Unsupervised transfer learning for structural health monitoring of urban pedestrian bridges

IF 3.6 2区 工程技术 Q1 ENGINEERING, CIVIL Journal of Civil Structural Health Monitoring Pub Date : 2024-04-14 DOI:10.1007/s13349-024-00786-w
Giulia Marasco, Ionut Moldovan, Eloi Figueiredo, Bernardino Chiaia
{"title":"Unsupervised transfer learning for structural health monitoring of urban pedestrian bridges","authors":"Giulia Marasco, Ionut Moldovan, Eloi Figueiredo, Bernardino Chiaia","doi":"10.1007/s13349-024-00786-w","DOIUrl":null,"url":null,"abstract":"<p>Bridge authorities have been reticent to integrate structural health monitoring into their bridge management systems, as they do not have the financial and technical resources to collect long-term monitoring data from every bridge. As bridge authorities normally own huge amount of similar bridges, like the pedestrian ones, the ability to transfer knowledge from one or a small group of well-known bridges to help make more effective decisions in new bridges and environments has gained relevance. In that sense, transfer learning, a subfield of machine learning, offers a novel solution to periodically evaluate the structural condition of all pedestrian bridges using long-term monitoring data from one or more pedestrian bridges. In this paper, the applicability of unsupervised transfer learning is firstly shown on data from numerical models and then on data from two similar pedestrian prestressed concrete bridges. Two domain adaptation techniques are used for transfer learning, where a classifier has access to unlabeled training data (source domain) from a reference bridge (or a small set of reference bridges) and unlabeled monitoring test data (target domain) from another bridge, assuming that both domains are from similar but statistically different distributions. This type of mapping is expected to improve the classification accuracy for the target domain compared to a procedure that does not implement domain adaptation, as a result of reducing distributions mismatch between source and target domains.</p>","PeriodicalId":48582,"journal":{"name":"Journal of Civil Structural Health Monitoring","volume":null,"pages":null},"PeriodicalIF":3.6000,"publicationDate":"2024-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Civil Structural Health Monitoring","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s13349-024-00786-w","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}
引用次数: 0

Abstract

Bridge authorities have been reticent to integrate structural health monitoring into their bridge management systems, as they do not have the financial and technical resources to collect long-term monitoring data from every bridge. As bridge authorities normally own huge amount of similar bridges, like the pedestrian ones, the ability to transfer knowledge from one or a small group of well-known bridges to help make more effective decisions in new bridges and environments has gained relevance. In that sense, transfer learning, a subfield of machine learning, offers a novel solution to periodically evaluate the structural condition of all pedestrian bridges using long-term monitoring data from one or more pedestrian bridges. In this paper, the applicability of unsupervised transfer learning is firstly shown on data from numerical models and then on data from two similar pedestrian prestressed concrete bridges. Two domain adaptation techniques are used for transfer learning, where a classifier has access to unlabeled training data (source domain) from a reference bridge (or a small set of reference bridges) and unlabeled monitoring test data (target domain) from another bridge, assuming that both domains are from similar but statistically different distributions. This type of mapping is expected to improve the classification accuracy for the target domain compared to a procedure that does not implement domain adaptation, as a result of reducing distributions mismatch between source and target domains.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于城市人行天桥结构健康监测的无监督迁移学习
桥梁管理部门一直不愿将结构健康监测纳入其桥梁管理系统,因为他们没有财力和技术资源来收集每座桥梁的长期监测数据。由于桥梁管理部门通常拥有大量类似的桥梁(如人行天桥),因此从一座或一小部分知名桥梁中迁移知识,以帮助在新桥梁和新环境中做出更有效决策的能力已变得越来越重要。从这个意义上说,机器学习的一个子领域--迁移学习提供了一种新颖的解决方案,即利用一座或多座人行天桥的长期监测数据,定期评估所有人行天桥的结构状况。本文首先在数值模型数据上展示了无监督迁移学习的适用性,然后在两座类似的预应力混凝土人行天桥数据上展示了无监督迁移学习的适用性。迁移学习使用了两种域适应技术,其中分类器可以访问来自参考桥梁(或一小组参考桥梁)的无标记训练数据(源域)和来自另一座桥梁的无标记监测测试数据(目标域),假设这两个域来自相似但统计上不同的分布。与不实施域适应的程序相比,这种映射方式可减少源域和目标域之间的分布不匹配,从而提高目标域的分类准确性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Journal of Civil Structural Health Monitoring
Journal of Civil Structural Health Monitoring Engineering-Safety, Risk, Reliability and Quality
CiteScore
8.10
自引率
11.40%
发文量
105
期刊介绍: The Journal of Civil Structural Health Monitoring (JCSHM) publishes articles to advance the understanding and the application of health monitoring methods for the condition assessment and management of civil infrastructure systems. JCSHM serves as a focal point for sharing knowledge and experience in technologies impacting the discipline of Civionics and Civil Structural Health Monitoring, especially in terms of load capacity ratings and service life estimation.
期刊最新文献
Development and implementation of medium-fidelity physics-based model for hybrid digital twin-based damage identification of piping structures Innovated bridge health diagnosis model using bridge critical frequency ratio R–C–C fusion classifier for automatic damage detection of heritage building using 3D laser scanning An AIoT system for real-time monitoring and forecasting of railway temperature Environmental effects on the experimental modal parameters of masonry buildings: experiences from the Italian Seismic Observatory of Structures (OSS) network
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1