On 08/08/2014 04:31 PM, Geoff Wright wrote:
Hi Robert,
On Friday, August 8, 2014 4:33:03 AM UTC-4, Robert Cimrman wrote:
Hi Geoff,
On 08/07/2014 10:05 PM, Geoff Wright wrote:
Hi Robert,
I'm getting pretty good results now, even on my final model which is a bit more complex than what I've been showing you. In addition to the points your mentioned, which were very helpful, I also found that running the 'optimize 3D' option in gmsh does give a significant improvement in reducing the magnitude of this error. My hypothesis is that before running the mesh optimizer there may be a lot of very small cells which might have wild gradients. The 3D optimizer I assume does some sort of regularization, removing the small cells. After combining all these effects the loss in flux at the sink is approx 1%, which is acceptable for my purposes.
Glad to hear that! If it's not secret, I would like to see some nice figures :)
Unfortunately the full model is proprietary right now. Once it becomes public I will try and remember to come back here and add the figures!
No problem, just curious :)
Overall factors which helped:
- increasing mesh density around the areas of interest
- choosing the right solver
You are using pyamg, right?
Yes, I'm using pyamg as you described earlier in this thread
FYI: With the optimized mesh the pyamg solver converges even for approx_order = 2 (while for 3 it does not). You can safely reduce the integral order to be 2 * approx_order, to speed up assembling.
- using gmsh 3D optimizer to eliminate small cells
This has to be run from the GUI, or is there a command-line switch as well? I tried:
gmsh -3 sphere_disc_inside_sphere.geo -format mesh -optimize
Yes, I think that is equivalent to what I did in the GUI. Theres also the optimize_netgen option which I haven't experimented with.
ok, thanks!
r.