open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I am attempting to write a liquid boundary file for my model and vary velocity and surface elevation. I have managed to get the model to run with a varying surface elevation, but I am having issues with prescribing the velocities.

After reading the manual I am a bit confused as to what format it needs to be in. The Telemac 2D manual suggests that the velocity for the first boundary should be defined as 'VIT(1)' on page 32. But then on page 33 it says to use 'Q for flow rate, SL for water level, U and V for velocities etc.' I can't see how to distinguish between horizontal(x) and vertical(y) velocity.

I have tried using this format, where the first line of the file reads:

T SL(1) U(1) V(1)
s m m/s m/s
0 0 0 0
149 -0.043 -0.023 0
298 -0.086 -0.045 0
447 -0.129 -0.068 0

But this didn't work, Can anyone advise?

Thanks

Hi
not sure it will works with the file (probably only VIT)
Check in read_fic_frliq.f

Regards

Hi I have been trying to set up a simple simulation of a narrow channel with an open boundary and apply a sine wave for currents only in the v direction to simulate a tidal cycle using the LIQUID BOUNDARIES FILE option, however, i have the same issues when formatting the file as suggested in the manual (T U(1) V(1)).

The simulation fails at the first time step as it appears the subroutine vit.f does not recognize the format of velocities.

ITERATION 0 TIME: 0.0000 S
THE LIQUID BOUNDARIES FILE CONTAINS
51 LINES WITH:
U(1) V(1)

VIT : MORE PRESCRIBED VELOCITIES ARE REQUIRED
IN THE PARAMETER FILE
AT LEAST 1 MUST BE GIVEN


I attach all relevant files for you to analyze. Please could you let me know how to set up this simulation and in particular the boundaries file correctly for a time varying velocity input at the single boundary.

regards

M

Hi
You should let the keyword PRESCRIBED VELOCITY in your steering file even the value isn't taken in account.

Regards

I was getting the same issue and discovered the solution. Check the number of the boundary that you are applying the data to. Telemac has its own way of numbering the boundaries (starting at (0,0) and going in an anti-clockwise direction), the first boundary may be the one on the right of your domain.

A simple check is to open up the boundary file (.cli) in text editor and have a look at all the outer nodes. The first boundary (group of outer nodes) that are fully listed will be the first boundary (i.e. number 1)

What you may need to do is to change it to U(2) and V(2) if the boundary where you want to apply the flow is listed after.

I hope that makes sense.

Hi Both, thank you for your responses, however none of these suggestions solved my issue. Using 'PRESCRIBED VELOCITY' in your steering file overrides the LIQUID BOUNDARIES FILE option and simply gives a step change in velocity to the values prescribed at each boundary denoted. Further to that point I only had one boundary in my domain for simplicity and therefore the numbering of boundaries and how this was represented in the file was not my issue.

After delving into the vit.f and read_fic_frliq.f subroutines and playing around with the format of the boundaries file, I have found that the code did not recognize U(1) and V(1) as inputs, however if I label the time series of velocity VIT(1) it runs perfectly (i.e. the time series is input as magnitude with direction effectively = 0). This was ok for me as I have my domain set up so that velocity is prescribed at the channel boundary in a northerly direction (i.e. only in the v direction), therefore my u vector time series would be zero anyway and I simply input all the flow along the v vector, as one would per a magnitude. When I included a second boundary at the north end of the channel for which I wanted to be an open boundary without forcing so that flow could propagate away from the channel I introduced VIT(2) into the boundary file with all values set at zero. FYI my .cli file boundaries are 466 (forced) in the south and 444 (unforced) in the north.

However, this negates the fact that one cannot apply time varying U and V velocity vectors separately using the LIQUID BOUNDARIES FILE as the manual states? I am running v6p3r2 for information. I am unsure if this was rectified in later releases? r.clayton what version are you running and does it recognize the U and V notations in the header of your boundary file? Do you see distinct values in the U and V direction at the boundary in your simulations. I do near the edge of the domain where friction occurs and the flow diverts.

M