Test filter
 

Hello

I'm trying to implement a Wale model for which I need to use a test filter.

this model is implemented using a VMS approach where the scales are divided into large, small and unresolved. The original filter separates the unresolved scales, the second filter separates the Large and small scales in the following manner:

f'=f_bar-f_hat

where f' is the small scale, f_bar the original filtered and f_hat the large scale.

The way I'm trying to implement this is by creating a new vector field like this:

volVectorField up = U()-(filter_(U()));

however i'm not sure how to make it so the filter size is twice the original filter.

thanks in advance.

Look at how the filter operation is defined in the filter function, or in the dynLocOneEqEddy model, which uses double filtering, to have a better idea of how it works :-)

Hi

I can see in locDynOneEqEddy that a second filter is defined using the line

simpleFilter simpleFilter_;

however, I'm still not sure where the filter lenght is defined.

Hey David. Were you able to successfully implement the WALE model? And was it the incompressible version? I am currently trying to implement it.

Cheers,
Deji

Hi deji, I did implement the WALE model, I based the code on the dyn smagorinsky model and modified it accordingly, however i didn't implement the VSM approach on the model.

Oh really. Can I have a copy of the code? I implemented it as well, but not the way yours was coded.


Cheers,
Deji

LES WALE incompressible model
 

Dear Deji,
Could you please send me your LES WALE incompressible model? I would like to see the impact of LES WALE predictions for my wall bounded flows.
Regards.
M K

Dear Deji,
Could you please send me your LES WALE incompressible model? I would like to see the impact of LES WALE predictions for my wall bounded flows.
Regards.
M K

Hello MK. The code is on the forum somewhere, just do a search for WALE model. I did upload the code on one of the threads and if you cannot find, I can send you another.

Cheers,
Deji

Hi Deji,
I can see that you have mentioned in your post that it is compressible. I am wondering if it is also suitable for incompressible.
M K

It's essentially the same code, you just have to make some minor changes to the code. For example, we use mu_sgs but your your code will have nu_sgs; therefore, go through the code and make the appropriate changes. If you have anymore questions, kindly inform me.

Cheers,
Deji