How does DualSPHysics handle the free surface?
Could someone explain it to me, even if shortly and point me in the right direction for ressources?
I know boundaries are using DBC or mDBC, but I actually do not know what is done at free surfaces. In multi-phase simulation I suppose it is handled as a phase interface or something like that.
Kind regards
Comments
The reason I find it difficult to understand what actually happens, is that the more I read about it, the only thing I find is that at the free surface, one just applies kernel correction and then everything is ok? Atleast for a mono-phase simulation
Seems kind of easy compared to FVM where you have to use Volume of Fluid methods etc., so I just would like to be sure I actually understand it.
Kind regards
Hi,
Please have a read through the Monaghan papers.
Very quickly and in a simplistic way to get you started:
SPH, and more precisely WCSPH, is quite good at that. At the free surface, your density is (near) equal to the reference density. The equation of state will give you a pressure of P = 0, handy.
Now, all the gradient operators are truncated at the free surface which results in a "small" error. We choose not to correct the SPH operators at the free surface (with the exemption of the density diffusion term). I can live with that.
Thus, you have a pressure (zero), a velocity, a density. etc!
Things are different with FVM. We use Lagrangian particles that move in space. The magic of WCSPH!
Regards,
George