Support vector machine based active damage detection method

R. Taniguchi, Akira Mita

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

An active damage detection method using Support Vector Machine (SVM) is proposed. The SVM is known as a powerful pattern recognition tool that can be applied to complicated classification problems. An active sensing diagnostic technique using PZT elements was adopted to characterize damages in bolted joints. Built-in PZTs were used to predict damages by using differences of Lamb wave signals recorded before and after damages. The wavelet transform was applied for a time-frequency analysis of the Lamb waves to characterize the feature vectors for feeding to SVMs. It can correctly estimate the location and the degree of damages. Applicability of proposed damage detection method was successfully demonstrated.

Original languageEnglish
Title of host publicationStructural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003
PublisherDEStech Publications
Pages749-756
Number of pages8
ISBN (Print)1932078207, 9781932078206
Publication statusPublished - 2003
Event4th International Workshop on Structural Health Monitoring: From Diagnostics and Prognostics to Structural Health Management, IWSHM 2003 - Stanford, United States
Duration: 2003 Sep 152003 Sep 17

Other

Other4th International Workshop on Structural Health Monitoring: From Diagnostics and Prognostics to Structural Health Management, IWSHM 2003
CountryUnited States
CityStanford
Period03/9/1503/9/17

Fingerprint

Damage detection
Surface waves
Support vector machines
Bolted joints
Wavelet Analysis
Wavelet transforms
Pattern recognition
Joints
Support Vector Machine

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Information Management

Cite this

Taniguchi, R., & Mita, A. (2003). Support vector machine based active damage detection method. In Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003 (pp. 749-756). DEStech Publications.

Support vector machine based active damage detection method. / Taniguchi, R.; Mita, Akira.

Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003. DEStech Publications, 2003. p. 749-756.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Taniguchi, R & Mita, A 2003, Support vector machine based active damage detection method. in Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003. DEStech Publications, pp. 749-756, 4th International Workshop on Structural Health Monitoring: From Diagnostics and Prognostics to Structural Health Management, IWSHM 2003, Stanford, United States, 03/9/15.
Taniguchi R, Mita A. Support vector machine based active damage detection method. In Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003. DEStech Publications. 2003. p. 749-756
Taniguchi, R. ; Mita, Akira. / Support vector machine based active damage detection method. Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003. DEStech Publications, 2003. pp. 749-756
@inproceedings{4e9d642bdf56439a9208731ca130d70e,
title = "Support vector machine based active damage detection method",
abstract = "An active damage detection method using Support Vector Machine (SVM) is proposed. The SVM is known as a powerful pattern recognition tool that can be applied to complicated classification problems. An active sensing diagnostic technique using PZT elements was adopted to characterize damages in bolted joints. Built-in PZTs were used to predict damages by using differences of Lamb wave signals recorded before and after damages. The wavelet transform was applied for a time-frequency analysis of the Lamb waves to characterize the feature vectors for feeding to SVMs. It can correctly estimate the location and the degree of damages. Applicability of proposed damage detection method was successfully demonstrated.",
author = "R. Taniguchi and Akira Mita",
year = "2003",
language = "English",
isbn = "1932078207",
pages = "749--756",
booktitle = "Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003",
publisher = "DEStech Publications",

}

TY - GEN

T1 - Support vector machine based active damage detection method

AU - Taniguchi, R.

AU - Mita, Akira

PY - 2003

Y1 - 2003

N2 - An active damage detection method using Support Vector Machine (SVM) is proposed. The SVM is known as a powerful pattern recognition tool that can be applied to complicated classification problems. An active sensing diagnostic technique using PZT elements was adopted to characterize damages in bolted joints. Built-in PZTs were used to predict damages by using differences of Lamb wave signals recorded before and after damages. The wavelet transform was applied for a time-frequency analysis of the Lamb waves to characterize the feature vectors for feeding to SVMs. It can correctly estimate the location and the degree of damages. Applicability of proposed damage detection method was successfully demonstrated.

AB - An active damage detection method using Support Vector Machine (SVM) is proposed. The SVM is known as a powerful pattern recognition tool that can be applied to complicated classification problems. An active sensing diagnostic technique using PZT elements was adopted to characterize damages in bolted joints. Built-in PZTs were used to predict damages by using differences of Lamb wave signals recorded before and after damages. The wavelet transform was applied for a time-frequency analysis of the Lamb waves to characterize the feature vectors for feeding to SVMs. It can correctly estimate the location and the degree of damages. Applicability of proposed damage detection method was successfully demonstrated.

UR - http://www.scopus.com/inward/record.url?scp=10044274525&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=10044274525&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:10044274525

SN - 1932078207

SN - 9781932078206

SP - 749

EP - 756

BT - Structural Health Monitoring 2003: From Diagnostics and Prognostics to Structural Health Management - Proceedings of the 4th International Workshop on Structural Health Monitoring, IWSHM 2003

PB - DEStech Publications

ER -