2D EBSD map to 3D abaqus model

[陈泊]

Hi everyone,

I am totally new to Dream3D, and now I am working on crystal plasticity finite element simulation. 

Currently, I want to generate a 3D model based on the obtained 2D EBSD map, is it possible to transform 2 dimensional EBSD map to 3D abaqus model using Dream.3D software? (assume the grains are directional-solidified along z direction, so the grains on the back surface have the same crystal orientations as the front surface.)

Very grateful for your kindly response.

Best regards,

Bo

[Sean Donegan]

Hi Bo,

There is not currently an easy way to get a 3D model directly from a 2D EBSD map.  What you can do is "stack" the 2D map along the z-direction to get a pseudo-3D model; the grains in this case would essentially just be extruded along the z-axis.  This is done by using the Copy Data Container and Append Z-Slice (Image Geometry) filters.  I have attached an example pipeline (the JSON file) that shows how this process works by stacking a single slice from the example small IN100 data set to get a 3D volume that is 4 slices thick.  The pipeline then proceeds with the regular EBSD cleanup and reconstruction.  I've also attached an image of the result.  You could then use the Abaqus Hexahedron Writer filter to get Abaqus input files for the structure.  

You should be able to open the pipeline (in DREAM.3D, File > Open...) and run it immediately in the latest release of DREAM.3D, assuming you have the example data.  If you don't have the small IN100 data, you can download it here:

http://dream3d.bluequartz.net/Data/SmallIN100.zip

Then, unzip the archive; inside should be two folders, one called "Image" and one called "SmallIN100".  Place both these folders inside the DREAM.3D Data directory, which should be at the same location as the DREAM.3D executable.  As long as the data is in that location, the attached pipeline should work fine; it will also place its output in this Data/ directory.

hope that helps,

-Sean

[陈泊]

Great thanks Sean, I am sorry for the very late reply, cause I am busy with my experiment work these several months. Now I am back to modelling again, I will give it a go anyway. 

Bo