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

Neumann Boundary for Delft-Flow model ?

TS
Thomas SAILLOUR, modified 5 Years ago.

Neumann Boundary for Delft-Flow model ?

Youngling Posts: 24 Join Date: 4/6/14 Recent Posts
Hi everyone,

I have an overall model, constrained everywhere (North, south, west and east) by astronomical boundary conditions.
Then I am using a nested model thus the value of the time series on the boundary conditions come from the overall model.

WL1 = ?
||--------
|| |
|| |
|| |
|| | WL2 (Time Series)
|| |
|| |
|| |
||--------
WL3 = ?

The WL2 uses the time series extracted from the overall model, and the model did not work when WL1 and WL3 where using time series from the overall model as well. (Too much constraints in the model maybe)

When I used Neumann Boundary conditions (still for WL1 and WL3) for the nested model I had two kind of results:

1) Using the time series of the overall model for the nested model --> resulted in a huge tidal wave for the first time steps.
2) Putting zeros for WL1 and WL3 --> resulted in the same results as a model with WL2 (Time Series) and No other boundary, i.e.: totally loose for the North and South boundaries.
--> results also in poor calibration of the model.

my detailed model is something like that drawn by Bert in an another thread but instead of river its a land area, someone actually had the same problem but wasn't answered.

Please guide me on what appropriate boundary condition I must use.

Many thanks

Thomas
MV
Maarten van Ormondt, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ? (Answer)

Padawan Posts: 45 Join Date: 3/11/11 Recent Posts
Hi Thomas,

When using the Neumann boundaries (which are water level gradient boundaries), did you compute the gradients across your boundaries WL1 and WL3? It's the gradients that should be in the bct file, and not the water levels, when using Neumann boundaries. The easiest way to compute the gradients is to compute the difference between water levels at both ends of WL2, and divide by the length of WL2. Note the sign of the gradients depends on the grid (i.e. m and n) orientation. The process of computing Neumann boundaries usually takes some matlab scripting but can also be done using Excel, or something similar.

Running the model with 3 water level boundaries should also work, but may give problems near the shore when wave and wind set-up are also computed within the model, due to a mismatch between boundary values and value computed within the domain. Did you check the time series of the bct file, in particular the values for the landward ends of WL1 and WL3? Sometimes these look funky when points in the overall model, that are used in the nesting procedure, fall dry during the simulation.

Regards,
Maarten van Ormondt
TS
Thomas SAILLOUR, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Youngling Posts: 24 Join Date: 4/6/14 Recent Posts
Hi Maarten,
Thanks for your answer,
I will try to compute the simulation with gradients calculated with the water levels extracted from the overall model.

Running the model with 3 water level boundaries should also work, but may give problems near the shore when wave and wind set-up are also computed within the model, due to a mismatch between boundary values and value computed within the domain. Did you check the time series of the bct file, in particular the values for the landward ends of WL1 and WL3? Sometimes these look funky when points in the overall model, that are used in the nesting procedure, fall dry during the simulation.


And yes, actually, there are some funny stuff happening near the boundaries (North and South mostly) when I implement the wind into the model, that is why I have to work around the model with Water Levels boundaries and give less constraints to the model.

I will update you as soon as I get results with Neumann boundaries using gradients from the time series.

Cheers!

Thomas
TS
Thomas SAILLOUR, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Youngling Posts: 24 Join Date: 4/6/14 Recent Posts
Hi Maarten,

I finally had some results with the Neumann Boundaries.
The model is stable, but there are still tiny wobbling stuff coming from the boundaries.
The model I work on may be too constrained (Meteo files, Water levels for the East bnd plus Neumann bnds for S and N).

The simplest and smoothest way is to set the N and S bnds loose, the results are identical anyway with or without Neumann bnds.

Thanks for your help.
Cheers!
SS
Saeed Shaeri, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ? (Answer)

Youngling Posts: 6 Join Date: 10/7/12 Recent Posts
Hi Thomas

Thank you for sharing your experience with us.

In this regard, I would like to ask what you mean by making the boundaries 'loose'.
My understanding is that if you want to have an open side for your domain without any boundary condition, it is not possible. You either can introduce a boundary or the side would act as it is closed.

Looking forward hearing from you,
Regards,
Saeed

Thomas SAILLOUR:
Hi Maarten,

I finally had some results with the Neumann Boundaries.
The model is stable, but there are still tiny wobbling stuff coming from the boundaries.
The model I work on may be too constrained (Meteo files, Water levels for the East bnd plus Neumann bnds for S and N).

The simplest and smoothest way is to set the N and S bnds loose, the results are identical anyway with or without Neumann bnds.

Thanks for your help.
Cheers!
TS
Thomas SAILLOUR, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Youngling Posts: 24 Join Date: 4/6/14 Recent Posts
Hi Saeed,

Indeed, you are right, the most stable model is with only one open boundary (East) with water levels conditions.
All the other boundaries are closed then.
Applying either Neumann or Water levels conditions to these boundaries would make the model less stable or even would make it crash.

Regards,

Thomas
Qinghua Ye, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ? (Answer)

Jedi Council Member Posts: 612 Join Date: 3/2/11 Recent Posts
Hi, Saeed,

See the attached files.

Qinghua
YW
Yun Wang, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Youngling Posts: 9 Join Date: 9/10/14 Recent Posts
Hi,Qinghua Ye
My name is Yun Wang. I have always seeing your reply to someone's problem and i know you are helpful person.
I'm a beginner of the Delft3D. I felt confused about the tagged versions and I don't know which one is the latest version .For example ,among the tagged versions 2367,2399,4.00.00.0061,4.00.01.0265,5.00.10.1983 and 5.01.00.2163,which one is latest?What different between the 2367 of the same kind and 4.00.00.0061 of the same kind?
What's the meaning of the introduce in the page of Download Source code that"The log of the tag contains a full list of version numbers, for example, the log of tag "2367" reads:
DELWAQ Version 5.00.00.2367
Delft3D-FLOW Version 6.00.00.2367
Delft3D-WAVE Version 3.04.01.2367"?
Would you very kind to tell me?
Thank you very much in advance.
Qinghua Ye, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Jedi Council Member Posts: 612 Join Date: 3/2/11 Recent Posts
Dear Yun,

So I believe you got your answer from Richard's reply. Sorry for late. Thank you, Richard.

Regards,

Qinghua
YW
Yun Wang, modified 5 Years ago.

RE: Neumann Boundary for Delft-Flow model ?

Youngling Posts: 9 Join Date: 9/10/14 Recent Posts
Hi Qinghua,

Thank you all the same.

Regards,

Yun