CFD Online Logo CFD Online URL
Home > Forums > OpenFOAM Running, Solving & CFD

Implementation of adaptive Meshrefinement for 2D for rhoCentralFoam

Register Blogs Members List Search Today's Posts Mark Forums Read

LinkBack Thread Tools Display Modes
Old   September 13, 2013, 07:33
Default Implementation of adaptive Meshrefinement for 2D for rhoCentralFoam
Join Date: Nov 2012
Posts: 71
Rep Power: 6
Henning86 is on a distinguished road

i managed to implement 2D adaptive Meshrefinement (only hex) for rhoCentralFoam.

The Implementation is based on the meshcutter. So the cells get first split in one direction and than in the other direction. (same idea as in refineMesh, multiDirRefinment)

This works great in the forward step tutorial of rhoCentralFoam - Refinement and Unrefinement. I already implemented it in parallel. I tested it on 12 processors and i get no errors. i had to change the meshcutter-class do be able to work in parallel.

The problem is if i cut a not perfect mesh with hex oriented in different directions. Some cells are not cut which results in a crash. Some of the missed cuts can be explained that there is no "splitHexCellLooper" but this can be implemented in one week.

Problem: the first refinement every cell are cut in both directions. But the second time the cells should be split results in a crash. Not only Splithex are not cut correct but hex as well, which isn't expected.

But if i restart the simulation the refinement works where it crashed in the first attempted.

I am really confused! i already checked CreateMesh.H but i find no hint in the constructors that the mesh is some how renumbered or something like that.

Any help is welcome.
Henning86 is offline   Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

All times are GMT -4. The time now is 20:19.