Article(id=1156264262138974844, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156264148657886112, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403476, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1715356800000, receivedDateStr=2024-05-11, revisedDate=1734364800000, revisedDateStr=2024-12-17, acceptedDate=null, acceptedDateStr=null, onlineDate=1753604482444, onlineDateStr=2025-07-27, pubDate=1740672000000, pubDateStr=2025-02-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1753604482444, onlineIssueDateStr=2025-07-27, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1753604482444, creator=13701087609, updateTime=1753604482444, updator=13701087609, issue=Issue{id=1156264148657886112, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='6', pageStart='2193', pageEnd='2636', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1753604455388, creator=13701087609, updateTime=1753771257443, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1156963767234945803, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156264148657886112, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1156963767234945804, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156264148657886112, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=2548, endPage=2553, ext={EN=ArticleExt(id=1156264263418237571, articleId=1156264262138974844, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Prediction of Platform Door Structure Deformation of High-Speed Railway Based on Artificial Intelligence, columnId=1156262728772735295, journalTitle=Science Technology and Engineering, columnName=Papers·Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

In order to solve the problem of real-time monitoring and accurate prediction of structural deformation of platform doors on high-speed railway lines, an artificial intelligence-based neural network method was used. Structural deformation data of platform doors, involving 210 different conditions of train length, blocking ratio, installation distance, and speed, were selected as training samples for the network model. Two neural network models, CNN(convolutional neural network) and K-Fold(K-Fold cross-validation) optimized GRNN(general regression neural network), were used to establish predictive models for platform door structural deformation under different working conditions of high-speed railways. These models were compared and verified with the remaining sample data. The research shows that both models effectively predict the operation and maintenance data of railway platform door structures. The K-Fold optimized GRNN model is superior to the CNN model in prediction accuracy. The Mean Square Error of the K-Fold optimized GRNN model is maintained within 0.22, and theRoot Mean Square Error is within 0.27, which is at the leading level in the field. The K-Fold optimized GRNN model better predicts the structural deformation of platform doors when trains pass, providing data references for the design and maintenance of high-speed railway platform doors.

, correspAuthors=Zhi-fei 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=Bo-xuan YANG, Zhi-fei WANG, Fan LI, Hao-dong WANG, Shuai LI), CN=ArticleExt(id=1156264334020956451, articleId=1156264262138974844, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于人工智能的高速铁路站台门结构变形的预测, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

为了解决运营线路站台门结构变形难以实时监控的问题,精准预测列车过站时高速铁路站台门的结构变形数据,采用基于人工智能的神经网络方法,选取210种不同车长、阻塞比、安装距离和车速的站台门结构变形数据作为网络模型训练样本,运用CNN(convolutional neural network)和基于K-Fold(K-fold cross-validation)的GRNN(general regression neural network)两种神经网络模型,建立了不同工况下的高速铁路站台门结构变形的预测模型,并与剩余样本数据进行对比验证。研究表明,两种模型均可有效预测铁路站台门结构运维数据,在预测精度上,基于K-Fold优化的GRNN模型优于CNN模型,基于K-Fold优化的GRNN模型的预测均方差能够维持在0.22之内,均方根误差维持在0.27之内,处于研究领域的领先水平。基于K-Fold优化的GRNN模型能够较好预测列车过站时的站台门结构形变量,为高速铁路站台门的设计与运维提供数据参考。

, correspAuthors=王志飞, authorNote=null, correspAuthorsNote=
* 王志飞(1981—),男,汉族,内蒙古东胜人,博士,研究员。研究方向:交通运输工程与控制。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=8joCoIStNgnMozEJjKjZqg==, magXml=wX7mFvu5f+eFwlm1LRUfWQ==, pdfUrl=null, pdf=oHFYvTAmBch62rfHzNl7xw==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=9jRPDuMNBr22ieou0xzrWw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=197WPlE/q0cWT3HWeBhmFA==, mapNumber=null, authorCompany=null, fund=null, authors=

杨博璇(1999—),女,汉族,河北石家庄人,硕士研究生。研究方向:智能轨道交通与安全防护技术。E-mail:

, authorsList=杨博璇, 王志飞, 李樊, 王浩东, 李帅)}, authors=[Author(id=1233422548059353940, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=yyy202302@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1233422548340372326, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422548059353940, language=EN, stringName=Bo-xuan YANG, firstName=Bo-xuan, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Graduate Department, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1233422548470395769, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422548059353940, language=CN, stringName=杨博璇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中国铁道科学研究院研究生部, 北京 100081, bio={"content":"

杨博璇(1999—),女,汉族,河北石家庄人,硕士研究生。研究方向:智能轨道交通与安全防护技术。E-mail:

