Efficient model updating using Bayesian probabilistic framework based on measured vibratory response

K. Zhou, G. Liang, J. Tang

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

3 Citations (Scopus)

Abstract

Currently, the deviation between the model and an actual structure is generally identified through a so-called model updating process, in which a set of experimental measurement of structural dynamic response is used in combination with the model prediction to facilitate an inverse analysis that is usually deterministic. In reality, however, structural properties, such as mass and stiffness, are inevitably subject to variation/uncertainties. As such, the identification of property variations in a probabilistic manner can truly reveal the underlying physical characteristics of the structure involved. In this research, we adopt the Bayesian probabilistic framework to conduct stochastic model updating using measured vibratory response. Furthermore, this paper proposes an efficient scheme to facilitate such procedures by incorporating the Gaussian process and Markov Chain Monte Carlo (MCMC) into the Bayesian framework. The feasibility of this presented methodology is validated by case studies.

Original languageEnglish
Title of host publicationNondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2014
PublisherSPIE
ISBN (Print)9780819499899
DOIs
Publication statusPublished - 2014
Externally publishedYes
EventNondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2014 - San Diego, CA, United States
Duration: 10 Mar 201413 Mar 2014

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9063
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceNondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2014
Country/TerritoryUnited States
CitySan Diego, CA
Period10/03/1413/03/14

Keywords

  • Bayesian probabilistic framework
  • Gaussian process
  • Markov Chain Monte Carlo
  • Measured vibratory response
  • Model updating
  • Uncertainties

ASJC Scopus subject areas

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

Fingerprint

Dive into the research topics of 'Efficient model updating using Bayesian probabilistic framework based on measured vibratory response'. Together they form a unique fingerprint.

Cite this