Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring

This thesis advances the field of anomaly detection in multivariate time series by addressing key challenges in anomaly detection, localization, and severity assessment. Through the development of EdgeConvFormer and U-GraphFormer, this research offers robust, interpretable, and efficient solutions a...

Full description

Bibliographic Details
Main Author: Liu, Jie
Format: Thesis
Published: Curtin University 2024
Online Access:http://hdl.handle.net/20.500.11937/96603
_version_ 1848766176503529472
author Liu, Jie
author_facet Liu, Jie
author_sort Liu, Jie
building Curtin Institutional Repository
collection Online Access
description This thesis advances the field of anomaly detection in multivariate time series by addressing key challenges in anomaly detection, localization, and severity assessment. Through the development of EdgeConvFormer and U-GraphFormer, this research offers robust, interpretable, and efficient solutions applicable across various domains, with a particular focus on SHM. Extensive evaluations across diverse multivariate time series datasets and real-world scenarios demonstrate the potential of these models to enhance the monitoring and maintenance of critical systems, ensuring their safety and longevity.
first_indexed 2025-11-14T11:46:59Z
format Thesis
id curtin-20.500.11937-96603
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T11:46:59Z
publishDate 2024
publisher Curtin University
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-966032024-12-17T00:20:24Z Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring Liu, Jie This thesis advances the field of anomaly detection in multivariate time series by addressing key challenges in anomaly detection, localization, and severity assessment. Through the development of EdgeConvFormer and U-GraphFormer, this research offers robust, interpretable, and efficient solutions applicable across various domains, with a particular focus on SHM. Extensive evaluations across diverse multivariate time series datasets and real-world scenarios demonstrate the potential of these models to enhance the monitoring and maintenance of critical systems, ensuring their safety and longevity. 2024 Thesis http://hdl.handle.net/20.500.11937/96603 Curtin University restricted
spellingShingle Liu, Jie
Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title_full Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title_fullStr Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title_full_unstemmed Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title_short Unsupervised Anomaly Detection and Localization for Multivariate Time Series and Their Applications in Structural Health Monitoring
title_sort unsupervised anomaly detection and localization for multivariate time series and their applications in structural health monitoring
url http://hdl.handle.net/20.500.11937/96603