A hybrid data-physics framework with conformal GNN for enhanced damage identification

[EN] Structural damage identification is crucial for ensuring safety, yet existing data-driven and physics-based methods often suffer from accuracy and computational limitations. To address these issues, we propose a hybrid framework that integrates Graph Neural Networks (GNNs) with a physics-based...

ver descrição completa

Detalhes bibliográficos
Autores: Eslamlou, Armin Dadras, Ghasemlou, Arshia, Riveiro, Belén, Barros-González, Brais|||0000-0001-7132-5951
Formato: artículo
Fecha de publicación:2025
País:España
Recursos:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/225949
Acesso em linha:https://riunet.upv.es/handle/10251/225949
Access Level:acceso abierto
Palavra-chave:Structural health monitoring
Graph neural networks
Modal data
Damage identification
AutoML
id ES_f63ff6fcfed501962862ecc55eb9eb6f
oai_identifier_str oai:riunet.upv.es:10251/225949
network_acronym_str ES
network_name_str España
repository_id_str
spelling A hybrid data-physics framework with conformal GNN for enhanced damage identificationEslamlou, Armin DadrasGhasemlou, ArshiaRiveiro, BelénBarros-González, Brais|||0000-0001-7132-5951Structural health monitoringGraph neural networksModal dataDamage identificationAutoML[EN] Structural damage identification is crucial for ensuring safety, yet existing data-driven and physics-based methods often suffer from accuracy and computational limitations. To address these issues, we propose a hybrid framework that integrates Graph Neural Networks (GNNs) with a physics-based Finite Element (FE) model updating approach. The first module employs a GNN trained on modal data from FE simulations to estimate the location and severity of structural damage, with an evolutionary AutoML framework optimizing the GNN's architecture and hyperparameters. In the second module, a conformal prediction technique quantifies uncertainty in the GNN's predictions, ensuring robust confidence bounds for damage estimations. These uncertainty-aware predictions initialize a warm-started FE model updating workflow, where the Water Strider Algorithm (WSA) efficiently minimizes a cost function based on limited modal data. The proposed methodology has been validated on benchmark structures, including the Louisville bridge, IASC-ASCE building and a dome structure, demonstrating a remarkable increase in damage identification accuracy compared to conventional approaches. Unlike pure data-driven and physics-based methods, this hybrid framework leverages their strengths while integrating uncertainty quantification, enhancing their efficiency. This hybrid approach is scalable to various structural configurations, making it a promising solution for enhanced structural health monitoring.The authors wish to acknowledge the grant awarded for the Pont3 project (ref: PID2021-124236OB-C33) funded by MCIN/AEI/10.13039/501100011033 and "ERDF A Way of Making Europe."ElsevierInstituto Universitario de Investigación de Ciencia y Tecnología del HormigónAgencia Estatal de InvestigaciónEuropean Regional Development FundRepositorio Institucional de la Universitat Politècnica de València Riunet20252025-11-01journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://riunet.upv.es/handle/10251/225949reponame:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valénciainstname:Universitat Politècnica de València (UPV)InglésengAgencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 PID2021-124236OB-C33 ENFOQUE INTERDISCIPLINAR EFICIENTE PARA ANTICIPAR LA PROPAGACION DE FALLOS EN PUENTES QUE SOBREPASAN SU VIDA UTIL: COMPUTACION SURROGADA Y BASADA EN DATOSopen accesshttp://purl.org/coar/access_right/c_abf2Reconocimiento - No comercial - Sin obra derivada (by-nc-nd) http://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessoai:riunet.upv.es:10251/2259492026-06-13T07:49:27Z
dc.title.none.fl_str_mv A hybrid data-physics framework with conformal GNN for enhanced damage identification
title A hybrid data-physics framework with conformal GNN for enhanced damage identification
spellingShingle A hybrid data-physics framework with conformal GNN for enhanced damage identification
Eslamlou, Armin Dadras
Structural health monitoring
Graph neural networks
Modal data
Damage identification
AutoML
title_short A hybrid data-physics framework with conformal GNN for enhanced damage identification
title_full A hybrid data-physics framework with conformal GNN for enhanced damage identification
title_fullStr A hybrid data-physics framework with conformal GNN for enhanced damage identification
title_full_unstemmed A hybrid data-physics framework with conformal GNN for enhanced damage identification
title_sort A hybrid data-physics framework with conformal GNN for enhanced damage identification
dc.creator.none.fl_str_mv Eslamlou, Armin Dadras
Ghasemlou, Arshia
Riveiro, Belén
Barros-González, Brais|||0000-0001-7132-5951
author Eslamlou, Armin Dadras
author_facet Eslamlou, Armin Dadras
Ghasemlou, Arshia
Riveiro, Belén
Barros-González, Brais|||0000-0001-7132-5951
author_role author
author2 Ghasemlou, Arshia
Riveiro, Belén
Barros-González, Brais|||0000-0001-7132-5951
author2_role author
author
author
dc.contributor.none.fl_str_mv Instituto Universitario de Investigación de Ciencia y Tecnología del Hormigón
Agencia Estatal de Investigación
European Regional Development Fund
Repositorio Institucional de la Universitat Politècnica de València Riunet
dc.subject.none.fl_str_mv Structural health monitoring
Graph neural networks
Modal data
Damage identification
AutoML
topic Structural health monitoring
Graph neural networks
Modal data
Damage identification
AutoML
description [EN] Structural damage identification is crucial for ensuring safety, yet existing data-driven and physics-based methods often suffer from accuracy and computational limitations. To address these issues, we propose a hybrid framework that integrates Graph Neural Networks (GNNs) with a physics-based Finite Element (FE) model updating approach. The first module employs a GNN trained on modal data from FE simulations to estimate the location and severity of structural damage, with an evolutionary AutoML framework optimizing the GNN's architecture and hyperparameters. In the second module, a conformal prediction technique quantifies uncertainty in the GNN's predictions, ensuring robust confidence bounds for damage estimations. These uncertainty-aware predictions initialize a warm-started FE model updating workflow, where the Water Strider Algorithm (WSA) efficiently minimizes a cost function based on limited modal data. The proposed methodology has been validated on benchmark structures, including the Louisville bridge, IASC-ASCE building and a dome structure, demonstrating a remarkable increase in damage identification accuracy compared to conventional approaches. Unlike pure data-driven and physics-based methods, this hybrid framework leverages their strengths while integrating uncertainty quantification, enhancing their efficiency. This hybrid approach is scalable to various structural configurations, making it a promising solution for enhanced structural health monitoring.
publishDate 2025
dc.date.none.fl_str_mv 2025
2025-11-01
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://riunet.upv.es/handle/10251/225949
url https://riunet.upv.es/handle/10251/225949
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.relation.none.fl_str_mv Agencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 PID2021-124236OB-C33 ENFOQUE INTERDISCIPLINAR EFICIENTE PARA ANTICIPAR LA PROPAGACION DE FALLOS EN PUENTES QUE SOBREPASAN SU VIDA UTIL: COMPUTACION SURROGADA Y BASADA EN DATOS
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
Reconocimiento - No comercial - Sin obra derivada (by-nc-nd)
http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
Reconocimiento - No comercial - Sin obra derivada (by-nc-nd)
http://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
instname:Universitat Politècnica de València (UPV)
instname_str Universitat Politècnica de València (UPV)
reponame_str RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
collection RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869424718901149697
score 15,812429

Registos relacionados