Multiscale computational homogenization of heterogeneous shells at small strains with extensions to finite displacements and buckling - Archive ouverte HAL Access content directly
Journal Articles International Journal for Numerical Methods in Engineering Year : 2015

Multiscale computational homogenization of heterogeneous shells at small strains with extensions to finite displacements and buckling

Abstract

In this paper, a framework for computational homogenization of shell structures is proposed in the context of small strain elastostatics, with extensions to large displacements and large rotations. At the macroscopic scale, heterogeneous thin structures are modeled using a homogenized shell model, based on a versatile three-dimensional 7-parameter shell formulation, incorporating a through-thickness and pre-integrated constitutive relationship. In the context of small strains, we show that the local solution on the elementary cell can be decomposed into 6 strain and 6 strain gradient modes, associated with corresponding boundary conditions. The heterogeneities can have arbitrary morphology, but are assumed to be periodically distributed in the tangential direction of the shell. We then propose an extension of the small strain framework to geometrical nonlinearities. The procedure is purely sequential and does not involve coupling between scales. The homogenization method is validated and illustrated through examples involving large displacements and buckling of heterogeneous plates and shells.
Fichier principal
Vignette du fichier
Multiscale_computational_homogeneization_Cong_al_2015 (1).pdf (4.12 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01165988 , version 1 (10-03-2016)

Identifiers

Cite

Yu Cong, Saeid Nezamabadi, Hamid Zahrouni, Julien Yvonnet. Multiscale computational homogenization of heterogeneous shells at small strains with extensions to finite displacements and buckling. International Journal for Numerical Methods in Engineering, 2015, 104 (4), pp.236-259. ⟨10.1002/nme.4927⟩. ⟨hal-01165988⟩
157 View
756 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More