Learning Forum

[h395523899]

I am studying the ddpm model recently.

I built a model with a size of 200*100*10mm. The simulated water carries the flow of sand.

My settings are as shown below, the simulation step size is 0.0001, but the simulation process always goes wrong, is it because my step size is too big?

[Rob]

Can you elaborate on "goes wrong"? 

[h395523899]

[Amine Ben Hadj Ali]

Try to start conservative and use algebraic granular fluctuation

[h395523899]

Do you mean the following settings?

[Amine Ben Hadj Ali]

Yes. Euler Granular might be a cheaper alternative.

Ensure that on start you are not at the friction limit.

[h395523899]

After the settings you said, the problem still exists.

 

 

[Amine Ben Hadj Ali]

Can you try with smaller particle time step size?

[h395523899]

As shown in the figure, do you mean that the parameter at mark 1 is adjusted?

My understanding of the two markers marked is:

1 represents the number of iterations of the particle phase on each fluid iteration step

2 indicates the time step of particle injection

I don't know if my understanding is right.

[Amine Ben Hadj Ali]

Settings 2 to use smaller particle time step size than the flow time step size. By the way have you tried using Euler Granular instead of DDPM? That is what I recommend especially if you are near packing limit and particles have single size.

[h395523899]

I have used Euler Granular before, but I didn't get the results I wanted.

My aim is to simulate water carrying sand.

No sand water and sand water are alternately injected. But the simulation result is that the sand is piled up together. The actual situation in the model should be a section of sand, a section of no sand.

[Amine Ben Hadj Ali]

Good Luck.