3. Modeling B: crack X- FEM#

In this modeling, the crack is no longer meshed, but it is represented by level sets: \(\mathit{LSN}\mathrm{=}z\mathrm{-}1\mathrm{/}2\) and \(\mathit{LST}\mathrm{=}\mathrm{-}y\mathrm{-}x\mathrm{/}2+1\).

3.1. Characteristics of the mesh#

The structure is modelled by a mesh composed of 15872 tetrahedra (see [Figure 2.1-a]).

_images/10000000000003BA000002950DC2C5BFF7EDF7F2.png

Figure 2.1-a: Healthy mesh

3.2. Tested sizes and results#

The values of \(\mathit{K1}\) are tested on the second point and on the last three points of the crack bottom. Indeed, the orientation of the crack implies that \(\mathit{K1}\) cannot be calculated on certain points. We test the points concerned to verify that Code_Aster assigns them the value of the nearest neighbor node or the calculation of \(\mathit{K1}\) could be performed.

Identification

Reference Type

Reference Value

Point 2

“NON_REGRESSION”

1324885.20838

Dot 22

“NON_REGRESSION”

851382.586304

Dot 23

“NON_REGRESSION”

851382.586304

Dot 24

“NON_REGRESSION”

851382.586304

We also test the values of \(\mathit{COOR}\text{\_}X\), \(\mathit{COOR}\text{\_}Y\), \(\mathit{COOR}\text{\_}Z\), \(\mathit{TEMP}\) and \(\mathit{NEUT}1\) at node 1 in order to verify that these components are indeed present in the output table for the case X- FEM.