Stochastic reduced order models for uncertain cantilevered plates in large deformations
Résumé
The consideration of structural uncertainty, including both data and model uncertainty, has been the focus of a series of recent investigations in the context of linear structural dynamic models. The extension of these efforts to nonlinear structural models, e.g. beams and plates in large deformations, has only very recently been addressed through the combination of a dedicated reduced order modeling approach and a nonparametric uncertainty characterization strategy relying on the maximization of the entropy. A preliminary validation of these concepts, achieved on the basis of a beam clamped on both of its side, demonstrated the strong potential of the approach. In this light, the goal of the present paper is to further validate the methodology to a significantly more complex system, i.e. a cantilevered plate of irregular geometry. The cantilevered aspect of this system implies a dramatic less severe nonlinearity than observed in the clampedclamped beam and thus much larger displacements are considered.