Article(id=1281203008876810774, tenantId=1146029695717560320, journalId=1240685776644648972, issueId=1281202552578478607, articleNumber=null, orderNo=null, doi=10.3969/j.issn.1007-7294.2026.05.010, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1757520000000, receivedDateStr=2025-09-11, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1783392200916, onlineDateStr=2026-07-07, pubDate=1778774400000, pubDateStr=2026-05-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1783392200916, onlineIssueDateStr=2026-07-07, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1783392200916, creator=13041195026, updateTime=1783392200916, updator=13041195026, issue=Issue{id=1281202552578478607, tenantId=1146029695717560320, journalId=1240685776644648972, year='2026', volume='30', issue='5', pageStart='659', pageEnd='842', issueExtLink='null', onlineDate='null', pubDate='1778774400000', pubDateStr='2026-05-15', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1783392092127, creator='13041195026', updateTime=1783395243852, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1281215774769525418, tenantId=1146029695717560320, journalId=1240685776644648972, issueId=1281202552578478607, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1281215775176372907, tenantId=1146029695717560320, journalId=1240685776644648972, issueId=1281202552578478607, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=779, endPage=790, ext={EN=ArticleExt(id=1281203010982351384, articleId=1281203008876810774, tenantId=1146029695717560320, journalId=1240685776644648972, language=EN, title=Delamination damage identification of thermoplastic composite pipe based on residual attention Brownian covariance network, columnId=1242129251223274417, journalTitle=Journal of Ship Mechanics, columnName=Structural Mechanics, runingTitle=null, highlight=null, articleAbstract=

Thermoplastic composite pipes (TCP) have been widely used in marine structures. In this paper, a residual attention Brownian covariance neural network is established to study the damage identification of TCP composite delamination. Firstly, the curvature modes of multiple groups of thermoplastic composite tubes with single damage, multiple damages and different damage degrees were calculated using the finite element method. Then, the delamination damage identification method of thermoplastic composite tubes was discussed. Finally, the residual attention Brownian covariance network model was constructed using the curvature modes as input parameters to identify the delamination damage location and damage degree of TCP. The results show that the damage identification model based on residual attention Brownian covariance network can identify the location and degree of damage. The accuracy of damage location identification is 100%, and the error of damage degree identification is less than 6%. The research results provide a reference for non-destructive testing of marine engineering structures.

, authors=Chuan-yuan PENG1, Qi-ming SHU2, You-wei DU1, Wen SHEN1, authorsList=Chuan-yuan PENG, Qi-ming SHU, You-wei DU, Wen SHEN, authorCompany=null, correspAuthors=Qi-ming SHU, authorNote=null, correspAuthorsNote=null, copyrightStatement=Copyright ©2026 Journal of Ship Mechanics. All rights reserved., 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, fund=null), CN=ArticleExt(id=1281203118566249167, articleId=1281203008876810774, tenantId=1146029695717560320, journalId=1240685776644648972, language=CN, title=基于残差注意力布朗协方差网络的热塑性复合材料管脱层损伤识别, columnId=1241023038926410098, journalTitle=船舶力学, columnName=结构力学, runingTitle=null, highlight=null, articleAbstract=

热塑性复合材料管(Thermoplastic Composite Pipes, TCP)结构在海洋结构物中得到了广泛的应用,本文建立残差注意力布朗协方差神经网络进行TCP复合材料层间脱层的损伤识别研究。首先,通过有限元方法计算多组具有单损伤、多损伤和不同损伤程度的热塑性复合材料管的曲率模态;然后,讨论热塑性复合材料管脱层损伤识别方法;最后,将曲率模态作为输入参数,构建残差注意力布朗协方差网络模型,对热塑性复合材料管脱层损伤位置和损伤程度进行识别。结果表明,本文提出的基于残差注意力布朗协方差网络的管道损伤识别模型能够识别损伤位置和大小。对损伤位置的识别准确度为100%,损伤程度识别误差低于6%,研究结果可为海洋工程结构物无损检测提供参考。

, authors=彭传远1, 舒启明2, 杜友威1, 申雯1, authorsList=彭传远, 舒启明, 杜友威, 申雯, authorCompany=null, correspAuthors=舒启明, authorNote=

彭传远(1997–),男,硕士,讲师,E-mail:

, correspAuthorsNote=
舒启明(1997–),男,博士研究生,通讯作者,E-mail:
, copyrightStatement=版权所有©《船舶力学》编辑部2026, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=VZG84tzqlnwvS3/90u6RTA==, magXml=6ma16iky/uQjukepY+C/rQ==, pdfUrl=null, pdf=XEiugqEZ98Io1qhS6n2RQg==, pdfFileSize=2972682, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=rQRzBKWAFUogIH9HKd8K/A==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=IVm9P9s1PpWOgD9PiBh6ug==, mapNumber=null, fund=null)}, authors=[Author(id=1281203122966074071, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=pcy01131234@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1281203125080003289, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203122966074071, language=EN, stringName=Chuan-yuan PENG, firstName=Chuan-yuan, middleName=null, lastName=PENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203125482656474, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203122966074071, language=CN, stringName=彭传远, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio={"content":"

彭传远(1997–),男,硕士,讲师,E-mail:

