Multiscale computational strategy for the design of composite structures with material and geometric uncertainties
Stratégie de calcul multi-échelle pour la conception de structures composites avec prise en compte d'incertitudes
Résumé
The design of composite structures necessitates the use of multiscale simulations to take into account their hierarchical structure. To account for aleatory uncertainties due to the inherent scattering on material data and the geometrical tolerance, such simulations must be integrated with statistical uncertainty quantification (UQ) and propagation (UP) methods. In this work, a non-intrusive approach is proposed for multiscale UQ and UP. The different steps of the approach are illustrated by investigating the influence of structural uncertainties on the aeroelastic flutter of graphite/epoxy cantilevered plates with varying amounts of bendingtorsion stiffness coupling. Monte Carlo (MC) hierarchical sensitivity analysis is performed to identify the most important parameters at the micro-mechanical, layer, and laminate levels having first order effects on aeroelastic phenomena and quantities. Attention is paid to probabilistic modelling, including dependence, of each uncertain parameter of interest, at the different scales of the material. A surrogate model, constructed on data points obtained using high fidelity aeroelastic models, is employed for achieving computational efficiency in quantifying uncertainty. The objective is also to evaluate how uncertainties influence the design analysis, from the behaviour of the laminate up to final failure.
Origine : Fichiers éditeurs autorisés sur une archive ouverte