Damage assessment of bending structures using support vector machine

M. Shimada, A. Mita

Research output: Contribution to journalConference articlepeer-review

12 Citations (Scopus)


A damage detection method utilizing Support Vector Machine (SVM) for bending structures is proposed. The SVM was recently proposed as a new technique for pattern recognition. The SVM is a powerful pattern recognition tool applicable to complicated classification problems and is effectively utilized in the method. Based on the modal frequency changes, the damage location and its severity are defined by the SVM. In our previous studies, it was shown that our proposed method worked very well for structures modeled by shear frames. However, this modeling is only appropriate for lowrise building structures and is not appropriate for tall buildings. Therefore, it is our purpose here to extend the method to bending frames that are appropriate models for tall buildings. In the analytical evaluation, we constructed the finite element models to represent bending structures. Then, we conducted a series of experiments for verification. We could show that the damage detection method using SVM was also possible and effective for bending structures.

Original languageEnglish
Article number103
Pages (from-to)923-930
Number of pages8
JournalProceedings of SPIE - The International Society for Optical Engineering
Issue numberPART 2
Publication statusPublished - 2005
Externally publishedYes
EventSmart Structures and Materials 2005 - Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems - San Diego, CA, United States
Duration: 2005 Mar 72005 Mar 10


  • Damage detection
  • Modal analysis
  • Support vector machine
  • System identification

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering


Dive into the research topics of 'Damage assessment of bending structures using support vector machine'. Together they form a unique fingerprint.

Cite this