"}, bioImg=null, bioContent=

彭传远(1997–),男,硕士,讲师,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])]), Author(id=1281203125637845724, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=shuqm@hust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1281203126338294494, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203125637845724, language=EN, stringName=Qi-ming SHU, firstName=Qi-ming, middleName=null, lastName=SHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203126883554015, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203125637845724, language=CN, stringName=舒启明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.华中科技大学 船舶与海洋工程学院,武汉 430074, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203122164962003, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=2., ext=[AuthorCompanyExt(id=1281203122219487956, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China), AuthorCompanyExt(id=1281203122546643669, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.华中科技大学 船舶与海洋工程学院,武汉 430074)])]), Author(id=1281203127596585697, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, 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=1281203129467245284, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203127596585697, language=EN, stringName=You-wei DU, firstName=You-wei, middleName=null, lastName=DU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203129781818085, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203127596585697, language=CN, stringName=杜友威, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])]), Author(id=1281203130524209895, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, 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=1281203131048497897, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203130524209895, language=EN, stringName=Wen SHEN, firstName=Wen, middleName=null, lastName=SHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203131514065642, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203130524209895, language=CN, stringName=申雯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])])], keywords=[Keyword(id=1281203133099512555, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, orderNo=1, keyword=thermoplastic composite pipe), Keyword(id=1281203133556691692, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, orderNo=2, keyword=curvature mode), Keyword(id=1281203134030648045, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, orderNo=3, keyword=damage identification), Keyword(id=1281203134840148718, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, orderNo=4, keyword=residual attention network), Keyword(id=1281203135213441775, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, orderNo=1, keyword=热塑性复合材料管), Keyword(id=1281203135658038001, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, orderNo=2, keyword=曲率模态), Keyword(id=1281203136085857010, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, orderNo=3, keyword=损伤识别), Keyword(id=1281203138103317237, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, orderNo=4, keyword=残差注意力网络)], refs=[Reference(id=1281203167434085148, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=7, pageStart=104, pageEnd=109, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=彭传远, 付新钰, 师梦科, journalName=工程塑料应用, refType=null, unstructuredReference=彭传远, 付新钰, 师梦科, . 大口径热塑性复合材料管及接头内压失效分析[J]. 工程塑料应用, 2022, 50(7): 104‒109., articleTitle=大口径热塑性复合材料管及接头内压失效分析, refAbstract=null), Reference(id=1281203167912235805, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=7, pageStart=104, pageEnd=109, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=Peng C Y, Fu X Y, Shi M K, journalName=Application of Engineering Plastics, refType=null, unstructuredReference=Peng C Y, Fu X Y, Shi M K, et al. Failure analysis of internal pressure of large-diameter thermoplastic composite pipes and joints[J]. Application of Engineering Plastics, 2022, 50(7): 104‒109. (in Chinese), articleTitle=Failure analysis of internal pressure of large-diameter thermoplastic composite pipes and joints, refAbstract=null), Reference(id=1281203171271873310, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=6, pageStart=180, pageEnd=185, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=潘静雯, 袁浩凡, 张芝芳, journalName=噪声与振动控制, refType=null, unstructuredReference=潘静雯, 袁浩凡, 张芝芳. 基于频率识别纤维增强复合材料弧形板分层损伤[J]. 噪声与振动控制, 2017, 37(6): 180‒185., articleTitle=基于频率识别纤维增强复合材料弧形板分层损伤, refAbstract=null), Reference(id=1281203171758412575, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=6, pageStart=180, pageEnd=185, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=Pan J W, Yuan H F, Zhang Z F, journalName=Noise and Vibration Control, refType=null, unstructuredReference=Pan J W, Yuan H F, Zhang Z F. Hierarchical damage identification of fiber-reinforced composite arc-shaped plates based on frequency[J]. Noise and Vibration Control, 2017, 37(6): 180‒185. (in Chinese), articleTitle=Hierarchical damage identification of fiber-reinforced composite arc-shaped plates based on frequency, refAbstract=null), Reference(id=1281203174132388640, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2019, volume=36, issue=11, pageStart=2614, pageEnd=2627, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=梁智洪, 詹 超, 张芝芳, journalName=复合材料学报, refType=null, unstructuredReference=梁智洪, 詹 超, 张芝芳. 基于频率识别纤维增强树脂复合材料加筋板的分层损伤[J]. 复合材料学报, 2019, 36(11): 2614‒2627., articleTitle=基于频率识别纤维增强树脂复合材料加筋板的分层损伤, refAbstract=null), Reference(id=1281203174556013345, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2019, volume=36, issue=11, pageStart=2614, pageEnd=2627, url=null, language=null, rfNumber=3, rfOrder=5, authorNames=Liang Z L, Zhan C, Zhang Z F, journalName=Journal of Composite Materials, refType=null, unstructuredReference=Liang Z L, Zhan C, Zhang Z F. Delamination damage of fiber-reinforced resin composite reinforced plates based on frequency identification[J]. Journal of Composite Materials, 2019, 36(11): 2614‒2627. (in Chinese), articleTitle=Delamination damage of fiber-reinforced resin composite reinforced plates based on frequency identification, refAbstract=null), Reference(id=1281203175059329826, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=10, pageStart=81, pageEnd=87, url=null, language=null, rfNumber=4, rfOrder=6, authorNames=刘 雨, 夏张松, 宫兆斌, journalName=无损检测, refType=null, unstructuredReference=刘 雨, 夏张松, 宫兆斌, . 复合材料修理区域的超声检测[J]. 无损检测, 2024, 46(10): 81‒87., articleTitle=复合材料修理区域的超声检测, refAbstract=null), Reference(id=1281203175390679843, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=10, pageStart=81, pageEnd=87, url=null, language=null, rfNumber=4, rfOrder=7, authorNames=Liu Y, Xia Z S, Gong Z B, journalName=Nondestructive Testing, refType=null, unstructuredReference=Liu Y, Xia Z S, Gong Z B, et al. Ultrasonic testing of the composite material repair area[J]. Nondestructive Testing, 2024, 46(10): 81‒87. (in Chinese), articleTitle=Ultrasonic testing of the composite material repair area, refAbstract=null), Reference(id=1281203175902384932, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=8, authorNames=王克凡, journalName=null, refType=null, unstructuredReference=王克凡. 基于电容成像的复合材料无损检测技术研究[D]. 青岛: 中国石油大学(华东), 2020., articleTitle=基于电容成像的复合材料无损检测技术研究, refAbstract=null), Reference(id=1281203176804160293, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=9, authorNames=Wang K F, journalName=null, refType=null, unstructuredReference=Wang K F. Research on non-destructive testing technology of composite materials based on capacitance imaging technique[D]. Qingdao: China University of Petroleum, 2020. (in Chinese), articleTitle=Research on non-destructive testing technology of composite materials based on capacitance imaging technique, refAbstract=null), Reference(id=1281203178490270502, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=14, issue=4, pageStart=786, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=10, authorNames=Sofer M, Jakub C, Martin F, journalName=Materials, refType=null, unstructuredReference=Sofer M, Jakub C, Martin F, et al. Damage analysis of composite CFRP tubes using acoustic emission monitoring and pattern recognition approach[J]. Materials, 2021, 14(4): 786., articleTitle=Damage analysis of composite CFRP tubes using acoustic emission monitoring and pattern recognition approach, refAbstract=null), Reference(id=1281203178834203431, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=6, pageStart=2005, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=11, authorNames=Scholz V, Winkler P, Andreas H, journalName=Sensors, refType=null, unstructuredReference=Scholz V, Winkler P, Andreas H, et al. Structural damage identification of composite rotors based on fully connected neural networks and convolutional neural networks[J]. Sensors, 2021, 21(6): 2005., articleTitle=Structural damage identification of composite rotors based on fully connected neural networks and convolutional neural networks, refAbstract=null), Reference(id=1281203179270411048, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=7, pageStart=44, pageEnd=49, url=null, language=null, rfNumber=8, rfOrder=12, authorNames=吴 磊, 梅江涛, 赵 硕, journalName=实验室研究与探索, refType=null, unstructuredReference=吴 磊, 梅江涛, 赵 硕. 基于IPSO-BP神经网络的管道损伤检测方法[J]. 实验室研究与探索, 2022, 41(7): 44‒49., articleTitle=基于IPSO-BP神经网络的管道损伤检测方法, refAbstract=null), Reference(id=1281203179819864873, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=7, pageStart=44, pageEnd=49, url=null, language=null, rfNumber=8, rfOrder=13, authorNames=Wu L, Mei J T, Zhao S, journalName=Laboratory Research and Exploration, refType=null, unstructuredReference=Wu L, Mei J T, Zhao S. Pipeline damage detection method based on IPSO-BP neural network[J]. Laboratory Research and Exploration, 2022, 41(7): 44‒49. (in Chinese), articleTitle=Pipeline damage detection method based on IPSO-BP neural network, refAbstract=null), Reference(id=1281203180197352235, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=10, pageStart=12937, pageEnd=12954, url=null, language=null, rfNumber=9, rfOrder=14, authorNames=Wu L, Mei J T, Zhao S, journalName=Applied Intelligence, refType=null, unstructuredReference=Wu L, Mei J T, Zhao S, et al. Pipeline damage identification based on an optimized back-propagation neural network improved by whale optimization algorithm[J]. Applied Intelligence, 2023, 53(10): 12937‒12954., articleTitle=Pipeline damage identification based on an optimized back-propagation neural network improved by whale optimization algorithm, refAbstract=null), Reference(id=1281203180469981996, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2023, volume=273, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=15, authorNames=Mei J T, Wu L, Chen E Q, journalName=Knowledge-based Systems, refType=null, unstructuredReference=Mei J T, Wu L, Chen E Q, et al. A novel structural damage detection method using a hybrid IDE-BP model[J]. Knowledge-based Systems, 2023, 273: 110606., articleTitle=A novel structural damage detection method using a hybrid IDE-BP model, refAbstract=null), Reference(id=1281203180943938349, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2023, volume=15, issue=6, pageStart=1520, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=16, authorNames=Zhao J Y, Xie W H, Yu D, journalName=Polymers, refType=null, unstructuredReference=Zhao J Y, Xie W H, Yu D, et al. Deep transfer learning approach for localization of damage area in composite laminates using acoustic emission signal[J]. Polymers, 2023, 15(6): 1520., articleTitle=Deep transfer learning approach for localization of damage area in composite laminates using acoustic emission signal, refAbstract=null), Reference(id=1281203181283676974, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2018, volume=170, issue=null, pageStart=171, pageEnd=185, url=null, language=null, rfNumber=12, rfOrder=17, authorNames=Liu K, Yan R J, Guedes S, journalName=Ocean Engineering, refType=null, unstructuredReference=Liu K, Yan R J, Guedes S, et al. Damage identification in offshore jacket structures based on modal flexibility[J]. Ocean Engineering, 2018, 170: 171‒185., articleTitle=Damage identification in offshore jacket structures based on modal flexibility, refAbstract=null), Reference(id=1281203183133365039, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=51, issue=2, pageStart=81, pageEnd=86, url=null, language=null, rfNumber=13, rfOrder=18, authorNames=杜 宇, 杨 涛, 何梅洪, journalName=宇航材料工艺, refType=null, unstructuredReference=杜 宇, 杨 涛, 何梅洪. 