Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns

Akira Mita, Hiromi Hagiwara

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

16 Citations (Scopus)

Abstract

A method using the support vector machine (SVM) to detect local damages in a building structure with the limited number of sensors is proposed. The SVM is a powerful pattern recognition tool applicable to complicated classification problems. The method is verified to have capability to identify not only the location of damage but also the magnitude of damage with satisfactory accuracy. In our proposed method, feature vectors derived from the modal frequency patterns are used after proper normalization. The feature vectors contain the information on the location and magnitude of damages. As the method does not require modal shapes, typically only two vibration sensors are enough for detecting input and output signals to obtain the modal frequencies. The support vector machines trained for single damage is also effective for detecting damage in multiple stories.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Pages118-125
Number of pages8
Volume5057
DOIs
Publication statusPublished - 2003
EventSmart Structures and Materials 2003: Smart Systems and Nondestructive Evaluation for Civil Infrastructures - San Diego, CA, United States
Duration: 2003 Mar 32003 Mar 6

Other

OtherSmart Structures and Materials 2003: Smart Systems and Nondestructive Evaluation for Civil Infrastructures
CountryUnited States
CitySan Diego, CA
Period03/3/303/3/6

Fingerprint

Support vector machines
damage
Sensors
Pattern recognition
sensors
pattern recognition
vibration
output

Keywords

  • Damage detection
  • Health monitoring
  • Modal frequency change
  • Support vector machine

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Mita, A., & Hagiwara, H. (2003). Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5057, pp. 118-125) https://doi.org/10.1117/12.482705

Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns. / Mita, Akira; Hagiwara, Hiromi.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5057 2003. p. 118-125.

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

Mita, A & Hagiwara, H 2003, Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 5057, pp. 118-125, Smart Structures and Materials 2003: Smart Systems and Nondestructive Evaluation for Civil Infrastructures, San Diego, CA, United States, 03/3/3. https://doi.org/10.1117/12.482705
Mita A, Hagiwara H. Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5057. 2003. p. 118-125 https://doi.org/10.1117/12.482705
Mita, Akira ; Hagiwara, Hiromi. / Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5057 2003. pp. 118-125
@inproceedings{96fca93ea0124ac2932fb7b6e4004834,
title = "Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns",
abstract = "A method using the support vector machine (SVM) to detect local damages in a building structure with the limited number of sensors is proposed. The SVM is a powerful pattern recognition tool applicable to complicated classification problems. The method is verified to have capability to identify not only the location of damage but also the magnitude of damage with satisfactory accuracy. In our proposed method, feature vectors derived from the modal frequency patterns are used after proper normalization. The feature vectors contain the information on the location and magnitude of damages. As the method does not require modal shapes, typically only two vibration sensors are enough for detecting input and output signals to obtain the modal frequencies. The support vector machines trained for single damage is also effective for detecting damage in multiple stories.",
keywords = "Damage detection, Health monitoring, Modal frequency change, Support vector machine",
author = "Akira Mita and Hiromi Hagiwara",
year = "2003",
doi = "10.1117/12.482705",
language = "English",
volume = "5057",
pages = "118--125",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Damage Diagnosis of a Building Structure Using Support Vector Machine and Modal Frequency Patterns

AU - Mita, Akira

AU - Hagiwara, Hiromi

PY - 2003

Y1 - 2003

N2 - A method using the support vector machine (SVM) to detect local damages in a building structure with the limited number of sensors is proposed. The SVM is a powerful pattern recognition tool applicable to complicated classification problems. The method is verified to have capability to identify not only the location of damage but also the magnitude of damage with satisfactory accuracy. In our proposed method, feature vectors derived from the modal frequency patterns are used after proper normalization. The feature vectors contain the information on the location and magnitude of damages. As the method does not require modal shapes, typically only two vibration sensors are enough for detecting input and output signals to obtain the modal frequencies. The support vector machines trained for single damage is also effective for detecting damage in multiple stories.

AB - A method using the support vector machine (SVM) to detect local damages in a building structure with the limited number of sensors is proposed. The SVM is a powerful pattern recognition tool applicable to complicated classification problems. The method is verified to have capability to identify not only the location of damage but also the magnitude of damage with satisfactory accuracy. In our proposed method, feature vectors derived from the modal frequency patterns are used after proper normalization. The feature vectors contain the information on the location and magnitude of damages. As the method does not require modal shapes, typically only two vibration sensors are enough for detecting input and output signals to obtain the modal frequencies. The support vector machines trained for single damage is also effective for detecting damage in multiple stories.

KW - Damage detection

KW - Health monitoring

KW - Modal frequency change

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=0242636137&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242636137&partnerID=8YFLogxK

U2 - 10.1117/12.482705

DO - 10.1117/12.482705

M3 - Conference contribution

VL - 5057

SP - 118

EP - 125

BT - Proceedings of SPIE - The International Society for Optical Engineering

ER -