"}, bioImg=null, bioContent=

杨博璇(1999—),女,汉族,河北石家庄人,硕士研究生。研究方向:智能轨道交通与安全防护技术。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1233422547652506409, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=1, ext=[AuthorCompanyExt(id=1233422547698643758, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547652506409, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Graduate Department, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547744781111, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547652506409, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国铁道科学研究院研究生部, 北京 100081)])]), Author(id=1233422548650750861, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wfbuaa@126.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1233422548805940121, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422548650750861, language=EN, stringName=Zhi-fei WANG, firstName=Zhi-fei, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1233422550332666800, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422548650750861, language=CN, stringName=王志飞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1233422547899970372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=2, ext=[AuthorCompanyExt(id=1233422547904164678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547920941894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1233422550475273150, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, 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=1233422550596907981, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422550475273150, language=EN, stringName=Fan LI, firstName=Fan, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1233422550710154202, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422550475273150, language=CN, stringName=李樊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1233422547899970372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=2, ext=[AuthorCompanyExt(id=1233422547904164678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547920941894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1233422550848566255, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, 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=1233422550974395383, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422550848566255, language=EN, stringName=Hao-dong WANG, firstName=Hao-dong, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1233422551112806406, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422550848566255, language=CN, stringName=王浩东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1233422547899970372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=2, ext=[AuthorCompanyExt(id=1233422547904164678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547920941894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])]), Author(id=1233422551247024149, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, 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=1233422551398019109, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422551247024149, language=EN, stringName=Shuai LI, firstName=Shuai, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1233422551666454582, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, authorId=1233422551247024149, language=CN, stringName=李帅, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1233422547899970372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=2, ext=[AuthorCompanyExt(id=1233422547904164678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547920941894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])])], keywords=[Keyword(id=1233422551997804634, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, orderNo=1, keyword=neural network), Keyword(id=1233422552119439464, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, orderNo=2, keyword=regression prediction), Keyword(id=1233422552211714161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, orderNo=3, keyword=platform door operation and maintenance), Keyword(id=1233422552312377467, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, orderNo=4, keyword=high-speed railway platform doors), Keyword(id=1233422552438206604, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, orderNo=1, keyword=神经网络), Keyword(id=1233422552572424339, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, orderNo=2, keyword=回归预测), Keyword(id=1233422552702447778, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, orderNo=3, keyword=站台门运维), Keyword(id=1233422552861831342, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, orderNo=4, keyword=高速铁路站台门)], refs=[Reference(id=1233422560008925741, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=9, pageStart=103, pageEnd=111, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=李燕, journalName=铁道运输与经济, refType=null, unstructuredReference=李燕. PHM技术在京张高速铁路运用检修中的应用研究[J]. 铁道运输与经济, 2022, 44(9): 103-111., articleTitle=PHM技术在京张高速铁路运用检修中的应用研究, refAbstract=null), Reference(id=1233422560189280820, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=9, pageStart=103, pageEnd=111, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Li Yan, journalName=Railway Transport and Economy, refType=null, unstructuredReference=Li Yan. Research on the application of PHM technology in the maintenance of Beijing-Zhangjiakou high-speed railway[J]. Railway Transport and Economy, 2022, 44(9): 103-111., articleTitle=Research on the application of PHM technology in the maintenance of Beijing-Zhangjiakou high-speed railway, refAbstract=null), Reference(id=1233422560352858683, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2018, volume=39, issue=6, pageStart=104, pageEnd=111, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=韩运动, 陈大伟, 刘韶庆, journalName=中国铁道科学, refType=null, unstructuredReference=韩运动, 陈大伟, 刘韶庆, 等. 不同风速风向条件下的列车风特性[J]. 中国铁道科学, 2018, 39(6): 104-111., articleTitle=不同风速风向条件下的列车风特性, refAbstract=null), Reference(id=1233422560440939073, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2018, volume=39, issue=6, pageStart=104, pageEnd=111, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Han Yundong, Chen Dawei, Liu Shaoqing, journalName=Chinese Railway Science, refType=null, unstructuredReference=Han Yundong, Chen Dawei, Liu Shaoqing, et al. Train aerodyna-mic characteristics under different wind speeds and directions[J]. Chinese Railway Science, 2018, 39(6): 104-111., articleTitle=Train aerodyna-mic characteristics under different wind speeds and directions, refAbstract=null), Reference(id=1233422560524825161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2017, volume=17, issue=27, pageStart=53, pageEnd=58, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=李新民, 聂小春, 晏致涛, journalName=科学技术与工程, refType=null, unstructuredReference=李新民, 聂小春, 晏致涛, 等.1 520 mm2大截面两分裂新月形覆冰导线空气动力特性试验研究[J]. 科学技术与工程, 2017, 17(27): 53-58., articleTitle=mm2大截面两分裂新月形覆冰导线空气动力特性试验研究, refAbstract=null), Reference(id=1233422560654848595, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2017, volume=17, issue=27, pageStart=53, pageEnd=58, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Li Xinmin, Nie Xiaochun, Yan Zhitao, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Xinmin, Nie Xiaochun, Yan Zhitao, et al. Experimental study on the aerodynamic characteristics of 1 520 mm2 large cross-section bifurcated crescent-shaped iced conductor[J]. Science Technology and Engineering, 2017, 17(27): 53-58., articleTitle=Experimental study on the aerodynamic characteristics of 1 520 mm2 large cross-section bifurcated crescent-shaped iced conductor, refAbstract=null), Reference(id=1233422560755511896, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=白雨坤, journalName=城市轨道交通站台门系统的控制与数据处理, refType=null, unstructuredReference=白雨坤. 城市轨道交通站台门系统的控制与数据处理[D]. 北京: 中国铁道科学研究院, 2014., articleTitle=null, refAbstract=null), Reference(id=1233422560872952414, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Bai Yukun, journalName=Control and data processing of platform screen door systems in urban rail transit, refType=null, unstructuredReference=Bai Yukun. Control and data processing of platform screen door systems in urban rail transit[D]. Bei’jing: China Academy of Railway Sciences, 2014., articleTitle=null, refAbstract=null), Reference(id=1233422560990392935, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2019, volume=15, issue=S1, pageStart=85, pageEnd=89, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=胡振亚, 罗威, journalName=中国安全生产科学技术, refType=null, unstructuredReference=胡振亚, 罗威. 轨道交通站台门轨旁安全设计[J]. 中国安全生产科学技术, 2019, 15(S1): 85-89., articleTitle=轨道交通站台门轨旁安全设计, refAbstract=null), Reference(id=1233422561107833456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2019, volume=15, issue=S1, pageStart=85, pageEnd=89, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Hu Zhenya, Luo Wei, journalName=Science Technology and EngineerinJournal of Safety Science and Technology, refType=null, unstructuredReference=Hu Zhenya, Luo Wei. Safety design for platform screen door trackside in rail transit railways[J]. Science Technology and EngineerinJournal of Safety Science and Technology, 2019, 15(S1): 85-89., articleTitle=Safety design for platform screen door trackside in rail transit railways, refAbstract=null), Reference(id=1233422561233662587, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=12, pageStart=134, pageEnd=137, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=王志飞, journalName=机械设计与制造, refType=null, unstructuredReference=王志飞. 列车高速过站风压对站台门结构特性的影响[J]. 机械设计与制造, 2020(12): 134-137, 141., articleTitle=列车高速过站风压对站台门结构特性的影响, refAbstract=null), Reference(id=1233422561359491715, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=12, pageStart=134, pageEnd=137, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=Wang Zhifei, journalName=Machine-ry Design & Manufacture, refType=null, unstructuredReference=Wang Zhifei. Impact of train pass-by wind pressure on the structural characteristics of platform screen doors at high speed[J]. Machine-ry Design & Manufacture, 2020(12): 134-137, 141., articleTitle=Impact of train pass-by wind pressure on the structural characteristics of platform screen doors at high speed, refAbstract=null), Reference(id=1233422561535652493, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=3, pageStart=25, pageEnd=29, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=周梅, 毛良, journalName=现代城市轨道交通, refType=null, unstructuredReference=周梅, 毛良. 城际铁路列车通过地下站站台门时的风致效应仿真与试验研究[J]. 