基于运行模态的复合材料梁脱层损伤识别[J]. 宇航材料工艺, 2021, 51(2): 81‒86., articleTitle=基于运行模态的复合材料梁脱层损伤识别, refAbstract=null), Reference(id=1281203183569572656, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=51, issue=2, pageStart=81, pageEnd=86, url=null, language=null, rfNumber=13, rfOrder=19, authorNames=Du Y, Yang T, He M H, journalName=Aerospace Materials & Technology, refType=null, unstructuredReference=Du Y, Yang T, He M H. Delaminate damage identification of composite beam based on operational modal[J]. Aerospace Materials & Technology, 2021, 51(2): 81‒86. (in Chinese), articleTitle=Delaminate damage identification of composite beam based on operational modal, refAbstract=null), Reference(id=1281203184295187249, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=44, issue=3, pageStart=385, pageEnd=389, url=null, language=null, rfNumber=14, rfOrder=20, authorNames=杜 宇, 何梅洪, 李菊峰, journalName=固体火箭技术, refType=null, unstructuredReference=杜 宇, 何梅洪, 李菊峰. 基于曲率模态变化率的复合材料梁脱层损伤识别[J]. 固体火箭技术, 2021, 44(3): 385‒389., articleTitle=基于曲率模态变化率的复合材料梁脱层损伤识别, refAbstract=null), Reference(id=1281203184668480306, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2021, volume=44, issue=3, pageStart=385, pageEnd=389, url=null, language=null, rfNumber=14, rfOrder=21, authorNames=Du Y, He M L, Li J F, journalName=Solid Rocket Technology, refType=null, unstructuredReference=Du Y, He M L, Li J F. Delamination damage diagnosis of composite beams based on curvature mode change rate[J]. Solid Rocket Technology, 2021, 44(3): 385‒389. (in Chinese), articleTitle=Delamination damage diagnosis of composite beams based on curvature mode change rate, refAbstract=null), Reference(id=1281203184819475251, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2022, volume=129, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=22, authorNames=Zhang Y, Guo J, Xie Y G, journalName=Applied Ocean Research, refType=null, unstructuredReference=Zhang Y, Guo J, Xie Y G, et al. Warship damage identification using mode curvature shapes method[J]. Applied Ocean Research, 2022, 129: 103396., articleTitle=Warship damage identification using mode curvature shapes method, refAbstract=null), Reference(id=1281203185482175284, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2025, volume=null, issue=null, pageStart=328, pageEnd=336, url=null, language=null, rfNumber=16, rfOrder=23, authorNames=Zhou F, Cicone A, Zhou H M, journalName=Pattern Recognition Letters, refType=null, unstructuredReference=Zhou F, Cicone A, Zhou H M, et al. An enhanced Iterative Residual Convolutional Neural Network for non-stationary signal decomposition[J]. Pattern Recognition Letters, 2025: 328‒336., articleTitle=An enhanced Iterative Residual Convolutional Neural Network for non-stationary signal decomposition, refAbstract=null), Reference(id=1281203185935160117, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2024, volume=41, issue=7, pageStart=2229, pageEnd=2234, url=null, language=null, rfNumber=17, rfOrder=24, authorNames=苟光磊, 朱东华, 李小菲, journalName=计算机应用研究, refType=null, unstructuredReference=苟光磊, 朱东华, 李小菲, . 深度掩膜布朗距离协方差小样本分类方法[J]. 计算机应用研究, 2024, 41(7): 2229‒2234., articleTitle=深度掩膜布朗距离协方差小样本分类方法, refAbstract=null), Reference(id=1281203187575132982, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2024, volume=41, issue=7, pageStart=2229, pageEnd=2234, url=null, language=null, rfNumber=17, rfOrder=25, authorNames=Gou G L, Zhu D H, Li X F, journalName=Application Research of Computers, refType=null, unstructuredReference=Gou G L, Zhu D H, Li X F, et al. Deep mask Brownian distance covariance for few-shot classification[J]. Application Research of Computers, 2024, 41(7): 2229‒2234. (in Chinese), articleTitle=Deep mask Brownian distance covariance for few-shot classification, refAbstract=null), Reference(id=1281203187688379191, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=770, pageEnd=778, url=null, language=null, rfNumber=18, rfOrder=26, authorNames=He K M, Zhang X Y, Ren S Q, journalName=Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[J]. Conference on Computer Vision and Pattern Recognition, 2016: 770‒778., articleTitle=Deep residual learning for image recognition, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)]), AuthorCompany(id=1281203122164962003, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=2., ext=[AuthorCompanyExt(id=1281203122219487956, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China), AuthorCompanyExt(id=1281203122546643669, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.华中科技大学 船舶与海洋工程学院,武汉 430074)])], figs=[ArticleFig(id=1281203138912817910, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.1, caption=Typical TCP structure, figureFileSmall=3JPXBzyzl9hdglBJTm08CQ==, figureFileBig=rQRzBKWAFUogIH9HKd8K/A==, tableContent=null), ArticleFig(id=1281203139353219831, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图1, caption=典型TCP结构, figureFileSmall=3JPXBzyzl9hdglBJTm08CQ==, figureFileBig=rQRzBKWAFUogIH9HKd8K/A==, tableContent=null), ArticleFig(id=1281203142746411769, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.2, caption=The overall architecture of Residual Attention Brownian Covariance Network (RABCN), figureFileSmall=sLk3+rzRCBkti3cPuie8ag==, figureFileBig=TZCVFIC2KiP09jJiO1f+Xg==, tableContent=null), ArticleFig(id=1281203143656575738, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图2, caption=残差注意力布朗协方差网络总体结构图, figureFileSmall=sLk3+rzRCBkti3cPuie8ag==, figureFileBig=TZCVFIC2KiP09jJiO1f+Xg==, tableContent=null), ArticleFig(id=1281203144352830203, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.3, caption=The architecture of Channel Attention, figureFileSmall=p+i2bzG0VmC+M6CcIUYb9Q==, figureFileBig=xSemTlUIN2UUwhAVTOE6nA==, tableContent=null), ArticleFig(id=1281203144520602364, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图3, caption=通道注意力结构图, figureFileSmall=p+i2bzG0VmC+M6CcIUYb9Q==, figureFileBig=xSemTlUIN2UUwhAVTOE6nA==, tableContent=null), ArticleFig(id=1281203144851952381, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.4, caption=Spatial attention module, figureFileSmall=w3i8348OBJU0FDO1D5B4mQ==, figureFileBig=61TIWjomZF64oYMBI2QY4g==, tableContent=null), ArticleFig(id=1281203145237828350, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图4, caption=空间注意力模块, figureFileSmall=w3i8348OBJU0FDO1D5B4mQ==, figureFileBig=61TIWjomZF64oYMBI2QY4g==, tableContent=null), ArticleFig(id=1281203146999436031, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.5, caption=Process of damage identification, figureFileSmall=s3WTrzKK8lia7IF/V16rqQ==, figureFileBig=AW6LzbELpzisGsUxuC7GuQ==, tableContent=null), ArticleFig(id=1281203147276260096, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图5, caption=损伤识别过程, figureFileSmall=s3WTrzKK8lia7IF/V16rqQ==, figureFileBig=AW6LzbELpzisGsUxuC7GuQ==, tableContent=null), ArticleFig(id=1281203147834102529, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.6, caption=TCP meshing, figureFileSmall=VynIDBe/jh1ML08FwCEGJQ==, figureFileBig=904iWY74Ymvbw3NqF3Xq6w==, tableContent=null), ArticleFig(id=1281203151701250818, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图6, caption=TCP网格划分, figureFileSmall=VynIDBe/jh1ML08FwCEGJQ==, figureFileBig=904iWY74Ymvbw3NqF3Xq6w==, tableContent=null), ArticleFig(id=1281203152129069828, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.7, caption=The first 3 displacement modes, figureFileSmall=7jKKlUlR8scZDLHO38nJsQ==, figureFileBig=sSusMz8ZvKIWO7P6757Dlg==, tableContent=null), ArticleFig(id=1281203152754021125, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图7, caption=前3阶位移模态, figureFileSmall=7jKKlUlR8scZDLHO38nJsQ==, figureFileBig=sSusMz8ZvKIWO7P6757Dlg==, tableContent=null), ArticleFig(id=1281203153148285702, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.8, caption=TCP curvature mode with different damage degrees, figureFileSmall=yDzPxbd33BPiAhGPcHNn3g==, figureFileBig=D5J8YrzwelL0hoUqw8Xl8g==, tableContent=null), ArticleFig(id=1281203153299280647, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图8, caption=不同损伤程度TCP曲率模态, figureFileSmall=yDzPxbd33BPiAhGPcHNn3g==, figureFileBig=D5J8YrzwelL0hoUqw8Xl8g==, tableContent=null), ArticleFig(id=1281203154016506632, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.9, caption=TCP curvature mode with different damage locations, figureFileSmall=V5dmcO+PprzEo9MGZyXicA==, figureFileBig=NzDHbK/7/puyehhFFxlcxg==, tableContent=null), ArticleFig(id=1281203154482074377, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图9, caption=不同损伤位置TCP曲率模态, figureFileSmall=V5dmcO+PprzEo9MGZyXicA==, figureFileBig=NzDHbK/7/puyehhFFxlcxg==, tableContent=null), ArticleFig(id=1281203156122047243, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.10, caption=Relation between the number of iterations and the loss function, figureFileSmall=UhErl49JK/NVtwKg+W5Bog==, figureFileBig=l92N9SYfEWpL53dIy0OErg==, tableContent=null), ArticleFig(id=1281203156256264972, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图10, caption=迭代次数与损失函数关系, figureFileSmall=UhErl49JK/NVtwKg+W5Bog==, figureFileBig=l92N9SYfEWpL53dIy0OErg==, tableContent=null), ArticleFig(id=1281203156566643469, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Fig.11, caption=Error rate of damage degree identification, figureFileSmall=IYdB4ns8MAakgs5OFEOvOw==, figureFileBig=BwKVfZ02wZXz0N/qYcT9IQ==, tableContent=null), ArticleFig(id=1281203157497778958, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=图11, caption=损伤程度识别误差率, figureFileSmall=IYdB4ns8MAakgs5OFEOvOw==, figureFileBig=BwKVfZ02wZXz0N/qYcT9IQ==, tableContent=null), ArticleFig(id=1281203157946569487, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.1, caption=

