search

ukaea / PROCESS

PROCESS is a systems code at UKAEA that calculates in a self-consistent manner the parameters of a fusion power plant with a specified performance, ensuring that its operating limits are not violated, and with the option to optimise to a given function of these parameters.
https://ukaea.github.io/PROCESS/
MIT License
34 stars 11 forks source link

L-H threshold for ST #818

Closed jonmaddock closed 1 year ago

jonmaddock commented 5 years ago

In GitLab by @skahn on Apr 30, 2019, 11:07

@stuartmuldrew @hlux

The martin L-H threshold scaling seems completely off for STs. This can be observed in the figure 1 from the attached paper.

One of the reasons might be that explicit database cuts has been used for the fit to match with ITER as much as possible. The other reason also might also be the complete change of physics of the ST especially concerning the rotational properties driving the pedestal physics ...

I think we should not use it for ST. Is there any alternatives available more suitable for ST ?

Cheers Seb

Martin2008.pdf

jonmaddock commented 5 years ago

In GitLab by @skahn on Apr 30, 2019, 11:21

@process

One of the Data Base conditions for the scaling fit also worries me for DEMO : "The radiated fraction should be less that 50%" This condition is not observed in the 2018 baseline for instance where

Psep = 172 MW

Prad = Pline + Psync + Pbrem = 261 MW

jonmaddock commented 5 years ago

In GitLab by @skahn on Apr 30, 2019, 11:23

@ajpearcey @rkemp

The data base also select single null only configurations which might be good to keep in mind for the double-null DEMO designs currently in the pipeline.

Cheers Seb

jonmaddock commented 5 years ago

In GitLab by @hlux on Apr 30, 2019, 11:37

Yes, we are well aware that the LH-threshold scaling is not relvant for High radiation scenarios and metal wall machines.

jonmaddock commented 5 years ago

In GitLab by @stuartmuldrew on Apr 30, 2019, 11:38

Figure 1 does suggest that Martin et al. is an underestimate, however I would caution that there are only 23/7700 spherical tokamak points in Figure 1. It is also a log-log plot and all the spherical values are small, so is exploding a very small absolute difference. I'm not aware of an ST specific L-H threshold; we could check Ono & Kaita or go find a MAST person to ask.

jonmaddock commented 5 years ago

In GitLab by @stuartmuldrew on Apr 30, 2019, 11:44

@skahn As an aside, you may find this paper interesting: https://iopscience.iop.org/article/10.1088/0029-5515/55/11/113019/meta It looks at ways to fit the database. It doesn't solve any of our ST problems though!

jonmaddock commented 5 years ago

In GitLab by @stuartmuldrew on Apr 30, 2019, 11:59

Page 36 of Ono & Kaita does have some discussion of ST and double-null, however they conclude:

Overall, while there has been very good progress in obtaining an extensive H-mode experimental data base for STs and tokamaks, the fundamental understanding needed to explain the observed H-mode power threshold and develop an associated predictive capability is still lacking.

jonmaddock commented 5 years ago

In GitLab by @rkemp on Apr 30, 2019, 12:05

There’s an additional factor here which is that – IIRC – H-mode depends to a certain extent on rotational shear in the pedestal region, and STs tend to have much higher rotation than conventional tokamaks, so there are reasons to believe behaviour might differ. But as Stuart’s reference points out, this is still not well-understood.

jonmaddock commented 5 years ago

In GitLab by @skahn on Apr 30, 2019, 13:00

Thanks for your inputs !

jonmaddock commented 5 years ago

In GitLab by @skahn on Jun 5, 2019, 10:02

As this is just a discussion and the issue is clearely tracked by STEP experts (including Stuart hehe)

I am closing this issue

jonmaddock commented 5 years ago

In GitLab by @skahn on Jun 5, 2019, 10:02

closed