README.md 2.04 KB
Newer Older
ylan's avatar
ylan committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
## 11145 (non-touching) pebbles in a cylinder (Last updated: Jun. 12, 2020)
cyl11145 N, Version 3.0, E3,575,076

This mesh is not very stable for Re10k. There is (at least) a known location having high distorted Cell.  
Run dt=5e-4 is fine for the first 14000 steps.   
After that dt=4e-4 will have CFL too high stop at 14000+5996th step (time=9.99800000e+00)   
We never manage to run over an convection unit  

This is a better mesh using improved techniques developed within last few months.  
Please note that the improvement has not finished.  

- New features:
  - A more robust and cleaner code with lots of user intreface improvement
  - Add Taubin smoothing before edge collapse
  - Switch to a passive tessellation if the first attempt didn't do well. 
  - Add local optimization for better performance.
  - dump co2 directly  
  - DEM results from project chrono on nurburg gpu
  - Fix some bugs as well  

- Geometry and BC:
  - Sphere distribution: flat-free
  - Sphere radius = 1.0 (Wall)
  - Cylinder radius = 40.672 (SYM), (convert SYM to Wall in usr file)
  - Zbot of the cylinder =  14.655 (Outflow)
  - Ztop of the cylinder = -12.627 (Inflow)
  - We use dimless scale to run Re10000 with vz=1.0

- Case files:
    - .re2, a curved (projected at lx1=3) hex20 re2 mesh, converted by a hex20 rea: xn3.rea+reatore2, R\_peb=1.0
      ```
        xyz min    -40.672      -40.672      -12.627    
        xyz max     40.672       40.672       14.655
      
        mesh metrics:
        GLL grid spacing min/max    : 3.67E-04 3.32E-01
        scaled Jacobian  min/max/avg: 1.52E-02 9.97E-01 3.65E-01
        aspect ratio     min/max/avg: 1.06E+00 6.06E+01 1.27E+01
      ```
    - Manage to run to time=7.0 with dt=5e-4 and the follow commit of NekRS
      ```
        commit c26b570af4a85e8ef8425b7e7057f0926f5aaace
        Author: Stefan Kerkemeier <stefanke@jlselogin2.ftm.alcf.anl.gov>
        Date:   Fri Aug 7 12:06:07 2020 +0000
        
            add missing file and fix gsh for scalars
      ```
    - peb.11k.dat, locations of the pebble centers under R\_peb=1.0