open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello all, I did a simple test to generate boundary conditions file with both Fudda and Blue Kenue. I used an example geometry file uploaded by one user www.opentelemac.org/index.php/kunena/16-...quid-boundaries-file (first post). To my surprise the node numbering is different! The numbering for the nodes along external boundary is the same between Fudda and Blue Kenue. However, the island ones are different. This could be an issue to determine the order of boundaries when I need inflow at an island boundary for some reason (eg hot water return from a power plant).

Could anyone please suggest why is it? It looks to me telemac can work with both boundary conditions files. So I am a bit confused what should be the correct numbering for boundary nodes.

Again any inputs are greatly appreciated,

Hailiang

Hi
in fact, there is no formal rules for the boundary condition file except:
- you should start with the external boundary before describe the islands
- you should entirely describe the boundary before moving to other boundary
- the external is describe anticlockwise
- the islands are describe clockwise
The order of appearance of the liquid part in the boundary file give the order for the values in the steering file

By convention, it was decided long time ago that the first boundary node will be the node with the minimum value of x+y.
This rule is probably implemented in Fudaa and BK so the external boundary description is the same.
But Telemac will works if you decide to start the boundary file from another node and if you take care of the numbering of liquid boundaries...

But as far as I know, there was no decision taken for island order... this could explain the difference...

To determine the order of boundaries, there is some tricks.
In Fudaa, you could enter a discharge value which is just an id so in the steering file you could identify the order.
In BK, you could give a name to the liquid boundary and this name is written in the boundary file so you just should open this file and look at the order of appearance

Hope this helps

This is excellent explanation Coulet!

I also would like to point out that the 1st liquid node(s) (eg lines 1-5) may belong to the same open boundary as the liquid nodes at the bottom of a boundary conditions file. So when one sees three liquid node groups in the file, it may in reality be just two open boundaries - I have seen such cases. But with the Fudda and BK tricks as mentioned by Coulet, these are handled.

Hailiang