Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting
Bridges are built to last more than 100 years, spanning many human generations. Throughout their lifetime, their service requirements may change, or they age and often suffer a material degradation process that can lead to the need of retrofitting. In bridge engineering, retrofitting refers to the s...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-09-01
|
Series: | Buildings |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-5309/13/9/2323 |
_version_ | 1797580928329973760 |
---|---|
author | Marcus Omori Yano Eloi Figueiredo Samuel da Silva Alexandre Cury Ionut Moldovan |
author_facet | Marcus Omori Yano Eloi Figueiredo Samuel da Silva Alexandre Cury Ionut Moldovan |
author_sort | Marcus Omori Yano |
collection | DOAJ |
description | Bridges are built to last more than 100 years, spanning many human generations. Throughout their lifetime, their service requirements may change, or they age and often suffer a material degradation process that can lead to the need of retrofitting. In bridge engineering, retrofitting refers to the strengthening of existing structures to make them more resistant and to increase the lifespan of bridges. Retrofitting normally increases the stiffness of bridge components, which can cause significant changes in the global modal properties. In the context of structural health monitoring, a classifier trained with datasets before retrofitting will most likely output many outliers after retrofitting, based on the premise that the new observations do not share the same underlying distribution. Therefore, how can long-term monitoring data from one bridge (labeled source domain) be reused to create a classifier that generalizes to the same bridge after retrofitting (unlabeled target domain)? This paper presents a novel approach based on transfer learning in the context of domain adaptation on datasets from two real bridges subjected to retrofit and under-monitoring programs. Based on the assumption that both bridges are undamaged before retrofitting, the results show that transfer learning can support the long-term damage detection process based on a classification using an outlier detection strategy. |
first_indexed | 2024-03-10T22:58:04Z |
format | Article |
id | doaj.art-33648137a4c04d02a81d2756220d4176 |
institution | Directory Open Access Journal |
issn | 2075-5309 |
language | English |
last_indexed | 2024-03-10T22:58:04Z |
publishDate | 2023-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Buildings |
spelling | doaj.art-33648137a4c04d02a81d2756220d41762023-11-19T09:52:28ZengMDPI AGBuildings2075-53092023-09-01139232310.3390/buildings13092323Transfer Learning for Structural Health Monitoring in Bridges That Underwent RetrofittingMarcus Omori Yano0Eloi Figueiredo1Samuel da Silva2Alexandre Cury3Ionut Moldovan4Department of Mechanical Engineering, UNESP—Universidade Estadual Paulista, Ilha Solteira 15385-000, BrazilFaculty of Engineering, Lusófona University, 1749-024 Lisbon, PortugalDepartment of Mechanical Engineering, UNESP—Universidade Estadual Paulista, Ilha Solteira 15385-000, BrazilGraduate Program in Civil Engineering, Federal University of Juiz de Fora, Juiz de Fora 36036-900, BrazilFaculty of Engineering, Lusófona University, 1749-024 Lisbon, PortugalBridges are built to last more than 100 years, spanning many human generations. Throughout their lifetime, their service requirements may change, or they age and often suffer a material degradation process that can lead to the need of retrofitting. In bridge engineering, retrofitting refers to the strengthening of existing structures to make them more resistant and to increase the lifespan of bridges. Retrofitting normally increases the stiffness of bridge components, which can cause significant changes in the global modal properties. In the context of structural health monitoring, a classifier trained with datasets before retrofitting will most likely output many outliers after retrofitting, based on the premise that the new observations do not share the same underlying distribution. Therefore, how can long-term monitoring data from one bridge (labeled source domain) be reused to create a classifier that generalizes to the same bridge after retrofitting (unlabeled target domain)? This paper presents a novel approach based on transfer learning in the context of domain adaptation on datasets from two real bridges subjected to retrofit and under-monitoring programs. Based on the assumption that both bridges are undamaged before retrofitting, the results show that transfer learning can support the long-term damage detection process based on a classification using an outlier detection strategy.https://www.mdpi.com/2075-5309/13/9/2323transfer learningstructural health monitoringjoint distribution adaptationdomain adaptationbridges |
spellingShingle | Marcus Omori Yano Eloi Figueiredo Samuel da Silva Alexandre Cury Ionut Moldovan Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting Buildings transfer learning structural health monitoring joint distribution adaptation domain adaptation bridges |
title | Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting |
title_full | Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting |
title_fullStr | Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting |
title_full_unstemmed | Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting |
title_short | Transfer Learning for Structural Health Monitoring in Bridges That Underwent Retrofitting |
title_sort | transfer learning for structural health monitoring in bridges that underwent retrofitting |
topic | transfer learning structural health monitoring joint distribution adaptation domain adaptation bridges |
url | https://www.mdpi.com/2075-5309/13/9/2323 |
work_keys_str_mv | AT marcusomoriyano transferlearningforstructuralhealthmonitoringinbridgesthatunderwentretrofitting AT eloifigueiredo transferlearningforstructuralhealthmonitoringinbridgesthatunderwentretrofitting AT samueldasilva transferlearningforstructuralhealthmonitoringinbridgesthatunderwentretrofitting AT alexandrecury transferlearningforstructuralhealthmonitoringinbridgesthatunderwentretrofitting AT ionutmoldovan transferlearningforstructuralhealthmonitoringinbridgesthatunderwentretrofitting |