Unsupervised machine and deep learning methods for structural damage detection: A comparative study

Abstract While many structural damage detection methods have been developed in recent decades, few data‐driven methods in unsupervised learning mode have been developed to solve the practical difficulties in data acquisition for civil infrastructures in different scenarios. To address such a challen...

Full description

Saved in:
Bibliographic Details
Main Authors: Zilong Wang, Young‐Jin Cha
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:Engineering Reports
Subjects:
Online Access:https://doi.org/10.1002/eng2.12551
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832576598612639744
author Zilong Wang
Young‐Jin Cha
author_facet Zilong Wang
Young‐Jin Cha
author_sort Zilong Wang
collection DOAJ
description Abstract While many structural damage detection methods have been developed in recent decades, few data‐driven methods in unsupervised learning mode have been developed to solve the practical difficulties in data acquisition for civil infrastructures in different scenarios. To address such a challenge, this article proposes a number of improved unsupervised novelty detection methods and conducts extensive comparative studies on a laboratory scale steel bridge to examine their performances of damage detection. The key concept behind unsupervised novelty detection in this article is that only normal data from undamaged/baseline structural scenarios are required to train statistical models with these methods. Then, these trained models are used to identify abnormal testing data from damaged scenarios. To detect structural damage in the form of loosening bolts in the steel bridge, four machine‐learning methods (i.e., K‐nearest neighbors method, Gaussian mixture models, one‐class support vector machines, density peaks‐based fast clustering method) and one deep learning method using a deep auto‐encoder are selected. Meanwhile, some modifications and improvements are made to enable these methods to detect structural damage in unsupervised novelty detection mode. In their comparative studies, the advantages and disadvantages of these methods are analyzed based on their results of structural damage detection.
format Article
id doaj-art-189a1b03946e4ea088118bdbba174ded
institution Kabale University
issn 2577-8196
language English
publishDate 2025-01-01
publisher Wiley
record_format Article
series Engineering Reports
spelling doaj-art-189a1b03946e4ea088118bdbba174ded2025-01-31T00:22:48ZengWileyEngineering Reports2577-81962025-01-0171n/an/a10.1002/eng2.12551Unsupervised machine and deep learning methods for structural damage detection: A comparative studyZilong Wang0Young‐Jin Cha1Suzhou Institute of Building Science Group Suzhou Jiangsu ChinaDepartment of Civil Engineering University of Manitoba Winnipeg Manitoba CanadaAbstract While many structural damage detection methods have been developed in recent decades, few data‐driven methods in unsupervised learning mode have been developed to solve the practical difficulties in data acquisition for civil infrastructures in different scenarios. To address such a challenge, this article proposes a number of improved unsupervised novelty detection methods and conducts extensive comparative studies on a laboratory scale steel bridge to examine their performances of damage detection. The key concept behind unsupervised novelty detection in this article is that only normal data from undamaged/baseline structural scenarios are required to train statistical models with these methods. Then, these trained models are used to identify abnormal testing data from damaged scenarios. To detect structural damage in the form of loosening bolts in the steel bridge, four machine‐learning methods (i.e., K‐nearest neighbors method, Gaussian mixture models, one‐class support vector machines, density peaks‐based fast clustering method) and one deep learning method using a deep auto‐encoder are selected. Meanwhile, some modifications and improvements are made to enable these methods to detect structural damage in unsupervised novelty detection mode. In their comparative studies, the advantages and disadvantages of these methods are analyzed based on their results of structural damage detection.https://doi.org/10.1002/eng2.12551deep learningfast clusteringmachine‐learningstructural damage detectionunsupervised novelty detection
spellingShingle Zilong Wang
Young‐Jin Cha
Unsupervised machine and deep learning methods for structural damage detection: A comparative study
Engineering Reports
deep learning
fast clustering
machine‐learning
structural damage detection
unsupervised novelty detection
title Unsupervised machine and deep learning methods for structural damage detection: A comparative study
title_full Unsupervised machine and deep learning methods for structural damage detection: A comparative study
title_fullStr Unsupervised machine and deep learning methods for structural damage detection: A comparative study
title_full_unstemmed Unsupervised machine and deep learning methods for structural damage detection: A comparative study
title_short Unsupervised machine and deep learning methods for structural damage detection: A comparative study
title_sort unsupervised machine and deep learning methods for structural damage detection a comparative study
topic deep learning
fast clustering
machine‐learning
structural damage detection
unsupervised novelty detection
url https://doi.org/10.1002/eng2.12551
work_keys_str_mv AT zilongwang unsupervisedmachineanddeeplearningmethodsforstructuraldamagedetectionacomparativestudy
AT youngjincha unsupervisedmachineanddeeplearningmethodsforstructuraldamagedetectionacomparativestudy