Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring

Detalhes bibliográficos
Autor(a) principal: Silva, Samuel da [UNESP]
Data de Publicação: 2021
Outros Autores: Yano, Marcus Omori [UNESP], Gonsalez-Bueno, Camila Gianini [UNESP]
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://dx.doi.org/10.1007/s10921-021-00794-6
http://hdl.handle.net/11449/221928
Resumo: The effects of temperature fluctuations in the impedance measurements’ spectral estimates confuse the procedures to distinguish actual states’ classification, demanding compensation. The present paper demonstrates a new method to achieve temperature compensation based on a Transfer Component Analysis (TCA), a subtype of transfer learning, of the features from a source domain (in a well-known labeled condition) to another target domain (in an unknown condition). This procedure assumes only the labeled features data in the healthy condition (baseline) and damaged state in a specific known temperature as source data. The features computed are the Root Mean Square Deviation (RMSD) indices of the real and imaginary impedance signals. A machine-learning algorithm based on Mahalanobis squared distance (D2) is trained using the features computed from the baseline condition in the reference temperature. Also, the other temperature and structural conditions data are assumed as testing data of the target condition. TCA’s main idea is mapping the features from the original features space to a new subspace where the detection becomes possible using the same training data in the source domain. The results performed in a testbench with a piezoelectric element (PZT) bonded under a set of temperatures monitored, and simulated damage confirmed that the proposed method could recognize the real states correctly by transferring the knowledge from the features of the source domain into the target domain, assuming different temperatures.
id UNSP_d1060f5de7b2dd9d45f9886bc8d6f2e0
oai_identifier_str oai:repositorio.unesp.br:11449/221928
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health MonitoringDomain adaptationElectromechanical impedanceTemperature effectsTransfer component analysisThe effects of temperature fluctuations in the impedance measurements’ spectral estimates confuse the procedures to distinguish actual states’ classification, demanding compensation. The present paper demonstrates a new method to achieve temperature compensation based on a Transfer Component Analysis (TCA), a subtype of transfer learning, of the features from a source domain (in a well-known labeled condition) to another target domain (in an unknown condition). This procedure assumes only the labeled features data in the healthy condition (baseline) and damaged state in a specific known temperature as source data. The features computed are the Root Mean Square Deviation (RMSD) indices of the real and imaginary impedance signals. A machine-learning algorithm based on Mahalanobis squared distance (D2) is trained using the features computed from the baseline condition in the reference temperature. Also, the other temperature and structural conditions data are assumed as testing data of the target condition. TCA’s main idea is mapping the features from the original features space to a new subspace where the detection becomes possible using the same training data in the source domain. The results performed in a testbench with a piezoelectric element (PZT) bonded under a set of temperatures monitored, and simulated damage confirmed that the proposed method could recognize the real states correctly by transferring the knowledge from the features of the source domain into the target domain, assuming different temperatures.Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)Department of Mechanical Engineering São Paulo State University (UNESP), Avenida Brasil 56Department of Mechanical Engineering São Paulo State University (UNESP), Avenida Brasil 56CNPq: 306526/2019-0CAPES: 88882.433643/2019-01Universidade Estadual Paulista (UNESP)Silva, Samuel da [UNESP]Yano, Marcus Omori [UNESP]Gonsalez-Bueno, Camila Gianini [UNESP]2022-04-28T19:41:26Z2022-04-28T19:41:26Z2021-09-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1007/s10921-021-00794-6Journal of Nondestructive Evaluation, v. 40, n. 3, 2021.1573-48620195-9298http://hdl.handle.net/11449/22192810.1007/s10921-021-00794-62-s2.0-85109590371Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengJournal of Nondestructive Evaluationinfo:eu-repo/semantics/openAccess2022-04-28T19:41:26Zoai:repositorio.unesp.br:11449/221928Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462022-04-28T19:41:26Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
title Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
spellingShingle Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
Silva, Samuel da [UNESP]
Domain adaptation
Electromechanical impedance
Temperature effects
Transfer component analysis
title_short Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
title_full Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
title_fullStr Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
title_full_unstemmed Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
title_sort Transfer Component Analysis for Compensation of Temperature Effects on the Impedance-Based Structural Health Monitoring
author Silva, Samuel da [UNESP]
author_facet Silva, Samuel da [UNESP]
Yano, Marcus Omori [UNESP]
Gonsalez-Bueno, Camila Gianini [UNESP]
author_role author
author2 Yano, Marcus Omori [UNESP]
Gonsalez-Bueno, Camila Gianini [UNESP]
author2_role author
author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (UNESP)
dc.contributor.author.fl_str_mv Silva, Samuel da [UNESP]
Yano, Marcus Omori [UNESP]
Gonsalez-Bueno, Camila Gianini [UNESP]
dc.subject.por.fl_str_mv Domain adaptation
Electromechanical impedance
Temperature effects
Transfer component analysis
topic Domain adaptation
Electromechanical impedance
Temperature effects
Transfer component analysis
description The effects of temperature fluctuations in the impedance measurements’ spectral estimates confuse the procedures to distinguish actual states’ classification, demanding compensation. The present paper demonstrates a new method to achieve temperature compensation based on a Transfer Component Analysis (TCA), a subtype of transfer learning, of the features from a source domain (in a well-known labeled condition) to another target domain (in an unknown condition). This procedure assumes only the labeled features data in the healthy condition (baseline) and damaged state in a specific known temperature as source data. The features computed are the Root Mean Square Deviation (RMSD) indices of the real and imaginary impedance signals. A machine-learning algorithm based on Mahalanobis squared distance (D2) is trained using the features computed from the baseline condition in the reference temperature. Also, the other temperature and structural conditions data are assumed as testing data of the target condition. TCA’s main idea is mapping the features from the original features space to a new subspace where the detection becomes possible using the same training data in the source domain. The results performed in a testbench with a piezoelectric element (PZT) bonded under a set of temperatures monitored, and simulated damage confirmed that the proposed method could recognize the real states correctly by transferring the knowledge from the features of the source domain into the target domain, assuming different temperatures.
publishDate 2021
dc.date.none.fl_str_mv 2021-09-01
2022-04-28T19:41:26Z
2022-04-28T19:41:26Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://dx.doi.org/10.1007/s10921-021-00794-6
Journal of Nondestructive Evaluation, v. 40, n. 3, 2021.
1573-4862
0195-9298
http://hdl.handle.net/11449/221928
10.1007/s10921-021-00794-6
2-s2.0-85109590371
url http://dx.doi.org/10.1007/s10921-021-00794-6
http://hdl.handle.net/11449/221928
identifier_str_mv Journal of Nondestructive Evaluation, v. 40, n. 3, 2021.
1573-4862
0195-9298
10.1007/s10921-021-00794-6
2-s2.0-85109590371
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Journal of Nondestructive Evaluation
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.source.none.fl_str_mv Scopus
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1797790257642471424