Loading…
This event has ended. Visit the official site or create your own event on Sched.
Welcome to ICSV26!
Back To Schedule
Wednesday, July 10 • 17:10 - 17:30
A BAYESIAN APPROACH FOR IDENTIFICATION OF FORCES CONTAINING TREND COMPONENTS

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
In this paper, we propose a new force history identification method based on Bayesian regularization. To-be-determined variables are incorporated into a hierarchical Bayesian framework. As a major difference from previous Bayesian approaches, a mean term is considered in the prior distribution of force history, and is assumed as low-order polynomials. Adaptability and better regularization performance are achieved in cases where notable trend components exist in force histories. The posterior distributions of unknown variables including force history, precision parameters, the number and values of polynomial coefficients are sampled with a trans-dimensional reversible-jump Markov chain Monte Carlo method. The expectations of unknown variables and their credible intervals are given in the post-evaluation. A numerical study is carried out to validate the proposed method.

Moderators
AR

Annie Ross

Professor, Polytechnique Montreal

Authors

Wednesday July 10, 2019 17:10 - 17:30 EDT
Westmount 3
  T10 Sig. Proc. & nonlin. mthds., RS04 Signal process in ac & vibr