intro story D-Flow FM

 

D-Flow Flexible Mesh

D-Flow Flexible Mesh (D-Flow FM) is the new software engine for hydrodynamical simulations on unstructured grids in 1D-2D-3D. Together with the familiar curvilinear meshes from Delft3D 4, the unstructured grid can consist of triangles, pentagons (etc.) and 1D channel networks, all in one single mesh. It combines proven technology from the hydrodynamic engines of Delft3D 4 and SOBEK 2 and adds flexible administration, resulting in:

  • Easier 1D-2D-3D model coupling, intuitive setup of boundary conditions and meteorological forcings (amongst others).
  • More flexible 2D gridding in delta regions, river junctions, harbours, intertidal flats and more.
  • High performance by smart use of multicore architectures, and grid computing clusters.
An overview of the current developments can be found here.
 
The D-Flow FM - team would be delighted if you would participate in discussions on the generation of meshes, the specification of boundary conditions, the running of computations, and all kinds of other relevant topics. Feel free to share your smart questions and/or brilliant solutions! 

 

=======================================================
We have launched a new website (still under construction so expect continuous improvements) and a new forum dedicated to Delft3D Flexible Mesh.

Please follow this link to the new forum: 
/web/delft3dfm/forum

Post your questions, issues, suggestions, difficulties related to our Delft3D Flexible Mesh Suite on the new forum.

=======================================================

** PLEASE TAG YOUR POST! **

 

 

Sub groups
D-Flow Flexible Mesh
DELWAQ
Cohesive sediments & muddy systems

 


Message Boards

FLOW, Domain Decomposition, and Space Varying Wind Errors

JH
Julia Hopkins, modified 4 Years ago.

FLOW, Domain Decomposition, and Space Varying Wind Errors

Youngling Posts: 14 Join Date: 1/1/13 Recent Posts
Hello,

I am a bit lost as to the right way to input space-varying wind conditions in a FLOW run with domain decomposition.

I have made the required .amu/v/p files, and I know how to initialize them with "Additional Parameters" keywords ("Filwu"/"-v"/"-p"). I am not convinced this alone is doing anything, however, so I have tried turning on the Wind component of the model, which requires me to feed the model some sort of dummy spatially varying wind file (.wnd).

The moment the Wind parameter is on in the FLOW model, I am unable to run the model. It begins initializing, and then exits abruptly with the error:

"This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information."

Does anyone have an idea of what is going wrong? Conversely, any advice on how to implement spatially varying winds in FLOW?

Your help and advice would be most appreciated.

Sincerely,

Julia
Emiel Moerman, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Padawan Posts: 84 Join Date: 3/10/11 Recent Posts
Hi Julia,

It is difficult to judge what is going wrong. Could you upload your files so I can have a more detailed look?
Note also the additional keyword AirOut = YES to write wind to the output files. This allows you to check the wind speed also once the model is running.

Regards,

Emiel
JH
Julia Hopkins, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Youngling Posts: 14 Join Date: 1/1/13 Recent Posts
Thank you for the prompt response!

I have uploaded the outermost grid of my domain decomposition (one of four) for the purposes of debugging. I have also included all of the wind conditions I am using. At the moment, FLOW is not coupled to WAVE since I am seeing the error both with the coupling and without.

The model has sediment transport initialized, as well, which could be complicating matters.

Let me know what you think -- if you need one grid that runs without the domain decomposition, I can supply that simplified model set of files as well!
JH
Julia Hopkins, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Youngling Posts: 14 Join Date: 1/1/13 Recent Posts
I must amend some parts of the question -- I've gotten to the point where I think I have narrowed down the issue to the way domain decomposition interfaces with wind in FLOW.

The "fakewind.wnd" file above was causing me problems because it was not in the right format for a space-varying wind file (similar to the .amu and .amv file formats). I fixed that (under the assumption that this .wnd grid is subsequently ignored by the model).

My error now has to do with the grid on which wind is defined. I am trying to apply wind conditions to a DD model, which means that my outer grids necessarily have "holes" in them for the inner grids. I get the feeling this is wreaking havoc with the wind declaration...the model does not seem to be able to find the points it expects to have wind values in them.

Does anyone with experience applying wind to a DD FLOW grid know the appropriate grid on which to define wind conditions, and how to recognize this grid in the .mdf file?

Thank you!
Adri Mourits, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Yoda Posts: 1224 Join Date: 1/3/11 Recent Posts
Hi Julia,

See the manual, section A.2.8 on wind input formats.

You should avoid "Wind on the computational grid", because then you will have to specify wind values on each individual FLOW grid cell of each subdomain.

The easiest way is to define one coarse equidistant wind grid, covering all subdomains, with values specified on that grid. An example testcase (without DomainDecomposition) is attached to this post.

It should be possible to use this one set of wind files (amp, amu, amv) on all subdomains by copying the Filwp/Filwu/Filwv lines into all mdf-files. Unfortunately, some compilers do not allow multiple access to input files. To solve that, you have to create a copy set of amp/amu/amv file for each subdomain and let each mdf-file point to it's own unique set.

Regards,

Adri
JH
Julia Hopkins, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Youngling Posts: 14 Join Date: 1/1/13 Recent Posts
Thank you so much for the help! I will try it out and see what happens...I think I might be looking at the second part of your solution (copy sets of the atmospheric files). I will see if this works!

Sincerely,

Julia
JH
Julia Hopkins, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors

Youngling Posts: 14 Join Date: 1/1/13 Recent Posts
Ah, I am remembering why I decided not to use the "equidistant grid" option:

My model is in degrees, and I cannot find a reasonable combination of "xllcorner"/"yllcorner"/dx/dy values which will keep the equidistant grid seen by Delft3D in the confines of my actual computational space.

My longitudinal coordinates are negative numbers, which might have something to do with it. Should I only be working in a coordinate system with positive longitudinal coordinates? Should I be specifying a negative "dx" in degrees to make sure it reads the grid in the right direction?

I've been going through many iterations of all of these, but it would really help if someone could direct me to the solution for a grid based on degree units instead of meters.
Adri Mourits, modified 4 Years ago.

RE: FLOW, Domain Decomposition, and Space Varying Wind Errors (Answer)

Yoda Posts: 1224 Join Date: 1/3/11 Recent Posts
Hi Julia,

Negative Longitudinal coordinates should work. dx should be positive.

Can you zip your full set of input files and attach it to a post on this forum? Ideally the model is as much simplified as possible and still showing the problem.

It might be informative if you try to move your model to positive lat/lon coordinates to check whether that does work.

Regards,

Adri