Structure parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
名称/单位数值名称/单位数值
总长 L /mm1000外保护层厚度 t3 /mm2.5
内径 R /mm50玻纤增强层层数/层10
内衬层厚度 t1 /mm3铺层方向(交叉铺层) /°±55
玻纤增强层厚度 t2 /mm3
), ArticleFig(id=1281203158751875856, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表1, caption=

结构参数

, figureFileSmall=null, figureFileBig=null, tableContent=
名称/单位数值名称/单位数值
总长 L /mm1000外保护层厚度 t3 /mm2.5
内径 R /mm50玻纤增强层层数/层10
内衬层厚度 t1 /mm3铺层方向(交叉铺层) /°±55
玻纤增强层厚度 t2 /mm3
), ArticleFig(id=1281203160727393041, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.2, caption=

Material parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
部件属性/单位参数
GFRP弹性模量 Eij /MPaE11 = 28000, E22 = E33 = 3200
剪切模量 Eij /GPaG12 = G13 = 2750, G23 = 1230
泊松比 vijv12 = v13 = 0.32, v23 = 0.33
HDPE弹性模量 E /MPa1423
泊松比 v0.38
), ArticleFig(id=1281203160882582290, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表2, caption=

材料参数

, figureFileSmall=null, figureFileBig=null, tableContent=
部件属性/单位参数
GFRP弹性模量 Eij /MPaE11 = 28000, E22 = E33 = 3200
剪切模量 Eij /GPaG12 = G13 = 2750, G23 = 1230
泊松比 vijv12 = v13 = 0.32, v23 = 0.33
HDPE弹性模量 E /MPa1423
泊松比 v0.38
), ArticleFig(id=1281203161536893715, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.3, caption=

Typical TCP damage condition

, figureFileSmall=null, figureFileBig=null, tableContent=
工况编号损伤节点编号损伤程度(层数)工况编号损伤节点编号损伤程度(层数)
11515103
21536203
31557303
41578403
), ArticleFig(id=1281203162182816532, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表3, caption=

典型TCP损伤工况

, figureFileSmall=null, figureFileBig=null, tableContent=
工况编号损伤节点编号损伤程度(层数)工况编号损伤节点编号损伤程度(层数)
11515103
21536203
31557303
41578403
), ArticleFig(id=1281203165471150870, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.4, caption=

Data of partial TCP curvature mode

, figureFileSmall=null, figureFileBig=null, tableContent=
样本
编号
脱层节点
编号
脱层
层数
损伤
程度
节点曲率模态
234484950
训练样本1210.1528.531066.621590.94−1576.91−1059.54−555.01
2230.3552.691284.991634.05−1574.89−1058.18−554.28
3250..5573.171531.051688.45−1572.61−1056.67−553.46
2455020.2551.211052.711566.47−1805.04−2180.50−726.42
测试样本246520.2556.351055.831574.22−1576.34−1059.14−554.80
247580.8554.701058.991576.091574.92−1058.19−554.30
2614080.8554.951059.421576.73−1576.59−1059.45−554.69
), ArticleFig(id=1281203165915747095, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表4, caption=

部分TCP曲率模态数据

, figureFileSmall=null, figureFileBig=null, tableContent=
样本
编号
脱层节点
编号
脱层
层数
损伤
程度
节点曲率模态
234484950
训练样本1210.1528.531066.621590.94−1576.91−1059.54−555.01
2230.3552.691284.991634.05−1574.89−1058.18−554.28
3250..5573.171531.051688.45−1572.61−1056.67−553.46
2455020.2551.211052.711566.47−1805.04−2180.50−726.42
测试样本246520.2556.351055.831574.22−1576.34−1059.14−554.80
247580.8554.701058.991576.091574.92−1058.19−554.30
2614080.8554.951059.421576.73−1576.59−1059.45−554.69
), ArticleFig(id=1281203166247097112, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.5, caption=

Comparison of model performance

, figureFileSmall=null, figureFileBig=null, tableContent=
任务模型MAEMAPE/%MSER2
位置识别ResNet1.21058.0727.840.8605
RABCN0001
程度识别ResNet0.083617.090.01890.6223
RABCN0.01723.730.00040.9924
), ArticleFig(id=1281203166582641433, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表5, caption=

模型性能对比

, figureFileSmall=null, figureFileBig=null, tableContent=
任务模型MAEMAPE/%MSER2
位置识别ResNet1.21058.0727.840.8605
RABCN0001
程度识别ResNet0.083617.090.01890.6223
RABCN0.01723.730.00040.9924
), ArticleFig(id=1281203166683304730, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=EN, label=Tab.6, caption=

Comparison of damage identification results

, figureFileSmall=null, figureFileBig=null, tableContent=
样本实际损伤状态(mt,nt预测损伤状态(mp,np样本实际损伤状态(mt,nt预测损伤状态(mp,np
15, 0.25, 0.188921, 0.621,0.629
25, 0.85, 0.8451024, 0.224,0.19
37, 0.27, 0.1951127, 0.827,0.828
48, 0.68, 0.6081233, 0.433,0.381
510, 0.210, 0.1891334, 0.834,0.808
614, 0.414, 0.4161437, 0.637,0.614
714, 0.614, 0.6351540, 0.440,0.405
819, 0.419, 0.4201640, 0.840,0.790
), ArticleFig(id=1281203167064986395, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, language=CN, label=表6, caption=

损伤识别结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
样本实际损伤状态(mt,nt预测损伤状态(mp,np样本实际损伤状态(mt,nt预测损伤状态(mp,np
15, 0.25, 0.188921, 0.621,0.629
25, 0.85, 0.8451024, 0.224,0.19
37, 0.27, 0.1951127, 0.827,0.828
48, 0.68, 0.6081233, 0.433,0.381
510, 0.210, 0.1891334, 0.834,0.808
614, 0.414, 0.4161437, 0.637,0.614
714, 0.614, 0.6351540, 0.440,0.405
819, 0.419, 0.4201640, 0.840,0.790
)], attaches=null, journal=Journal(id=1240685663704625163, delFlag=0, nameCn=船舶力学, nameEn=Journal of Ship Mechanics, nameHistory1=null, nameHistory2=null, issn=1007-7294, eissn=null, cn=32-1468/U, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=8VABIsgSw5I4HF/4goaIlA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Ship Mechanics, journalRemark=null, publicationField=null, createdTime=1773732113054, updatedTime=1784019383023, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=8VABIsgSw5I4HF/4goaIlA==, picEn=Ve/8A/FRYceS/z8o7GCJ5g==, jcr=null, cjcr=null, exts=[JournalExt(id=1283833602848834028, language=CN, name=船舶力学, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784019383394, updatedTime=1784019383394, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=1, submissionEditorUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=3, submissionReviewUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1283833602983051757, language=EN, name=Journal of Ship Mechanics, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784019383426, updatedTime=1784019383426, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=1, submissionEditorUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=3, submissionReviewUrl=https://cblx.cbpt.cnki.net/EditorD3N/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1240685776644648972, websiteList=[Website(id=1240686760611607145, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1240685776644648972, 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/cblx/CN, language=CN, createTime=1773732374576, createBy=18614031015, updateTime=1773732394922, updateBy=18614031015, name=船舶力学-中文, tplId=1146099689490845704, title=船舶力学, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1240687166741869216, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=articleTextType, value=kx, createTime=1773732471404, updateTime=1773732471404, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166725091997, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=banner, value=null, createTime=1773732471400, updateTime=1773732471400, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166758646435, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=grayFlag, value=0, createTime=1773732471408, updateTime=1773732471408, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166716703388, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=logo, value=https://castjournals.cast.org.cn/joweb/cblx/CN/file/pic?fileId=/ZVjg3ao2vf5wYxw5Kn1Uw==, createTime=1773732471398, updateTime=1773732471398, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166771229349, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=minRunFlag, value=0, createTime=1773732471411, updateTime=1773732471411, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166737674911, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/cblx/CN/file/pic, createTime=1773732471403, updateTime=1773732471403, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166767035044, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=silenceFlag, value=0, createTime=1773732471410, updateTime=1773732471410, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166729286302, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1773732471401, updateTime=1773732471401, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166750257825, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=themeColor, value=null, createTime=1773732471406, updateTime=1773732471406, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687166754452130, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760611607145, code=themeStyle, value=null, createTime=1773732471407, updateTime=1773732471407, creator=18614031015, updator=18614031015)]), Website(id=1240686760674521707, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1240685776644648972, 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/cblx/EN, language=EN, createTime=1773732374590, createBy=18614031015, updateTime=1773732413338, updateBy=18614031015, name=船舶力学-英文, tplId=1146101810881728533, title=Journal of Ship Mechanics, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1240687194269078438, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=articleTextType, value=kx, createTime=1773732477967, updateTime=1773732477967, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194248106915, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=banner, value=null, createTime=1773732477962, updateTime=1773732477962, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194285855657, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=grayFlag, value=0, createTime=1773732477971, updateTime=1773732477971, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194239718306, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=logo, value=https://castjournals.cast.org.cn/joweb/cblx/EN/file/pic?fileId=/ZVjg3ao2vf5wYxw5Kn1Uw==, createTime=1773732477960, updateTime=1773732477960, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194302632875, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=minRunFlag, value=0, createTime=1773732477975, updateTime=1773732477975, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194260689829, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/cblx/EN/file/pic, createTime=1773732477965, updateTime=1773732477965, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194294244266, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=silenceFlag, value=0, createTime=1773732477973, updateTime=1773732477973, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194256495524, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1773732477964, updateTime=1773732477964, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194277467047, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=themeColor, value=null, createTime=1773732477969, updateTime=1773732477969, creator=18614031015, updator=18614031015), WebsiteProps(id=1240687194281661352, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1240686760674521707, code=themeStyle, value=null, createTime=1773732477970, updateTime=1773732477970, creator=18614031015, updator=18614031015)])], journalTitle=船舶力学, weixinUrl=null, journalUrl=https://cblx.cbpt.cnki.net/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Ship Mechanics, journalPhotoCn=8VABIsgSw5I4HF/4goaIlA==, journalPhotoEn=Ve/8A/FRYceS/z8o7GCJ5g==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/cblx/CN/10.3969/j.issn.1007-7294.2026.05.010, detailUrlEn=https://castjournals.cast.org.cn/joweb/cblx/EN/10.3969/j.issn.1007-7294.2026.05.010, pdfUrlCn=https://castjournals.cast.org.cn/joweb/cblx/CN/PDF/10.3969/j.issn.1007-7294.2026.05.010, pdfUrlEn=https://castjournals.cast.org.cn/joweb/cblx/EN/PDF/10.3969/j.issn.1007-7294.2026.05.010, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1778774400000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于残差注意力布朗协方差网络的热塑性复合材料管脱层损伤识别
收藏切换
PDF下载
彭传远 1 , 舒启明 2 , 杜友威 1 , 申雯 1
船舶力学 | 结构力学 2026,30(5): 779-790
收起
收藏切换
船舶力学 |结构力学 2026 , 30 (5) : 779 -790
基于残差注意力布朗协方差网络的热塑性复合材料管脱层损伤识别
全屏
[Author(id=1281203122966074071, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=pcy01131234@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1281203125080003289, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203122966074071, language=EN, stringName=Chuan-yuan PENG, firstName=Chuan-yuan, middleName=null, lastName=PENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203125482656474, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203122966074071, language=CN, stringName=彭传远, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio={"content":"

