| 1. |
- Polindara, César, et al.
(author)
-
A computational framework for modelling damage-induced softening in fibre-reinforced materials - Application to balloon angioplasty
- 2017
-
In: International Journal of Solids and Structures. - : Elsevier BV. - 0020-7683. ; 118-119, s. 235-256
-
Journal article (peer-reviewed)abstract
- A computational framework for modelling damage-induced softening in fibre-reinforced materials is presented. The main aspect of this framework is the proposed non-local gradient-enhanced continuum damage formulation. At the material level, the elastic constitutive behavior is defined by a hyperelastic functional including a volumetric and an isochoric contribution. The isochoric contribution is subdivided into three contributions associated to three different phases i=0,1,2. Phase 0 is represented by an incompressible neo-Hookean material, whereas phases 1 and 2 are represented by an exponential format that accounts for the stretching along two preferred anisotropy directions, i.e. two fibre families. Furthermore, a 1-di-type damage function, is introduced to reproduce the loss of stiffness in each phase i. Following the ideas discussed in Dimitrijeciv and Hackl (2008) Waffenschmidt et al. (2014), and references cited therein, the model is built around the enhancement of the local free energy function by means of terms that contain the referential gradients of the non-local damage variables ϕ(symbol)i The inclusion of these terms ensures an implicit regularisation of the finite element implementation. A finite element implementation of the non-local gradient-enhanced continuum damage model is presented. To this end we develop an 8-noded Q1Q1P0 hexahedral element following a variational approach, in order to efficiently model the quasi-incompressible behaviour of the hyperelastic material. This element is implemented in Abaqus by means of a user subroutine UEL. Three boundary value problems are studied: an anisotropic plate with a hole, a balloon angioplasty and a full-3D artery-like tube. These computational experiments serve to illustrate the main capabilities of the proposed model.
|
|
