open TELEMAC-MASCARET The mathematically superior suite of solvers

Hey @ all,

after couple of hours of mesh creation process i finally got my first T2D-model to work. Right now my problem is that I think I have a well descretized mesh but I have to use 0.1s as Timestep for the simulation. This results in a very long computation time for my steady state case.

The automatic stop when steady state conditions are reached are not working because it always beaks after about 8 min (to early) an I don't know why.

Because the parallel version is not working I thought maybe somebody can verify my choice of parameters.

As mentioned I just have steady state cases (6) that I have to model but its very time consuming.

Maybe somebody can check how fast the computation time is on his/her pc-environment. (if he/she has available cpu-time )

In the attachement I added my t2d files.

Thanks in advance

Mirko

Hi
Before looking on computation time, you should look to your initial condition because "zero elevation" is a hard initial condition when you impose fix discharge and water level at boundaries.

Hope this helps

I know but with a constant water depth or height I'll have the problem that dry areas, which are not directly hydraulic connected, are filled water.

is there a way just to fill the main channel with water as initial boundary?

To get an overview I added a link for my slf-file (11MB)

Part of the message is hidden for the guests. Please log in or register to see it.

Mirko

You could start with a linear variation of discharge from 0 to the requested value
Same for the downstream water level

Otherwise, you could try to build an initial condition with BK or Fudaa by giving Water height in the main channel

Regards

OK the first thing I don't understand how to accomplish that. Is there something written in the users manual? keyword?

The second option I understand but maybe you can verify my idea to implement that:

1) create linear 3D Surface between In- and Outflow in ArcGIS and save it as 3D-Polygon

2) import it into BK and add it to my input slf-file as free surface

3) say T2D that i have initial conditions in my slf-inputfile

how can i achieve no. 3 in may *.cas-file ?

Mirko

About the initial condition, look at previous computation file and associated keywords

About the variation of discharge, yes you just have to create a boundary liquid file to do that.

For the time step, hard to say because it's necessarily linked to your mesh...

regards

Ok never mind, I got it with the first suggestion, you mean I should increase the discharge by time from zero to my end value in linear steps, right?

Can I reduce the Timestep when I do it that way? What experience do you have when doing it that way?

regrads

Mirko