open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I created mesh for the study area with the same edge length. I would like to create fine mesh at one location as well as coarser. Is that possible to create a same type a mesh for a particular depth (e.g -10m-fine mesh) and coarse for particular depth(away from the shore). Otherwise is it possible to create different type of mesh region by region in a particular study area depends on the bathymetry data.

-regards,
smk.svks

Hi,

The easiest way in Bluekenue to get different size mesh, as for you example where you want fine mesh at coast and it get coarser as it goes in the ocean, is to get boundary nodes near the coast to be with less distance spacing and get the open ocean nodes at wider distance for the closed line defined in Bluekenue.

You may check the file which I have attached and try to open it in bluekenue.




Try to make a mesh with the sample file which I have given.
in mesh generation properties, untick "Resample Outline", and "Resample lines only", keep default edge length as 500. 500 is the open ocean boundary spacing, and 50 is the spacing along the coastal boundary. Try generating various mesh with different edge growth ratio, default is 1.2, you will notice that as you lower the edge growth ratio the mesh elements sizes will be less as along the open ocean.

You may use hard lines, sub-meshes, define density criteria using hard points, to make various other refining of the mesh. Refer to the Bluekenue manual, in section 2.2.3.1.

With Best Regards,
Gourish

Hi,

May I ask you to have a look at this very useful BlueKenue tutorial at this post: #6481

Paragraph 5.1 describes creating variable density meshes.

What I usually do in my meshes is that I depend the density of the mesh on the water depth, so that the greater the depth, the greater the element area. To do this, I create a base uniform mesh of my area, assign bathymetry to the mesh and then I use the calculator tool on the mesh (Tools->Calculator) with an expression to transform depths to element size. Assign the bathymetry variable to "A" and use an expression like this:That multiplies bathymetries to make them positive (because element size is positive) and then add or subtract a number (this case 1) depending on how fine or coarse mesh is desired. This is a simple linear expression that works fine if you don't have positive elevations, but even then you can work it out.

The final step is to create a new mesh, adding the mesh created by the calculator as a density object in your mesher. The new mesh will have smooth element sizing depending on the depth.

It may look confusing but if you get your way around Blue Kenue, it will be a straight forward process.

Regards,
Costas

Hi Coasts,
As you suggested i tried for #6481 but i am not able to open without password, pl tell if it is any other way to download. And trying to do the mesh as per your instruction

-regards,
smk.svks

Hi
As indicated in Martin’s post #8088, the BlueKenue tutorial is also available in both French and English on the NRC web site:

ftp://kenueftp.chc.nrc.ca/BlueKenue/Tutorial

(Do not click but copy and paste the address)

Thierry

Hi

Once you have downloaded the tutorial in pdf and the fichier_ascii.zip files, unzip the file. It will provide 2 directories:

Chapter3 contains the files which are called for in chapter 3 of the tutorial (basic mesher). When you read the instructions, you will be able to load the various files into BK on your own machine, and reproduce the steps required for the creation of the mesh.

Section 3.4 of the tutorial shows how to define locally a mesh with a different size.

Chapter6 contains the files for the mesh editor of the tutorial.

Thierry

Hi,
I tried what you mentioned but i am not able to make it. I assigned the bathy values to mesh then i check the calculator in that variable i can only select the generated mesh anyway i tried the expression too but there is no change in mesh element size. Is there anything i missed out?


regards,
smk.svks

Hi,

What are doing with this method is to create a new mesh, with the same element structure but with an artificial "bathymetry" that will be used to control the density of the final mesh that you want to produce.

Have you done that?

Costas

Hi,
Yes i did that and saw the density change too. I also tried the following to produce dense to course mesh depends on the bathymetry. I untick the resample outline and reduce the edge growth ratio to 1.05 (trail and error if i gone below 1 it produce much more nodes), it gave the reasonable nodes and then i tried calculator method too(i detect 7000 from bathy -just a trail) but when i run the model it shows some nodes are connected to 8 or 9 adjacent nodes. I am attaching my mesh file just have a look at it and give your feedback.

attachment,



-regards,
smk.svks