Dealing with complex geometry/extending lbm tutorial 4
(I assume that this is the correct place to ask question about using waLBerla. If not, apologies!)
I'm a master student in Mathematics at TUM, writing my master thesis on using LBM to simulate the blood flow near cerebral aneurysms, and intend on using waLBerla as my LBM library, as I really like what I've seen so far.
First question beforehand: What is nowadays the preferred way to use waLBerla? I've read in some other gitlab issues that using lbmpy and pystencils are usually the way to go, either to generate C++ code or run 'natively' in python, but didn't see any documentation on the website on when/how to make this choice.
To do so, I have some .obj file with a triangle mesh of an artery near a cerebral aneurysm. When I simply use this object instead of, for example, the bunny and make some other small changes (i.e. exclude the scaling of the domain as done in the tutorial) and compile/run the program, I get the warning message that my mesh is not watertight and that is has boundary edges. As my mesh indeed is not watertight, I adjusted that in Blender, see also the attached pictures. The warning remains, so it has to do with the boundary edges.
My simulation should be relatively straightforward, I have one inflow, two outflows and on the geometry a noslip boundary condition. I understand that I must let the library know where these boundaries are, but I can't find how to do this in the documentation and all of the example applications don't do similar things.
I have found that the 2014 master thesis by Al Hasnine and the 2013 waLBerla paper about blood flow simulations on the coronary artery tree are very similar to my topics, but was unable to find the code corresponding to these simulations.
Are there any examples available on how to do such things? Or what would be the best strategy?
As I assume more people wanting to experiment with waLBerla may have similar questions, so I have no problem working the answer to this question out into a (small) tutorial!
Many thanks in advance!