Draft: Implement the multi-ion MHD system of Toth #1427
Conversation
…rsion with only conservative flux and estimates of wavespeeds
…e on-species case
…n LLF variant that is consistent with single species MHD)
There was a problem hiding this comment.
My first impression is that there are a lot of type instabilities in lower-level functions due to the approach to multiple components. Some ways to resolve them are
- use
MVectors not crossing function borders - use
ntuplewithVal{ncomponents(equations)}() - use StrideArrays.jl as we do in our manual SIMD optimizations
- use
Base.Cartesian - last resort:
@generatedcode
src/Trixi.jl
Outdated
| @@ -174,14 +175,14 @@ export boundary_condition_do_nothing, | |||
| boundary_condition_wall, | |||
| BoundaryConditionNavierStokesWall, NoSlip, Adiabatic, Isothermal | |||
|
|
|||
| export initial_condition_convergence_test, source_terms_convergence_test | |||
| export initial_condition_convergence_test, source_terms_convergence_test, source_terms_standard | |||
There was a problem hiding this comment.
This is a very general name that can lead to a lot of confusion
|
It would be great to see some timer output to verify whether that's the case |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1427 +/- ##
==========================================
+ Coverage 96.37% 96.39% +0.02%
==========================================
Files 486 488 +2
Lines 39186 40090 +904
==========================================
+ Hits 37764 38644 +880
- Misses 1422 1446 +24
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
|
|
||
| cons = ("B1", "B2", "B3") | ||
| for i in eachcomponent(equations) | ||
| cons = (cons..., tuple("rho_" * string(i),"rho_v1_" * string(i), "rho_v2_" * string(i), "rho_v3_" * string(i), "rho_e_" * string(i))...) |
There was a problem hiding this comment.
This is probably not type stable, thus I wonder if this needs to be fixed. It would probably be safer to do so.
|
|
||
| prim = ("B1", "B2", "B3") | ||
| for i in eachcomponent(equations) | ||
| prim = (prim..., tuple("rho_" * string(i),"v1_" * string(i), "v2_" * string(i), "v3_" * string(i), "p_" * string(i))...) |
Co-authored-by: Hendrik Ranocha <[email protected]>
Co-authored-by: Michael Schlottke-Lakemper <[email protected]>
Co-authored-by: Michael Schlottke-Lakemper <[email protected]>
…st_wave compatible with single precision
… and removed duplicated code
This PR implements the non-conservative multi-ion MHD system of Toth. The multi-ion MHD system describes the motion of multi-species plasmas, where each ion species has its own heat capacity ratio and charge, which leads to independent mass, momentum, and energy equations for each ion species. See, e.g.,
In this PR, we include entropy consistent discretizations in 1D and 2D. We neglect the electron pressure for the moment, and do not consider collision or reaction source terms. No divergence-cleaning technique is added.