现代城市轨道交通, 2023(3): 25-29., articleTitle=城际铁路列车通过地下站站台门时的风致效应仿真与试验研究, refAbstract=null), Reference(id=1233422561703424668, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=3, pageStart=25, pageEnd=29, url=null, language=null, rfNumber=[7], rfOrder=13, authorNames=Zhou Mei, Mao Liang, journalName=Modern Urban Transit, refType=null, unstructuredReference=Zhou Mei, Mao Liang. Simulation and experimental study on the wind-induced effects of intercity railway trains passing platform screen doors at underground stations[J]. Modern Urban Transit, 2023(3): 25-29., articleTitle=Simulation and experimental study on the wind-induced effects of intercity railway trains passing platform screen doors at underground stations, refAbstract=null), Reference(id=1233422561850225314, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2017, volume=184, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=Ali J M, Omar A A, Ali M’B A, journalName=IOP Conference, refType=null, unstructuredReference=Ali J M, Omar A A, Ali M’B A, et al. Numerical investigation of aerodynamic characteristics of high speed trail[J]. IOP Conference, 2017, 184(1).DOI:10.1088/1757-899X/184/1/012015., articleTitle=Numerical investigation of aerodynamic characteristics of high speed trail, refAbstract=null), Reference(id=1233422562080912046, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=24, pageStart=9930, pageEnd=9936, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=肖艳, 王建东, 王振, journalName=科学技术与工程, refType=null, unstructuredReference=肖艳, 王建东, 王振. 基于亲信度的过程信号与报警信号相关性分析[J]. 科学技术与工程, 2020, 20(24): 9930-9936., articleTitle=基于亲信度的过程信号与报警信号相关性分析, refAbstract=null), Reference(id=1233422562198352560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=24, pageStart=9930, pageEnd=9936, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=Xiao Yan, Wang Jiandong, Wang Zhen, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Xiao Yan, Wang Jiandong, Wang Zhen. Correlation analysis of process signals and alarm signals based on credibility[J]. Science Technology and Engineering, 2020, 20(24): 9930-9936., articleTitle=Correlation analysis of process signals and alarm signals based on credibility, refAbstract=null), Reference(id=1233422563691524800, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2022, volume=56, issue=9, pageStart=1256, pageEnd=1261, url=null, language=null, rfNumber=[10], rfOrder=17, authorNames=曾国治, 魏子清, 岳宝, journalName=上海交通大学学报, refType=null, unstructuredReference=曾国治, 魏子清, 岳宝, 等. 基于CNN-RNN组合模型的办公建筑能耗预测[J]. 上海交通大学学报, 2022, 56(9): 1256-1261., articleTitle=基于CNN-RNN组合模型的办公建筑能耗预测, refAbstract=null), Reference(id=1233422563808965317, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2022, volume=56, issue=9, pageStart=1256, pageEnd=1261, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=Zeng Guozhi, Wei Ziqing, Yue Bao, journalName=Journal of Shanghai Jiao Tong University, refType=null, unstructuredReference=Zeng Guozhi, Wei Ziqing, Yue Bao, et al. Energy consumption prediction for office buildings based on a combined CNN-RNN model[J]. Journal of Shanghai Jiao Tong University, 2022, 56(9): 1256-1261., articleTitle=Energy consumption prediction for office buildings based on a combined CNN-RNN model, refAbstract=null), Reference(id=1233422563913822925, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=9, pageStart=3737, pageEnd=3746, url=null, language=null, rfNumber=[11], rfOrder=19, authorNames=江佳霖, 钟宝荣, journalName=科学技术与工程, refType=null, unstructuredReference=江佳霖, 钟宝荣. 基于改进Mask R-CNN的高密度砂岩颗粒的分割识别[J]. 科学技术与工程, 2024, 24(9): 3737-3746., articleTitle=基于改进Mask R-CNN的高密度砂岩颗粒的分割识别, refAbstract=null), Reference(id=1233422564077400788, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=9, pageStart=3737, pageEnd=3746, url=null, language=null, rfNumber=[11], rfOrder=20, authorNames=Jiang Jialin, Zhong Baorong, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Jiang Jialin, Zhong Baorong. Segmentation and identification of high-density sandstone particles based on improved mask R-CNN[J]. Science Technology and Engineering, 2024, 24(9): 3737-3746., articleTitle=Segmentation and identification of high-density sandstone particles based on improved mask R-CNN, refAbstract=null), Reference(id=1233422564190647003, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=20, pageStart=8719, pageEnd=8726, url=null, language=null, rfNumber=[12], rfOrder=21, authorNames=杨雅萍, 张文莲, 孙晓云, journalName=科学技术与工程, refType=null, unstructuredReference=杨雅萍, 张文莲, 孙晓云. 基于改进蝙蝠算法优化广义回归神经网络的岩质边坡稳定性预测[J]. 科学技术与工程, 2021, 21(20): 8719-8726., articleTitle=基于改进蝙蝠算法优化广义回归神经网络的岩质边坡稳定性预测, refAbstract=null), Reference(id=1233422564345836257, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=20, pageStart=8719, pageEnd=8726, url=null, language=null, rfNumber=[12], rfOrder=22, authorNames=Yang Yaping, Zhang Wenlian, Sun Xiaoyun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Yang Yaping, Zhang Wenlian, Sun Xiaoyun. Prediction of rock slope stability based on generalized regression neural network optimized by improved bat algorithm[J]. Science Technology and Engineering, 2021, 21(20): 8719-8726., articleTitle=Prediction of rock slope stability based on generalized regression neural network optimized by improved bat algorithm, refAbstract=null), Reference(id=1233422564505219815, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, doi=null, pmid=null, pmcid=null, year=2024, volume=59, issue=1, pageStart=11, pageEnd=14, url=null, language=null, rfNumber=[13], rfOrder=23, authorNames=Makomere S R, Koech L, Rutto L H, journalName=Journal of Environmental Science and Health: Part A, refType=null, unstructuredReference=Makomere S R, Koech L, Rutto L H, et al. Precision forecasting of spray-dry desulfurization using Gaussian noise data augmentation and K-fold cross-validation optimized neural computing[J]. Journal of Environmental Science and Health: Part A, 2024, 59(1): 11-14., articleTitle=Precision forecasting of spray-dry desulfurization using Gaussian noise data augmentation and K-fold cross-validation optimized neural computing, refAbstract=null)], funds=[Fund(id=1233422559547552267, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, awardId=U21A201646, language=CN, fundingSource=国家自然科学基金(U21A201646), fundOrder=null, country=null), Fund(id=1233422559715324438, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, awardId=2023YJ129, language=CN, fundingSource=中国铁道科学研究院集团有限公司重点项目(2023YJ129), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1233422547652506409, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=1, ext=[AuthorCompanyExt(id=1233422547698643758, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547652506409, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Graduate Department, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547744781111, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547652506409, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国铁道科学研究院研究生部, 北京 100081)]), AuthorCompany(id=1233422547899970372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, xref=2, ext=[AuthorCompanyExt(id=1233422547904164678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China), AuthorCompanyExt(id=1233422547920941894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, companyId=1233422547899970372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081)])], figs=[ArticleFig(id=1233422553184792780, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.1, caption=CNN neural network structure diagram, figureFileSmall=9ja/axGjuCwWl6t3GqYm1w==, figureFileBig=FK256s9Ir8f/2bVADJf7/w==, tableContent=null), ArticleFig(id=1233422553289650394, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图1, caption=CNN神经网络结构图, figureFileSmall=9ja/axGjuCwWl6t3GqYm1w==, figureFileBig=FK256s9Ir8f/2bVADJf7/w==, tableContent=null), ArticleFig(id=1233422554770239723, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.2, caption=GRNN neural network structure diagram, figureFileSmall=SwwlTFKj0QkAPbA14IwiCw==, figureFileBig=SsdazsGtCYw7VOO6DVasog==, tableContent=null), ArticleFig(id=1233422554875097336, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图2, caption=GRNN神经网络结构图, figureFileSmall=SwwlTFKj0QkAPbA14IwiCw==, figureFileBig=SsdazsGtCYw7VOO6DVasog==, tableContent=null), ArticleFig(id=1233422555055452417, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.3, caption=K-Fold cross verification diagram, figureFileSmall=7MDi7Bu60kd/C6uSmIVqGQ==, figureFileBig=QZ/HVXRMxM/Ej6kefi5o5g==, tableContent=null), ArticleFig(id=1233422555214835984, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图3, caption=K重交叉验证示意图, figureFileSmall=7MDi7Bu60kd/C6uSmIVqGQ==, figureFileBig=QZ/HVXRMxM/Ej6kefi5o5g==, tableContent=null), ArticleFig(id=1233422555332276506, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.4, caption=Relation between mean square error and number of iterations, figureFileSmall=kmuOusJmLgmKs6i2FLMj/w==, figureFileBig=eWKy45jRPoE5LlvI1Y5PRw==, tableContent=null), ArticleFig(id=1233422555487465773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图4, caption=均方误差和迭代次数的关系, figureFileSmall=kmuOusJmLgmKs6i2FLMj/w==, figureFileBig=eWKy45jRPoE5LlvI1Y5PRw==, tableContent=null), ArticleFig(id=1233422555684598075, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.5, caption=Comparison between the real value and the predicted value of CNN model, figureFileSmall=W7fPRQgXoqHPZJSqKdCxtQ==, figureFileBig=tY9PS3HZ6C3h1Ng8M6WWag==, tableContent=null), ArticleFig(id=1233422555835593030, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图5, caption=真实值与CNN模型预测值对比, figureFileSmall=W7fPRQgXoqHPZJSqKdCxtQ==, figureFileBig=tY9PS3HZ6C3h1Ng8M6WWag==, tableContent=null), ArticleFig(id=1233422555974005073, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.6, caption=Comparison between the real value and the predicted value of GRNN model based on K-Fold optimization, figureFileSmall=M3cZwRpifWfESudy/EWuww==, figureFileBig=Fq0/K9w36GfVuUV1zfgRhg==, tableContent=null), ArticleFig(id=1233422556112417120, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图6, caption=真实值与基于K-Fold优化的GRNN模型预测值对比, figureFileSmall=M3cZwRpifWfESudy/EWuww==, figureFileBig=Fq0/K9w36GfVuUV1zfgRhg==, tableContent=null), ArticleFig(id=1233422556259217770, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Fig.7, caption=Comparison of actual values with model predictions, figureFileSmall=j+FlS5F+UureBrwpoBMBSw==, figureFileBig=JSG8HFTDiMYQMTqGOfd19w==, tableContent=null), ArticleFig(id=1233422556401824119, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=图7, caption=实际值与模型预测值对比, figureFileSmall=j+FlS5F+UureBrwpoBMBSw==, figureFileBig=JSG8HFTDiMYQMTqGOfd19w==, tableContent=null), ArticleFig(id=1233422556494098812, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 1, caption=

