open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everyone

I got serious issues to make the mesh in this wide size area. I have a maritime structures (length order of magnitude is 10 cm. But I you see the computation domain is very large Left bank to right bank is approximatively 3km.
I tried multiple densities for making the mesh around my area of interest (structures). But it seems like the magnitude of the strucutures is too small compare to the domain size. At the end, at the resuslt I get a terrible mesh. I think maybe BlueKenue is not appropriate for my mesh or maybe do I need do decrease de domain size ?

Did someone had this issue before ? and how did you handled that ?

Maybe if you share your data someone might be able to try a quick test to help you out? And/or they might be more willing to give you some pointers...

Kindly,
Alan

Hi
According to your domain, you will probably genrate a huge mesh but there is no limitations about the mesh size.
What is a terrible mesh for you? Could you show an example?
Regards

Hi,

Here is the .T3M mesh generator.

Thanks

No file!
Join a zip file
If too large use another transfer solution

Sorry here is the .zip file

Regards

Sorry

Saw the attached file

Sometimes (when you have such a large range of densities) it is better to create the detailed mesh separately and include it as a submesh to the final mesh.

Also (depending on what you are trying to model) in this case it
might be advisable to run this as 2 separate models where the river and "offshore" are a coarser model that provides you with boundary conditions for your detailed harbour.

This looks like the Saquenay fiord to me so by "offshore" I mean the St. Lawrence. Are you planning to study this in 3D?

Cheers... Martin

HI Martin

Thanks for your reply.

Exactly the domain is the Saguenay Fjord and the St-Laurence estuarune. it is quiest a big and complex model (the stratification, temperature and salinity are so important in this area of interest. My study is mainly on the Saguenany River, 97km far from the outlet of the river (cf the files). But even at that distance from de St-Laurence, its effects are non-negligible. What exactly do you mind about considering two submeshs ?

Regards

I agree with Martin. If you run such a large model with tiny elements, the time step will be very small and the model will take ages to run.

First run a coarse model without the details of the platform and the piles (some of them could even be omitted completely if they won't affect the flow a few km up and downstream).

Then in this model you extract levels and/or flows in 2 sections up and downstream of your study area.

Then you use these sections as boundary conditions for a second more detailed model which covers only the study area between the 2 sections.

Thierry