Problem Wiht Gap Between Fluid And DEM Particle

mtmtmt

Hi,I Working With Case_DEM And I Built Rock Slope, But I Have 2 Problem,
1) When DEM Particle Fall Under their Weight,DEM Particle Prevent water penetration.and i know that this problem to come from 2D Simulation But Is there any way to resolve this?
2) when the fluid particle interact with DEM Particle,There is a gap between Fluid Particle And DEM Particle. How to reduce this distance?

Asalih3d

Hi

1) As you mention it is a limit of the 2D simulation type, so you can't really resolve it.

2) Have an even smaller particle spacing and you should be able to resolve it. There will ALWAYS be a gap in v4.2 - the only way to decrease the gap size, is by adding more particles. In next version some bound correction will be added, which will be able to resolve this problem (atleast for fixed/moving geometries, not sure about floatings).

As long as variable particle spacing isn't available this is your best bet, to just add more particles.

Kind regards

mtmtmt

Thanks dear @Asalih3d
Do you know where the interaction between a float (Or DEM) object and fluid particles are calculated in the code? I can not find it in the code. Is it possible to decrease the gap without using smaller dp?
Kind regards

Asalih3d

No, I do not know where it is located in code unfortunately, I haven't really had a good look into the code.

No, it is not possible to decrease the gap, since the gap appears because of the dynamic boundary condition - due to changes in density of a solid particle a repulsive force will appear to push away fluid particles. This force is too big and therefore this gap appears - reducing the gap can only be done using a different boundary condition, which is currently not available in DualSPHysics official release.

Kind regards

mtmtmt

Dear @Asalih3d Thanks For Your Attention,Your Advice Was very helpful.
Kind Regards

zilonglee

@mtmtmt What I know is ghost boundary condition having no problem with gap. But I am not sure whether DualSPHysics would implement this boundary condition, and it can be very helpful to your problem if the developers have the plan to implement it.

mtmtmt

Dear @zilonglee Thanks For Your Attention,Your Advice Was very helpful.
Kind Regards

kevinxmu

@zilonglee Is that complicate to implement the ghost boundary condition? I encountered similar problems for the floating body / fluid interactions. This is expected to be fixed in version 4.4, but not sure when 4.4 can be expected.

zilonglee

@kevinxmu If you are quite familiar with C++ and DualSPHysics, you can try to implement it youself. A good reference paper is "A generalized wall boundary condition for smoothed particle hydrodynamics" (S.Adami, X.Y. Hu, and N.A. Adams).

dong

@zilonglee I have a question that if I implement the correction of boundary and remove the gap successfully , can I get more accurate result with the same value of dp?
Because now I have to set a small dp to get accurate result with so much time.
If the correction can improve the accuracy, it is so great!
regards

zilonglee

@dong Theoretically it will get more accurate result with the same value of dp after you implement the correction of boundary.

dong

@zilonglee Thanks for your reply. That is great.But the correction is a little difficult for me.
Best regards
dong