Floating body: explosions in the sky
 

Hi all,

I am playing around with the floating body capabilities of interDyMFoam. So far I have carried out the floating body tutorial and it runs flawlessly, I then added some waves and eventually it crashed due to the drifting of the body making the mesh horribly distorted, as expected.

I now decided to go for a more complex geometry and created something that represents an offshore platform in a *very* simplified way. The mesh is structured and orthogonal and has hex cells only. Other than that, the BCs, schemes and tolerances are the very same from the floating body tutorial, with updated mass, CoM and inertias.

The case blows up due to non-physical velocities around the edges located in the air area, check the image:

http://i.imgur.com/VLE1XMO.png


For the record, I know this change is not correct at all but the same case with zeroGradient instead of movingWallVelocity in U BC for the structure is running. Watching the video makes you see that the results are unphysical because water presents a weird behaviour, but it runs...

I have tried other options, none of which works:
- Having additional refinement levels around the structure
- Creating a mesh with cell aspect ratio = 1
- Changing the div(rho*phi,U) scheme to Gauss limitedLinearV 1 and others, known to be more robust
- Selecting very conservative Co numbers or smaller initial deltaT
- Selectively relaxing large air velocities

Any further recommendations?

Thanks!

Pablo

Some news:

I have managed to take care of the well-known pressure fluctuations adapting some techniques from the shipFoam (OF 1.6-ext) solver and others. Finally the case runs, but I feel that OpenFOAM is not the model to simulate floating bodies out of the box at this stage. Anyway, this is not my field and I was just playing...

You can check the results here:

http://www.youtube.com/watch?v=OnjGPbCNF7g

Best,

Pablo