open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,there
I have done an experiment about temperature stratification,this stratified is mainly caused by convection. The results are as follows:
t=10min


t=70min


Now .I am trying to simulate it by TELEMAC-3D ,and the results are as follows:
t=10min


t=70min


In my simulation result ,there is no stable stratification and termocline. I set the turbulence model as k-epsilon .Compared with experiment results,the vertical diffusion is faster while the horizontal diffusion is slower . I hope someone could help me solve this problem.Thanks in advance!

Best regards,
Andy

Hello Andy

The stratification is a tricky phenomenon and needs lots attention in order to be correctly obtained. I have a glance on your case, and you can start by increase the accuracy odf tracer's solver to 10-10 for instance, tehn you have to play with the vertical transformation of mesh and the number of planes and their vertical distribution.
then you can test more options for vertical turbulence model.

with my best regards
Riadh

Hello Riadh


Thank for very much you advice!
Just as you said:firstly, i increase the accuracy of tracer's solver to 10 -10;secondly ,i regenerated the mesh and vertical distribution of mesh.In order to get horizontal mesh plane,I simplified the side wall to be vertical.And i redefine the VERTICAL VELOCITY PROFILE and TRACERS VERTICAL PROFILE in my fortran file (see file andy.f).





The result are as follows:
t=10min




Compare with previous simulations, this result is much more better.But the result around the inlet and outlet is not so satisfactory ,i found the vertical velocity component around the inlet areas is weird(in some point the vertical is too big).
vertical velocity component around the inlet areas:



vertical velocity component around the inlet areas and the mesh:



I am confused why there is so strange vertical velocity around inlet,i hope you can give me some suggestion.

With my best regards,
Andy