SOLPS-NN integration

[sballinger-cfs]

Hi all, starting this thread to discuss the nuts and bolts of integrating SOLPS-NN in TORAX. To reiterate some info from the meeting, here are the lists of parameters each code can provide to the other, in order of priority:

SOLPS-NN → TORAX

1. nsep
2. 1D profile of fueling source
3. 1D profile of heat sinks in core (CX, ionization)

(Coupled quantities TORAX can get from XL:)
4. Separatrix impurity concentration
5. Tesep, Tisep

(Post-processed quantities TORAX can get from XL:)
6. Target temperature/parallel heat flux
7. pdiv

TORAX → SOLPS-NN

1. Core boundary heat flux, Psep
2. Main ion particle flux from core boundary (more uncertainty than Psep)
3. Transport coefficients from core boundary to separatrix
4. Impurity concentration (at least for comparison)

As a first step it would be good to see the current interface to edge models (and quantities in the TORAX → SOLPS-NN list) in TORAX and discuss any modifications needed for SOLPS-NN.