Identification of elastic and damping properties of material from a Bayesian formulation of the Force Analysis Technique
Résumé
A good knowledge of elastic and damping properties of material is required to predict the dynamic and vibroacoustic behavior of structures, because of their great influence. Recently, an approach derived from the Force Analysis Technique (FAT), initially used to identify external efforts, has been proposed to identify these material properties. The purpose of this work is to present an approach in the same spirit, where the identification of material properties is considered from a Bayesian point of view. The principle of the inverse problem is first exposed in a probabilistic framework where a priori and likelihood function are defined precisely. The solving is then performed by using a Monte-Carlo Markov Chain (MCMC) method allowing one to obtain values and uncertainties of results thanks to their probability densities assessments. After the explanation of all steps of the proposed method, some examples are shown, where adjustments are discussed and where elastic and damping material properties are identified on analytically modeled structures.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...