open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I want to customize the initial suspended sediment concentration when using GAIA, for this I modified the user_condi3d_trac.f. My intention was to introduce an external txt file(named data2.txt here) to define the initial tracer values in the full calculation domain, and the number of TA values in this txt file has been set to the same number as my grid.
But when debugging, the following Segmentation fault occurs. I think this is caused by the wrong location of my data file.
At the moment I put it in the upper level of user_fortran to call, but I am not sure where I should put the txt file to run successfully. Or maybe I am using it this way is wrong, please help me out.

Hi, everyone,
I made some changes to my program to remove the use of "END", the program no longer has the SIGSEGV problem, but there is an iteration limit problem, does anyone know the solution to this problem? Thank you!shihangliu wrote:

Read the user manuals particularly the parts on formatted file. THis will allow you to manage your file in a simpler way
Regards

Hello coulet,

Thanks for your reply, I read the mteo.f file in the wind_txy case and modified my program after its read-in, but a new error appeared: MURD3D: ITERATION NO. REACHED 100 , STOP. just like the file I uploaded in the second post above (#42232).
I wasn't able to find a description of the relevant keywords in the user manual, or should I find the fortran program that changed the error message in the sources/gaia folder?
I upload the error again below.

The problem is not in the file but probably in the values of the initial suspended sediment concentration...
Before more investigation, try to test your program with 0 value at every node (like a default initialisation and see if the computation starts well.

Hi coulet,

If all the initial values are set to zero, it is possible to run. When I add values, the program will stop unless all the values are small (close to 1e-100). Currently, the calculation can only be done by entering the distribution function into the user_condi3d_trac.f, but this is still limited and not my original intention.

I don't know what kind of initial value you would like to impose but as there is a problem when you impose your values, this mean there is probably a consistency problem in your values.
As this is a 3D computation, think the points are numbered by planes, starting at the bottom...
Hope this helps