Boundary fliud separation problem

Rieck1991

Dear all,
I have been simulating wave structure interaction in a 3D wave flume and there is a big gap between fliud and boudary and I can not simulate the phenomenon described by Dr C.Altomare. I don't know why.
Wave flume size: 4.7m*0.15m*1.0m, initial particle distance dp=0.008m. Any advice will be appreciated.

Best regards,

Rieck

Alex

The gap is related with the boundary conditions treatment implemented in DualSPHysics that we called Dynamic Boundary Conditions (DBC).
In this case a gap of 1.2 times "h" (smoothing length) is usually observed when fluid particles interact with boundary particles not in contact first with fluid!

In order to reduce this gap you have serveral solutions:
- reducing h by reducing dp (higher resolution)
- reducing h by reducing coefh (but always higher than 0.85)
- during visualistion do not use Surface created by IsoSurface since the observed gap is higher than plotting vtk of particles.

I hope this helps, if not please email us with more information.

Regards

Rieck1991

Thank you for your reply. Is there existing same problem in Corrado Altomare's paper (2014) when simulating armour block run up?

Alex

Yes, in that case we created antifer blocks in a different way.
Boundary particles of the antifers were created 1.2*h far from the boundary layer of each antifer, so that, the influence of the boundary was working ok while we could get enough number of particles moving through the blocks.

Rieck1991

How to implement the boundary particles created 1.2*h far from the boundary layer of each antifer detailedly, I really am interested in it. Thank you very much.

Rieck1991

I found there existed the same bondary fulid separation phenomenon in Corrado's paper.

Alex

We created a STL with original dimensions that we import in GenCase to create particles. But to create particles 1.2*h far from boundary layer, we just created a new STL with those smaller dimensions...

Rieck1991

Dear Alex, I have sent a copy of my executive file.xml. Would you like to help me take a look at it and give some advice detailedly. Thank you very much.