彭传远(1997–),男,硕士,讲师,E-mail:

"}, bioImg=null, bioContent=

彭传远(1997–),男,硕士,讲师,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])]), Author(id=1281203125637845724, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=shuqm@hust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1281203126338294494, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203125637845724, language=EN, stringName=Qi-ming SHU, firstName=Qi-ming, middleName=null, lastName=SHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203126883554015, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203125637845724, language=CN, stringName=舒启明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.华中科技大学 船舶与海洋工程学院,武汉 430074, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203122164962003, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=2., ext=[AuthorCompanyExt(id=1281203122219487956, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China), AuthorCompanyExt(id=1281203122546643669, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203122164962003, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.华中科技大学 船舶与海洋工程学院,武汉 430074)])]), Author(id=1281203127596585697, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, 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=1281203129467245284, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203127596585697, language=EN, stringName=You-wei DU, firstName=You-wei, middleName=null, lastName=DU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203129781818085, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203127596585697, language=CN, stringName=杜友威, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])]), Author(id=1281203130524209895, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, 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=1281203131048497897, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203130524209895, language=EN, stringName=Wen SHEN, firstName=Wen, middleName=null, lastName=SHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1281203131514065642, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, authorId=1281203130524209895, language=CN, stringName=申雯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.青岛黄海学院 智能制造学院,山东 青岛 266427, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1281203121703588560, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, xref=1., ext=[AuthorCompanyExt(id=1281203121779086033, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China), AuthorCompanyExt(id=1281203121800057554, tenantId=1146029695717560320, journalId=1240685776644648972, articleId=1281203008876810774, companyId=1281203121703588560, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.青岛黄海学院 智能制造学院,山东 青岛 266427)])])]
彭传远1 , 舒启明2 , 杜友威1, 申雯1
作者信息
  • 1.青岛黄海学院 智能制造学院,山东 青岛 266427
  • 2.华中科技大学 船舶与海洋工程学院,武汉 430074
通讯作者:
舒启明(1997–),男,博士研究生,通讯作者,E-mail:
作者简介:

彭传远(1997–),男,硕士,讲师,E-mail:

Delamination damage identification of thermoplastic composite pipe based on residual attention Brownian covariance network
Chuan-yuan PENG1 , Qi-ming SHU2 , You-wei DU1, Wen SHEN1
Affiliations
  • 1.Intelligent Manufacturing College, Qingdao Huanghai University, Qingdao 266427, China
  • 2.School of Shipbuilding and Ocean Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
出版时间: 2026-05-15 doi: 10.3969/j.issn.1007-7294.2026.05.010
文章导航
收藏切换

热塑性复合材料管(Thermoplastic Composite Pipes, TCP)结构在海洋结构物中得到了广泛的应用,本文建立残差注意力布朗协方差神经网络进行TCP复合材料层间脱层的损伤识别研究。首先,通过有限元方法计算多组具有单损伤、多损伤和不同损伤程度的热塑性复合材料管的曲率模态;然后,讨论热塑性复合材料管脱层损伤识别方法;最后,将曲率模态作为输入参数,构建残差注意力布朗协方差网络模型,对热塑性复合材料管脱层损伤位置和损伤程度进行识别。结果表明,本文提出的基于残差注意力布朗协方差网络的管道损伤识别模型能够识别损伤位置和大小。对损伤位置的识别准确度为100%,损伤程度识别误差低于6%,研究结果可为海洋工程结构物无损检测提供参考。

热塑性复合材料管  /  曲率模态  /  损伤识别  /  残差注意力网络

Thermoplastic composite pipes (TCP) have been widely used in marine structures. In this paper, a residual attention Brownian covariance neural network is established to study the damage identification of TCP composite delamination. Firstly, the curvature modes of multiple groups of thermoplastic composite tubes with single damage, multiple damages and different damage degrees were calculated using the finite element method. Then, the delamination damage identification method of thermoplastic composite tubes was discussed. Finally, the residual attention Brownian covariance network model was constructed using the curvature modes as input parameters to identify the delamination damage location and damage degree of TCP. The results show that the damage identification model based on residual attention Brownian covariance network can identify the location and degree of damage. The accuracy of damage location identification is 100%, and the error of damage degree identification is less than 6%. The research results provide a reference for non-destructive testing of marine engineering structures.

