Eventos Anais de eventos
COBEM 2017
24th ABCM International Congress of Mechanical Engineering
BAYESIAN DYNAMIC MODEL UPDATING USING MARKOV CHAIN MONTE CARLO METHODS
Submission Author:
Rafael Beal Macedo , SP
Co-Authors:
Airton Nabarrete, José Antonio Hernandes, Rafael Beal Macedo
Presenter: Rafael Beal Macedo
doi://10.26678/ABCM.COBEM2017.COB17-1646
Abstract
The application of Bayesian model updating techniques such as Markov Chain Monte Carlo to structural dynamic finite element models is a growing interest of the industry, including applications requiring the determination of damping. Viscoelastic materials, such as the skin damping that is typically used on aircraft fuselages to attenuate the propagation of noise and vibration to the cabin, present non-linear properties that are difficult to model, as well as high damping values that are not always possible to determine with classical methods. Using the dynamic model of a beam with viscoelastic material, correlating model data to experimental results, this paper implements an adaptive Metropolis-Hastings (AMH) method to the determination of damping at the frequency domain. The method is shown to be capable of determining the parameters accurately, allowing the determination of damping of the viscoelastic skin damping.
Keywords
bayesian model updating, markov chain monte carlo, Viscoelastic, non-linear
