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

Courant number and instability

OL
Orpita Laz, modified 6 Years ago.

Courant number and instability

Youngling Posts: 3 Join Date: 6/11/12 Recent Posts
Hi,

I am doing a research on morphological assessment of Jamuna River in Bangladesh by using Delft3D. I have many questions about Delft3D. Could anyone please help me regarding the following issues:

- I have generated a grid by fulfilling all the requirements (such as orthogonality, aspect ratio) but courant numbers are shown only for positive depth values. Most of the z values in my sample file are negative, so when I am interpolating by triangulation it is showing the negative values in depth file. Please find the attached the mdf, grid, depth, sample and land boundary file.

- I gave a morphological run of sediment transport for 2 years and after extracting the cross-sections’ bed level, a continuous up and down appears in the cross-section. Do they indicate instability of the model? Please find some example images attached herewith.

- Some warnings has been shown in tri-diag file which I couldn’t understand


********************************************************************************
***
*** Deltares, Delft3D-FLOW Version 4.00.01.000000, Mar 15 2011, 10:57:16
***
*** runid : jamuna_feb_2010
*** date,time : 2012-10-24, 12:11:13
***
********************************************************************************

*** MESSAGE 0 bed load fraction(s) found in sediment file: Jamuna.sed
*** MESSAGE Double precision computation using reals of kind 8
*** Start of User Defined Model description
-
- 1 October Run

*** WARNING (Part of) Cross-section (121,42)..(121,145) lies on inactive points
(m1,n) = 121 42
(m1,n) = 121 145
*** WARNING Bed change exceeds 5.0 % of waterdepth after 721 timesteps, location (m,n) = (186,116)
***Sediment erosion shortage at NM 27606 Fraction: 1 Mass available : -0.4362E-01 Mass to be eroded: -0.8596E-01
*** WARNING Source and sink term sediment 1 reduced with factor 100.27 (m,n)=(1,3), after 53724 timesteps



Thanks and Regards
Orpita Urmi Laz
Richard Measures, modified 7 Years ago.

RE: Courant number and instability

Jedi Knight Posts: 178 Join Date: 3/23/11 Recent Posts
Hi Orpita,

That looks like an interesting model. Looking at the model results there is definitely something strange going on - the channels appear to be incising vertically (until they reach the limit of the bed thickness), there is almost no lateral movement/migration, and the instability you mention is resulting in straight single grid cell width channels. I've had a quick look at your model and list some key properties below for others who are reading this:
  • Grid resolution ~150m
  • Grid dimensions = 145 x 298 cells
  • Timestep = 1min
  • Model simulation = 4 months (120 days)
  • Flow = varying but generally increasing from 10,000 to 65,000 m3/s over the duration of the simulation
  • Single sediment fraction - fine sand (0.2mm)
  • bed thickness = 5m
  • STFormula = default (Van Rijn)
  • MORFAC = 8.25

Without looking into it in more detail some ideas I had for things you might want to investigate are:
  • Grid resolution - your grid looks quite coarse - do you need a finer grid to be able to reproduce the processes correctly?
  • Bank erosion, secondary flow and transverse bed slope effects - These processes tend to encourage lateral channel movement and wider shallower channels, potentially reducing channel incision and encouraging channel migration.
  • Sediment feed - if there is insufficient sediment feed this could cause your channels to incise, especially at the upstream part of the model. I see you are using equilibrium boundary condition for sediment at your upstream boundary but depending on the local flow conditions at the boundary this does not necessary supply the correct amount of sediment - maybe check the cumulative transport rate at you different monitoring cross-sections to see hor the sediment input at the upstream boundary compares to transported load in the model.
  • Sediment formula - consider if other sediment transport formulations might be moire suitable? Maybe test something like Meyer-Peter-Muller to see if it gives a better result?
  • Bed armouring - if your river surface is coarser than the sub-surface or there are significant mixed size sediment effects (hiding/sheltering) you may need to use multiple size fraction sediments and multiple bed layers.

