open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear moderators and Telemac users
I ran many simulations with different number of sources : from 1 to 235 points. Errors occur in the case of more than 15 sources . These error grow up with the numer of sources (0.2 m3/s for 15 sources and -10 m3/s for 235 sources).
Does there any explanation? I set the solver accuracy to 1.E-8.
Any remark or suggestion is welcome
Thank you in advance
Mourad

I wasn't precise in my question. I mean each listing step TELELAMC2D posts the Water balance (fluxes up and downstream Q(1) and Q(2)and additional volume due to the sources Vsce).

Logically Vsce should be equal to the total discharge of the sources times the time step : Vsce = dt * sum(Qi).
In my case, I have the discharge of each source point and its evolution thoughout time. I compared the total volume of sources (Vsce) which I computed and the one posted by TELEMAC on the screen each listing of the Water balance.
when source points number is lower than 15 , TELEMAC displays the right value of Vsce (accuracy : 10^-5)
However Error commited by TELEMAC increase with the number of sources: from 0.2 m3 (Nsce = 15) to 10m3(Nsce = 235).
Greetings

Hello Mourad

Yes the total volume computed by Telemac includes :
- initial volumes
- water in (with positive discharges)
- water out (with negative discharges)
- added water from sources (with positive sources)
- sinked water from negative sources.

There is no reason that the balance is not verified since there is no numerical losses.
Please verify your own computations with what is done in subroutine bilan.f (sources/telemac2d) and then if you can send me your case and I will check what is the problem
with my best regards

Riadh

Thank you Riadh for you support.
Please find in the attached the following folder (zipped with winrar) :
1-SHARED : GEO.slf, CLI.cli, LIQUID and POROSITE.f files. These files are common for the all the simulations : copy/paste them in each simulation folder.
2- TEST_10, TEST_20, TEST_50, TEST_100 and TEST_200 : folders were CAS files are stored and eventually sources files too. For example, TEST_50 deals with the simulation of 50 sources points.
3- RESULTS : results of simulations and their log files (i.e for 10 source points, you will find R2D_10.slf and R2D_10.out).

And now TELEMAC2D resutls :
Each source point has a constant discharge of 0.2 m3/s. Note that time step dt = 1s
10 sources : Volume added by sources = 2 m3 , this is correct cause 0.2*10*dt =2m3
20 sources : Volume added by sources = 4 m3 → this is correct too.
50 sources : Volume added by sources = 8.16 m3 → this is false : 0.2*50*dt = 10m3
100 sources : Volume added by sources = 17 m3 → TELEMAC must find 20 m3
200 sources : Volume added by sources = 34 m3 → TELEMAC must find 40 m3.
Finally, the error grows up with the number of sources.
I have checked both bilan.f and difsou.f but still have no idea about why telemac did these errors on the volume added by sources in each time step.
Riadh, any remark or indication from you will be appreciated. Hope find the mistake my be in my steering file or in difsou.f

Here the results folder

The results files are too big, I can't send them to you however I sent the log files.
Greetings

Results

Hi Riadh,
I sent you the results and all the necessary files to run simulations on your computer. Did you see the error telemac made when computing the added water volume due to sources ( during a time-step) and how it increase with the numbre of sources? Did I make mistakes?
Thank you in advance,
Greetings