open TELEMAC-MASCARET The mathematically superior suite of solvers

I am a new user of Telemac. I ran my simulation for a river model and after 4 hours, the simulation failed with an error=-1. I have searched hard but I can't find the meaning of this error. Has anyone encountered a similar error? I put a capture of the error in attachment.

Best regard

The number -1 is not really important.
With only these capture I would say the problem is a parallel problem but hard to say more without any other information.
Seems one of the host has a problem (not necessary the master) so you should also investigate in the listing of the slave processor...
Hope this helps

Hello Christophe,

I finally fixed my error by reinstalling Telemac and it works.

Thanks for the feedback

Hi Carole,
I think I may have encountered a similar problem in the past, this could be due a loss in Internet connection whilst running a simulation in PARALLEL. Try running a much shorter simulation to see if parallel processes work fine (getting the my work is done message). If everything seems fine then you can always merge result log files and use the resulting selafin file as previous computation file, in case you don't want to waste that 4 hour simulation. Hope this helps.

Cheers,
Nayef

Hi Nayef,

You may be right because it is a parallel simulation i was doing. Thanks for the information because I didn't know that parallel simulations required a working internet connection. But how to merge result log files and use the resulting selafin file as previous computation file to save time?.

Thanks for the feedback

Hi Carole,

There are several ways to merge failed simulation files, the most forward one would be copying the executable gretel.exe -attached in this post- to the folder where failed simulation files are (usually in the same directory as your steering file), then by running the executable you will be asked to enter the name of geometry file, boundary conditions file, number of processors... etc. The result of this process will be a selafin file of your simulation. Attached figures show the failed simulation folder and what you see when open the executable. Note that the name of geometry file, boundary conditions..etc should reflect the names in the failed simulation folder (not the original name of your cli or slf files), so the name for your boundary conditions file would be T2DCLI if you're running a telemac2d simulation. Hope this helps.

Cheers,
Nayef

Well I can't post exe files up here, but you can find all executables in the Builds folder(E.g., C:\opentelemac-mascaret\v8p0r0\builds\wing64mpi\bin)

Cheers,
Nayef

Thank you Nayef my simulation is running well. However in my results my courant number exceeds 1 even if I prescribe a value of 1 with the command DESIRED COURANT NUMBER and the type of advection is 1;5. Is there a method to limitate the courant number to 1?

Hi Carole,

I am not particularly an expert on this topic, but reading through the manual you can find the following:

"The stability of the N and the PSI scheme (type of advection 4 and 5) is conditioned by a Courant number lower than 1. When using these schemes, TELEMAC-2D verifies the Courant number for each point at each time step. If the Courant number is greater than 1, the software will automatically execute intermediate time steps in order to satisfy the stability condition. However, if the number of sub-iterations reaches 200, TELEMAC-2D will consider that solving the advection term is no longer possible and the computation is stopped with an appropriate error message printed in the output listing."

This suggests that if the simulation is working then the problem has been taken care of, why do you think your results were affected by this? you can try setting the keyword DESIRED COURANT NUMBER to 0.9 if problem still exists..

Cheers,
Nayef

Hello Carole,

If you use default choice EQUATIONS = 'SAINT-VENANT FE' (Finite Elements, not Finite Volumes), the keyword DESIRED COURANT NUMBER is only taken into account if VARIABLE TIME-STEP = YES.

Anyway, using VARIABLE TIME-STEP = YES with Finite Elements does not always work correctly in TELEMAC-2D.

I would first suggest you to decrease the time step if your results are not good (but using characteristics i.e. TYPE OF ADVECTION = 1;5, there are less constraints with Courant number).

Hope this helps,

Chi-Tuan