thermoplastic composite pipe  /  curvature mode  /  damage identification  /  residual attention network
彭传远, 舒启明, 杜友威, 申雯. 基于残差注意力布朗协方差网络的热塑性复合材料管脱层损伤识别. 船舶力学, 2026 , 30 (5) : 779 -790 . DOI: 10.3969/j.issn.1007-7294.2026.05.010
Chuan-yuan PENG, Qi-ming SHU, You-wei DU, Wen SHEN. Delamination damage identification of thermoplastic composite pipe based on residual attention Brownian covariance network[J]. Journal of Ship Mechanics, 2026 , 30 (5) : 779 -790 . DOI: 10.3969/j.issn.1007-7294.2026.05.010
热塑性复合材料管是由热塑性聚合物挤出管与多层玻纤增强复合材料带缠绕而成的新型管道,具有强度高、质量轻、设计性强等优点,已在海洋工程、土木工程等领域广泛应用。管道由内向外依次为内衬层、玻纤增强层和外保护层,典型TCP(Thermoplastic Composite Pipes)结构如图1所示。
热塑性复合材料管的内衬层和外保护层为高密度聚乙烯材料HDPE(High Density Polyethylene),玻璃纤维增强层GFRP(Glass Fiber Reinforced Composite)由玻璃纤维和热塑性树脂基体复合而成。其中玻璃纤维为增强材料,为管道承载内压和环境载荷提供强度与刚度,HDPE为基体材料,起到保护和固定纤维的作用[1]。管道通过加热熔融形成一体式的管壁结构,具有较高的层间剪切强度,能够有效降低在位工况下管道的泄漏发生率。
热塑性复合材料管的复合材料层损伤包括基体开裂、纤维断裂和层间脱层等多种形式,复合管在生产、安装、服役过程中容易受到温度变化、应力集中、腐蚀等因素影响,进而发生脱层。据统计,脱层损伤占所有损伤类型的50%以上[2]。因此,如何有效地识别TCP的脱层损伤是一个具有挑战性的问题。
目前,针对工程结构的损伤检测方法主要包括超声检测法、电容成像法、声发射检测法[3]。刘雨等[4]利用超声检测的方法开展纤维增强复合材料损伤识别的实验研究,实验结果表明,超声检测对复合材料修理区域的分层缺陷具有良好的识别效果。王克凡[5]基于电容成像检测原理对蜂窝复合材料和复合材料抽油杆进行缺陷检测,研究发现电容成像可以实现复合材料完整性检测。Sofer等[6]研究了声发射信号在复合材料中的衰减、模态转换、散射等传播特性,发现该方法可识别基体裂纹、纤维断裂和脱粘等失效机制。然而,上述方法要求高精度的仪器设备,且依赖人工经验,面对复杂工程结构的损伤检测效果较差。
近年来,深度学习正逐渐成为一种解决上述问题的合适方案。典型的深度学习模型包括卷积神经网络和人工神经网络等,它们可以自动提取损伤相关的特征,大大减少了对专业知识的依赖。例如,Scholz等[7]应用全连接神经网络和卷积神经网络对复合材料转子的结构振动响应进行识别、定位和量化研究。Wu等[89]构建了IPSO-BP神经网络模型针对钢制管道的腐蚀进行损伤识别,预测效果良好。以此为基础,Mei等[10]应用多种神经网络模型对单损伤、多损伤及不同损伤位置的管道进行损伤预测,发现所提出的神经网络模型具有良好的检测精度。Zhao等[11]采用深度迁移学习实现损伤区域的实时定位,通过连续小波变换将声发射时域信号变换为输入图像,基于卷积神经网络的模型,通过从输入图像中提取特征来自动定位受损区域。
综上,研究复合材料在加工、装配、服役过程中的脱层损伤问题,并对损伤发生位置以及损伤程度进行识别至关重要。基于此背景,本文提出残差注意力布朗协方差网络进行TCP层间脱层损伤位置和损伤程度判定,为保障管道安全运行提供技术支撑。
结构位移模态包含结构的振型信息,在实验过程中高阶位移模态往往难以获取,因此常用的位移模态一般为1~3阶。一般情况下,结构损伤会引起结构的动态参数变化,当管道某个部位发生损伤时,会对该部位的刚度产生影响,导致结构位移模态发生变化,进而影响结构曲率[1214]
将热塑性复合材料管看成一维无阻尼梁式结构,梁的截面抗弯刚度和曲率关系为
$ k(x)=\frac{1}{\rho (x)}=\frac{M}{EI(x)} $
式中:$ k\left(x\right) $为曲率;$ \rho \left(x\right) $为曲率半径;$ M $为截面弯矩;EIx)为截面抗弯刚度。
由式(1)可知,曲率模态随着结构的抗弯刚度的变化而变化,当TCP中存在损伤时,损伤位置处的弯曲刚度会降低,结构曲率模态会在该位置发生突变。因此,结构曲率模态可以作为TCP结构的损伤识别的标准。利用中心差分法,TCP结构上任意节点的曲率模态可以被表示为
$ {C}_{r}(i)=\frac{{\varphi }_{r}{}_{(i+1)}+{\varphi }_{r}{}_{(i-1)}-2{\varphi }_{r}{}_{(i)}}{{l}^{2}} $
式中:r为模态阶数;i为节点编号;Cri)为i节点处的曲率模态;$ {\varphi }_{r}\left(i\right) $i节点在第r阶模态的横向位移;l为相邻节点间距离。
复合材料脱层会导致脱层位置的弯曲刚度发生变化,从而导致TCP的曲率发生改变,本文将通过对比不同脱层损伤位置的结构曲率模态进行TCP损伤位置与损伤程度的判定[15]
本文提出的残差注意力布朗协方差网络RABCN(Residual Attention Brownian Covariance Network)模型通过学习模态数据来实现TCP损伤位置与损伤程度的预测。假设TCP的训练数据集中位移模态数据为$ {D}_{{\mathrm{d}}}=\left({x}_{i,j},m,n\right) $。其中,m代表损伤位置,n代表损伤程度。TCP损伤识别由损伤位置识别的分类任务1和损伤程度识别的回归任务2组成。两个任务的具体细节如下:
在损伤位置识别的分类过程中,对数据进行预处理,位移模态数据$ {D}_{{\mathrm{d}}}=\left({x}_{i,j},i,j\right) $被转换为曲率模态数据$ {D}_{{\mathrm{c}}}=({X}_{ij},i,j) $。构建损伤位置识别数据集$ {D}_{1}=\left({X}_{i,j},i\right) $被输入到RABCN模型中进行训练。将测试数据送到训练好的RABCN模型,获得预测的损伤位置值。
在损伤程度识别的回归过程中,对数据进行预处理,位移模态数据$ {D}_{{\mathrm{d}}}=\left({x}_{i,j},i,j\right) $被转换为曲率模态数据$ {D}_{{\mathrm{c}}}=({X}_{ij},i,j) $。获取的曲率模态数据$ {D}_{2}=\left({X}_{i,j},j\right) $被输入到RABCN模型中进行训练。测试数据被送到训练好的RABCN模型,获得预测的损伤程度值。值得注意的是,在分类任务中,全连接层的输出为位置类别,回归任务中全连接层的输出为1。
针对热塑性复合材料管的损伤精准识别这一难点问题,残差注意力网络凭借通道与空间注意力机制,能够显著增强对局部故障特征的提取能力[16];同时,布朗协方差模型以其对高阶统计关联的敏感性,能够有效捕捉变量间的复杂相关性[17]。本文融合二者优势,提出了一种结合残差注意力与布朗协方差匹配的网络模型(RABCN),以揭示热塑性复合材料管损伤特征与结构状态之间的深层非线性关联,实现复杂工况下的高精度损伤识别,其结构如图2所示。
在残差注意力模块中,首先对监测信号进行初步的特征提取。采用卷积层和残差连接机制,将监测信号转化为高维特征。在这个过程中,空间注意力与通道注意力被加入到残差连接路径中,以增强对故障特征的全局关注。接着,高维特征被输入到布朗协方差模块,通过随机观测和矩阵转换,生成布朗协方差BDC(Brownian Distance Covariance)矩阵。最后,将BDC矩阵输入到全连接层进行诊断。
残差注意力模块的作用是提取监测信号中与故障相关的高维特征。作为残差网络ResNet (Residual Network)的一种变体,残差注意力模块在残差连接中加入通道注意力机制CAM (Channel Attention Mechanism)和空间注意力机制SAM (Spatial Attention Mechanism),提升模块对监测信号中隐藏特征的全局注意力,进而提取具有故障信息的全局特征。
CAM采用3D排列来保留跨三维的特征信息。随后,利用一个双层感知机MLP(Multilayer Perceptron),模拟通道之间的非线性关系。其中,MLP是一个编码器-解码器结构,具有缩减比r,sigmoid为激活函数。通过CAM,卷积层提取的特征F1被转化为特征F2,该过程可公式化为
$ {F}_{2}=\text{CAM}\left({F}_{1}\right)\otimes {F}_{1} $
式中:$ \text{CAM}\left(\cdot \right) $是通道注意力函数,表达式为:$ {\mathrm{CAM}}({F}_{1})={\mathrm{sigmoid}}({\mathrm{MLP}}({\mathrm{Permute}}({F}_{1}))) $,Permute将维度从(C,H,W)置换为(H,W,C),$ \otimes $是逐个元素相乘运算。通道注意力模块具体结构如图3所示。
SAM使用两个卷积层进行空间信息融合,从而捕捉空间信息。值得注意的是,由于SAM不包含传统注意力机制中的池化层,不改变特征张量的空间尺寸。因此,输出张量包含了三个维度的相互作用。通过SAM,提取的特征F2被转化为特征F3,该过程可公式化为
$ {F}_{3}=S\text{AM}\left({F}_{2}\right)\otimes {F}_{2} $
式中:$ \text{SAM}\left(\cdot \right) $是空间注意力函数,表达式为:$ {\mathrm{SAM}}({F}_{2})={\mathrm{sigmoid}}({\mathrm{Conv2}}({\mathrm{Conv1}}({F}_{2}))) $$ {\mathrm{Conv2}} $$ {\mathrm{Conv1}} $是两个卷积层,$ \otimes $是逐个元素相乘运算。空间注意力模块如图4所示。
最后,残差注意力模块通过残差连接机制生成特征F4,该过程可公式化为
$ {F}_{4}={\mathrm{RC}}\left({F}_{1},{F}_{3}\right) $
式中:$ \text{RC}\left(\cdot \right) $为残差连接函数,表达式为:$ y=F(x,\{{W}_{i}\})+x $x是输入向量,y为输出向量,$ \{{W}_{i}\} $指的是网络层的权重集合。
残差注意力模块的输出特征F4进一步输入布朗协方差模块以消除干扰信息,获得BDC矩阵F5。基于BDC度量理论,BDC矩阵可以根据边际与联合特征函数的乘积来度量两个随机向量之间的距离。两个随机向量之间的BDC度量表示为
$ \rho \left(X,Y\right)={\int }_{{{}_{{{R}^{p}}}}}{\int }_{{{}_{{{R}^{q}}}}}\frac{{\left| {\varnothing }_{XY}\left(\boldsymbol{t},\boldsymbol{s}\right)-{\varnothing }_{X}\left(\boldsymbol{t}\right){\varnothing }_{Y}\left(\boldsymbol{s}\right)\right| }^{2}}{{c}_{p}{c}_{q}\parallel \boldsymbol{t}{\parallel }^{1+p}\parallel \boldsymbol{s}{\parallel }^{1+q}}{\mathrm{d}}\boldsymbol{t}{\mathrm{d}}\boldsymbol{s} $
其中,$ X\in {R}^{p} $$ Y\in {R}^{q} $分别表示p维和q维欧几里得空间的随机向量。$ \rho \left(\cdot \right) $表示度量函数,$ \left|\left|\cdot \right|\right| $表示欧几里德范数,$ {c}_{p}={{\text{π}} }^{\tfrac{1+p}{2}}/\mathit{\Gamma }\left(\dfrac{1+p}{2}\right) $$ {c}_{q}={{\text{π}} }^{\tfrac{1+q}{2}}/\mathit{\Gamma }\left(\dfrac{1+q}{2}\right) $$ {\varnothing }_{XY}\left(\boldsymbol{t},\boldsymbol{s}\right) $$ X $$ Y $的联合特征函数,$ {\varnothing }_{X}\left(\boldsymbol{t}\right) $$ {\varnothing }_{Y}\left(\boldsymbol{s}\right) $分别是$ X $$ Y $的边缘特征函数。
