Uncertain quantification in nonlinear dynamics with an high-dimensional computational model
Résumé
The present work concerns the dynamical analysis of an uncertain structure in the context of nonlinear dynamics. The structure is assumed to undergo large displacements and large deformations although the constitutive equations remain linearly elastic. The proposed strategy is compatible with the use of high-dimensional computational models, requiring to compute the random dynamical response from a stochastic nonlinear reduced-order model expressed in the time domain. With the proposed method, the uncertainty is introduced by replacing a deterministic chosen reduced-order basis with a stochastic projection basis, for which a new nonparametric probabilistic approach is used so that each realization of the projection basis respects some mathematical properties linked to the available information. The methodology is then applied on a computational model of a bi-clamped tridimensional beam structure.
Fichier principal
conference-2018-ISMA-Leuven-sept-17-19-capiez-soize-preprint.pdf (1.48 Mo)
Télécharger le fichier
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...