Inverse quantification of epistemic uncertainty under scarce data: Bayesian or Interval approach?

Matthias Faes, Matteo Broggi, Edoardo Patelli, Yves Govers, John Mottershead, Michael Beer, David Moens

Research output: Contribution to conferencePaper

4 Downloads (Pure)

Abstract

This paper introduces a practical comparison of a newly introduced inverse method for the quantification of epistemically uncertain model parameters with the well-established probabilistic framework of Bayesian model updating via Transitional Markov Chain Monte Carlo. The paper gives a concise overview of both techniques, and both methods are applied to the quantification of a set of parameters in the well-known DLR Airmod test structure. Specifically, the case where only a very scarce set of experimentally obtained eigenfrequencies and eigenmodes are available is considered. It is shown that for such scarce data, the interval method provides more objective and robust bounds on the uncertain parameters than the Bayesian method, since no prior definition of the uncertainty is required, albeit at the cost that less information on parameter dependency or relative plausibility of different parameter values is obtained.

Original languageEnglish
Number of pages8
Publication statusPublished - 26 May 2019
Event13th International Conference on Applications of Statistics and Probability in Civil Engineering - Seoul, Korea, Republic of
Duration: 26 May 201930 May 2019

Conference

Conference13th International Conference on Applications of Statistics and Probability in Civil Engineering
Abbreviated titleICASP 13
CountryKorea, Republic of
CitySeoul
Period26/05/1930/05/19

Keywords

  • inverse method
  • Bayesian model
  • Transitional Markov Chain Monte Carlo

Fingerprint Dive into the research topics of 'Inverse quantification of epistemic uncertainty under scarce data: Bayesian or Interval approach?'. Together they form a unique fingerprint.

  • Cite this

    Faes, M., Broggi, M., Patelli, E., Govers, Y., Mottershead, J., Beer, M., & Moens, D. (2019). Inverse quantification of epistemic uncertainty under scarce data: Bayesian or Interval approach?. Paper presented at 13th International Conference on Applications of Statistics and Probability in Civil Engineering, Seoul, Korea, Republic of.