open TELEMAC-MASCARET The mathematically superior suite of solvers

HI
I do a simulation by telemac 2d, using data from an experimental test in a rectangular channel with uniform flow regime (I took this simple example to test a model). The problem at the channel entrance, I find water depth results are far from measured results(the measured values are between 0.09 and 0.07)), I changed the velocity diffusivity, the results are improved but still far from the measured, I didn't understand why.
the friction coefficient is taken based on the expérimental tests with uniform flow
in the attached document you find the results I found ainisi the case file.

thank you in advince

Hello,

You have a Strickler coefficient of 37, which corresponds to a grain size of about 11 cm, higher than your depth of 0.07 m, so there must be some misunderstanding on the real friction coefficient to be applied in your physical model.

With best regards,

JMH

P.S. my formula (Strickler's actually...) is:

grain size in metres = (26/Strickler)**6

Hello JMH
thank so much for your answer

I repeat the calculation with ks= 39.73 but I didn't find a good results and this the results , the level of the water stay high in the inlet of the channel !!!

Hello,
you prescribe a water level of 0.0785 m for the outlet. If the bottom height there is 0.0 something must be wrong according to your picture.
You should try to set OPTION POUR LES FRONTIERES LIQUIDES = 1

Hope this helps,
Clemens

Hello Clemens
thank you for your suggestion, I changed the OPTION POUR LES FRONTIERES LIQUIDES = 1
but i didn't find a good results, I repeat also the mesh , I didn't understand the problem

maybe should change the the velocity diffusivity or the turbulence model !!!

best regards

MORRI

Hello,

I don't think that changing the turbulence model or the eddy viscosity will give you much other results.

You said that in the experimental test there was uniform flow regime. This means that your prescribed water level at the outlet must correspond to the normal depth.
I can only advise to calculate, based on the roughness, width and slope, the normal depth, the critical depth, and using the Bernoulli equation along with your boundary conditions also the water surface line. This will maybe bring some insights into your model.

Hope this helps,
Clemens

Hello,

Please look at my formula : grain size in metres = (26/Strickler)**6,
your new value of Strickler (39.73 ???) corresponds to a diameter of sediment of 7,85 cm, it changes nothing compared to 11 cm, is it the size of the sediment in your physical model ? Try Strickler = 100 and see what happens.

Regards,

JMH

Hello Clemens and JMH

Thank you so much for your important suggestions
Clemens, I did all the calculation (I calculate the hn) .
JMH, thank you forthis formula I used it to determine the roughness coefficient, in my case I don’t have a sediment but I have a rigid vegetation (the high of these cylinder which was installed in the bottom of the channel was 0.015m), I take ks equal to 52.35.
I repeat the mesh and used the data based on your comments and I find that the problem in the choose of the mesh size when I used all the data with the pervious mesh I didn’t get a good results, when I changed the mesh size, I find an amelioration in the results

Thank you so much
Best regards
MORRI