Characterisation of composite elastic properties by means of a multi-scale two-level inverse approach - Archive ouverte HAL
Article Dans Une Revue Composite Structures Année : 2018

Characterisation of composite elastic properties by means of a multi-scale two-level inverse approach

Résumé

This work deals with the problem of characterising the elastic properties of a composite material at both me- soscopic (ply-level) and microscopic (constitutive phases-level) scales. This goal is attained by means of an adequate multi-scale identification strategy (MSIS) which aims at identifying the constitutive properties, at each relevant scale, by exploiting the information restrained in the macroscopic dynamic response of the composite. In this background, the multi-scale identification problem is split into two interdependent sub-problems which are stated, at both levels, as constrained minimisation problems. At the first level the goal is the characterisation of the lamina properties by minimising the distance between the numerical and the reference harmonic re- sponses of the composite. The second level problem aims at identifying the elastic properties of both fibre and matrix by minimising the distance between the effective elastic properties evaluated through a homogenisation process and those provided by the first-level inverse problem. The MSIS is based on a special global hybrid optimisation tool and on the strain energy homogenisation method of periodic media. Its effectiveness is proven through a meaningful benchmark.
Fichier principal
Vignette du fichier
I2M_CS_2018_DAU.pdf (2.47 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02318801 , version 1 (05-02-2020)

Identifiants

Citer

Lorenzo Cappelli, Marco Montemurro, Frédéric Dau, Laurent Guillaumat. Characterisation of composite elastic properties by means of a multi-scale two-level inverse approach. Composite Structures, 2018, 204, pp.767-777. ⟨10.1016/j.compstruct.2018.08.007⟩. ⟨hal-02318801⟩
30 Consultations
161 Téléchargements

Altmetric

Partager

More