ALE restart option

[mehdi.04]

Hello,

I'm trying to use Saturne for a FSI problem.

To avoid "wasting" time at each simulation initializing fluid I tried to use the restart option, so I launch the study without using the ALE module. When the calculation is done, I add the ALE module ( mesh viscosity and a small number of steps for fluid initialization...).
When the configuration of the paramaters is done, I launch the calculatioin using the restart option, but what I get in the listing file and the output result is that the mesh viscosity isn't considered ( it's 1 in all the domain). I tried different cases, different meshes... I got the same result.

Is it normal ? what could be the problem ?

Thanks

[Xepls]

Hello.

You need start your case with ALE necessarily.

Other limitation in case of restart, is the number max structure is limited to 99 (I isnt sure that you can use 100)

if you dont like that your structure is moving, only put all force in the structure with 0.

[mehdi.04]

Hi,

Thank you for your answer, I tried launching the case with ALE, with setting a number of iterations for fluid initialization equal to the total number of iterations, so the mesh doesn't move in the first calculation, but when I launch the 2nd calculation ( refining the time step, and changing the output) I get an error, which you can see in the listing file

listing.txt

(19.81 KiB) Downloaded 264 times

.

ps: I didn't set a 0 force on the structure in this case.

Best regards,

[Xepls]

Hello.

I thinks that you forgot to select the checkpoint folder. I can see that you dont generate anything restart file (4 restart file for default in CS, search in your result folder if is)

Or maybe you dont select correctly checkpoint folder, only you have that select this folder (no file).

I recomend you to read this tutorial (page 65).

Remember setup more iterations in the new simulation (greater old simulation).

[mehdi.04]

Hello
Thank you for your answer, that's one of the 4 problems I get when I use the restart option :
- CS doesn't read the checkpoint file.
- CS reads it but can't read the mesh coordinates even though the mesh hadn't moved yet (ileaux = 0)

listing_ileaux.txt

(19.81 KiB) Downloaded 235 times

- CS gives me an error that I have 200 structures, but in my case there is only one structure.

listing_NFMTST.txt

(68.48 KiB) Downloaded 231 times

- CS reads the checkpoint file but doesn't use the mesh_viscosity that I applied.

listing_visc.txt

(62.92 KiB) Downloaded 259 times

I'm using the GUI, so can anyone see what's wrong with my inputs ?

Regards,

[Yvan Fournier]

Hello,

The message about the structures seems to be just a warning.

For viscosity, this seems strange, but may be related to sub-options. Do you have a small test case you could post here ?

Regards,

Yvan

[mehdi.04]

Thank you Yvan for your help, as you can see in the listing file, it shows the warning, the calculation is launched, but with default values for mesh viscosity, the one with ILEAUX, I don't really understand why CS can't read the mesh nodes.

Here I have the case I'm working on, even though it's not really the same as the one in the listing here (I changed the domain and the mesh) but I get the same problems on it and with/without restart option, I get a negative volume when the ALE starts.

Is it all you need ? or you need the mesh too ?

Regards,

[Yvan Fournier]

Hello,

I'll need the mesh to do any testing on my side.

Regards,

Yvan

[mehdi.04]

Hello,
Here is the mesh I used, and the case I'm working on.

I uploaded it on Wetransfer because I couldn't post it here

https://we.tl/FHUS9dc8HU

Regards,

Mehdi

[Edjuanma]

Hi did you resolve your issue?

Something similar happen to me with the ILEAUX. To solve it, turns out in the GUI you have to go to the Start/Restart menu and in "Advance Options" check the option "Read the auxiliary restart file at the begin of the calculation"


Regards

Edgar,