SHM data anomaly classification using machine learning strategies: a comparative study

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, su...

Full description

Saved in:
Bibliographic Details
Main Authors: Chou, Jau-Yu, Fu, Yuguang, Huang, Shieh-Kung, Chang, Chia-Ming
Other Authors: School of Civil and Environmental Engineering
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/161492
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-161492
record_format dspace
spelling sg-ntu-dr.10356-1614922022-09-06T01:45:00Z SHM data anomaly classification using machine learning strategies: a comparative study Chou, Jau-Yu Fu, Yuguang Huang, Shieh-Kung Chang, Chia-Ming School of Civil and Environmental Engineering Engineering::Civil engineering Data Anomaly Classification Ensemble Neural Network Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time. 2022-09-06T01:45:00Z 2022-09-06T01:45:00Z 2022 Journal Article Chou, J., Fu, Y., Huang, S. & Chang, C. (2022). SHM data anomaly classification using machine learning strategies: a comparative study. Smart Structures and Systems, 29(1), 77-91. https://dx.doi.org/10.12989/sss.2022.29.1.077 1738-1584 https://hdl.handle.net/10356/161492 10.12989/sss.2022.29.1.077 2-s2.0-85129158782 1 29 77 91 en Smart Structures and Systems © 2022 Techno-Press, Ltd. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Civil engineering
Data Anomaly Classification
Ensemble Neural Network
spellingShingle Engineering::Civil engineering
Data Anomaly Classification
Ensemble Neural Network
Chou, Jau-Yu
Fu, Yuguang
Huang, Shieh-Kung
Chang, Chia-Ming
SHM data anomaly classification using machine learning strategies: a comparative study
description Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.
author2 School of Civil and Environmental Engineering
author_facet School of Civil and Environmental Engineering
Chou, Jau-Yu
Fu, Yuguang
Huang, Shieh-Kung
Chang, Chia-Ming
format Article
author Chou, Jau-Yu
Fu, Yuguang
Huang, Shieh-Kung
Chang, Chia-Ming
author_sort Chou, Jau-Yu
title SHM data anomaly classification using machine learning strategies: a comparative study
title_short SHM data anomaly classification using machine learning strategies: a comparative study
title_full SHM data anomaly classification using machine learning strategies: a comparative study
title_fullStr SHM data anomaly classification using machine learning strategies: a comparative study
title_full_unstemmed SHM data anomaly classification using machine learning strategies: a comparative study
title_sort shm data anomaly classification using machine learning strategies: a comparative study
publishDate 2022
url https://hdl.handle.net/10356/161492
_version_ 1744365406818140160