Revolve 2D MAST-U mesh toroidally to get a 3D mesh

UKAEA-Edge-Code / EdgeCodeRequirements

MIT License

1 stars 0 forks source link

Revolve 2D MAST-U mesh toroidally to get a 3D mesh #21

Open jwscook opened 3 weeks ago

jwscook commented 1 week ago

@futtersack please feel free to add here context around inaccurate 3D interpolation of the 2D magnetic field.

futtersack commented 1 week ago

Tried to draw something if it makes things more clear ? We have B finely defined on RZ plane. Currently I export the projections onto xyz planes in a B.pts file at 24 toroidal locations (500Mo). but the cells between those slices get interpolated in a "straight" line (beside taking 5mn per composant Bx, By, Bz from the large file). 2DequTo3D

JamesEdgeley commented 3 days ago

This is the full mesh that Romain produced

I've also made a NekMesh module to revolve a 2D Nektar++ mesh toroidally. This is the edge region from the 2D MASTU example:

cmacmackin commented 3 days ago

@futtersack Does this address your issues, or would it still be helpful for me to modify your script?

JamesEdgeley commented 3 days ago

Link to NekMesh revolve MR

jwscook commented 2 days ago

Now we need to run a simulation, ideally (ambipolar) diffusion on these meshes.