Sample validation data table

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 车长/
m
阻塞比 安装距离/
mm
车速/
(km·h-1)
形变量/
mm
1 101.4 5.5 700 200 6.99
2 151.4 5.78 1 000 100 2.12
3 151.4 5.78 1 000 120 2.69
4 187.0 5.93 700 200 8.29
5 201.4 6.25 1 500 140 4.13
6 201.4 6.25 1 500 160 4.68
7 187.0 6.24 1 000 140 3.83
8 95.8 6.45 1 200 120 4.93
), ArticleFig(id=1233422556624122245, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表1, caption=

样本验证数据表

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 车长/
m
阻塞比 安装距离/
mm
车速/
(km·h-1)
形变量/
mm
1 101.4 5.5 700 200 6.99
2 151.4 5.78 1 000 100 2.12
3 151.4 5.78 1 000 120 2.69
4 187.0 5.93 700 200 8.29
5 201.4 6.25 1 500 140 4.13
6 201.4 6.25 1 500 160 4.68
7 187.0 6.24 1 000 140 3.83
8 95.8 6.45 1 200 120 4.93
), ArticleFig(id=1233422556770922899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 2, caption=

CNN model parameter values

, figureFileSmall=null, figureFileBig=null, tableContent=
参数更新
步长
最大迭
代次数
小批量
数据
Shuffle/
every-epoch
0.1 120 32 1
), ArticleFig(id=1233422556921917852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表2, caption=

CNN模型参数取值

, figureFileSmall=null, figureFileBig=null, tableContent=
参数更新
步长
最大迭
代次数
小批量
数据
Shuffle/
every-epoch
0.1 120 32 1
), ArticleFig(id=1233422557035164073, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 3, caption=

K-Fold parameter selection

, figureFileSmall=null, figureFileBig=null, tableContent=
σ取值区间 σ变化步长
[-1,1] 0.1
), ArticleFig(id=1233422557135827377, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表3, caption=

