Damage classification and estimation in experimental structures using time series analysis and pattern recognition

Oliver R. de Lautour, Piotr Omenzetter

Research output: Contribution to journalArticle

70 Citations (Scopus)
17 Downloads (Pure)

Abstract

Developed for studying long sequences of regularly sampled data, time series analysis methods are being increasingly investigated for the use of Structural Health Monitoring (SHM). In this research, Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures: a 3-storey bookshelf structure and the ASCE Phase II Experimental SHM Benchmark Structure, in undamaged and limited number of damaged states. The coefficients of the AR models were considered to be damage-sensitive features and used as input into an Artificial Neural Network (ANN). The ANN was trained to classify damage cases or estimate remaining structural stiffness. The results showed that the combination of AR models and ANNs are efficient tools for damage classification and estimation, and perform well using small number of damage-sensitive features and limited sensors.
Original languageEnglish
Pages (from-to)1556-1569
Number of pages14
JournalMechanical Systems and Signal Processing
Volume24
Issue number5
Early online date7 Jan 2010
DOIs
Publication statusPublished - Jul 2010

Fingerprint

Time series analysis
Pattern recognition
Structural health monitoring
Neural networks
Stiffness
Sensors

Cite this

Damage classification and estimation in experimental structures using time series analysis and pattern recognition. / de Lautour, Oliver R.; Omenzetter, Piotr.

In: Mechanical Systems and Signal Processing, Vol. 24, No. 5, 07.2010, p. 1556-1569.

Research output: Contribution to journalArticle

@article{7cfc92c763904d4fbe5ddb318728535b,
title = "Damage classification and estimation in experimental structures using time series analysis and pattern recognition",
abstract = "Developed for studying long sequences of regularly sampled data, time series analysis methods are being increasingly investigated for the use of Structural Health Monitoring (SHM). In this research, Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures: a 3-storey bookshelf structure and the ASCE Phase II Experimental SHM Benchmark Structure, in undamaged and limited number of damaged states. The coefficients of the AR models were considered to be damage-sensitive features and used as input into an Artificial Neural Network (ANN). The ANN was trained to classify damage cases or estimate remaining structural stiffness. The results showed that the combination of AR models and ANNs are efficient tools for damage classification and estimation, and perform well using small number of damage-sensitive features and limited sensors.",
author = "{de Lautour}, {Oliver R.} and Piotr Omenzetter",
year = "2010",
month = "7",
doi = "10.1016/j.ymssp.2009.12.008",
language = "English",
volume = "24",
pages = "1556--1569",
journal = "Mechanical Systems and Signal Processing",
issn = "0888-3270",
publisher = "Academic Press Inc.",
number = "5",

}

TY - JOUR

T1 - Damage classification and estimation in experimental structures using time series analysis and pattern recognition

AU - de Lautour, Oliver R.

AU - Omenzetter, Piotr

PY - 2010/7

Y1 - 2010/7

N2 - Developed for studying long sequences of regularly sampled data, time series analysis methods are being increasingly investigated for the use of Structural Health Monitoring (SHM). In this research, Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures: a 3-storey bookshelf structure and the ASCE Phase II Experimental SHM Benchmark Structure, in undamaged and limited number of damaged states. The coefficients of the AR models were considered to be damage-sensitive features and used as input into an Artificial Neural Network (ANN). The ANN was trained to classify damage cases or estimate remaining structural stiffness. The results showed that the combination of AR models and ANNs are efficient tools for damage classification and estimation, and perform well using small number of damage-sensitive features and limited sensors.

AB - Developed for studying long sequences of regularly sampled data, time series analysis methods are being increasingly investigated for the use of Structural Health Monitoring (SHM). In this research, Autoregressive (AR) models were used to fit the acceleration time histories obtained from two experimental structures: a 3-storey bookshelf structure and the ASCE Phase II Experimental SHM Benchmark Structure, in undamaged and limited number of damaged states. The coefficients of the AR models were considered to be damage-sensitive features and used as input into an Artificial Neural Network (ANN). The ANN was trained to classify damage cases or estimate remaining structural stiffness. The results showed that the combination of AR models and ANNs are efficient tools for damage classification and estimation, and perform well using small number of damage-sensitive features and limited sensors.

U2 - 10.1016/j.ymssp.2009.12.008

DO - 10.1016/j.ymssp.2009.12.008

M3 - Article

VL - 24

SP - 1556

EP - 1569

JO - Mechanical Systems and Signal Processing

JF - Mechanical Systems and Signal Processing

SN - 0888-3270

IS - 5

ER -