setup for commissionning is slighly different from setup for one simulation:
see README.md from python_magnetworkflows branch refactor
method: depends on geometry cfpdes with thmag_hcurl for Axi, "CG" for 3D (not implemented)
params:
- mdata: a json containing dict per magnet
- dict: value, step, type, filter*, relax param**, flow
- cooling methods: "mean" "meanH" "grad" "gradH" "gradHZ"
- hcorrelations: "Montgomery"
- frictions: "Constant"
- Tin: must be added (value + step)
params to be selected by user: cooling_methods, hcorrelations, frictions, (Tin, step) , (Inominal, step,flow) per magnet
structure of the archive would be different if we use only one coolings / hcorrelations / frictions
You need a tree structure like :
M9Bitters
|_ mean/M9Bitters-cfpdes-thmag_hcurl-nonlinear-Axi-sim.cfg
|_ meanH/M9Bitters-cfpdes-thmag_hcurl-nonlinear-Axi-sim.cfg
|_ grad/M9Bitters-cfpdes-thmag_hcurl-nonlinear-Axi-sim.cfg
|_ gradH/M9Bitters-cfpdes-thmag_hcurl-nonlinear-Axi-sim.cfg
|_ gradHZ/M9Bitters-cfpdes-thmag_hcurl-nonlinear-Axi-sim.cfg
- filter: needed when dealing with a site ()
- relax param: 0 (from 0 to 1 non inclus) - normalement on en a pas besoin
Pour setup_cmds:
utiliser python_magnetworkflows