K-Fold参数选取

, figureFileSmall=null, figureFileBig=null, tableContent=
σ取值区间 σ变化步长
[-1,1] 0.1
), ArticleFig(id=1233422557299405250, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 4, caption=

Comparison of prediction errors of CNN and K-Fold GRNN

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 真实值/
mm
CNN K-Fold GRNN
预测值/
mm
误差率 预测值/
mm
误差率
1 6.99 6.64 -0.050 7.07 0.011
2 2.12 2.08 -0.018 2.40 0.132
3 2.69 2.97 0.104 3.03 0.126
4 8.29 7.31 -0.118 7.98 -0.037
5 4.13 3.74 -0.094 3.85 -0.067
6 4.68 4.69 0.002 4.76 0.017
7 3.83 4.15 0.083 3.97 0.036
8 4.93 5.09 0.032 4.89 -0.008
), ArticleFig(id=1233422557408457161, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表4, caption=

CNN与K-Fold GRNN预测误差对比

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 真实值/
mm
CNN K-Fold GRNN
预测值/
mm
误差率 预测值/
mm
误差率
1 6.99 6.64 -0.050 7.07 0.011
2 2.12 2.08 -0.018 2.40 0.132
3 2.69 2.97 0.104 3.03 0.126
4 8.29 7.31 -0.118 7.98 -0.037
5 4.13 3.74 -0.094 3.85 -0.067
6 4.68 4.69 0.002 4.76 0.017
7 3.83 4.15 0.083 3.97 0.036
8 4.93 5.09 0.032 4.89 -0.008
), ArticleFig(id=1233422557530091989, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 5, caption=

Comparison of MEA and RMSE of two models

, figureFileSmall=null, figureFileBig=null, tableContent=
神经网络模型 均方差MEA 均方根误差RMSE
CNN 0.337 3 0.434 3
K-Fold GRNN 0.223 9 0.265 2
), ArticleFig(id=1233422557672698334, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表5, caption=

两种模型的MEA与RMSE对比

, figureFileSmall=null, figureFileBig=null, tableContent=
神经网络模型 均方差MEA 均方根误差RMSE
CNN 0.337 3 0.434 3
K-Fold GRNN 0.223 9 0.265 2
), ArticleFig(id=1233422557806916074, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=EN, label=Table 6, caption=

Comparison of the optimal accuracy of two prediction methods

, figureFileSmall=null, figureFileBig=null, tableContent=
预测方法 均方差
MEA
均方根误差
RMSE
有限元仿真(城际铁路)[7] 0.43 0.31
有限元仿真(高速铁路)[6] 0.38 0.28
机器学习(高速铁路) 0.22 0.27
), ArticleFig(id=1233422559279116788, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156264262138974844, language=CN, label=表6, caption=

两种预测方法最优精度对比

, figureFileSmall=null, figureFileBig=null, tableContent=
预测方法 均方差
MEA
均方根误差
RMSE
有限元仿真(城际铁路)[7] 0.43 0.31
有限元仿真(高速铁路)[6] 0.38 0.28
机器学习(高速铁路) 0.22 0.27
)], 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.1671-1815.2403476, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403476, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403476, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403476, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于人工智能的高速铁路站台门结构变形的预测
收藏切换
PDF下载
杨博璇 1 , 王志飞 2, * , 李樊 2 , 王浩东 2 , 李帅 2
科学技术与工程 | 论文·交通运输 2025,25(6): 2548-2553
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(6): 2548-2553
基于人工智能的高速铁路站台门结构变形的预测
全屏
杨博璇1 , 王志飞2, * , 李樊2, 王浩东2, 李帅2
作者信息
  • 1 中国铁道科学研究院研究生部, 北京 100081
  • 2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081
  • 杨博璇(1999—),女,汉族,河北石家庄人,硕士研究生。研究方向:智能轨道交通与安全防护技术。E-mail:

通讯作者:

* 王志飞(1981—),男,汉族,内蒙古东胜人,博士,研究员。研究方向:交通运输工程与控制。E-mail:
Prediction of Platform Door Structure Deformation of High-Speed Railway Based on Artificial Intelligence
Bo-xuan YANG1 , Zhi-fei WANG2, * , Fan LI2, Hao-dong WANG2, Shuai LI2
Affiliations
  • 1 Graduate Department, China Academy of Railway Sciences, Beijing 100081, China
  • 2 Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China
出版时间: 2025-02-28 doi: 10.12404/j.issn.1671-1815.2403476
文章导航
收藏切换

为了解决运营线路站台门结构变形难以实时监控的问题,精准预测列车过站时高速铁路站台门的结构变形数据,采用基于人工智能的神经网络方法,选取210种不同车长、阻塞比、安装距离和车速的站台门结构变形数据作为网络模型训练样本,运用CNN(convolutional neural network)和基于K-Fold(K-fold cross-validation)的GRNN(general regression neural network)两种神经网络模型,建立了不同工况下的高速铁路站台门结构变形的预测模型,并与剩余样本数据进行对比验证。研究表明,两种模型均可有效预测铁路站台门结构运维数据,在预测精度上,基于K-Fold优化的GRNN模型优于CNN模型,基于K-Fold优化的GRNN模型的预测均方差能够维持在0.22之内,均方根误差维持在0.27之内,处于研究领域的领先水平。基于K-Fold优化的GRNN模型能够较好预测列车过站时的站台门结构形变量,为高速铁路站台门的设计与运维提供数据参考。

神经网络  /  回归预测  /  站台门运维  /  高速铁路站台门

In order to solve the problem of real-time monitoring and accurate prediction of structural deformation of platform doors on high-speed railway lines, an artificial intelligence-based neural network method was used. Structural deformation data of platform doors, involving 210 different conditions of train length, blocking ratio, installation distance, and speed, were selected as training samples for the network model. Two neural network models, CNN(convolutional neural network) and K-Fold(K-Fold cross-validation) optimized GRNN(general regression neural network), were used to establish predictive models for platform door structural deformation under different working conditions of high-speed railways. These models were compared and verified with the remaining sample data. The research shows that both models effectively predict the operation and maintenance data of railway platform door structures. The K-Fold optimized GRNN model is superior to the CNN model in prediction accuracy. The Mean Square Error of the K-Fold optimized GRNN model is maintained within 0.22, and theRoot Mean Square Error is within 0.27, which is at the leading level in the field. The K-Fold optimized GRNN model better predicts the structural deformation of platform doors when trains pass, providing data references for the design and maintenance of high-speed railway platform doors.

