Accès direct au contenu


Version anglaise


Version francaise > ÉVÈNEMENTS > Séminaires

Recherche - Valorisation

Adaptive surrogate strategies for investigating mechanical instability

le 28 novembre 2019
à 13h30

Amélie FAU, (Secteur Structures et Système)

Amélie FAU

Amélie FAU

Résumé :

Cumbersome parametric studies are generally necessary for detecting instability regimes of complex systems. These analyses may lead to high computational cost, if a fine exploration of parametric space is needed. To tackle that problem, an innovative adaptive scheme based on kriging metamodel has been proposed to guide the observations in the most beneficial positions in order to detect accurately instability subdomains based on a reduced number of observations [1]. That numerical scheme has appeared appealing for analysing the risk of stick-slip behavior for dynamic systems [2].

[1]       Fuhg, J.N., Fau, A., An innovative adaptive kriging approach for efficient binary classification of mechanical problems, arXiv preprint, 2019.

[2]       Fuhg, J.N., Fau, A., Surrogate model approach for investigating the stability of a friction-induced oscillator of Duffing's type, Nonlinear Dynamics, 1-21, 2019.

This research activity lies in the international framework of the French-German doctoral college between LMT and Leibniz Unviersität Hannover
Type :
Séminaires - conférences
Lieu(x) :
Campus de Cachan
ENS Paris-Saclay - Bâtiment Léonard de Vinci - Amphithéâtre E-Media

Recherche d'une actualité

Recherche d'une actualité