Bug Fix in the additiveManager routine:

In pointAdditiveManager.cpp, some boundary conditions were not correctly deactivated even when they were not connected to active volume elements. I modified the code so that when the routine deactivates volume elements before the simulation it will automatically deactivate connected boundary conditions if they are not on the edge of the geometry. This only affects simulations where the boundary conditions where not explicitly given to the routine for deactivation. (i.e. not given via pointAdditiveManager.addBoundaryCondition(BC)).

In short, the boundary conditions will now correctly be deactivated at the beginning of a simulation if :

1) They are given to pointAdditiveManager via pointAdditiveManager.addBoundaryCondition(BC)

2) They are not on the edge of the geometry in the “to be activated” area.

Corrected test parameters

In chiumenti, I found a small error in the laser path where an extra 0.0875s per layer was spend between layers. The differences on the results are highlighted here after.

Creation of lagamine mesh importer (only for BLZ2T lagamine elements with CONRA lagamine BCs)

I created a Lagamine geometry importer, although it only works for 2 specific types of Lagamine elements at the moment. The importer is situated in toolbox\lag2py.py . The importer reads the “testName.lag” source file and translates the geometry in a metafor input file “testName_lag.py”. More specifically:

• The geometry (points/lines/quads) is created. The point numbering is the same as in the lagamine file. The initial temperature is also set at each point.
• The points are all sorted into metafor groups, 1 group per lagamine material law respecting the lagamine material law number. This allows to push this group in a FieldApplicator to created all elements.
• For each of the CONRA boundary conditions, a list of metafor groups is created, 1 group = 2 points connecting a CONRA element. This allows to push those groups in a LoadingInteraction to create boundary condition elements.

For an example of use, check additiveM\withAm\tests\HashemiAndJardin\jardin.py

The rest of the test is then created by hand in the metafor input file, but this allows to use the same mesh in metafor and lagamine.

The comparison of the Chiumenti test with and without the laser path error is given hereafter.

CHANGE HERE

With this last modification in the code and the addition of the Lagamine importer I was finally able to reproduce the results from Jardin:

The results are given hereafter:

I added battery tests …

— Cédric Laruelle 2019/01/07