Cheers,

Richard
OL
Orpita Laz, modified 6 Years ago.

RE: Courant number and instability

Youngling Posts: 3 Join Date: 6/11/12 Recent Posts
Hi Richard,

Thank you very much for your quick response. But I have a major problem, that is when I am checking the courant number the grid cells in my model showing the value of the courant number for the positive depth values. The grid cells having negative depth value are not showing the shade for courant number. I have attached relevant files. Could you or anyone please check and help me in this regard.

Thanks

Orpita Urmi Laz
Richard Measures, modified 6 Years ago.

RE: Courant number and instability

Jedi Knight Posts: 178 Join Date: 3/23/11 Recent Posts
Hi,

To calculate Courant number you need to know the water depth. In QUICKIN it assumes water level = 0 when it estimates Courant number. This works fine for a coastal model but not for a river model (i.e. where water level will not be horizontal - this is the problem you have). You can estimate the maximum Courant number based on estimated maximum water depth and minimum grid size (M or N direction, not M*N) using equation 4.1, page 40 of the manual - this is generally what I do for my river models. You can also use QuickPlot to calculate Courant number using your model results (trim file) - I think this is described in the QuickPlot manual.

Hope this helps,

Richard
TL
Tony Lee, modified 2 Years ago.

RE: Courant number and instability

Youngling Posts: 9 Join Date: 4/1/17 Recent Posts
Hi Richard,
As the flow manual reads, the CFL number is about the characteristic value (in
many cases the minimal value) of the grid spacing in either direction.

But how can we know the grid size from delft3d ?
cfl
Richard Measures, modified 2 Years ago.

RE: Courant number and instability

Jedi Knight Posts: 178 Join Date: 3/23/11 Recent Posts
Hi Tony,

The grid size is output in the model results. It can also be calculated/displayed from the model grid using RGFGRID.

Cheers,
Richard
YS
Yuwana Sriraharjo, modified 2 Years ago.

RE: Courant number and instability

Youngling Posts: 2 Join Date: 7/29/16 Recent Posts
Dear Richard,

I have problems for modelling transport sediment on near coast.

Firstly, I want to generate longshore current driven by wave. I use both FLOW and WAVE module to simulate. In flow model there is only Neuman Boundary and Water Level with 0 value. Parameter in the wave module based on significant wave hindcast model. Instead of getting longshore current on near shore, I get unusual result in velocity, some vectors velocity shows higher value in offshore.

Secondly, when I add sediment in the model, with morphological factor 365, sedimentation only occur at shore area. I hope get erosion near the structures.
Some warning about "Source and sink term sediment" occurred but I can't figure it out
"*** WARNING Source and sink term sediment 1 reduced with factor 198.81 (m,
*** n)=(45,1), after 236 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 500.31 (m,
*** n)=(45,1), after 236 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 1496.84 (m,
*** n)=(45,1), after 237 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 5637.99 (m,
*** n)=(45,1), after 237 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 32162.42 (m,
*** n)=(45,1), after 238 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 185.62 (m,
*** n)=(46,1), after 238 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 408.75 (m,
*** n)=(46,1), after 239 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 1073.96 (m,
*** n)=(46,1), after 239 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 36476.17 (m,
*** n)=(45,1), after 240 timesteps.
*** WARNING Source and sink term sediment 1 reduced with factor 3360.13 (m,
*** n)=(46,1), after 240 timesteps.

Can you check on my setup model, so I can get better result

Here also I attached my model setup

Thank you
MA
Md Bin, modified 6 Years ago.

RE: Courant number and instability

Padawan Posts: 42 Join Date: 12/17/12 Recent Posts
Hi Orpita I want to prepare Jamuna Model. For this reason I want to work with you. I am from IWFM,BUET, Dhaka. If you have interest please mail me.

Md. Khalid Bin Siddique
Mobile: 01717353211
Research Associate
Mail:mrkhalidiwfm@gmail.com
ESPA Deltas Grant Consortium Project(modelling)
Institute of Water and Flood Management
BUET