNTpuH21vLZb9e+iO6/RA==, pdfFileSize=4186754, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=gptcFZYiXbXvMIGwDC89Ag==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=7lX9YOym4Kr8w8t+ee6Z5A==, mapNumber=null, authorCompany=null, fund=null, authors=

王忠美(1984—),男,汉族,湖北荆州人,博士,讲师。研究方向:智能信息处理,智能诊断与健康监测。E-mail:

, authorsList=王忠美, 邓玮, 刘建华, 聂芃轩, 吴海波, 王文昆)}, authors=[Author(id=1218525108658950569, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=ldwangzm2008@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1218525108822528441, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525108658950569, language=EN, stringName=Zhong-mei WANG, firstName=Zhong-mei, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525108965134788, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525108658950569, language=CN, stringName=王忠美, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 湖南工业大学轨道交通学院, 株洲 412000, bio={"content":"

王忠美(1984—),男,汉族,湖北荆州人,博士,讲师。研究方向:智能信息处理,智能诊断与健康监测。E-mail:

"}, bioImg=null, bioContent=

王忠美(1984—),男,汉族,湖北荆州人,博士,讲师。研究方向:智能信息处理,智能诊断与健康监测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)])]), Author(id=1218525109090963927, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, 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=1218525109200015841, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109090963927, language=EN, stringName=Wei DENG, firstName=Wei, middleName=null, lastName=DENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525109325844969, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109090963927, language=CN, stringName=邓玮, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 湖南工业大学轨道交通学院, 株洲 412000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)])]), Author(id=1218525109460062714, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, 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=1218525109615251979, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109460062714, language=EN, stringName=Jian-hua LIU, firstName=Jian-hua, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525109732692504, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109460062714, language=CN, stringName=刘建华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 湖南工业大学轨道交通学院, 株洲 412000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)])]), Author(id=1218525109837550119, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, orderNo=3, 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=1218525109954990643, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109837550119, language=EN, stringName=Peng-xuan NIE, firstName=Peng-xuan, middleName=null, lastName=NIE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525110093402691, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525109837550119, language=CN, stringName=聂芃轩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 湖南工业大学轨道交通学院, 株洲 412000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)])]), Author(id=1218525110328283727, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, orderNo=4, 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=1218525110475084380, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525110328283727, language=EN, stringName=Hai-bo WU, firstName=Hai-bo, middleName=null, lastName=WU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525110605107811, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525110328283727, language=CN, stringName=吴海波, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 湖南工业大学轨道交通学院, 株洲 412000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)])]), Author(id=1218525110735131249, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, orderNo=5, 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=1218525110881931908, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525110735131249, language=EN, stringName=Wen-kun WANG, firstName=Wen-kun, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Zhuzhou Times Electronic Technology Co., Ltd., Zhuzhou 412007, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525111016149650, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, authorId=1218525110735131249, language=CN, stringName=王文昆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 株洲时代电子技术有限公司, 株洲 412007, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525108340183434, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=2, ext=[AuthorCompanyExt(id=1218525108474401179, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108340183434, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhuzhou Times Electronic Technology Co., Ltd., Zhuzhou 412007, China), AuthorCompanyExt(id=1218525108499567006, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108340183434, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 株洲时代电子技术有限公司, 株洲 412007)])])], keywords=[Keyword(id=1218525111217476257, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, orderNo=1, keyword=rail wave grinding), Keyword(id=1218525111339111085, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, orderNo=2, keyword=injury evolution law), Keyword(id=1218525111452357305, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, orderNo=3, keyword=degradation trend characteristics), Keyword(id=1218525111645295307, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, orderNo=4, keyword=t-SNE), Keyword(id=1218525111771124438, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, orderNo=5, keyword=damage stages), Keyword(id=1218525111943090912, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, orderNo=1, keyword=钢轨波磨), Keyword(id=1218525112064725743, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, orderNo=2, keyword=伤损演化规律), Keyword(id=1218525112165389051, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, orderNo=3, keyword=退化趋势特征), Keyword(id=1218525112261858052, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, orderNo=4, keyword=t-SNE), Keyword(id=1218525112400270090, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, orderNo=5, keyword=伤损阶段)], refs=[Reference(id=1218525115545997316, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=21, issue=9, pageStart=3595, pageEnd=3607, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=汪梦寒, 顾晓菡, 王安斌, journalName=铁道科学与工程学报, refType=null, unstructuredReference=汪梦寒, 顾晓菡, 王安斌. 科隆蛋扣件直线路段钢轨波磨形成机理研究[J]. 铁道科学与工程学报, 2024, 21(9): 3595-3607., articleTitle=科隆蛋扣件直线路段钢轨波磨形成机理研究, refAbstract=null), Reference(id=1218525115642466316, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=21, issue=9, pageStart=3595, pageEnd=3607, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Wang Menghan, Gu Xiaohan, Wang Anbin, journalName=Journal of Railway Science and Engineering, refType=null, unstructuredReference=Wang Menghan, Gu Xiaohan, Wang Anbin. Research on the mechanism of rail corrugation formation in straight track sections with Cologne-egg fasteners[J]. Journal of Railway Science and Engineering, 2024, 21(9): 3595-3607., articleTitle=Research on the mechanism of rail corrugation formation in straight track sections with Cologne-egg fasteners, refAbstract=null), Reference(id=1218525115743129623, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=2, pageStart=766, pageEnd=772, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=陈金明, 钟盛华, 刘逸平, journalName=科学技术与工程, refType=null, unstructuredReference=陈金明, 钟盛华, 刘逸平, 等. 基于车轨耦合动力学分析的地铁e型弹条扣件疲劳寿命预测[J]. 科学技术与工程, 2024, 24(2): 766-772., articleTitle=基于车轨耦合动力学分析的地铁e型弹条扣件疲劳寿命预测, refAbstract=null), Reference(id=1218525115835404317, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=2, pageStart=766, pageEnd=772, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Chen Jinming, Zhong Shenghua, Liu Yiping, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Chen Jinming, Zhong Shenghua, Liu Yiping. Fatigue life prediction of e-type clip fastener in metro based on vehicle-rail coupling dynamicsanalysis[J]. Science Technology and Engineering, 2024, 24(2): 766-772., articleTitle=Fatigue life prediction of e-type clip fastener in metro based on vehicle-rail coupling dynamicsanalysis, refAbstract=null), Reference(id=1218525115961233445, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=16, pageStart=7080, pageEnd=7086, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=王思维, 姚德臣, 杨建伟, journalName=科学技术与工程, refType=null, unstructuredReference=王思维, 姚德臣, 杨建伟, 等. 变速工况下基于WSST的地铁钢轨波磨检测[J]. 