neural network  /  regression prediction  /  platform door operation and maintenance  /  high-speed railway platform doors
杨博璇, 王志飞, 李樊, 王浩东, 李帅. 基于人工智能的高速铁路站台门结构变形的预测. 科学技术与工程, 2025 , 25 (6) : 2548 -2553 . DOI: 10.12404/j.issn.1671-1815.2403476
Bo-xuan YANG, Zhi-fei WANG, Fan LI, Hao-dong WANG, Shuai LI. Prediction of Platform Door Structure Deformation of High-Speed Railway Based on Artificial Intelligence[J]. Science Technology and Engineering, 2025 , 25 (6) : 2548 -2553 . DOI: 10.12404/j.issn.1671-1815.2403476
目前,中国高速铁路最高运营速度350 km/h,高速列车在过站时,由于空气的黏性作用使周围的空气被列车表面带动并随之一起作用,产生了列车风[1]。随着高铁的不断发展,列车的运营速度朝着400 km/h级发展,列车风的影响将更加突出[2]。站台门通过实现乘客与列车行轨的物理隔离,降低了列车风对站台的影响,同时也给自身结构的负载能力带来了挑战[3]。由于站台门安装在站台边缘,发生较大的结构变形会导致门体侵入限界,甚至脱落掉入轨道,引发重大安全事故。除此之外,站台门的结构形变还可能进一步导致控制系统失灵等问题,容易引发站台乘客人身伤亡等重大安全事故的发生[4]
随着全国高速铁路网的逐步形成和发展、对站台门的需求量不断增加,总体运维成本也在迅速增长。文献[5]对站台门轨旁事故进行分析,强调了站台门结构变形侵入限界的风险;文献[6-7]针对运营线路上站台门结构形变难以实时监控的问题,提出了高铁站台门有限元模型加载方法。文献[8]对高速列车的气动效应进行了数值研究,部分学者对站台门的使用寿命产生了预测,没有对站台门形变量预测的相关研究。站台门一旦在列车正常运行过程中出现故障,轻则影响乘客的正常出行,重则产生重大安全事故,严重损害乘客的生命财产安全;同时由于其检修项目种类繁多、技术门槛较高,人力资源花费较大[9],定时对站台门进行安全检修不仅不能完全防范故障的产生,更会消费巨大的经济与人力成本。为了预防列车过站时站台门结构变形产生的安全事故、为站台门的前期设计提供结构强度数据参考、节省高速铁路站台门的运维费用,解决运营线路上站台门结构形变难以实时监控的问题,基于人工智能的站台门结构变形的研究迫在眉睫。
为了更有效地预防站台门结构变形引发的安全事故,降低站台门故障发生率、提升站台门形变量预测的准确性,本文研究提出基于人工智能的高速铁路站台门结构变形的预测方法,利用结构变形数据对CNN(convolutional neural network)网络模型和基于K-Fold(K-Fold cross-validation)算法进行参数优化的GRNN(general regression neural network)网络模型进行训练,最后将两种网络模型的预测值与真实值进行对比。
CNN神经网络是一种深度学习模型,通过卷积层和池化层有效处理图像数据,广泛用于图像分类、目标检测等领域。其层次结构包括输入层、多个卷积-池化层和全连接层[10]。CNN神经网络结构如图1所示。图1w为卷积核的权重参数,用于与输入特征图进行卷积操作,以提取不同层次的特征。通过网络训练会被不断优化,以增强对目标数据的识别和分类能力。
卷积层是神经网络的核心组件,通过卷积运算提取输入数据中的特征,通用的卷积公式为
$s(i,j)=\left(X\mathrm{*}W\right)(i,j)+b$
式(1)中:i为输入特征图的行索引;j为输入特征图的列索引;s(i,j)为卷积结果中位置(i,j)的值;W为卷积核;X为输入数据;W的维度与输入数据向量维度K保持一致;b为偏置项。
数据从卷积层传输到池化层还需经过激励函数处理,CNN网络选用ReLU函数作为激活函数,函数一般形式为
$f\left(x\right)=\frac{1}{1+{\mathrm{e}}^{-x}}$
式(2)中:x为某一层神经网络的输出。
池化层分为用于减少卷积层输出特征向量、防止过拟合的平均池化层和最大池化层,一般采用最大值max函数或平均值average函数进行池化操作,在本文中使用max函数来压缩池化层数据。
全连接层将之前卷积层和池化层提取的特征进行收集,再选用Softmax函数将输出值进行分类[11]
在对铁路站台门结构运维数据进行试验与预测时,算法目标不再是预测输入图像的类别,而是预测一个连续的目标值。因此将算法的最后一层全连接层的输出形式修改为一个单一的连续值,然后使用一个回归损失函数来训练网络。
常见的损失函数有平均误差损失函数L1、均方误差损失函数L2、函数SSIM(structural similarity index)等,考虑到输入的数据形式一般采用均方误差函数损失函数,函数形式为
${L}_{2}=\frac{1}{N}\stackrel{N}{\sum _{a=1}}({x}_{a}-{y}_{a}{)}^{2}$
式(3)中:N为数据的样本数量;xa为第a个样本的真实值;ya为第a个样本的预测值。
GRNN属于径向基神经网络的一种变形形式,该网络模型没有训练过程,主要通过对模式层的平滑因子进行优化来获得好的输出结果[12],GRNN的神经网络结构图如图2所示。
输入层实现数据的输入过程,并将输入数据传递至模式层,该层的节点数为输入数据的特征维度。
模式层一般用高斯函数对输入数据进行处理,节点数为训练样本的个数,具体计算公式为
${g}_{i}=\mathrm{e}\mathrm{x}\mathrm{p}\left(-\frac{\left({x}_{i}-x\right)}{2{\sigma }^{2}}\right)$
式(4)中:xi为训练样本;x为学习样本;σ为平滑因子;gi为训练样本与学习样本之间的相似度。
假设输出样本维度为j,则求和层节点数为j+1,其中一个节点输出sd,为模式层输出的算术和[13],其余节点输出sni均为模式层输出的加权和,计算公式为
${s}_{\mathrm{d}}=\stackrel{n}{\sum _{i=1}}{g}_{i}$
${s}_{ni}=\stackrel{n}{\sum _{i=1}}{w}_{ij}{g}_{i}$
式中:wij为加权系数。
输出层的节点数为输出样本维度Oj,主要根据求和层求出的算术和与加权和进行计算输出,计算公式为
${O}_{j}=\frac{{s}_{ni}}{{s}_{\mathrm{d}}}$
影响GRNN模型准确度的因素主要是模式层的平滑因子σ,σ取值偏大或偏小会导致模型出现欠拟合现象或过拟合现象。对于不同的训练样本数据,σ的最佳取值差异较大。在GRNN模型训练中σ的最优值通常难以寻求,为尽量提升GRNN模型的准确度,本文研究选用K-Fold算法对σ进行优化。
交叉验证即将数据集Data划分成训练集和验证集两部分[13],首先用训练集训练模型,然后用测试集验证模型,以此来作为评价模型的性能指标。
K重交叉验证的示意图如图3所示,将训练集数据Data分为k份,用k-1份数据进行模型训练,剩余一份数据对模型进行评估,则共有k种分割方式。对k种训练集数据分别进行训练,取评估时模型准确度最高的分割方式和σ取值作为Data数据集的最优划分,用最优划分数据训练出来的模型可视为K重交叉验证的最优模型,此时的参数取值可视为当前训练集数据下的最优值。
