|
[Sponsors] |
March 28, 2013, 05:47 |
Problem with mapFields - map 2d to 3d
|
#1 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 342
Rep Power: 28 |
Hello,
I want to map the solution of a 2d problem to a 3d mesh. The images show the blocks of the 2d and the 3d meshes. The 2d mesh has 3600 cells, the 3d has 36000 cells (it is 10 cells deep). If I use mapFields with an empty mapFieldsDict Code:
patchMap ( ); cuttingPatches ( ); How can I map the 2d fields to the 3d mesh in all depth. I need my 3d mesh to be 3 blocks in depth. |
|
March 4, 2014, 14:00 |
|
#2 |
Senior Member
Andrea Ferrari
Join Date: Dec 2010
Posts: 319
Rep Power: 16 |
Hi Gerhard,
did you solve the problem? I also have to map a 2-D solution onto the corresponding 3-d mesh (basically i just "refined" the mesh in 1 direction, from 1 cell depth to 10 cells depth). best andrea |
|
March 4, 2014, 14:26 |
|
#3 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 342
Rep Power: 28 |
I only came up with a quick and dirty workaround, nothing more.
If your 2D domain is just a thinner version of your 3D domain, you can use transformPoints to "inflate" the mesh so that the domain has the same dimensions as the 3D domain. transformPoints applies either a uniform or a non-uniform scaling factor to the point coordinates. So, to "inflate" the mesh, I applied a scaling factor greater than one in z-direction and a factor of 1 in the other directions. Then I could use mapFields. However, this applies only if the 2D mesh is just a thinner version of the 3D mesh. |
|
March 5, 2014, 05:47 |
|
#4 |
Senior Member
Andrea Ferrari
Join Date: Dec 2010
Posts: 319
Rep Power: 16 |
This is exactly what i have.
Following your suggestions, i used transformpoint to "inflate" the 2-d geoemtry to correspond to the 3-d. Then i used the option -consistent and it works...but...is it normal that is so slow? it run all the night without doing anything. It is stuck here Create databases as time Source time: 0 Target time: 0 Create meshes Source mesh size: 4066574 Target mesh size: 20332870 the target mesh is quite big..so maybe its normal. Can it run in parallel to speed up the process? thanks andrea |
|
March 5, 2014, 07:00 |
|
#5 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 342
Rep Power: 28 |
I have no experience with mapping from a 4M cells case to a 20M cell case. You should check whether you run out of memory. You need to load both meshes in order to map fields between them. You also need to load all fields. So the mapping operation you describe sounds very memory hungry.
|
|
March 5, 2014, 07:22 |
|
#6 |
Senior Member
Andrea Ferrari
Join Date: Dec 2010
Posts: 319
Rep Power: 16 |
probably it is because the mesh is really big. Anyway i solved it in another way...using matlab
thanks again andrea |
|
July 15, 2020, 10:50 |
|
#7 | |
New Member
Arash
Join Date: May 2017
Posts: 17
Rep Power: 9 |
Quote:
Can you please share your idea of doing it with MATALB? Of course, a brute force way is to look at the coordinates of each mesh element and match it with its 2D correspondence. This is , however, by no means scalable, especially for the cases where both 2D and 3D cases are renumbered. Thanks for your help, Arash |
||
Tags |
mapfields, preprocessor |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
UDF compiling problem | Wouter | Fluent UDF and Scheme Programming | 6 | June 6, 2012 05:43 |
Problem Importing Geometry ProE to CFX | fatb0y | CFX | 3 | January 14, 2012 20:42 |
fluent add additional zones for the mesh file | SSL | FLUENT | 2 | January 26, 2008 12:55 |
natural convection problem for a CHT problem | Se-Hee | CFX | 2 | June 10, 2007 07:29 |
Adiabatic and Rotating wall (Convection problem) | ParodDav | CFX | 5 | April 29, 2007 20:13 |