Braginskii equations + Braginskii closure

[jwscook]

Equations being implemented in https://github.com/ExCALIBUR-NEPTUNE/NESO-Tokamak/tree/Turbulence

Need to write Nektar++ Riemann solver that can handle multiple different advection velocities. <- DONE

Partly blocked by DG discontinuity issue https://github.com/ExCALIBUR-NEPTUNE/nektar-diffusion/tree/je/DG_discontinuity

Let's fork NESO to make repaid changes for NESO-Tokamak and then push those changes when they are stable.
New NESO fork to have source and sink term changes driven by changes to NESO-Tokamak

Blocked by MPI boundary bug fix from DM

[jwscook]

Work has begun on an implicit version, to avoid the explicit MPI boundary bug.

[jwscook]

Braginskii equations are close to working. Drift velocities are almost working.

[JamesEdgeley]

Approximately working

braginskii.mp4

[jwscook]

Plan is to omit vorticity and drifts to verify the foundational layer of implementation.

[jwscook]

perhaps verify vs mms test or compare against Hermes-3 in slab.

• requires anomalous cross field transport

[jwscook]

Waiting for the diffusion fix from Nektar team to solve the MPI rank discontinuity for LDG (local DG) formulation.

[jwscook]

We can use CG for now but need DG for non-conforming elements

[jwscook]

Braginskii closure has been refactored into a Closure object.

[jwscook]

Waiting for the diffusion fix from Nektar team to solve the MPI rank discontinuity for LDG (local DG) formulation.

Increasing a numerical stability coefficient, LDGC11, solves this