open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear Qilong

I read with great interest your questions and comments. As we are now working on a quite similar case, I would be happy to get your (final?) conclusions about the choice of BC for k and epsilon at the free surface.

Another point about the plots you sent, your legend indicates "Initialization using ML theory". Did you initialize the theoretical ML properties in the whole channel, or in that for you entrance boundary conditions, or anything else ?

Thanks a lot,

Damien

Dear Damien,

Thanks for your interest in my test cases on k-epsilon model.

We tested the behavior of k-epsilon model especially near the bottom region and we found the non-uniform distribution of vertical layers are not idea in terms of getting good profile. Instead, a uniform distributed sigma layers seem to be better.

If you're interested, you can find some details about my tests in the attached document (www.dropbox.com/s/5u6nnj1nu7mbi2s/K-epsi...-3D%202016.docx?dl=0).

Hope this helps!

Kind regards,
Qilong

Thanks, that helps a lot ! We are working on quite similar cases, I will keep you in touch.

Damien