Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques

P. Omenzetter, O. R. De Lautour

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)

2 Citations (Scopus)
5 Downloads (Pure)

Abstract

Developed for studying long, periodic records of various measured quantities, time series analysis methods are inherently suited and offer interesting possibilities for Structural Health Monitoring (SHM) applications. However, their use in SHM can still be regarded as an emerging application and deserves more studies. In this research, Autoregressive (AR) models were used to fit experimental acceleration time histories from two experimental structural systems, a 3- storey bookshelf-type laboratory structure and the ASCE Phase II SHM Benchmark Structure, in healthy and several damaged states. The coefficients of the AR models were chosen as damage sensitive features. Preliminary visual inspection of the large, multidimensional sets of AR coefficients to check the presence of clusters corresponding to different damage severities was achieved using Sammon mapping - an efficient nonlinear data compression technique. Systematic classification of damage into states based on the analysis of the AR coefficients was achieved using two supervised classification techniques: Nearest Neighbor Classification (NNC) and Learning Vector Quantization (LVQ), and one unsupervised technique: Self-organizing Maps (SOM). This paper discusses the performance of AR coefficients as damage sensitive features and compares the efficiency of the three classification techniques using experimental data.
Original languageEnglish
Title of host publicationSensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010
Subtitle of host publicationProceedings of SPIE
EditorsMasayoshi Tomizuka
PublisherSPIE Press
Volume7647
ISBN (Print)9780819480620
DOIs
Publication statusPublished - 31 Mar 2010

Fingerprint

Time series analysis
Pattern recognition
Structural health monitoring
Vector quantization
Self organizing maps
Data compression
Inspection

Cite this

Omenzetter, P., & De Lautour, O. R. (2010). Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques. In M. Tomizuka (Ed.), Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010: Proceedings of SPIE (Vol. 7647). [76474S] SPIE Press. https://doi.org/10.1117/12.852573

Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques. / Omenzetter, P.; De Lautour, O. R.

Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010: Proceedings of SPIE. ed. / Masayoshi Tomizuka. Vol. 7647 SPIE Press, 2010. 76474S.

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)

Omenzetter, P & De Lautour, OR 2010, Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques. in M Tomizuka (ed.), Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010: Proceedings of SPIE. vol. 7647, 76474S, SPIE Press. https://doi.org/10.1117/12.852573
Omenzetter P, De Lautour OR. Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques. In Tomizuka M, editor, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010: Proceedings of SPIE. Vol. 7647. SPIE Press. 2010. 76474S https://doi.org/10.1117/12.852573
Omenzetter, P. ; De Lautour, O. R. / Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques. Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010: Proceedings of SPIE. editor / Masayoshi Tomizuka. Vol. 7647 SPIE Press, 2010.
@inbook{84f76adc31c849ae8514b747be6ceff0,
title = "Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques",
abstract = "Developed for studying long, periodic records of various measured quantities, time series analysis methods are inherently suited and offer interesting possibilities for Structural Health Monitoring (SHM) applications. However, their use in SHM can still be regarded as an emerging application and deserves more studies. In this research, Autoregressive (AR) models were used to fit experimental acceleration time histories from two experimental structural systems, a 3- storey bookshelf-type laboratory structure and the ASCE Phase II SHM Benchmark Structure, in healthy and several damaged states. The coefficients of the AR models were chosen as damage sensitive features. Preliminary visual inspection of the large, multidimensional sets of AR coefficients to check the presence of clusters corresponding to different damage severities was achieved using Sammon mapping - an efficient nonlinear data compression technique. Systematic classification of damage into states based on the analysis of the AR coefficients was achieved using two supervised classification techniques: Nearest Neighbor Classification (NNC) and Learning Vector Quantization (LVQ), and one unsupervised technique: Self-organizing Maps (SOM). This paper discusses the performance of AR coefficients as damage sensitive features and compares the efficiency of the three classification techniques using experimental data.",
author = "P. Omenzetter and {De Lautour}, {O. R.}",
year = "2010",
month = "3",
day = "31",
doi = "10.1117/12.852573",
language = "English",
isbn = "9780819480620",
volume = "7647",
editor = "Masayoshi Tomizuka",
booktitle = "Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010",
publisher = "SPIE Press",

}

TY - CHAP

T1 - Classification of damage in structural systems using time series analysis and supervised and unsupervised pattern recognition techniques

AU - Omenzetter, P.

AU - De Lautour, O. R.

PY - 2010/3/31

Y1 - 2010/3/31

N2 - Developed for studying long, periodic records of various measured quantities, time series analysis methods are inherently suited and offer interesting possibilities for Structural Health Monitoring (SHM) applications. However, their use in SHM can still be regarded as an emerging application and deserves more studies. In this research, Autoregressive (AR) models were used to fit experimental acceleration time histories from two experimental structural systems, a 3- storey bookshelf-type laboratory structure and the ASCE Phase II SHM Benchmark Structure, in healthy and several damaged states. The coefficients of the AR models were chosen as damage sensitive features. Preliminary visual inspection of the large, multidimensional sets of AR coefficients to check the presence of clusters corresponding to different damage severities was achieved using Sammon mapping - an efficient nonlinear data compression technique. Systematic classification of damage into states based on the analysis of the AR coefficients was achieved using two supervised classification techniques: Nearest Neighbor Classification (NNC) and Learning Vector Quantization (LVQ), and one unsupervised technique: Self-organizing Maps (SOM). This paper discusses the performance of AR coefficients as damage sensitive features and compares the efficiency of the three classification techniques using experimental data.

AB - Developed for studying long, periodic records of various measured quantities, time series analysis methods are inherently suited and offer interesting possibilities for Structural Health Monitoring (SHM) applications. However, their use in SHM can still be regarded as an emerging application and deserves more studies. In this research, Autoregressive (AR) models were used to fit experimental acceleration time histories from two experimental structural systems, a 3- storey bookshelf-type laboratory structure and the ASCE Phase II SHM Benchmark Structure, in healthy and several damaged states. The coefficients of the AR models were chosen as damage sensitive features. Preliminary visual inspection of the large, multidimensional sets of AR coefficients to check the presence of clusters corresponding to different damage severities was achieved using Sammon mapping - an efficient nonlinear data compression technique. Systematic classification of damage into states based on the analysis of the AR coefficients was achieved using two supervised classification techniques: Nearest Neighbor Classification (NNC) and Learning Vector Quantization (LVQ), and one unsupervised technique: Self-organizing Maps (SOM). This paper discusses the performance of AR coefficients as damage sensitive features and compares the efficiency of the three classification techniques using experimental data.

UR - http://www.scopus.com/inward/record.url?scp=77953482262&partnerID=8YFLogxK

U2 - 10.1117/12.852573

DO - 10.1117/12.852573

M3 - Chapter (peer-reviewed)

AN - SCOPUS:77953482262

SN - 9780819480620

VL - 7647

BT - Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2010

A2 - Tomizuka, Masayoshi

PB - SPIE Press

ER -