FUSE meeting nots: May 14 2024

[orso82]

Orso

• Sherwood
• TJLF for SMARTS
• Reworked how layers outline build works. Now reliable also for negative triangularity.
• Costing of magnets meeting tomorrow
• Applied for a Stellar account (Tim as well)
• Trip3D, SURFMN to be rewritten in Julia by Tyler's student
• MFE reorg and IR&D reprioritization may bring $$ for DIII-D pulse design. What do we need?
  • priority 1
    • rabbit
    • limited plasma
    • breakdown, L, LH transition model (look at TSC, CORSICA, DINA, TokSys)
    • pulse_schedule, nbi
  • priority 2
    • free boundary equilibrium
  • priority 3
    • wall source
    • RMPs --> need a pumpout model, ask Nik Logan
    • sawtooth and/or anomalous flux pumping model for flux consumption
  • Need to pick a shot that we want to model
    • Fiducial DIII-D discharge
    • 163303 has a power step in it
    • 175865 has a torque-scan at fixed power

Tim

• Lots of fixes seen during optimization
• Added optimization of BOP (now we get reasonable numbers)
• Ran more optimizations
  • fixed issues with constraints
  • different optimizations for different LTS/HTS choices
  • results make a lot of sense now

Galina

• Failed on the FPP case
• Something failed

Adriana

• Helped Jackson setting up a skeleton for costing of BOP
• Wrapping rabbit
  • Still for D3D only
  • Prompt orbit losses on DIII-D seem high 50%

Nan

• TGYRO and FluxMatcher giving different results
  • Why?
  • TJLF seems to be doing ok
  • It's really TGLF that doesn't
  • Will run TGYRO, take output and compare with FUSE at that point
• Talked Tomas about impurity modeling
  • Agrees it should work! great!
  • Low Z affect fluxes
  • High Z affect radiation

Tom

• Will go on paternity leave
• Worked on DB generation for UKAEA/STEP
• Progress on QL-NN
• Should add the ability to generate database from ITBs
• Add extra particle outputs to NNs -- start with ITER sat1 em
• TJLF
  • Makes sense to partner with SMARTS for UQ. They'll throw in SciDAC resources.
  • TGLF-EP fails sometimes. Need to talk to Erik

[orso82]

From @bclyons12: Possible pulse design target: Shot 175865, from a torque-scan experiment in which the NBI power was held constant as the co- vs countercurrent NBI was varied.