open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I'm trying to run a simulation but the velocity results are no good. I have velocity value too high.
Please, someone can help me to understand the problem?
In attach the slf cli.cas.

Thanks so much

Cristina

Well for one thing, you should set
SPHERICAL COORDINATES = YES
SPATIAL PROJECTION TYPE =3


Hope this helps... Martin

Hello,

Looking quickly at your steering file, why do you use DIFFUSION OF VELOCITY = NO ? Have you changed subroutines other than CONDI in your steering file?

When you talk about high velocity, what do you have and what do you want?

Anyway, you can calibrate velocities with the keyword COEFFICIENT TO CALIBRATE TIDAL VELOCITIES (please read the user manual).

Hope this helps,

Chi-Tuan

Thanks so much to help me.
I done that you suggested to me, but still I have velocity value no reasonable. In attachment the new cas file.
When I told that I have high velocity, the problem is that I have velocity around 40m/s just in one point along the boundary (look screenshot), instead of velocity value around 2 m/s

Best regards,

Cristina

Hi Cristina,

Please look at:

LAW OF BOTTOM FRICTION = 2
FRICTION COEFFICIENT = 180

Because bottom friction coefficient can be used as a calibration parameter. You can adjust it to achieve a reasonable result.

I usually use Manning, so I would suggest you, for example:
LAW OF BOTTOM FRICTION = 4 (manning)
FRICTION COEFFICIENT = 0.02

To see if you can get lower velocity.

Also, you are running a simple case, but you put many parameters in your steering file. For instance: COEFFICIENT TO CALIBRATE TIDAL VELOCITIES = 999999. make it simpler.

You can try: COEFFICIENT TO CALIBRATE TIDAL VELOCITIES = 0.9

Hope it helps

Best Regards

Huy

Hello,

Your problem seems similar to the one on the slide 13 of :
www.opentelemac.org/media/kunena/attachm..._workshop_121017.pdf

Did you try the local solution Gulf of Mexico rather than the regional Atlantic Ocean solution? Do you have the same behaviour for the right upper corner? This may happen when the water depth calculated at the boundary conditions with the tidal database is small or equal to 0. As the 2nd binary database for tide contains the water depth times an horizontal component of velocity (= transport), to get the horizontal velocity, the operation is to divide the transport by the water depth.

Anyway, I have had a quick look at your mesh and close to the node which coordinates are around (-60.38 ; 43.96), the size of the elements are a little bit too different e.g.. The ratio between two neighbor elements should not be around 2. You may have to generate a new mesh again. Some transitions of mesh sizes in a few areas may be improved.

Moreover, to model tide, you should start with Finite Elements schemes (EQUATIONS = 'SAINT-VENANT EF') like the method of characteristics (TYPE OF ADVECTION = 1;5) which is robust and stable, SOLVER = 1 when using the pseudo-wave equation (TREATMENT OF THE LINEAR SYSTEM = 2), ADVECTION = YES (default value), SPHERICAL COORDINATES = YES, TIDE GENERATING FORCE = YES may be useful.

Do not forget that if you have the same keyword a few times in your steering file, it is the last one which is taken into account.

Hope this helps,

Chi-Tuan

Hello Hui,

I edit my case file how you advice to me, I got better results, but I still have high velocity along the boundary side


Hello Chi-Tuan,
I also tryed Gulf of Mexico rather than the regional Atlantic Ocean solution, but it doesn't work because my domain is bigger than Golfo of Mexico.
I'll fix the mesh, but in the main time I changed the case file how you suggest,
so I set (EQUATIONS = 'SAINT-VENANT EF') , (TYPE OF ADVECTION = 1;5), SOLVER = 1 , (TREATMENT OF THE LINEAR SYSTEM = 2), ADVECTION = YES (default value), I also set VARIABLE TIME-STEP = NO becouse I need to run with a fixed time step.
But the error that appears is in attach.
I'll fix the mesh and I'll let you know.
Maybe there is some problem on the tidal condition that I set? some thing wrong on the boundary conditions or cli file?

also with
SPHERICAL COORDINATES = YES, dosn't work.

Thanks so much for your help,

Best regards,

Cristina