Karhunen-Loève based sensitivity analysis

G. Perrin (1, 2, 3) , Christian Soize (1) , Denis Duhamel (2, 4) , C. Fünfschilling (3)

1 MSME - Laboratoire de Modélisation et Simulation Multi Echelle
2 navier umr 8205 - Laboratoire Navier
3 SNCF - Direction de l'Innovation et de la Recherche
4 Dynamique des structures et identification

The identification of the most dangerous combinations of excitations that a non-linear mechanical system can be confronted to is not an easy task. Indeed, in such cases, the link between the maximal values of the inputs and of the outputs is not direct, as the system can be more sensitive to a problematic succession of excitations of low amplitudes than to high amplitudes for each kind of excitations. This work presents therefore an innovative method to identify the combined shapes of excitations that are the most correlated to problematic responses of the studied mechanical system.

Domaines: Mécanique [physics.med-ph]; Probabilités [math.PR]; Statistiques [math.ST]

G. Perrin, Christian Soize, Denis Duhamel, C. Fünfschilling. Karhunen-Loève based sensitivity analysis. International Workshop on Uncertainty Quantification in Fluids Simulation (BOQUSE 2013), Dec 2013, Bordeaux, France. pp.1-3. ⟨hal-00940374⟩