WIMAS-2 Weekly Meeting

Wednesday 23 Sept 2020, 10:00 → 11:00 Europe/Berlin

https://kth-se.zoom.us/j/7764082514

Thomas Jonsson

WIMAS-2 Weekly Meeting are technical meetings to plan and coordinate the development of the ETS-6 workflow.

 

Remote participation:

https://kth-se.zoom.us/j/7764082514

1 Actions from previous meetings

https://docs.google.com/spreadsheets/d/1Ps-cGQyKNsYTObCr5Kqu58bXBNQdH45Fjpz5QD0hLB0/edit?usp=sharing

Speaker: Thomas Jonsson

Summary of progress on actions

The list of actions can be found here:

https://docs.google.com/spreadsheets/d/1Ps-cGQyKNsYTObCr5Kqu58bXBNQdH45Fjpz5QD0hLB0/edit?usp=sharing

34 - Done
35 - Done
36 - Done
(These 3 tested, need to be merged into develop branch)
37 - now low priority
39 - Done (needs adding to dressed Kepler)
40 - Testing planned for next week for ETS5. ~3 weeks time for IMAS compatibility
41 - Done
42 - Done
43 - proposals made in slides presented by TJ. DY pointed out the fractional solution (electron density and coronal) in ETS5
44 - proposals made in slides presented by TJ. DY asked about meaning of flat from BC. TJ to clarify that the value will be from the states not coronal distribution.

DY - Impurity densities can already be initialised to be in coronal equilibrium, but this needs testing (action on this)
AP - A useful initial profile would be better to start from zero, or a small impurity density (action on this).
DC - Options for initialising temperature? TJ - we take from input IDS. DC - suggest we take AP proposal further. Discussion on extending ETS_Init since a python GUI is not currently in scope/budget. 

Actions:

• In ets-init, add option to set impurity concentration w.r.t. electron density profiles; the charge state densities would be given by coronal
• Add option to set BCs for impurity density as a fraction of the electron density; the charge state densities would be given by coronal
• Add clarification of the BC called "flat_from_BC".
• Test that impurities can be initialised to be in coronal equilibrium
• Add option to initialise impurities with a fixed density, "flat_from_value".
• Long term: We should add the option to initialise temperatures and densities from analytical profiles.

TJ=Thomas Jonsson, DY=Dmitriy Yadykin, DC=David Coster, AP=Alexei Polevoi

2 ETS-6 News

Speaker: Thomas Jonsson

ETS-6- Suggested codeparameter updates.pptx

AGENDA

Branches merged into "develop"

• sprint1 - implementations of internal boundary conditions and hyperdiffusion, see next presentation.
• revise_impurity - developments from Denis and Irena on impurities. 
  • Much more work is needed on impurities.
  • So far I've shown that we can with impurities in a STATIC mode.
• coreprofData - Make sure the data stored in core_profiles are correct; deallocate instead of filling with zeros.

Branches under development/testing

• bulk_rotation - Testing not yet ready.
• codeparam - Update code parameter lists for impurities and transport solver as suggest at weekly meetings, see slides.
• outputHyperdiffusion - Store hyper-diffusion in core_transport as a transport model 

 

3 Summary of Sprint-1 on Internal Boundary Conditions and Hyperdiffusion

Speaker: Thomas Johnson (Fusion Plasma Physics (KTH/VR))

Sprint-1 Internal boundary and hyperdiffusion.pptx

Discussion on renaming ‘internal boundary’
DECISION - use ‘core-edge boundary’

TJ - Shall we use the same core-edge boundary for main ions and impurities?
 -   DC - Yes, for now. If we need to split them later we'll do so.

DC - Do we always want to keep the flux equation all the way to the seperatrix? Discussion follows.
AP - Can we allow different species to have different core-edge boundary?
DY - Not possible, since all ions are solved in one matrix equation.
DC - A possible update for the future. 

AP - option to specify time-dependant function for electron edge profiles?
TJ - think about for edge sprint

TJ - hyperdiffusion for impurities?
DECISION - yes with current implementation for now

TJ - How to store hyperdiffusion?
DECISION - Both options. May need to expand option two in future

TJ - How to calculate energy flux from hyperdiffusion? Discussion has been ongoing on Slack.
Flux multiplier for hyper diffusion?
DC - Proposed to set the flux multiplier to zero, such that one can isolate non-zero fluxes from the density and temperature equations.
AP - Note that with the flux multiplier set to zero you can have problem analysing energy conservation.

Discussion…
AP - asked about neutrals. 
TJ - Neutrals will be implemented by Irena.
DC - In principle, we could try to run an ion Z=0 to mimic a neutral.

Actions:

• Implement the same core-edge boundary for main ions and impurities.
• Implement the same hyper-diffusion for main ions and impurities.
• Store hyper-diffusion both as a separate model in core_transport and as a term in the "transport_solver" model.

TJ=Thomas Jonsson, DY=Dmitriy Yadykin, DC=David Coster, AP=Alexei Polevoi

4 Testing of internal boundary conditions

Speaker: Dmitriy Yadkin (Chalmers)

test_ibc_imas.pdf

DC - Temp drop comparing Te and Ti very different. Would have expected similar drops. Some discussion on this...
DC - initial conditions may explain difference.

5 Any other business