科学技术与工程, 2023, 23(16): 7080-7086., articleTitle=变速工况下基于WSST的地铁钢轨波磨检测, refAbstract=null), Reference(id=1218525116082868263, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=16, pageStart=7080, pageEnd=7086, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Wang Siwei, Yao Dechen, Yan Jianwei, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Siwei, Yao Dechen, Yan Jianwei, et al. Detection method of rail corugation in metro based on wavelet synchrosqueezing transform under variable speed[J]. Science Technology and Engineering, 2023, 23(16): 7080-7086., articleTitle=Detection method of rail corugation in metro based on wavelet synchrosqueezing transform under variable speed, refAbstract=null), Reference(id=1218525116238057520, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Bo W, Qing W, Shuihui L, journalName=Engineering Failure Analysis, refType=null, unstructuredReference=Bo W, Qing W, Shuihui L, et al. Investigation of plastic flow of rail vehicle wheels and its effects on rolling contact fatigue based on a dynamic method[J]. Engineering Failure Analysis, 2024.DOI: 10.1016/j.engfailanal.2024.108484., articleTitle=Investigation of plastic flow of rail vehicle wheels and its effects on rolling contact fatigue based on a dynamic method, refAbstract=null), Reference(id=1218525116359692340, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=13, pageStart=214, pageEnd=225, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=刘渊博, 陈相, 刘妤, journalName=振动与冲击, refType=null, unstructuredReference=刘渊博, 陈相, 刘妤. 基于卷积自编码器和时间卷积网络的轴承性能退化趋势预测[J]. 振动与冲击, 2023, 42(13): 214-225., articleTitle=基于卷积自编码器和时间卷积网络的轴承性能退化趋势预测, refAbstract=null), Reference(id=1218525116477132859, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=13, pageStart=214, pageEnd=225, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Liu Yuanbo, Chen Xiang, Liu Yu, journalName=Journal of Vibration and Shock, refType=null, unstructuredReference=Liu Yuanbo, Chen Xiang, Liu Yu. Prediction of bearing performance degradation trend based on CAE and TCN[J]. Journal of Vibration and Shock, 2023, 42(13): 214-225., articleTitle=Prediction of bearing performance degradation trend based on CAE and TCN, refAbstract=null), Reference(id=1218525116590379074, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=10, pageStart=3338, pageEnd=3349, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=吴梦蝶, 程龙生, 陈闻鹤, journalName=系统工程与电子技术, refType=null, unstructuredReference=吴梦蝶, 程龙生, 陈闻鹤. 基于自适应马氏空间与深度学习的滚动轴承退化趋势预测[J]. 系统工程与电子技术, 2023, 45(10): 3338-3349., articleTitle=基于自适应马氏空间与深度学习的滚动轴承退化趋势预测, refAbstract=null), Reference(id=1218525116691042376, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=10, pageStart=3338, pageEnd=3349, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=Wu Mengdie, Cheng Longsheng, Chen Wenhe, journalName=Systems Engineering and Electonics, refType=null, unstructuredReference=Wu Mengdie, Cheng Longsheng, Chen Wenhe. Degradation trend prediction of rolling bearing based on adaptive Mahalanobis space and deep learing[J]. Systems Engineering and Electonics, 2023, 45(10): 3338-3349., articleTitle=Degradation trend prediction of rolling bearing based on adaptive Mahalanobis space and deep learing, refAbstract=null), Reference(id=1218525116800094287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=10, pageStart=3941, pageEnd=3950, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=白雲杰, 贾希胜, 梁庆海, journalName=科学技术与工程, refType=null, unstructuredReference=白雲杰, 贾希胜, 梁庆海, 等. 基于深度学习的柴油机气门健康状态评估[J]. 科学技术与工程, 2022, 22(10): 3941-3950., articleTitle=基于深度学习的柴油机气门健康状态评估, refAbstract=null), Reference(id=1218525116892368980, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=10, pageStart=3941, pageEnd=3950, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=Bai Yunjie, Jia Xisheng, Liang Qinghai, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Bai Yunjie, Jia Xisheng, Liang Qinghai, et al. Evaluation of diesel engine valve health status based on deep learning[J]. Science Technology and Engineering, 2022, 22(10): 3941-3950., articleTitle=Evaluation of diesel engine valve health status based on deep learning, refAbstract=null), Reference(id=1218525117018198108, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=128, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=Wen L, Yang G, Hu L, journalName=Engineering Applications of Artificial Intelligence, refType=null, unstructuredReference=Wen L, Yang G, Hu L, et al. A new unsupervised health index estimation method for bearings early fault detection based on Gaussian mixture model[J]. Engineering Applications of Artificial Intelligence, 2024, 128.DOI: 10.1016/j.engappai.2023.107562., articleTitle=A new unsupervised health index estimation method for bearings early fault detection based on Gaussian mixture model, refAbstract=null), Reference(id=1218525117114667106, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=15, issue=21, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=Yuan X, Liu S, Feng W, journalName=Remote Sensing, refType=null, unstructuredReference=Yuan X, Liu S, Feng W, et al. Feature importance ranking of random forest-based end-to-end learning algorithm[J]. Remote Sensing, 2023, 15(21). DOI: 10.3390/rs15215203., articleTitle=Feature importance ranking of random forest-based end-to-end learning algorithm, refAbstract=null), Reference(id=1218525117227913321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=2023, issue=null, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=Yang J, Han S, Chen Y, journalName=Journal of Advanced Transportation, refType=null, unstructuredReference=Yang J, Han S, Chen Y. Prediction of traffic accident severity based on random forest[J]. Journal of Advanced Transportation, 2023, 2023: 1-8., articleTitle=Prediction of traffic accident severity based on random forest, refAbstract=null), Reference(id=1218525117336965231, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=15, pageStart=18715, pageEnd=18757, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=Mafarja M, Thaher T, Al-Betar M A, journalName=Applied Intelligence, refType=null, unstructuredReference=Mafarja M, Thaher T, Al-Betar M A, et al. Classification framework for faulty-software using enhanced exploratory whale optimizer-based feature selection scheme and random forest ensemble learning[J]. Applied Intelligence, 2023, 53(15): 18715-18757., articleTitle=Classification framework for faulty-software using enhanced exploratory whale optimizer-based feature selection scheme and random forest ensemble learning, refAbstract=null), Reference(id=1218525117425045620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=13, issue=12, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=Wang H, Li D, Li D, journalName=Applied Sciences, refType=null, unstructuredReference=Wang H, Li D, Li D, et al. Remaining useful life prediction of aircraft turbofan engine based on random forest feature selection and multi-layer perceptron[J]. Applied Sciences, 2023, 13(12). DOI: 10.3390/app13127186., articleTitle=Remaining useful life prediction of aircraft turbofan engine based on random forest feature selection and multi-layer perceptron, refAbstract=null), Reference(id=1218525117513126009, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=5, pageStart=973, pageEnd=991, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=Han Y, Song X, Shi J, journalName=Transactions of the Institute of Measurement and Control, refType=null, unstructuredReference=Han Y, Song X, Shi J, et al. KPCA-WPHM-SCNs-based remaining useful life prediction method for motor rolling bearings[J]. Transactions of the Institute of Measurement and Control, 2024, 46(5): 973-991., articleTitle=KPCA-WPHM-SCNs-based remaining useful life prediction method for motor rolling bearings, refAbstract=null), Reference(id=1218525117609595006, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=18, issue=5, pageStart=3839, pageEnd=3848, url=null, language=null, rfNumber=[14], rfOrder=19, authorNames=Zhao Y, Zhou M, Wang L, journalName=Journal of Electrical Engineering & Technology, refType=null, unstructuredReference=Zhao Y, Zhou M, Wang L, et al. Bearing fault diagnosis of single-channel data by a 3D dcn with bilinear lbp and modified KPCA[J]. Journal of Electrical Engineering & Technology, 2023, 18(5): 3839-3848., articleTitle=Bearing fault diagnosis of single-channel data by a 3D dcn with bilinear lbp and modified KPCA, refAbstract=null), Reference(id=1218525117710258304, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=11, pageStart=2222, pageEnd=2239, url=null, language=null, rfNumber=[15], rfOrder=20, authorNames=Pan Y, Wang H, Chen J, journalName=Transactions of the Institute of Measurement and Control, refType=null, unstructuredReference=Pan Y, Wang H, Chen J, et al. A new fault component selection strategy based on statistical detection for slewing bearing weak signal de-noising[J]. Transactions of the Institute of Measurement and Control, 2024, 46(11): 2222-2239., articleTitle=A new fault component selection strategy based on statistical detection for slewing bearing weak signal de-noising, refAbstract=null), Reference(id=1218525117810921605, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=34, issue=12, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=21, authorNames=Pan H, Jiao W, Yan T, journalName=Mea-surement Science and Technology, refType=null, unstructuredReference=Pan H, Jiao W, Yan T, et al. Combining kernel principal component analysis and spatial group-wise enhance convolutional neural network for fault recognition of rolling element bearings[J]. Mea-surement Science and Technology, 2023, 34(12). DOI: 10.1088/1361-6501/ace790., articleTitle=Combining kernel principal component analysis and spatial group-wise enhance convolutional neural network for fault recognition of rolling element bearings, refAbstract=null), Reference(id=1218525117911584906, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=11, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=22, authorNames=Su H, Wang Z, Cai Y, journalName=Machines, refType=null, unstructuredReference=Su H, Wang Z, Cai Y, et al. Refined composite Multiscale fluctuation dispersion entropy and supervised manifold mapping for planetary gearbox fault diagnosis[J]. Machines, 2023, 11(1). DOI: 10.3390/machines11010047., articleTitle=Refined composite Multiscale fluctuation dispersion entropy and supervised manifold mapping for planetary gearbox fault diagnosis, refAbstract=null), Reference(id=1218525118016442513, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=24, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=23, authorNames=Wang H, Yao L, Wang H, journalName=Sensors, refType=null, unstructuredReference=Wang H, Yao L, Wang H, et al. Supervised manifold learning based on multi-feature information discriminative fusion within an adaptive nearest neighbor strategy applied to rolling bearing fault diagnosis[J]. Sensors, 2023, 23(24). DOI: 10.3390/s23249820., articleTitle=Supervised manifold learning based on multi-feature information discriminative fusion within an adaptive nearest neighbor strategy applied to rolling bearing fault diagnosis, refAbstract=null), Reference(id=1218525118112911511, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=12, issue=3, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=24, authorNames=Lu W, Shi C, Fu H, journalName=Electronics, refType=null, unstructuredReference=Lu W, Shi C, Fu H, et al. A power transformer fault diagnosis method based on improved sand cat swarm optimization algorithm and bidirectional gated recurrent unit[J]. Electronics, 2023, 12(3). DOI: 10.3390/electronics12030672., articleTitle=A power transformer fault diagnosis method based on improved sand cat swarm optimization algorithm and bidirectional gated recurrent unit, refAbstract=null), Reference(id=1218525118226157722, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=17, issue=6, pageStart=773, pageEnd=787, url=null, language=null, rfNumber=[20], rfOrder=25, authorNames=Lu W, Shi C, Fu H, journalName=IET Electric Power Applications, refType=null, unstructuredReference=Lu W, Shi C, Fu H, et al. Research on transformer fault diagnosis based on ISOMAP and IChOA-LSSVM[J]. IET Electric Power Applications, 2023, 17(6): 773-787., articleTitle=Research on transformer fault diagnosis based on ISOMAP and IChOA-LSSVM, refAbstract=null), Reference(id=1218525118435872930, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2024, volume=18, issue=3, pageStart=297, pageEnd=311, url=null, language=null, rfNumber=[21], rfOrder=26, authorNames=Lee C Y, Maceren E D C, journalName=IET Electric Power Applications, refType=null, unstructuredReference=Lee C Y, Maceren E D C. Induction motor bearing fault classification using deep neural network with particle swarm optimization-extreme gradient boosting[J]. IET Electric Power Applications, 2024, 18(3): 297-311., articleTitle=Induction motor bearing fault classification using deep neural network with particle swarm optimization-extreme gradient boosting, refAbstract=null), Reference(id=1218525118528147623, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=7, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=27, authorNames=Zair M, Rahmoune C, Imane M, journalName=Journal of the Brazilian Society of Mechanical Sciences and Engineering, refType=null, unstructuredReference=Zair M, Rahmoune C, Imane M, et al. A new health indicator extracted by unsupervised learning using autoencoder in tandem with t-SNE and multi-kernel CNN to enhance the early detection and classification of bearings multi-faults[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2023, 45(7). DOI: 10.1007/s40430-023-04307-6., articleTitle=A new health indicator extracted by unsupervised learning using autoencoder in tandem with t-SNE and multi-kernel CNN to enhance the early detection and classification of bearings multi-faults, refAbstract=null), Reference(id=1218525118637199530, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, doi=null, pmid=null, pmcid=null, year=2022, volume=191, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=28, authorNames=Jiang W, Xu Y, Chen Z, journalName=Measurement, refType=null, unstructuredReference=Jiang W, Xu Y, Chen Z, et al. Fault diagnosis for rolling bearing using a hybrid hierarchical method based on scale-variable dispersion entropy and parametric t-SNE algorithm[J]. Measurement, 2022, 191.DOI: 10.1016/j.measurement.2022.110843., articleTitle=Fault diagnosis for rolling bearing using a hybrid hierarchical method based on scale-variable dispersion entropy and parametric t-SNE algorithm, refAbstract=null)], funds=[Fund(id=1218525115080430565, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, awardId=2021YFE05011, language=CN, fundingSource=国家重点研发计划(2021YFE05011), fundOrder=null, country=null), Fund(id=1218525115202065391, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, awardId=22B0586, language=CN, fundingSource=湖南省教育厅青年项目(22B0586), fundOrder=null, country=null), Fund(id=1218525115327894520, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, awardId=2022JGYB186, language=CN, fundingSource=湖南省教育厅教改项目(2022JGYB186), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1218525108201771387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=1, ext=[AuthorCompanyExt(id=1218525108218548606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China), AuthorCompanyExt(id=1218525108231131519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108201771387, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 湖南工业大学轨道交通学院, 株洲 412000)]), AuthorCompany(id=1218525108340183434, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, xref=2, ext=[AuthorCompanyExt(id=1218525108474401179, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108340183434, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhuzhou Times Electronic Technology Co., Ltd., Zhuzhou 412007, China), AuthorCompanyExt(id=1218525108499567006, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, companyId=1218525108340183434, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 株洲时代电子技术有限公司, 株洲 412007)])], figs=[ArticleFig(id=1218525112677094182, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Fig.1, caption=Ranking of feature importance in random forest, figureFileSmall=x4j6v7gkn3HhUntyw9Mx1Q==, figureFileBig=y24+sYPSgNJDJVjMyFikyQ==, tableContent=null), ArticleFig(id=1218525112794534705, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=图1, caption=随机森林特征重要性排名结果, figureFileSmall=x4j6v7gkn3HhUntyw9Mx1Q==, figureFileBig=y24+sYPSgNJDJVjMyFikyQ==, tableContent=null), ArticleFig(id=1218525112949723973, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Fig.2, caption=Visualization of the dimensionality reduction effect using KPCA, figureFileSmall=rXA5wo+DruE8+Tby6AJSOQ==, figureFileBig=dhUvbTy/cnyJmDHf3DCAig==, tableContent=null), ArticleFig(id=1218525113088136016, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=图2, caption=KPCA的降维效果可视化, figureFileSmall=rXA5wo+DruE8+Tby6AJSOQ==, figureFileBig=dhUvbTy/cnyJmDHf3DCAig==, tableContent=null), ArticleFig(id=1218525113226548064, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Fig.3, caption=Visualization of the dimensionality reduction effect using ISOMAP, figureFileSmall=KWtK4n9PcvIoeJt0JqUtJw==, figureFileBig=t0ErTAC+S2Qd7bOBvPT4IA==, tableContent=null), ArticleFig(id=1218525113360765802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=图3, caption=ISOMAP的处理效果可视化, figureFileSmall=KWtK4n9PcvIoeJt0JqUtJw==, figureFileBig=t0ErTAC+S2Qd7bOBvPT4IA==, tableContent=null), ArticleFig(id=1218525113469817717, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Fig.4, caption=Visualization of the dimensionality reduction effect using t-SNE, figureFileSmall=cjG1Fe0HTHsE5Fx/7grFJw==, figureFileBig=bgNoj2gVvfjMn5ifjGCgTg==, tableContent=null), ArticleFig(id=1218525113566286716, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=图4, caption=t-SNE的处理效果可视化, figureFileSmall=cjG1Fe0HTHsE5Fx/7grFJw==, figureFileBig=bgNoj2gVvfjMn5ifjGCgTg==, tableContent=null), ArticleFig(id=1218525113721475976, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Fig.5, caption=The final damage evolution metrics obtained using the method proposed in this article, figureFileSmall=mogQfm2Qmzuo58zfc1yJ5Q==, figureFileBig=99a/stzYNC4JZkGonDj94g==, tableContent=null), ArticleFig(id=1218525113947968404, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=图5, caption=使用本文方法得到的最终伤损演化指标, figureFileSmall=mogQfm2Qmzuo58zfc1yJ5Q==, figureFileBig=99a/stzYNC4JZkGonDj94g==, tableContent=null), ArticleFig(id=1218525114094769057, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Table 1, caption=

Theextracted Features from various domains in this article

, figureFileSmall=null, figureFileBig=null, tableContent=
指标分类 特征名称
统计学 最大值,最小值,均值,均方值,方差,方根幅值
时域 峰峰值,峭度,峭度因子,偏度,波形因子,峰值因子,脉冲因子,裕度因子,形状因子,谱峭度的均值,谱峭度的标准差,谱峭度的偏度,谱峭度的峭度
频域 中心频率,均方频率,均方根频率,频率方差,能量特征
熵特征 Tsallis熵,Shannon熵,Renyi熵
), ArticleFig(id=1218525114325455787, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=表1, caption=

本文提取的各域特征

, figureFileSmall=null, figureFileBig=null, tableContent=
指标分类 特征名称
统计学 最大值,最小值,均值,均方值,方差,方根幅值
时域 峰峰值,峭度,峭度因子,偏度,波形因子,峰值因子,脉冲因子,裕度因子,形状因子,谱峭度的均值,谱峭度的标准差,谱峭度的偏度,谱峭度的峭度
频域 中心频率,均方频率,均方根频率,频率方差,能量特征
熵特征 Tsallis熵,Shannon熵,Renyi熵
), ArticleFig(id=1218525114480645046, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Table 2, caption=

Parameter settings for three dimensionality reduction technique

, figureFileSmall=null, figureFileBig=null, tableContent=
降维方法 相关参数设置
t-SNE Perp=30
ISOMAP n_neighbors=30
KPCA Kernel=Gauss
), ArticleFig(id=1218525114606474179, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=表2, caption=

3种降维方法参数设置

, figureFileSmall=null, figureFileBig=null, tableContent=
降维方法 相关参数设置
t-SNE Perp=30
ISOMAP n_neighbors=30
KPCA Kernel=Gauss
), ArticleFig(id=1218525114715526089, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=EN, label=Table 3, caption=

Variation of indicators at each stage of damage evolution

, figureFileSmall=null, figureFileBig=null, tableContent=
演化阶段 采样区间 指标规律 指标分布区间 均值 标准差 平均绝对偏差
[1,6] 缓慢增长 [0.000,0.386] 0.167 0.130 0.103
[7,23] 平缓发展 [0.961,1.436] 1.167 0.202 0.189
[24,49] 平缓发展 [7.538,7.925] 7.709 0.138 0.132
[50,98] 迅速增长 [8.068,17.130] 13.374 2.907 2.633
), ArticleFig(id=1218525114849743829, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780467467116891, language=CN, label=表3, caption=

伤损演化各阶段的指标变化情况

, figureFileSmall=null, figureFileBig=null, tableContent=
演化阶段 采样区间 指标规律 指标分布区间 均值 标准差 平均绝对偏差
[1,6] 缓慢增长 [0.000,0.386] 0.167 0.130 0.103
[7,23] 平缓发展 [0.961,1.436] 1.167 0.202 0.189
[24,49] 平缓发展 [7.538,7.925] 7.709 0.138 0.132
[50,98] 迅速增长 [8.068,17.130] 13.374 2.907 2.633
)], 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.1671-1815.2403576, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403576, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403576, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403576, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于t-SNE的重载铁路钢轨波磨伤损演化指标
收藏切换
PDF下载
王忠美 1 , 邓玮 1 , 刘建华 1 , 聂芃轩 1 , 吴海波 1 , 王文昆 2
科学技术与工程 | 论文·自动化技术、计算机技术 2025,25(10): 4199-4205
收起
收藏切换
科学技术与工程 | 论文·自动化技术、计算机技术 2025, 25(10): 4199-4205
基于t-SNE的重载铁路钢轨波磨伤损演化指标
全屏
王忠美1 , 邓玮1, 刘建华1, 聂芃轩1, 吴海波1, 王文昆2
作者信息
  • 1 湖南工业大学轨道交通学院, 株洲 412000
  • 2 株洲时代电子技术有限公司, 株洲 412007
  • 王忠美(1984—),男,汉族,湖北荆州人,博士,讲师。研究方向:智能信息处理,智能诊断与健康监测。E-mail:

Evolution Indicators of Heavy-Duty Railway Rail Wave Wear Damage Based on t-SNE
Zhong-mei WANG1 , Wei DENG1, Jian-hua LIU1, Peng-xuan NIE1, Hai-bo WU1, Wen-kun WANG2
Affiliations
  • 1 College of Railway Transportation, Hunan University of Technology, Zhuzhou 412000, China
  • 2 Zhuzhou Times Electronic Technology Co., Ltd., Zhuzhou 412007, China
出版时间: 2025-04-08 doi: 10.12404/j.issn.1671-1815.2403576
文章导航
收藏切换

认识钢轨服役性能演化规律对降低重载铁路钢轨运维成本具有重要意义。针对钢轨运行环境复杂多变,难以构建科学有效的伤损演化指标以反映客观发展规律的问题,提出了基于t-分布随机近邻分布(t-distributed stochastic neighbor embedding,t-SNE)的波磨伤损演化规律构建方法。首先,对钢轨原始波磨振动信号提取时域、频域、统计学、熵的特征指标;然后,使用随机森林算法对特征进行特征重要性排名,选取排名靠前的特征构建特征矢量;接着使用t-SNE等方式降维,验证了t-SNE更具优势,使用欧氏距离度量和中值滤波法进行平滑处理得到最终的时序性伤损退化指标。结果表明本文方法对于伤损阶段的划分具有较好的区分度、抗干扰能力和工程实用性。

钢轨波磨  /  伤损演化规律  /  退化趋势特征  /  t-SNE  /  伤损阶段

Understanding the evolution law of rail service performance is of great significance for reducing the operation and maintenance costs of heavy-duty railway rails. Due to the complex and variable operating environment of rail tracks, which makes it difficult to construct scientifically effective damage evolution indicators to reflect objective development patterns, a method based on t-SNE(t-distributed stochastic neighbor embedding) was proposed for constructing the evolution law of corrugation damage. Firstly, the time-domain, frequency-domain, statistical, and entropy features were extracted from the original rail corrugation vibration signal. The random forest algorithm was then used to rank the features by importance, and the top-ranked features were selected to construct the feature vector. Dimensionality reduction was performed using t-SNE and other methods, and it is found that t-SNE demonstrates superior performance. The final temporal damage degradation index is obtained through Euclidean distance metric and median filtering for smoothing. The results indicate that this method provides good discrimination, anti-interference capability, and practical applicability for damage stages classification.

rail wave grinding  /  injury evolution law  /  degradation trend characteristics  /  t-SNE  /  damage stages
王忠美, 邓玮, 刘建华, 聂芃轩, 吴海波, 王文昆. 基于t-SNE的重载铁路钢轨波磨伤损演化指标. 科学技术与工程, 2025 , 25 (10) : 4199 -4205 . DOI: 10.12404/j.issn.1671-1815.2403576
Zhong-mei WANG, Wei DENG, Jian-hua LIU, Peng-xuan NIE, Hai-bo WU, Wen-kun WANG. Evolution Indicators of Heavy-Duty Railway Rail Wave Wear Damage Based on t-SNE[J]. Science Technology and Engineering, 2025 , 25 (10) : 4199 -4205 . DOI: 10.12404/j.issn.1671-1815.2403576
重载铁路凭借强大、高效、低成本的运输能力,成为国民经济中大宗散装货物的重要交通运输方式。铁路钢轨在列车总重大、运量大、行车密度和速度高、运行环境恶劣等严苛条件下长时间服役,不可避免地面临伤损劣化失效进而影响行车安全的问题。轻微伤损主要采取钢轨打磨,严重伤损往往需要更换钢轨。钢轨维护过早会浪费人力物力、推高运维成本,以至影响重载铁路竞争力;维护过晚则会导致安全事故造成巨大损失。深刻地研究钢轨的伤损演化规律进而采取合理的运维策略,对重载铁路的长远发展具有重要的意义。
目前对于工业原件伤损的研究方法主要分为基于机理建模和基于数据建模两大方向。机理模型或称物理模型,在解决工程伤损演化问题上被广泛地应用。机理模型是描述过程机制或原理的模型。不仅仅包括定量的数学模型,也应该包括定性的描述变量间的驱动关系模型。汪梦寒等[1]使用ABAQUS建立轮轨耦合有限元模型,分析轨道模态特性和轮轨瞬态滚动特性,发现波磨的产生主要与轨道结构中高频固有特性相关,高频模态特征被激发是导致和加剧波磨的根本原因;陈金明等[2]为研究地铁e型弹条扣件的疲劳性能,通过有限元软件ABAQUS仿真扣件系统进行疲劳寿命的预测分析;王思维等[3]使用多体动力学仿真软件SIMPACK构建了车辆动力学模型,对牵引及制动工况下的钢轨波磨进行诊断,实现故障频率的准确定位;Bo等[4]基于轮轨接触力学和多体动力学方法研究车轮的塑性滚动,指出缓解塑性滚动可以减少车轮的滚动接触疲劳损伤;基于机理建模深入探讨了工业元件退化的本质,理论上可以对元件的性能达到较好的预测精度,然而对于相互作用关系较为复杂的元件而言,建模过程将变得非常复杂、有非常严格的适用条件和较差的迁移性,需要研究人员对元件运行的深层机理有着较深的了解。微小的差异和扰动在经过多个环节的放大和积累后,都有可能使得模型和实际产生巨大的差距,造成模型失效。
基于数据建模的预测方法能够克服机理模型对研究者学术专业或生产经验要求较高、建模过程复杂等困难,直接从信号数据中获取设备的退化趋势与特征,运用机器学习预测设备的剩余使用寿命、演化规律。刘渊博等[5]提出了结合了核主成分分析(kernel principal component analysis, KPCA)和卷积自编码器构建健康指标;吴梦蝶等[6]通过集合经验模态分解分解振动信号获取特征、马氏距离表征退化趋势;白雲杰等[7]基于小波包分解和一维卷积神经网络构建了柴油发动机气门芯的健康评估指标;Wen等[8]利用高斯混合建模构建轴承健康指数、划分健康阶段和检测早期故障;基于数据驱动的伤损演化规律研究方法通过研究系统的输出响应信号以探索元件劣化衰退的内在规律,在避免了传统机理建模模型复杂、限制条件苛刻可移植性差、计算量巨大等缺陷的同时,取得了较好的拟合效果。然而,上述方法均存在一定缺陷:所提方法往往过于侧重于对伤损指标的趋势预测,对于构建的伤损指标本身往往缺乏较好的鲁棒性,不能体现元件伤损的客观发展规律,容易造成伤损的误诊断。
针对钢轨运行工况复杂,难以获取科学可靠的伤损演化特征难的问题,现提出基于随机森林的特征筛选以获取高维度的特征集、基于t-SNE的特征融合以获取高维特征集蕴含的低维度演化趋势;针对钢轨运行工况复杂多变、环境噪声干扰较大导致伤损指标的时序性较差的问题,提出结合欧氏距离和中值滤波法以获得伤损退化指标;使用实际钢轨波磨的振动数据评估波磨伤损的发展进程,证实本文方法在研究钢轨波磨伤损领域构建演化指标、评估伤损发展、探索演化规律等系列问题上的有效性和科学性。
为构建重载铁路钢轨波磨损伤的伤损指标,使用列车行驶通过伤损路段产生的振动数据作为输入,使用箱型图法剔除异常信号,特征提取和t-SNE降维处理获取高维特征空间蕴含的低维度内在的本质演化趋势属性,进而使用欧氏空间度量体现其时序性、使用中值滤波法削弱噪声波动获得最终的演化指标,进行演化规律的评估。
提取的特征表现不一,为了选择较好的特征,使用随机森林特征重要性排名对特征进行筛选。随机森林(random forest)由若干决策树(decision tree)组成。随机森林的主要步骤如下。
(1)采用抽样放回的方法从样本集中选取n个样本作为一个训练集。
(2)使用步骤(1)得到的样本集生成一颗决策树,在生成的每一节点上:随机不重复地选择d个特征,利用它们分别划分样本集,本文依据基尼指数作为特征划分的依据指标。
(3)重复上述步骤共k次,其中的k即为随机森林中决策树的个数。
(4)用训练得到的随机森林对测试样本进行预测,依据票选法决定最终的结果。
求得各个特征在所有决策树上所做的贡献量即求出各个特征在决策树节点上分枝前后的基尼指数(Gini index)的差值。将各特征基尼指数变化量除以全体特征基尼指数总和,以求出各特征在全体特征中的贡献量,最后再依据贡献量大小排序,并剔除在特征贡献度中排名靠后的特征。
设共有k个特征类别, pi表示第i个类别的概率估计值,得到基尼指数Gini(p)为
Gini(p)= i = 1 kpi(1-pi)=1- i = 1 k p i 2
变量在节点m的重要性,即节点m分枝前后,Gini指数变化量为
VIMjm=GIm-GIl-GIr
式(2)中:VIMjm为特征j在节点m上的基尼指数变化值;GIm为分枝前的基尼指数;GIl和GIr为节点m分枝后,产生的两个新节点的基尼指数。
特征j在决策树中出现在节点集合M中,由此得到特征j在第i棵决策树上的基尼指数变化量为
VIMij= m MVIMjm
假定随机森林中有n棵决策树,特征j的总基尼指数变化量为
VIMj= i = 1 nVIMij
特征j对于全体特征的贡献占比为
VIM'j= V I M j i = 1 c V I M i
式(5)中:VIM'j为特征j对于全体特征的贡献占比; i = 1 cVIMi为所有特征基尼指数总和。
本文研究使用基于随机森林特征重要性排名的特征筛选对提取的钢轨特征进行筛选以获取表现较好、能代表钢轨伤损劣化发展的时序性特征,这一方法能够较好地解决传统特征筛选方法抑制过拟合效果较差、不能较好体现特征之间相互关系、处理高维数据面临维度灾难等等困难。该方法在遥感[9]、交通安全预测[10]、故障检测[11]、剩余寿命预测[12]等领域都得到了广泛的应用。
X为高维空间数据集,维度为D,样本个数为n,Y为低维度特征数据集,维度为d,满足d<D,即:X={x1,x2,···,xn}⊆RD,Y={y1,y2,···,yn}⊆Rd若存在映射f(·)可以使∀yjY,f(yj)=xj,则可认为低维的Y可以表征高维的特征数据集X。这一过程被称为降维,能够去除数据的冗余信息、分析数据的内在规律和特征。本文研究通过对钢轨多种特征进行降维以获得钢轨随时间推进的演化特征。
重载铁路钢轨工作环境复杂严苛,加之研究对象振动信号本身具有波动性,毫无疑问会导致数据样本的高度非线性。传统上为了获取特征集的共有属性常采用的主成分分析(principal component analysis, PCA)线性降维方法不能够很好地获取高维复杂的数据样本所蕴含低维本质的退化趋势。为克服以上缺陷,学界提出了一系列改进或替代方案。作为PCA的改进方案之一,KPCA[13-16]在一些故障诊断实例中取得了较好的效果;等距特征映射(isometric feature mapping, ISOMAP)[17-20] 作为一种重要的流形学习方法也在伤损诊断领域得到了广泛的应用。
t-SNE在高维非线性数据的处理上得到了广泛应用。Lee等[21]通过深度神经网络自适应提取隐藏特征结合t-SNE提高了轴承故障分类的精度,在不同噪音水平下依然体现了较好的准确度和鲁棒性;Zair等[22]提出了基于自编码器、t-SNE和多核卷积神经网络的故障自动检测和分类;Jiang等[23]提出了一种基于尺度变量离散熵和参数t分布随机邻域嵌入算法的滚动轴承故障诊断的混合分层故障诊断方法,提高了诊断的准确性和速度。
t-SNE原理是将高维空间特征数据集X内的点通过条件概率的形式表示,基于“高维空间特征样本点X之间服从高斯分布、低维度特征样本点Y之间服从t分布”的假设之下,提出了以下算法原理。
(1)x·表示X中的一点,计算高维空间中xixj两两样本的条件概率密度函数 p j i
$p_{j \mid i}=\frac{\exp \frac{-\left\|x_{i}-x_{j}\right\|^{2}}{2 \sigma_{i}^{2}}}{\sum_{k \neq i} \exp \frac{-\left\|x_{i}-x_{k}\right\|^{2}}{2 \sigma_{i}^{2}}}$
式(6)中:σi为以xi为中心的高斯分布方差,可根据困惑度Prep和二分搜索确定,困惑度的公式为
$\text { Prep }=2^{-\sum_{j=1}^{n} P_{j \mid i} \log _{2} P_{j \mid i}}$
(2)计算xixj两两样本之间的联合概率密度函数。
$P_{i j}=\frac{P_{i \mid j}+P_{j \mid i}}{2 n}$
(3)构建低维数据样本概率分布。利用自由度为1的t分布计算xixj在低维空间中映射的样本yiyj的联合概率分布Qij
$Q_{i j}=\frac{\left(1+\left\|y_{i}-y_{j}\right\|^{2}\right)^{-1}}{\sum_{k \neq l}\left(1+\left\|y_{k}-y_{l}\right\|^{2}\right)^{-1}}$
(4)计算梯度 C Y的值。
$\frac{\partial C}{\partial Y}=4 \sum_{j=1}^{n}\left(P_{i j}-Q_{i j}\right)\left(y_{i}-y_{j}\right)\left(1+\left\|y_{i}-y_{j}\right\|^{2}\right)^{-1}$
式(10)中:C为KL散度,用以表示高维度P概率分布Pij与低维度Q概率分布Qij的相似度。
$C=\mathrm{KL}(P \| Q)=\sum_{i=1}^{n} \sum_{j=1}^{n} P_{i j} \ln \frac{P_{i j}}{Q_{i j}} $
(5)计算得到低维度映射数据。
$Y^{k}=Y^{k-1}+\alpha \frac{\partial C}{\partial Y}+m(k)\left(Y^{k-1}-Y^{k-2}\right)$
式(12)中:k为循环次数;α为学习效率;m为动量因子;Y(k)为迭代k次后的低维度映射数据。
(6)迭代循环步骤(3)~步骤(5)直至迭代次数达到设置要求。
t-SNE能够保留数据点的局部结构有利于获取数据中的局部聚类与结构,能够很好地处理钢轨波磨振动数据复杂的非线性因素,较好地解决传统降维方法特征点距离过近、相互叠压不利于伤损诊断的缺点。同时,作为一种无监督的机器学习算法,t-SNE能够更好地脱离人为主观判断对结果的影响,挖掘数据时序发展的内在规律,并且有更强的可移植性,能适应不同工况和环境的变化。
本次实验数据使用国内一家轨道交通装备电子公司160 m试验线路的轨检车,获取钢轨从健康到波磨伤损产生和发展直至钢轨失效的全过程振动信号。采样频率4 kHz,采样周期不变,采集共计98个振动数据样本,每个样本包含60 000个采样点。
通过箱型图法剔除各采样周期中的异常离群值,以使特征提取能够取得较好的效果。
特征提取具备降低数据维度去除冗余信息、提高计算效率、提高模型性能减少过拟合的风险、增强模型的可解释性和通用性等优点。
对剔除异常离群值后的振动信号进行特征提取,获取统计学、时域、频域、熵特征共计27种特征指标,如表1所示,处理后数据维度变成98×27。各个特征的幅值、量纲都不尽相同,对各个特征进行[0,1]归一化处理,以确保后续步骤的客观性。
对2.1节中获取的已完成归一化的27项特征进行随机森林特征重要性排名,结果如图1所示。
选取排名靠前的6种特征,构建高维特征矢量F,其维度为98×6。
F= F 1 F 2 F R= X A 1 X B 1 ··· X N 1 X A 2 X B 2 ··· X N 2 X A R X B R ··· X N R
式(13)中:的行向量[ X A i X B i ··· X N i]表示第i次采样的数据特征;列向量[ X j 1 X j 2 ··· X j R]T表示入选的第j种特征。
使用t-SNE、ISOMAP、KPCA对特征筛选后获得的高维特征矢量F降维至二维(数据维度为98×2),以便对降维效果实现可视化观察,表2为3种方法相关参数的设置值,通过观察降维后特征点的分布以评价降维效果的好坏。图2~图4为3种降维效果的结果,横轴纵轴分别表示降维后的二维特征数值(均为无单位的量),特征点旁的数字表示该点对应的采样顺序,取值为1~98按顺序对应98次采样点。
可以观察到,就降维后特征点的分布而言,ISOMAP的第一和第二维度的特征主要分别分布于区间[-0.5,0.5]和[-0.1,0.1];KPCA的两个维度的特征主要分别分布于区间[-0.1,0.3]和[-0.1,0.1];两者在降维后均出现了大量特征点在很小区间范围内密集混叠距离过近的现象,降维处理效果较差,处理的特征指标的值在数值上非常接近,这意味着指标对抗扰动的性能可能很差,稍有扰动可能会导致很严重的误诊断,不能对不同伤损阶段做出有效且有说服力的划分;而对于t-SNE,在降维后第一和第二维度的特征分别分布于区间[-6,4]和[-7.5,10]的广大区间内,特征点在分布上出现了几组聚类,而且采样序号接近的特征点相对距离较近,特征点采样序号较远的点距离相对较远,结果意味着:不同阶段的特征指标可以具有较好的区分度,即使出现一定的扰动也可以避免误诊断,使用该方法获取特征指标的效果明显好于前两种方法,在工程应用上具备较好的实用性。
对t-SNE处理后的样本采用欧氏空间度量,以首次采样点的数据作为基线数据,计算后续样本与基线数据在欧氏度量空间下的距离。任一特征点距离初始点的距离能够体现出研究对象随时间推进,当前和初始健康状态之间的特性差异和演化程度,即伤损指标的时序性。
主观上,研究对象重载铁路钢轨工作环境严苛复杂常常伴有各种干扰信号,并且采集的振动信号本身具有波动性;客观上,人工和设备在采集过程中也无可避免会产生测量误差。在各种复杂的因素作用下,伤损指标毫无疑问会出现频繁的波动,可能会为铁路运行维护带来误诊断甚至安全风险,为了尽可能消除波动,得到尽可能稳定平滑的指标,本文使用中值滤波法处理伤损指标,得到了最终无单位的钢轨波磨伤损演化指标(图5)。
依据伤损指标的发展,可以将钢轨的波磨伤损演化大致划分为4个阶段(表3),钢轨的波磨伤损演化第一阶段对应采样周期[1,6]伤损指标从0缓慢增长至0.386;第二阶段[7,23],指标迅速跃升激增至1.401,随后在区间[0.961,1.436]内平缓发展;进入第三阶段[24,39],指标再次迅速激增达到7.538,接着在区间[7.538,7.925]内波动;随着伤损进一步发展进入第四阶段[40,98],指标呈现阶梯状快速增长直至达到16~17,钢轨完全达到使用极限。
另一方面,就伤损指标的区分度而言,前三阶段伤损指标均值、分布区间,后一阶段都是前一阶段的数倍;同时在前三区间内部反映数据波动性的指标标准差、平均绝对误差都相对较小,表明指标有较好的稳定性;前三周期的伤损指标数值在不同区间之间区分度大、区间内部分布较为均匀稳定,体现出了本文方法可以对伤损等级进行准确的划分、有效避免误诊断并及时采取对应的措施。第四周期指标呈现阶梯状快速上升趋势,意味着伤损进入快速劣化阶段。
综上所述,本次实验下钢轨波磨伤损演化规律评估结果与实际使用情况相接近,证明了本文方法能够有地表征伤损演化的趋势。
钢轨作为重载铁路轮轨作用的重要组成部分,其健康状态对铁路安全运行具有关键意义。提出了基于t-SNE的钢轨波磨伤损演化指标构建方法,建立了波磨伤损的评估模型,并使用实际数据验证了本文方法的实用性和有效性。
(1)基于箱型图剔除信号中的异常离群值,保证后续特征筛选所得到的特征的规律性不受到干扰。
(2)特征提取和基于随机森林的特征筛选降低数据维度去除冗余信息、提高计算效率、提高模型性能减少过拟合的风险、增强了模型的可解释性和通用性。
(3)基于t-SNE的流形数据学习方法,从高维特征中获取了反映伤损演化趋势的低维特性,与其他的非线性降维方式相比,在伤损阶段划分上取得了良好的优势。同时t-SNE作为一种无监督学习的研究方法,能够避免人为主观判断对诊断结果的影响更加客观探索伤损本质。
(4)通过欧氏空间度量体现了伤损演化指标的时序性,通过中值滤波方法削弱了伤损指标的异常波动,获得了伤损演化的趋势,可以对伤损进行准确的划分,为分析伤损演化规律提供便利、有效避免误诊断并及时采取对应的运维措施。
为了全面认识钢轨伤损的演化规律尚需进行持续且系统性的监测与数据收集以消除偶然误差因素。然而,钢轨伤损的演化是一个长期缓慢的过程,一段线路的钢轨从伤损出现到报废可能需要耗时数月甚至数年,需要消耗大量的人力物力长期监测,对数据采集造成困难。重载线路所在坡度、曲率和重载列车性质、车轮状态等因素都会对伤损的演化发展造成或多或少的影响。因此,全面地研究伤损的演化规律、制定更科学有效的钢轨维护和更新策略,确保铁路运营的安全和高效,尚需时间更密集、工况更全面的数据。
  • 国家重点研发计划(2021YFE05011)
  • 湖南省教育厅青年项目(22B0586)
  • 湖南省教育厅教改项目(2022JGYB186)
参考文献 引证文献
排序方式:
[1]
汪梦寒, 顾晓菡, 王安斌. 科隆蛋扣件直线路段钢轨波磨形成机理研究[J]. 铁道科学与工程学报, 2024, 21(9): 3595-3607.
Wang Menghan, Gu Xiaohan, Wang Anbin. Research on the mechanism of rail corrugation formation in straight track sections with Cologne-egg fasteners[J]. Journal of Railway Science and Engineering, 2024, 21(9): 3595-3607.
[2]
陈金明, 钟盛华, 刘逸平, 等. 基于车轨耦合动力学分析的地铁e型弹条扣件疲劳寿命预测[J]. 科学技术与工程, 2024, 24(2): 766-772.
Chen Jinming, Zhong Shenghua, Liu Yiping. Fatigue life prediction of e-type clip fastener in metro based on vehicle-rail coupling dynamicsanalysis[J]. Science Technology and Engineering, 2024, 24(2): 766-772.
[3]
王思维, 姚德臣, 杨建伟, 等. 变速工况下基于WSST的地铁钢轨波磨检测[J]. 科学技术与工程, 2023, 23(16): 7080-7086.
Wang Siwei, Yao Dechen, Yan Jianwei, et al. Detection method of rail corugation in metro based on wavelet synchrosqueezing transform under variable speed[J]. Science Technology and Engineering, 2023, 23(16): 7080-7086.
[4]
Bo W, Qing W, Shuihui L, et al. Investigation of plastic flow of rail vehicle wheels and its effects on rolling contact fatigue based on a dynamic method[J]. Engineering Failure Analysis, 2024.DOI: 10.1016/j.engfailanal.2024.108484.
[5]
刘渊博, 陈相, 刘妤. 基于卷积自编码器和时间卷积网络的轴承性能退化趋势预测[J]. 振动与冲击, 2023, 42(13): 214-225.
Liu Yuanbo, Chen Xiang, Liu Yu. Prediction of bearing performance degradation trend based on CAE and TCN[J]. Journal of Vibration and Shock, 2023, 42(13): 214-225.
[6]
