Population-Based SHM Under Environmental Variability Using a Classifier for Unsupervised Damage Detection

Onderzoeksoutput: Conference paperResearch

30 Downloads (Pure)

Samenvatting

In this paper, we introduce a novel deep learning technique for anomaly detection in the context of Population-Based Structural Health Monitoring (PB-SHM). The proposed method eliminates manual feature engineering by utilizing Power Spectral Density (PSD) as input, allowing examination of the entire spectrum. It is based on an auxiliary classification task, wherein the model is trained to discriminate between different systems according to their dynamic response. The classifier confidence is then used during inference for damage detection. The neural network extracts discriminative features commonly impacted by damage, which are employed to create a normality model. The efficacy of our method is demonstrated on a simulated population of 20 individual 8-
DOF systems influenced by a latent environmental variables, emphasizing its potential for PB-SHM under diverse conditions. Our technique achieves performance comparable to resonance frequency-based methods while potentially exhibiting higher capability in complex structures with multiple modes. Anomalies caused by a 5% decrease in stiffness are successfully detected, yielding an AUC of 0.94.
Originele taal-2English
TitelSTRUCTURAL HEALTH MONITORING 2023
RedacteurenSaman Farhangdoust, Alfredo Guemes, Fu-Kuo Chang
UitgeverijDEStech Publications, Inc
HoofdstukSPECIAL SESSION: ARTIFICIAL INTELLIGENCE FOR SHM: MACHINE LEARNING APPROACHES
Pagina's1479-1488
Aantal pagina's10
ISBN van elektronische versie978-1-60595-693-0
StatusPublished - 14 sep 2023
Evenement14th International Workshop on Structural Health Monitoring : IWSHM - Stanford University, Stanford, CA, United States
Duur: 12 sep 202314 sep 2023
https://iwshm2023.stanford.edu/about-iwshm-2023

Publicatie series

NaamStructural Health Monitoring 2023: Designing SHM for Sustainability, Maintainability, and Reliability - Proceedings of the 14th International Workshop on Structural Health Monitoring

Conference

Conference14th International Workshop on Structural Health Monitoring
Verkorte titelIWSHM 2023
Land/RegioUnited States
StadStanford, CA
Periode12/09/2314/09/23
Internet adres

Bibliografische nota

Publisher Copyright:
© 2023 by DEStech Publi cations, Inc. All rights reserved

Vingerafdruk

Duik in de onderzoeksthema's van 'Population-Based SHM Under Environmental Variability Using a Classifier for Unsupervised Damage Detection'. Samen vormen ze een unieke vingerafdruk.

Citeer dit