open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I'm trying to run a simple steady state simulation.
It seems that my simulation is running but I don't get any fluxes into or out of the domain. I'd read some threads of the problem, but there isn't a good advice for my simulation:

FINAL BALANCE OF WATER VOLUME

RELATIVE ERROR CUMULATED ON VOLUME: NaN

INITIAL VOLUME : 1646114. M3
FINAL VOLUME : NaN M3
TOTAL VOLUME LOST : NaN M3
CALLING UTIMP_TELEMAC2D
BACK FROM UTIMP_TELEMAC2D

END OF TIME LOOP

CALL OF P_EXIT IN ITS VOID VERSION

CORRECT END OF RUN

ELAPSE TIME :
13 MINUTES
42 SECONDS


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

... handling result files

+> steuer.cas
moving: Ergebnise_Test_Janet


My work is done

The simulation should have a steady discharge (Q=324) at the inlet. The outlet has a waterheigt-discharge-relationship. Please look attached at my files for the simulation.
I hope someone could help me.

Sorry find attached my liquid boundaries and geometry file

Hello,

I tried to run your simulation and I arrive at the same. I changed a couple of things made it steady-state, as i'm not familiar with the stage-discharge. It seems to me you have your boundaries inverse. I used the following

PRESCRIBED ELEVATIONS =310.99;0.0
PRESCRIBED FLOWRATES =0.0;324.0

Also you specify constant depth, which means a constant layer of water on top of your bathymetry, with water elevation following your bathymetry. I suggest you use constant elevation same as your downstream boundary (311.0).

If you do this along with a variable time step you can create a hot start for your liquid boundary file and stage-discharge curve.

Hope this helps,
Patrick

Thank you for your help.
I tried your suggestions. The simulation lasted one second and I just have a picture not an animation. When I opened the results file in BK I could see th depths for the time step but at the free surface data only the highest elevations are seen. The water is completly distributed in the domain, how can I force the water to stay in the channel?
Please, can you describe me what do you mean by a hot start for liquid boundary and stage-discharge-curve? You can find my new steering file attached.
Please look attached at the output of TELEMAC:

---

BALANCE OF WATER VOLUME
VOLUME IN THE DOMAIN : NaN M3
FLUX BOUNDARY 1: NaN M3/S ( >0 : ENTERING <0 : EXITING )
FLUX BOUNDARY 2: 178.2000 M3/S ( >0 : ENTERING <0 : EXITING )
ABSOLUTE ERROR IN VOLUME AT T = 1.000 S: NaN
CALLING UTIMP_TELEMAC2D
BACK FROM UTIMP_TELEMAC2D

END OF TIME LOOP

CALL OF P_EXIT IN ITS VOID VERSION

CORRECT END OF RUN

ELAPSE TIME :
21 SECONDS


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

... handling result files

+> steuer.cas
moving: Ergebnise_Test_Janet


My work is done

Hi,

Not sure I can help you all the way, because I think this is related to your numerical settings. The quick test I did was actually with numerical settings from simulations I have done myself.
Find attached my adopted steering file and liquid boundary file. In the liquid boundary file you will see the with discharge is 0 and then goes upto 324m3/s in 5min. I do this because the water in the domain is at rest and this is reducing the shock to the system when you start of the simulation, you might want to increase this time.
Hope this helps

here the liquid file just change they extension to .liq

Now I can see some fluxes thank you, but why ?
You don't used the the relationship between elevation and discharge at the oulet or am I wrong? The elevation at the outlet is set to 307.7 m. Isn't there a possibility to use this relationship?
I increased the duration to 1000 to see more iterationsteps and to see the max level of the discharge.

If you go to 1000s, you will see velocities slowly increasing in the domain. At the end of the simulation the discharge at the downstream boundary will not be at 324 m3/s, but lower (around 100, I did a simulation of 900s).

If you want to continue your simulation after I suggest you use the results file as a 'hot start'.
in your cas file change the name of the results file and add the following 2 lines:

PREVIOUS COMPUTATION FILE = 'Ergebnise_Test_Janet4.slf'
COMPUTATION CONTINUED = YES

Comment the following 2 lines:
/INITIAL CONDITIONS ='CONSTANT ELEVATION'
/INITIAL ELEVATION =307.7

Now the simulation should start from 1000s onwards.

I didn't use the stage-discharge file because:
1- you have a constant discharge in your liquid boundary file
2- the model is at a stationary initial condition, this will introduce a lot of complications.

To answer your other question about water, in Telemac the water surface is always higher then the bottom. So the surface will follow topography on 'land' where it's dry. There is no flow in these cells.

Thank you very much for your help!
Please can you tell me what was wrong with my steering file and which keyword is the correct one to start the simulation?

I read the manual again and i found out, that you set the courant number to 0.9 because of the manual's recommendation. My simulation with the duration of 1000 iterations is at iteration 310 and the flux of boundary 2 is about 324,0. Why are the flux at boundary 1 negative?
I checked the steering file, isn't it possible to set the prescribed elevation to 307.7 because the elevation of 310.99 is for case of Q=1765?

Hi you're welcome,

All questions we go through when we start using TELEMAC (I did myself too). Hopefully my answers clear things up a bit.

Boundary 1 is negative because water is exiting the model, this one should eventually reach -324 m3/s so it equals the upstream discharge (Boundary 2 in your case).

prescribed elevation in your steering file is used when no liquid boundary file is specified. To start the model you would have initial elevation and prescribed elevations (or the value in the liquid boundary file) at the same. You can prescribe higher initial elevation, the boundary will start draining the filed from the beginning, this might help speeding up finding steady state.

I didn't get this question:
- Please can you tell me what was wrong with my steering file and which keyword is the correct one to start the simulation?

You mean you tried to 'hot start' after 1000s? Upload your cas file.