吴梦蝶, 程龙生, 陈闻鹤. 基于自适应马氏空间与深度学习的滚动轴承退化趋势预测[J]. 系统工程与电子技术, 2023, 45(10): 3338-3349.
Wu Mengdie, Cheng Longsheng, Chen Wenhe. Degradation trend prediction of rolling bearing based on adaptive Mahalanobis space and deep learing[J]. Systems Engineering and Electonics, 2023, 45(10): 3338-3349.
[7]
白雲杰, 贾希胜, 梁庆海, 等. 基于深度学习的柴油机气门健康状态评估[J]. 科学技术与工程, 2022, 22(10): 3941-3950.
Bai Yunjie, Jia Xisheng, Liang Qinghai, et al. Evaluation of diesel engine valve health status based on deep learning[J]. Science Technology and Engineering, 2022, 22(10): 3941-3950.
[8]
Wen L, Yang G, Hu L, et al. A new unsupervised health index estimation method for bearings early fault detection based on Gaussian mixture model[J]. Engineering Applications of Artificial Intelligence, 2024, 128.DOI: 10.1016/j.engappai.2023.107562.
[9]
Yuan X, Liu S, Feng W, et al. Feature importance ranking of random forest-based end-to-end learning algorithm[J]. Remote Sensing, 2023, 15(21). DOI: 10.3390/rs15215203.
[10]
Yang J, Han S, Chen Y. Prediction of traffic accident severity based on random forest[J]. Journal of Advanced Transportation, 2023, 2023: 1-8.
[11]
Mafarja M, Thaher T, Al-Betar M A, et al. Classification framework for faulty-software using enhanced exploratory whale optimizer-based feature selection scheme and random forest ensemble learning[J]. Applied Intelligence, 2023, 53(15): 18715-18757.
[12]
Wang H, Li D, Li D, et al. Remaining useful life prediction of aircraft turbofan engine based on random forest feature selection and multi-layer perceptron[J]. Applied Sciences, 2023, 13(12). DOI: 10.3390/app13127186.
[13]
Han Y, Song X, Shi J, et al. KPCA-WPHM-SCNs-based remaining useful life prediction method for motor rolling bearings[J]. Transactions of the Institute of Measurement and Control, 2024, 46(5): 973-991.
[14]
Zhao Y, Zhou M, Wang L, et al. Bearing fault diagnosis of single-channel data by a 3D dcn with bilinear lbp and modified KPCA[J]. Journal of Electrical Engineering & Technology, 2023, 18(5): 3839-3848.
[15]
Pan Y, Wang H, Chen J, et al. A new fault component selection strategy based on statistical detection for slewing bearing weak signal de-noising[J]. Transactions of the Institute of Measurement and Control, 2024, 46(11): 2222-2239.
[16]
Pan H, Jiao W, Yan T, et al. Combining kernel principal component analysis and spatial group-wise enhance convolutional neural network for fault recognition of rolling element bearings[J]. Mea-surement Science and Technology, 2023, 34(12). DOI: 10.1088/1361-6501/ace790.
[17]
Su H, Wang Z, Cai Y, et al. Refined composite Multiscale fluctuation dispersion entropy and supervised manifold mapping for planetary gearbox fault diagnosis[J]. Machines, 2023, 11(1). DOI: 10.3390/machines11010047.
[18]
Wang H, Yao L, Wang H, et al. Supervised manifold learning based on multi-feature information discriminative fusion within an adaptive nearest neighbor strategy applied to rolling bearing fault diagnosis[J]. Sensors, 2023, 23(24). DOI: 10.3390/s23249820.
[19]
Lu W, Shi C, Fu H, et al. A power transformer fault diagnosis method based on improved sand cat swarm optimization algorithm and bidirectional gated recurrent unit[J]. Electronics, 2023, 12(3). DOI: 10.3390/electronics12030672.
[20]
Lu W, Shi C, Fu H, et al. Research on transformer fault diagnosis based on ISOMAP and IChOA-LSSVM[J]. IET Electric Power Applications, 2023, 17(6): 773-787.
[21]
Lee C Y, Maceren E D C. Induction motor bearing fault classification using deep neural network with particle swarm optimization-extreme gradient boosting[J]. IET Electric Power Applications, 2024, 18(3): 297-311.
[22]
Zair M, Rahmoune C, Imane M, et al. A new health indicator extracted by unsupervised learning using autoencoder in tandem with t-SNE and multi-kernel CNN to enhance the early detection and classification of bearings multi-faults[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2023, 45(7). DOI: 10.1007/s40430-023-04307-6.
[23]
Jiang W, Xu Y, Chen Z, et al. Fault diagnosis for rolling bearing using a hybrid hierarchical method based on scale-variable dispersion entropy and parametric t-SNE algorithm[J]. Measurement, 2022, 191.DOI: 10.1016/j.measurement.2022.110843.
2025年第25卷第10期
PDF下载
310
121
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403576
  • 接收时间:2024-05-15
  • 首发时间:2025-07-09
  • 出版时间:2025-04-08
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-15
  • 修回日期:2025-01-10
基金
国家重点研发计划(2021YFE05011)
湖南省教育厅青年项目(22B0586)
湖南省教育厅教改项目(2022JGYB186)
作者信息
    1 湖南工业大学轨道交通学院, 株洲 412000
    2 株洲时代电子技术有限公司, 株洲 412007
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403576
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