假设得到一个$ {X}\in {R}^{hw\times d} $特征张量,两个随机观测$ {x}_{i}\in {R}^{hw} $$ {x}_{j}\in {R}^{d} $X中获得。其中,$ {x}_{i} $代表$ {X} $的每一列,$ {x}_{j} $代表$ {X} $的每一行。BDC矩阵$ \boldsymbol{A} $可以被获取
$ \boldsymbol{\tilde{A}}=2{\left(1\left({{X}}^{{\mathrm{T}}}{X}\circ \boldsymbol{I}\right)\right)}_{sym}-2{{X}}^{{\mathrm{T}}}{X} $
$ \boldsymbol{A}=\boldsymbol{\hat{A}}-\frac{2}{\text{d}}{(1\boldsymbol{\hat{A}})}_{{\mathrm{sym}}}+\frac{1}{{\text{d}}^{2}}1\boldsymbol{\hat{A}}1 $
其中,$ 1\in {{R}}^{d\times d} $代表每个元素为1的矩阵,$ \boldsymbol{I} $代表单位矩阵,$ \circ $代表Hadamard乘积,$ \boldsymbol{\hat{A}}=\sqrt{\boldsymbol{\tilde{A}}} $。因此,由于能够使用欧几里得距离捕获通道之间的非线性关系,BDC矩阵可以消除干扰信息,并提取深层的代表性特征。
全连接层FCL(Fully Connected Layer)的作用是将学习到的代表性特征进行显式表示。通过空间映射将代表性特征空间与样本空间进行关联,代表性特征被展开为一维向量,并与当前层的权重进行线性组合,再通过激活函数进行非线性变换,最终被映射为预测值。值得注意的是,在分类任务中,全连接层的机制可以被表达为
$ {\mathrm{FCL}}\left(x\right)={\mathrm{GELU}}\left({\boldsymbol{W}}x+b\right) $
其中,W是权重矩阵,b是偏差向量。GELU是激活函数,$ {\mathrm{GELU}}(y)=\dfrac{y}{2}\left(1+{\mathrm{erf}}\left(\dfrac{y}{\sqrt{2}}\right)\right) $,其中erf为误差函数,它被用来对特征进行非线性变换,增强模型的表征能力。
RABCN网络的损失函数是通过计算真实值和预测值之间的差值构成的。由于损伤识别包括损伤位置识别任务1和损伤程度识别任务2,因此有两个损伤函数。
$ {L}_{1}=\frac{1}{N}\sum \limits_{i=1}^{N}{\left({m}_{i}-{\hat{m}}_{i}\right)}^{2} $
$ {L}_{2}=\frac{1}{N}\sum \limits_{i=1}^{N}{\left({n}_{i}-{\hat{n}}_{i}\right)}^{2} $
其中,N是样本数量,mi是第$ i $个样本的损伤位置真实值,$ {\hat{m}}_{i} $是第$ i $个样本的损伤位置预测值,ni是第i个样本的损伤程度真实值,$ {\hat{n}}_{i} $是第i个样本的损伤程度预测值。
$ \vartheta $$ \theta $分别为残差注意力模块和全连接层的参数。公式(9)和(10)可以改写为
$ \delta _{\vartheta }^{*},\delta _{\theta }^{*}\leftarrow \underset{{\delta }_{\vartheta },{\delta }_{\theta }}{\text{argmin}}{L}_{{\mathrm{MSE}}}\left({\delta }_{\vartheta },{\delta }_{\theta }\right) $
其中,$ \delta _{\vartheta }^{*} $$ \delta _{\theta }^{*} $表示网络的优化参数。
采用提出的RABCN网络进行TCP损伤识别过程如图5所示,包含以下步骤:
(1)数据采集与划分:通过TCP仿真模型,获取位移模态数据,通过预处理,将位移模态数据转化为曲率模态数据。构建损伤位置识别任务和损伤程度识别任务,选取任一任务,按照一定比例,将处理后的数据分成训练集、验证集和测试集,用于模型训练和评估。
(2)RABCN模型构建与训练:使用划分好的训练集数据对RABCN模型进行训练,验证集数据对模型进行评估,并根据评估结果对模型进行调优,以提高预测准确率和稳定性。
(3)损伤识别与定位:使用测试集对训练好的RABCN模型进行评估,实现TCP损伤位置和损伤程度预测。
应用数值仿真软件Abaqus建立TCP模型进行模态分析,其结构参数和材料参数如表12所示。
管道两端分别创建参考点RP-1和RP-1与两端面建立6个自由度方向的耦合约束,管道两端简支。采用C3D8R单元对TCP进行网格划分,沿管道轴向布置51个等分节点,划分50个单元,离散后的网格和沿管道轴向的节点编号如图6所示。
采用共用节点的方法模拟复合材料层间粘接,脱层单元处不共用节点,脱层位置分别设置为第2个节点至第50个节点。通过有限元模态分析可以提取含脱层损伤TCP结构的固有频率和模态振型。为解释数据提取流程和处理方法,选择具有不同脱层位置和脱层层数的8种典型预设损伤工况进行对比分析,如表3所示。
通过有限元模态分析可以得到TCP的固有频率和振型参数,由于在工程分析中高阶模态的频率测量和提取较为困难,仅提取前3阶位移模态并进行归一化处理,以工况1为例,不同节点的前3阶位移模态数据如图7所示。
图7可知,发生脱层损伤和未发生脱层损伤节点的各阶位移模态曲线光滑,也就是说刚度变化对位移模态影响较小,脱层损伤对结构振型参数不敏感。
为提高损伤识别的稳定性,将位移模态数据处理成结构曲率模态数据,作为结构的损伤识别数据。将不同工况下的2阶模态位移参数代入公式(2),利用中心差分法将位移模态转换为曲率模态,得到管道各节点的曲率模态数据,为探讨曲率模态对TCP损伤程度识别的敏感性,计算工况1~4不同损伤程度的管道曲率模态进行对比,如图8所示。
图8可知,工况1~4在节点15附近出现显著的曲率模态突变峰,其突变极值位置与表3预设损伤坐标完全吻合。同时,管道曲率模态会在损伤区域与未损伤区域表现出显著差异:在完好区域呈现连续平滑特征,而损伤位置则出现明显的模态突变现象,且随着损伤层数的增加,损伤位置处的弯曲刚度减小,模态突变幅值显著增大。这一特征表明曲率模态对损伤程度识别具有良好的敏感性。
为对比曲率模态对TCP损伤位置识别的敏感性,对计算工况5~8不同损伤位置的管道曲率模态进行对比,如图9所示。
图9可知,在工况5~8中,预设损伤节点位置均呈现出显著的曲率模态突变峰特征,而未损伤区域则保持平滑的模态分布特性。值得注意的是,损伤位置与曲率模态突变峰在空间分布上具有一一对应关系。该现象充分验证了曲率模态参数对复合材料脱层损伤位置的表征灵敏度,表明曲率模态对损伤位置的识别同样具有良好的敏感性,为脱层损伤的定位检测提供了有效依据。
综上所述,使用管道曲率模态参数作为管道的损伤识别参数是可行的。曲率模态数据不仅能通过特征峰突变准确定位TCP的脱层损伤位置,还能有效区分损伤程度,实现TCP结构损伤位置和损伤程度的有效评估。
以此为基础,本文共构建261组TCP损伤工况作为网络的输入参数展开研究。为提高前后处理效率,基于Python语言编写参数化脚本进行有限元模型创建和结果数据提取。其中,训练样本245组,分别对应不同损伤位置和损伤程度下的管道参数;测试样本16组,损伤位置和损伤程度随机生成。部分TCP曲率模态数据如表4所示。
为了定量评估模型的性能,本文采用4个评价指标,分别是平均绝对误差MAE(Mean Absolute Error)、平均绝对百分比误差MAPE(Mean Absolute Percentage Error)、均方误差MSE(Mean Squared Error)和决定系数R2来评价预测值与真实值之间的误差。其中,MAE是预测值与真实值之差的绝对值,然后取平均值,反映了误差平均值的大小。计算公式如下
$ \text{MAE}=\frac{1}{N}\sum \limits_{i=1}^{N}\left| {y}_{i}-{\hat{y}}_{i}\right| $
其中,$ {y}_{i} $$ {\hat{y}}_{i} $分别代表预测值和真实值,N表示样本的数量。
MAPE是相对误差的平均值,它按比例反映了误差的大小。计算公式如下
$ \text{MAPE}=\frac{1}{N}\sum \limits_{i=1}^{N}\left| \frac{{\hat{y}}_{i}-{y}_{i}}{{y}_{i}}\right| $
MSE是每个误差平方和的平均值,用来衡量真实值与预测值之间的差值。计算公式如下
$ \text{MSE}=\frac{1}{N}\sum \limits_{i=1}^{N}{\left({y}_{i}-{\hat{y}}_{i}\right)}^{2} $
决定系数R2用于评估回归模型与数据的拟合程度。它测量的是因变量中可从自变量中预测的方差的比例。R2越高表示拟合越好。计算公式如下
$ {R}^{2}=1-\dfrac{\dfrac{1}{N}\displaystyle\sum \limits_{i=1}^{N}{\left({y}_{i}-{\hat{y}}_{i}\right)}^{2}}{\dfrac{1}{N}\displaystyle\sum \limits_{i=1}^{N}{\left({y}_{i}-\overline{y}\right)}^{2}} $
其中,$ \overline{y} $表示所有样本的平均值。
RABCN模型由Adam优化器训练,学习率为0.001,批大小为64。图10显示了RABCN模型的训练迭代过程。由于损伤识别包括损伤程度识别任务和损伤位置识别任务,因此RABCN模型有两个损失函数。在损伤位置识别的训练迭代过程中,RABCN模型在90次迭代时收敛。在损伤程度识别的训练迭代过程中,当迭代次数在30次左右时,RABCN模型趋于稳定,这意味着模型收敛速度较快,能够快速识别TCP脱层损伤的损伤位置和损伤程度。
表5显示了RABCN模型和ResNet模型的识别性能结果对比,可以发现,在损伤位置识别任务中,本文提出的RABCN模型在4项指标中均达到最优。在损伤程度识别任务中,RABCN模型的预测效果较好,MAE、MAPE、MSE和R2分别达到0.01719、0.03732、0.0004和0.9924。明显地,在TCP的脱层损伤识别中,本文提出的RABCN模型识别性能优于ResNet模型[18],这是因为提出的RABCN模型中的布朗协方差模块利用欧几里得度量函数捕获曲率模态与损伤位置、曲率模态与损伤程度之间的非线性关系,从而有效提取隐藏的损伤特征,进而实现良好的识别效果。
表6图11显示了布朗协方差网络损伤识别结果,其中mtnt分别为实际损伤的位置和程度,mpnp分别为预测损伤的位置和程度。无脱层损伤管道显示损伤程度为0,所有层均脱层显示损伤程度为1。根据误差率ER(Error Rate)对预测结果进行统计,计算公式为
$ \text{ER=}\frac{{y}_{i}-{\widehat{y}}_{i}}{{y}_{i}} $
16组检测样本中,损伤位置预测准确率达100%,损伤程度误差率均低于6%(最大值为6%,最小值为1%),其中6个检测样本的损伤程度误差率低于3%。本文提出的RABCN模型在结构损伤检测中展现出显著优势,不仅能准确识别损伤位置,还能对不同脱层损伤程度实现定量化判别,验证了该方法在复杂损伤场景下的鲁棒性与可靠性。
本文针对热塑性复合材料管道的脱层损伤检测展开研究,研究形成的“特征参数–深度学习”诊断体系,提升了复合材料管道损伤检测的工程适用性,获得结论如下:
(1)基于曲率模态的损伤识别方法发现,在损伤单元节点处曲率模态存在显著突变。该参数不仅可实现脱层损伤的精准定位,还能通过突变幅值与损伤面积的正相关特性实现损伤程度的量化,为TCP损伤识别提供了高灵敏度指标。
(2)构建基于残差注意力布朗协方差网络的智能识别模型,其在损伤位置识别中实现全样本准确判别,准确率达到100%,损伤程度的识别误差率低于6%,其优异的非线性特征提取能力可为海洋工程结构健康监测提供参考。

