Variation of the Smooth coefficient h
Hello everyone,
I have a question concerning the variation of the coefficient h in function of a zone.
I explain you why this question:
- I study at the moment a tank (10 m*10m*10m) filled with water, linked to a canal (L=10m and l=1m) where the water flows into and goes out into a second tank.
- using of GPU resources and with a dp of 0.01
- In the lateral wall of the canal, there are several little holes (10 cm) and in the reality the water should go through those holes but it's not the case.
As you can see the problem has size scales quite large.
Solutions:
- Impose a certain pressure; for example by adding a lateral slope
- close the canal.
However I loose all the interest of the method because I use it to study complex geometries and I have to do geometric simplifications of the problem to have some results...
- reduce the dp but it extends a lot the time of calculation.
- Another solution would be to define several coefficient h for different bunch of particles which are supposed to go into specific zones.
More generally, the question is: is it possible to vary the coefficient h with DSPHysics?
Thank you.
Thomas
I have a question concerning the variation of the coefficient h in function of a zone.
I explain you why this question:
- I study at the moment a tank (10 m*10m*10m) filled with water, linked to a canal (L=10m and l=1m) where the water flows into and goes out into a second tank.
- using of GPU resources and with a dp of 0.01
- In the lateral wall of the canal, there are several little holes (10 cm) and in the reality the water should go through those holes but it's not the case.
As you can see the problem has size scales quite large.
Solutions:
- Impose a certain pressure; for example by adding a lateral slope
- close the canal.
However I loose all the interest of the method because I use it to study complex geometries and I have to do geometric simplifications of the problem to have some results...
- reduce the dp but it extends a lot the time of calculation.
- Another solution would be to define several coefficient h for different bunch of particles which are supposed to go into specific zones.
More generally, the question is: is it possible to vary the coefficient h with DSPHysics?
Thank you.
Thomas
Comments
Please send to us (dualsphysics@gmail.com) the XML case and we will check it.
And about your question... yes it is possible to use a variable h, but the formulation becomes more difficult and we have not found the need on our applications yet to start to implement it. But you can find many information on published papers.
Regards