将列车长度、阻塞比、站台门安装距离、列车速度作为输入层数据,站台门结构形变量作为神经网络的输出数据。为保证训练模型的训练结果的准确性,从210组数据中选取189组作为训练数据,余下21组作为验证数据,训练与验证数据比例为9∶1。列取8组验证数据,如表1所示。
在CNN算法中,训练样本不能直接使用,需要经过归一化处理和数据维度转换后作为输入数据再进行模型的训练与验证。
首先采用最大最小归一化函数(mapminmax)对样本数据归一化,将其映射到[-1,1]区间(在后续步骤中,预测出的结果需要反归一化回原始数据的范围),使4种拥有不同单位和不同范围的输入数据(车长、阻塞比、站台门安装距离、列车速度)在模型训练中拥有相同权重。
其次,CNN模型训练采用的是四维数据,因此需使用MATLAB函数Reshape将用于训练的仿真二维数据转换为四维数据。
最后,利用MATLAB自带的Deep Learning Toolbox工具箱函数构建CNN模型,为保证模型的训练效果,需根据输入实际数据卷积核大小进行设置。经过反复多次试验与调整,最终确定模型参数如表2所示。
函数InitialLearnRate控制模型参数更新的步长,设置为0.001;最大迭代次数参数MaxEpochs设置为120,表明模型经过120次完整的训练;使用小批量数据进行训练对模型进行优化,设置小批量数据大小MiniBatchSize为32;数据混洗(Shuffle)、每个迭代重新混洗(every-epoch)参数设定训练数据的混洗策略,设置every-epoch参数在每次迭代开始时重新混洗训练数据,增加训练的随机性。
图4是利用表2各项参数调整后的训练过程,展示了均方根误差(root mean squared error,RMSE)和迭代次数的关系。可见当迭代次数达到400次时,平均方根误差曲线趋于稳定,此时达到训练要求。
通过输入验证数据对训练好的CNN模型进行准确度评估,使用MATLAB作图对比真实值和CNN模型的预测值,模型的直观预测效果如图5所示,可见真值曲线与CNN模拟曲线走向一致。
在预测站台门结构形变量数据时,为降低真实数据与模拟数据的误差,需要对GRNN模型的σ参数优化求解,本节采用K-Fold方法进行求解。
分析站台门型变量影响因素,取车长、阻塞比、车速和安装距离4项指标因素作为网络输入,以站台门型变量指标因素作为网络输出,构建GRNN模型网络。
对样本数据的前189组进行划分,取188组数据作为交叉验证的训练集,剩余1组作为验证集,共有189种划分方式。K-Fold方法的选取参数如表3所示,设置σ参数取值区间[-1,1],变化步长为0.1,通过循环求解真实数据与模拟数据误差最小的情况,此时的取值为最优值,利用该最优参数值和对应的划分方式训练得到的模型为最优模型。
训练样本同样需要用mapminmax函数经归一化后才能作为GRNN模型训练函数中作为函数输入使用。取K-Fold方法计算得到的最佳σ、最佳划分方式作为函数参数输入,命名得到的GRNN模型为net。
通过输入验证数据对训练好的GRNN模型进行准确度评估,图6为站台门结构形变量真实值与基于K-Fold算法的GRNN模型预测值的对比,可见预测曲线与模拟曲线走向一致。
站台门结构运维数据的真实值与两种网络模型的预测值的对比如图7所示。从细节看,两个模型均在个别相同点位出现了较明显的误差;从整体看,两条模拟曲线均贴合真实曲线。
2种模型对于8组数据的预测误差如表4所示,观察发现第3组数据的整体误差较大,考虑到两种模型的预测彼此接近,可判断为真实值记录时存在误差导致;在第二组数据的预测上,CNN模型拟合较好而GRNN模型误差较大。除模拟误差较大的第2组数据外其余6组数据的预测误差均保持在11%、4%和8%以内,推测可知GRNN模型在训练过程中偶尔出现样本分布不均匀导致极个别输入数据的预测中表现不佳的情况,CNN模型则较稳定。
分别计算两种模型的预测数据与仿真数据之间的均方差(mean absolute error,MEA)和均方根误差RMSE,计算结果如表5所示。
$MEA=\frac{1}{n}\stackrel{n}{\sum _{i=1}}\left|{x}_{i}-x\right|$
$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=\sqrt{\frac{1}{n}\stackrel{n}{\sum _{i=1}}({x}_{i}{-x)}^{2}}$
式中:n为样本数量;xi为预测值;x为真实值;i为样本次序。
利用K-Fold算法进行参数优化后的GRNN模型的预测均方差和均方根误差都最小,分别是0.223 9和0.265 2。从模拟结果上看基于K-Fold算法优化的GRNN模型的整体预测精度优于CNN模型。
国内站台门形变量预测以往通常采用流体力学、有限元分析和试验验证相结合的方式,通过模拟列车过站时的气动载荷进行结构变型的有限元分析,最终进行线路实验验证。当前已有的站台门型变量预测方式与机器学习方式的最优精度对比如表6所示,通过有限元分析的方式进行高铁站台门型变量预测的最大均方差维持在0.38内,均方根误差为0.28,城际列车站台门型变量预测最大均方差维持在0.43内,均方根误差为0.31[6-7],与国内站台门结构形变的已有研究成果对比,本文通过机器学习算法对站台门形变预测的最高精度模型为基于K-Fold的GRNN模型,预测均方差维持在0.23以内,均方根误差为0.27,低于有限元分析仿真算法,大幅提升了结构形变预测的准确度,为高速铁路站台门的前期设计与后期维修提供了数据参考和支撑。
将高速列车过站时在不同车长、速度,站台不同安装距离及不同阻塞比的210种工况及210种工况下的站台门结构运维数据作为样本数据,采用CNN神经网络和基于K-Fold算法优化的GRNN神经网络,分别建立站台门形变量预测模型,并进行试验和验证。得出如下结论。
(1)基于K-Fold优化的GRNN模型的均方差维持在0.22以内,均方根误差为0.27,优于CNN模型,两种模型均能较好的预测铁路站台门结构运维数据。
(2)基于K-Fold优化的GRNN模型精确度高于传统的有限元仿真分析方式,处于国内相关研究的领先水平,能够较好预测列车过站时的站台门结构形变量,为站台门前期结构设计与后期运维提供数据参考。
  • 国家自然科学基金(U21A201646)
  • 中国铁道科学研究院集团有限公司重点项目(2023YJ129)
参考文献 引证文献
排序方式:
[1]
李燕. PHM技术在京张高速铁路运用检修中的应用研究[J]. 铁道运输与经济, 2022, 44(9): 103-111.
Li Yan. Research on the application of PHM technology in the maintenance of Beijing-Zhangjiakou high-speed railway[J]. Railway Transport and Economy, 2022, 44(9): 103-111.
[2]
韩运动, 陈大伟, 刘韶庆, 等. 不同风速风向条件下的列车风特性[J]. 中国铁道科学, 2018, 39(6): 104-111.
Han Yundong, Chen Dawei, Liu Shaoqing, et al. Train aerodyna-mic characteristics under different wind speeds and directions[J]. Chinese Railway Science, 2018, 39(6): 104-111.
[3]
李新民, 聂小春, 晏致涛, 等.1 520 mm2大截面两分裂新月形覆冰导线空气动力特性试验研究[J]. 科学技术与工程, 2017, 17(27): 53-58.