参考文献 引证文献
排序方式:
1
彭传远, 付新钰, 师梦科, . 大口径热塑性复合材料管及接头内压失效分析[J]. 工程塑料应用, 2022, 50(7): 104‒109.
Peng C Y, Fu X Y, Shi M K, et al. Failure analysis of internal pressure of large-diameter thermoplastic composite pipes and joints[J]. Application of Engineering Plastics, 2022, 50(7): 104‒109. (in Chinese)
2
潘静雯, 袁浩凡, 张芝芳. 基于频率识别纤维增强复合材料弧形板分层损伤[J]. 噪声与振动控制, 2017, 37(6): 180‒185.
Pan J W, Yuan H F, Zhang Z F. Hierarchical damage identification of fiber-reinforced composite arc-shaped plates based on frequency[J]. Noise and Vibration Control, 2017, 37(6): 180‒185. (in Chinese)
3
梁智洪, 詹 超, 张芝芳. 基于频率识别纤维增强树脂复合材料加筋板的分层损伤[J]. 复合材料学报, 2019, 36(11): 2614‒2627.
Liang Z L, Zhan C, Zhang Z F. Delamination damage of fiber-reinforced resin composite reinforced plates based on frequency identification[J]. Journal of Composite Materials, 2019, 36(11): 2614‒2627. (in Chinese)
4
刘 雨, 夏张松, 宫兆斌, . 复合材料修理区域的超声检测[J]. 无损检测, 2024, 46(10): 81‒87.
Liu Y, Xia Z S, Gong Z B, et al. Ultrasonic testing of the composite material repair area[J]. Nondestructive Testing, 2024, 46(10): 81‒87. (in Chinese)
5
王克凡. 基于电容成像的复合材料无损检测技术研究[D]. 青岛: 中国石油大学(华东), 2020.
Wang K F. Research on non-destructive testing technology of composite materials based on capacitance imaging technique[D]. Qingdao: China University of Petroleum, 2020. (in Chinese)
6
Sofer M, Jakub C, Martin F, et al. Damage analysis of composite CFRP tubes using acoustic emission monitoring and pattern recognition approach[J]. Materials, 2021, 14(4): 786.
7
Scholz V, Winkler P, Andreas H, et al. Structural damage identification of composite rotors based on fully connected neural networks and convolutional neural networks[J]. Sensors, 2021, 21(6): 2005.
8
吴 磊, 梅江涛, 赵 硕. 基于IPSO-BP神经网络的管道损伤检测方法[J]. 实验室研究与探索, 2022, 41(7): 44‒49.
Wu L, Mei J T, Zhao S. Pipeline damage detection method based on IPSO-BP neural network[J]. Laboratory Research and Exploration, 2022, 41(7): 44‒49. (in Chinese)
9
Wu L, Mei J T, Zhao S, et al. Pipeline damage identification based on an optimized back-propagation neural network improved by whale optimization algorithm[J]. Applied Intelligence, 2023, 53(10): 12937‒12954.
10
Mei J T, Wu L, Chen E Q, et al. A novel structural damage detection method using a hybrid IDE-BP model[J]. Knowledge-based Systems, 2023, 273: 110606.
11
Zhao J Y, Xie W H, Yu D, et al. Deep transfer learning approach for localization of damage area in composite laminates using acoustic emission signal[J]. Polymers, 2023, 15(6): 1520.
12
Liu K, Yan R J, Guedes S, et al. Damage identification in offshore jacket structures based on modal flexibility[J]. Ocean Engineering, 2018, 170: 171‒185.
13
杜 宇, 杨 涛, 何梅洪. 基于运行模态的复合材料梁脱层损伤识别[J]. 宇航材料工艺, 2021, 51(2): 81‒86.
Du Y, Yang T, He M H. Delaminate damage identification of composite beam based on operational modal[J]. Aerospace Materials & Technology, 2021, 51(2): 81‒86. (in Chinese)
14
杜 宇, 何梅洪, 李菊峰. 基于曲率模态变化率的复合材料梁脱层损伤识别[J]. 固体火箭技术, 2021, 44(3): 385‒389.
Du Y, He M L, Li J F. Delamination damage diagnosis of composite beams based on curvature mode change rate[J]. Solid Rocket Technology, 2021, 44(3): 385‒389. (in Chinese)
15
Zhang Y, Guo J, Xie Y G, et al. Warship damage identification using mode curvature shapes method[J]. Applied Ocean Research, 2022, 129: 103396.
16
Zhou F, Cicone A, Zhou H M, et al. An enhanced Iterative Residual Convolutional Neural Network for non-stationary signal decomposition[J]. Pattern Recognition Letters, 2025: 328‒336.
17
苟光磊, 朱东华, 李小菲, . 深度掩膜布朗距离协方差小样本分类方法[J]. 计算机应用研究, 2024, 41(7): 2229‒2234.
Gou G L, Zhu D H, Li X F, et al. Deep mask Brownian distance covariance for few-shot classification[J]. Application Research of Computers, 2024, 41(7): 2229‒2234. (in Chinese)
18
He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[J]. Conference on Computer Vision and Pattern Recognition, 2016: 770‒778.
2026年第30卷第5期
PDF下载
170
89
引用本文
BibTeX
文章信息
doi: 10.3969/j.issn.1007-7294.2026.05.010
  • 接收时间:2025-09-11
  • 首发时间:2026-07-07
  • 出版时间:2026-05-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-09-11
基金
作者信息
    1.青岛黄海学院 智能制造学院,山东 青岛 266427
    2.华中科技大学 船舶与海洋工程学院,武汉 430074

通讯作者:

舒启明(1997–),男,博士研究生,通讯作者,E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/cblx/CN/10.3969/j.issn.1007-7294.2026.05.010
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