open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello Marleine,

.res extension should not work when uploading files. I am surprised that .slf do not also.

Perhaps decreasing STOP CRITERIA for water depth and/or horizontal velocity components may lead to better results? 1.E-5 or less?

Anyway, if what is entering in terms of flow rates is far from what is exiting in terms flow rates (see the output file), hard to say that your steady state is reached.

If you use release v8p4, you can use feature RESTART MODE to resume a previous computation without lack of accuracy, not to resume a computation from scratch or basic initial conditions.

Hope this helps,

Chi-Tuan

Hello Chi-Tuan,
Having feedback help me a lot. Thank you

I have got an automatic stop of running with STOP CRITERIA 1.E-3 but results extracted look wierd. I am going to try the values you suggest.

I take notice of your comments about the steady state.

I am using v7p2.

Regards,
Marleine.

Hello Marleine,

As written in other topics, please use the latest release available (currently v8p4r0) as many bug fixes have been solved and new features have been integrated.
v7p2 is 6 years old or more now.

Chi-Tuan

Hello Chi-Tuan,
I take notice. I already installed the version 8.4 and a simulation is in progress.
Thank you very much.

Best regards,
Marleine

Hello everyone,
I am still having trouble reaching the steady state.
I did a 13 days simulation and I got the outflow greater than the total inflow (picture attached).
A 3 days simulation with TREATMENT OF FLUXES AT THE BOUNDARIES = 1 is in progress but I am not really confident about the result.
What do you suggest please ?

Thank you and Best Regards ?

Hello Marleine,

I still cannot see any geometry file.

Anyway, I wonder if the boundary conditions you are prescribing are well designed to get a steady state.

To my understanding, you want to prescribe 300 m3/s at liquid boundary #3 "Oueme", 160 m3/s at liquid boundary #4 "So" and prescribe 0 m at liquid boundaries #1 "Sea" and #2 "Estern". Is this what you really want to prescribe?
Do the 0 m prescribing fit with the bathymetry of the model?

A few comments about your choices, one is not what is currently recommended (see the TELEMAC-2D user manual):
TREATMENT OF NEGATIVE DEPTHS = 1 with THRESHOLD FOR NEGATIVE DEPTHS = -0.01.
What is recommended is TREATMENT OF NEGATIVE DEPTHS = 2 with the combo:
- OPTION FOR THE TREATMENT OF TIDAL FLATS = 1 (default choice),
- MASS-LUMPING ON H = 1.,
- CONTINUITY CORRECTION = YES,
- SUPG OPTION = 2;0 or SUPG OPTION = 0;0.

What about the solving step, do you reach accuracy = 1.E-10? How many iterations are used (see and/or upload the listing file)? Are you sure the convergence is reached?
Perhaps relaxing a little bit is enough (e.g. 1.E-8, 1.E-6 or even 1.E-4 if it is difficult to reach the convergence of this step).

Hope this helps,

Chi-Tuan

Hi Chi-Tuan,
Thank you for your reply.

I tried to load the geometry file but it seems that it cannot be loaded. Is there any other way to post it apart from using "Add File"?

The prescribed values are as you have summarised.
The "Sea" boundary represents the point where the lake meets the ocean. But I placed the boundary about 5 km further south, i.e. in the ocean (bottom depth up to -18 m). The "Estern" border is across a channel with a bottom depth up to -4 m.

About the solving step :
- the accuracy is about 1.E-10, 1.E-9,
- I set the maximum number of iterations to 100 (I no longer have the listing file since it is automatically deleted once the result file is generated)
- convergence is not achieved (the output is greater than the input).

In the meantime I have redone the BC file (following the instructions on the Hydro-Informatics.com platform regarding the length of the boundaries, their position in the river bed, etc.) and, to make the model simpler, as a first approach I have closed the "Estern" boundary.

I will take your comments into account when I run a new simulation.

Best Regards,
Marleine

Hi Chi-Tuan,
I am still struggling to have a steady state. I have noticed that the total inflow through upstream control sections gets constant before reaching the prescribed discharge (prescribed discharge=160+300=460) ; moreover, this inflow is not smooth even if it gets constant (see the picture attached showing outflow when it gets constant).
I tried a 3 days simulation taking into account the comments in your last answer but it does not seem to improve the output.
Kindly find attached the simulation files. Apart from the files attached here, I have load the geometry file on the following wetransfer link : we.tl/t-2YdrVoYGCZ.
Thank for your help.

Best Regards,
Marleine

Hello Marleine,

Sorry, the delay has expired and I cannot succeeded in downloading your geometry file.
Anyway, I find strange the your "outflow" is around 339 m3/s? and not 460 m3/s. What about the absissae axis, is it time in seconds or something else?

I wonder if prescribing flowrates 160 + 300 m3/s really fit prescribing 0 m at the oulet.

Chi-Tuan

Hi Chi-Tuan,
Thank you for your answer.
I think there was something wrong with my control sections, reason why the outflow was not 460m3/s. I worked on it. Things seem to get better even if I am not completely satisfied yet. (See picture attached).
The absissae axis on the picture in my last post is just the number of line I extracted from the control section file to make the plot to illustrate what looks strange. (each line correspond to one time step (5 second), but the first line is just the first of my sample and not the first time step of the simulation).
I precribe 0 downstream as mean value of tide since at the outlet water flow to the ocean.

Thank you and Best Regards,
Marleine