Li Xinmin, Nie Xiaochun, Yan Zhitao, et al. Experimental study on the aerodynamic characteristics of 1 520 mm2 large cross-section bifurcated crescent-shaped iced conductor[J]. Science Technology and Engineering, 2017, 17(27): 53-58.
[4]
白雨坤. 城市轨道交通站台门系统的控制与数据处理[D]. 北京: 中国铁道科学研究院, 2014.
Bai Yukun. Control and data processing of platform screen door systems in urban rail transit[D]. Bei’jing: China Academy of Railway Sciences, 2014.
[5]
胡振亚, 罗威. 轨道交通站台门轨旁安全设计[J]. 中国安全生产科学技术, 2019, 15(S1): 85-89.
Hu Zhenya, Luo Wei. Safety design for platform screen door trackside in rail transit railways[J]. Science Technology and EngineerinJournal of Safety Science and Technology, 2019, 15(S1): 85-89.
[6]
王志飞. 列车高速过站风压对站台门结构特性的影响[J]. 机械设计与制造, 2020(12): 134-137, 141.
Wang Zhifei. Impact of train pass-by wind pressure on the structural characteristics of platform screen doors at high speed[J]. Machine-ry Design & Manufacture, 2020(12): 134-137, 141.
[7]
周梅, 毛良. 城际铁路列车通过地下站站台门时的风致效应仿真与试验研究[J]. 现代城市轨道交通, 2023(3): 25-29.
Zhou Mei, Mao Liang. Simulation and experimental study on the wind-induced effects of intercity railway trains passing platform screen doors at underground stations[J]. Modern Urban Transit, 2023(3): 25-29.
[8]
Ali J M, Omar A A, Ali M’B A, et al. Numerical investigation of aerodynamic characteristics of high speed trail[J]. IOP Conference, 2017, 184(1).DOI:10.1088/1757-899X/184/1/012015.
[9]
肖艳, 王建东, 王振. 基于亲信度的过程信号与报警信号相关性分析[J]. 科学技术与工程, 2020, 20(24): 9930-9936.
Xiao Yan, Wang Jiandong, Wang Zhen. Correlation analysis of process signals and alarm signals based on credibility[J]. Science Technology and Engineering, 2020, 20(24): 9930-9936.
[10]
曾国治, 魏子清, 岳宝, 等. 基于CNN-RNN组合模型的办公建筑能耗预测[J]. 上海交通大学学报, 2022, 56(9): 1256-1261.
Zeng Guozhi, Wei Ziqing, Yue Bao, et al. Energy consumption prediction for office buildings based on a combined CNN-RNN model[J]. Journal of Shanghai Jiao Tong University, 2022, 56(9): 1256-1261.
[11]
江佳霖, 钟宝荣. 基于改进Mask R-CNN的高密度砂岩颗粒的分割识别[J]. 科学技术与工程, 2024, 24(9): 3737-3746.
Jiang Jialin, Zhong Baorong. Segmentation and identification of high-density sandstone particles based on improved mask R-CNN[J]. Science Technology and Engineering, 2024, 24(9): 3737-3746.
[12]
杨雅萍, 张文莲, 孙晓云. 基于改进蝙蝠算法优化广义回归神经网络的岩质边坡稳定性预测[J]. 科学技术与工程, 2021, 21(20): 8719-8726.
Yang Yaping, Zhang Wenlian, Sun Xiaoyun. Prediction of rock slope stability based on generalized regression neural network optimized by improved bat algorithm[J]. Science Technology and Engineering, 2021, 21(20): 8719-8726.
[13]
Makomere S R, Koech L, Rutto L H, et al. Precision forecasting of spray-dry desulfurization using Gaussian noise data augmentation and K-fold cross-validation optimized neural computing[J]. Journal of Environmental Science and Health: Part A, 2024, 59(1): 11-14.
2025年第25卷第6期
PDF下载
220
67
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403476
  • 接收时间:2024-05-11
  • 首发时间:2025-07-27
  • 出版时间:2025-02-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-11
  • 修回日期:2024-12-17
基金
国家自然科学基金(U21A201646)
中国铁道科学研究院集团有限公司重点项目(2023YJ129)
作者信息
    1 中国铁道科学研究院研究生部, 北京 100081
    2 中国铁道科学研究院集团有限公司电子计算技术研究所, 北京 100081

通讯作者:

* 王志飞(1981—),男,汉族,内蒙古东胜人,博士,研究员。研究方向:交通运输工程与控制。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403476
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