open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi everyone,

First i would like to thank you for this forum and your website. It is very usefull when you manage to learn to use Telemac.

I’m coming here because I face some challenge with Telemac2D.

First let me explain what I’m trying to do. My initial problem is not really a hydrological problem but is very similar to the run-off of water on a specific mountain. Water “appears” on the top of the mountain and run-out in channels to the bottom (see the screenshot of my mesh).

In Telemac, I configure different sources, one with positive discharge on the highest point of my relief, and the other with negative discharges. In this way, I assume that the flow will go from the highest point to the other one through the different channels (even if, for now it’s not possible to reach all the destination nodes).

When I run the simulation with Telemac2d, I get very strange results: sometimes there are flows in the opposite direction of slopes.
I search some explanation on this forum. Maybe one can be that the slopes are too steep for some elements? So I wonder if you have any idea about this problem.

I attached the steering and the geometry file I use for my simulation.
Thanks in advance for your idea or explanation.

Hi
There is no screenshot in your message...
I'm not fully understand why you create source points with negative discharge downstream, maybe you could start with only the positive discharge upstream.
Could you illustrate your strange results?

regards

PS: think to update your profile

I'm sorry, part of files have not been taken into account since (wrong format i think).

The aim of point with negative discharge is to simulate an outlet (as if there is a hole in the domain and water can go in it). I'll without any negative discharge but i think it will not solve the initial problem.

I also join an example of these strange results. The black and white background is the topography, and i assumed that the flow will stay in the black parts (which are lower than white parts) because the source node is in this part.

Thanks for your answer !

The results seems strange you're right but the case itself is strange!
According to the steering file, there is no discharge so which is the discharge you imposed?
The mesh is based on a regular grid 1000m x 1000m, but you have a time step of 10s...
What about about the solver convergence?
No friction?

The power of unstructured mesh should allow you to better represent the channels and the ensure a proper result

regards

Thanks for your answer. I’ll try to give further explanations about my case.
The discharge I imposed is in a separate file (my apologies, I forgot it…).
So if I’m right, there is a relationship between the size of the grid and the time step ?

I also change the argument LAW OF BOTTOM FRICTION (from 0 to 3) and, yes results are better…

So let’s try to explain my case in details.

My study case is based on the following hypothesis: traffic flow is analogous to the flow of water on a mountain. That’s why, I try to simulate water flow on a topography I build thanks to socio-economic characteristics of the area (it’s not really the subject here). On this topography, the highest point is the origin of flows and the different lowest points are the destinations. The idea of my Telemac simulation is to generate water on this highest point, see how it behave on the topography and how it distribute on the several destinations where water is supposed to get out of the domain. That’s why I prescribe positive discharge on the highest point and negative ones on the other. But, honestly, I think I’m wrong…

I also try to keep the simulation with less parameters as possible, but maybe I missing some of them.

Regards