Skip to content

geonhong/mlcfd

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Resources of the machine learning for CFD

volfrac

Estimate the volume fraction of cells in a fixed space

airfoil

A sample case for predicting the aerodynamic performances of an airfoil. It contains a mesh generator for an airfoil which is written in .m4 macro (Please refer to airfoil/system/blockMeshDict.m4). To generate a mesh, you need a airfoil data with .dat file extension. It is recommended to locate the airfoil data file in constant/airfoilData. Once the airfoil data file is prepared, you can generate the mesh by running meshgen.

Prepare the mesh generation

Manipulate the airfoil data file

Suppose that we want to generate a mesh for a Clarky airfoil. The airfoil data is prepared at constant/airfoilData/clarky.dat. To generate a mesh with blockMesh utility, the airfoil should be divided into 3 sections: upper, front, lower sections. Breakpoints between those three sections should be given in the airfoil data file with a character 'b'. For example,

constant/airfoilData/clarky.dat

...
0.36 0.0916266
0.34 0.0915079
0.32 0.0911857 /* Upper section until this point /
0.3 0.0906804 b /
Break point here /
0.28 0.0900016 /
Front section from this point */
0.26 0.089084
0.24 0.0878308
...

Then the airfoil data will be separated and generate sectional data file, such as airfoil.upper.z0, airfoil.upper.z1, airfoil.endPoint.z0, etc. The z0 represents the vertices on z=0 plane while z1 corresponds to z=0.1 plane. These manipulate files will be included in the blockMeshDict.m4 macro file to generate a blockMeshDict. To manipulate the airfoil data, one should run constant/airfoilData/airfoilmanip.sh as following.

$ constant/airfoilData/airfoilmanip.sh constant/airfoilData/clarky.dat
As you can see you should give the airfoil data file after the airfoilmanip.sh command, or it will use constant/airfoilData/airfoil.dat as default.

About

Machine Learning for CFD

Resources

Stars

4 stars

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors