FUSE meeting notes: Jul 9 2024

[orso82]

Orso

• open FUSE
  • it's happening
  • lot of work ahead: public repo, documentation website, tutorials, press releases, presentations, and a reference paper
  • pay attention how you pitch new IR&Ds
    • ideas/models must stand on its own
    • IR&Ds need pipeline opportunities to target
• lots of work on pedestal_models branch across multiple repos
  • integrate EPED and WPED
  • faster transport and convergence
  • TEQUILA actually re-uses previous solution for stationary and dynamic plasma
  • faster IMASDD
  • handle GGD in IMASDD
  • profiled and speedup Hirshmar-Sigmar (was the bottleneck for transport calculations)

Long pulse discussion

• likely will be topical focused, not machine focused
• FUSE + control
• need to talk to Nick & Raffi
• present at the different facilities where we would like to target

Tom

• Opening up TJLF.jl (allowing Pablo & Erik to contribute)
• Nan + Tim worked on parametrizing ini to support ITBs

Adriana

• Spent time with Nan
  • Issue with NEO? No, issue with OMEGA worker node. Login node worked. Something changed in the OMEGA slurm. Tim and Nan have the same problem. Brendan also had issues unrelated to NEO.
  • now it works. They fixed something the background
  • Talked to Brendan about the risk
    • Need to improve the stability models
    • Adding density limit from a more recent paper
  • Adriana to look at DECAF code, see how it boils down different limits into a single disruptivity metric

Giacomo

• Working with Jerome on IR&D proposals
• Two objectives
  1. SOL
     • A 0D SOL model (like Nickols paper)
     • A 1D SOL model with onion skin (with cross-field transport via Perret's scale length)
  2. couple edge with SOL
     • how to connect the two?

Nan

• ion particle source calculation is missing in FUSE
• Tom will train a DIII-D neural net that includes the ion particle source
• New high-beta_p with Carbon and Nitrogen
  • Want to use this case to test multi-ion impurity transport
• Used TJLF.jl
• FUSE quite different from STEP. Needs more power.
• Need ExB shear calculation from NEO. Since rotation profile has big effect on transport. Must update NEO.jl to run vgen