open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I am trying to extract a temporal subset from a hydrodynamic file with BlueKenue.

My initial file is the result of a Telemac simulation, with, say, 500 frames, and I would like to use frames 200 to 500 to recreate a selafin hydrodynamic file that could be used for SISYPHE (uncoupled).

So far, I have extracted (with calculator) the subsets for each of the variables velocity U, velocity V, water depth, bottom and free surface, and created a new slf with these variables.

When I try to launch Sisyphe however, using such a file as 'HYDRODYNAMIC FILE', the program stops, as if not reading properly the hydrodynamic file. (see message below).

I am not sure if the problem is in the generation of the file, or in the reading by Sisyphe.

If anyone has some advice about this, it will be more than welcome!

Thank you,

Caroline

Logfile from Sisyphe computation:

...
TITLE OF PREVIOUS COMPUTATION: newSelafin

NAME: VELOCITY U UNIT: M/S
NAME: VELOCITY V UNIT: M/S
NAME: WATER DEPTH UNIT: M
NAME: FREE SURFACE UNIT: M
NAME: BOTTOM UNIT: M

SUITE_SERAFIN : READ OF RECORD 679


TIME OF RECORD: 610000.0 S

VARIABLE : VELOCITY U M/S
IS IN THE FILE BUT WILL NOT BE READ

VARIABLE : VELOCITY V M/S
IS IN THE FILE BUT WILL NOT BE READ

VARIABLE : WATER DEPTH M
IS IN THE FILE BUT WILL NOT BE READ

VARIABLE : FREE SURFACE M
IS IN THE FILE BUT WILL NOT BE READ

VARIABLE : BOTTOM M
IS IN THE FILE BUT WILL NOT BE READ
SORTIE DE BIEF_SUITE
LECTURE FICHIER HYDRODYNAMIQUE:

CAUTION : THE PERIOD OF COMPUTATION IS NOT A MULTIPLE
OF THE HYDRODYNAMIC FILE PRINTOUT PERIOD.
THE LENGTH OF COMPUTATION WILL THEREFORE BE -0.000000
SECONDS
BIEF_SUITE

SUITE_SERAFIN : READ OF RECORD *****
LIT : ABNORMAL END OF FILE
ONE INTENDED TO READ
A RECORD OF 1 VALUES
OF TYPE : R4
ON LOGICAL UNIT : 5


PLANTE: PROGRAM STOPPED AFTER AN ERROR
RETURNING EXIT CODE: 2
Duree du calcul : 2 secondes ( 0:0:2 ) (systeme=0.015 sec)

Hi
From the error message, it seems that the result file is not correctly recognize by Sisyphe. the velocity and free surface values are in the file but are not read.
Perhaps a problem of language?
Another possibility is to use Fudaa-Prepro to open your result fil and export it to another result file. In the dialog box, you could choose the variables and also the different time steps you want to extract. You don't have manipulate the variables of the result file in order to create a new selafin file.

Hope this helps

Hi,

I am trying to open my result file with Fudaa but I am very familiar with it: should I open the file with "analyser fichier de resultats" or with "postprocesseur rubens"?

Then, concerning the names of the variables, I just wanted to add that, whenever the name or unit of the variable was not typed properly, I got another type of error message from sisyphe that said "unknown variable". So I thought that "is in the file but will not be read" was actually quite encourageing.
I will try with the nomvar file though.

Thank you!

Caroline

HI
With fudaa, you just have to select the result file and choose Open (POST).
In the new screen, if you select the 2Dview of the result you could export it (you have to select the variables and the time step).

Regards

Hello,

I am not at work to check everything but in view of the listing, it seems that the name of variables has been changed by BlueKenue. This name should be in 32 characters including the unit: 16 for the name, 16 for the unit. It looks like what is in the file has been somewhat compacted, and it is not recognised when compared to the true name. One possibility would be (perhaps) to take nomvar_sisyphe.f in your Fortran file. There are two sets of names there, those for previous files, and those for files that are being written. It could be that by changing the names in one set in the English section it will be read correctly (I am not sure however, it depends on how the geometry file is read).

If I'm correct on BlueKenue changing spaces in the names, this is also a message for Martin Serrer...

With best regards,

Jean-Michel Hervouet

Blue Kenue writes the variable names and units into 32 characters.
16 characters for the name and 16 characters for the units.
Each of the 16 character strings are padded with ASCII space character 0x20.
I just confirmed this in the source code and by creating a new selafin file from the contents of an old geo file using Blue Kenue and then comparing the files with a WinHex. (A binary file editor)
The two files are identical.
There is something else going on here.

Can you post the offending file here and I'll have a look at it.

Martin

Hello,

I have tried to post the file but it seems to be too large, is there another way for me to send it to you?

Another detail about this: When I use this file as a "previous computation file" for Telemac, it seems to be read properly. So the problem seems to be related with Sisyphe.

Thank you,

Caroline

Hi Caroline,

If TELEMAC reads the file successfully then (as you say) I'll have to assume that it's a Sysiphe problem and Christophe or Jean-Michel and co. will have to look at it.

However you could try putting it on our ftp site.
ftp://kenueftp.chc.nrc.ca/incoming/Caroline

Martin

Hi Caroline,
Did you menage to solve the problem? It seems I have the same error message with passing tel3d_v6p1 results to sisyphe.


Regards,
Marcin Kawka

Hi Marcin,

No, I did not solve the problem.

What I eventually did is that I used the whole hydrodynamic file, but I modified the sisyphe function in order to start the bedload calculation only at the time I wanted (i.e. at the time corresponding to the beginning of the temporal subset that I did not manage to extract)
Have a look at message 3159 if this solution can help you.

Another thing that I have noticed is that, in your log file, you have the message "Variable ... unknown, will not be kept". In my case, I had such a message when the name of the variable was not typed properly, or when the unit was not written in upper case. Otherwise, I had the message "Variable ... is in the file but will not be read". So perhaps you could check the spelling and case of your variables?

Sorry not to be able to help,
Caroline