Updated ITPA confinement scalings: ITPA20

[mkovari]

Discussed in https://github.com/ukaea/PROCESS/discussions/2889

^{Originally posted by **jonmaddock** July 9, 2018} In GitLab by @mkovari on Jul 9, 2018, 11:36 The analysis of the updated energy confinement scaling is now published: [The updated ITPA global H-mode confinement database: description and analysis](https://iopscience.iop.org/article/10.1088/1741-4326/abdb91/meta) G Verdoolaege et al > Extrapolation of this scaling to ITER yields a somewhat lower confinement time compared to the IPB98(y,2) prediction. The paper also includes regression results for individual machines, including MAST and NSTX, which may be helpful for ST work. The engineering scaling, henceforth referred to as ITPA20 is  The original version of this issue quoted a slightly different equation, in which triangularity appeared as $\delta_{95}$, rather than $\delta$. However, Geer Verdoolaege confirmed that the LCFS values should be used, so the formula above from the Nuclear Fusion paper is correct. This is based on the entire set of ELMy H-modes in the database, not restricted to the ITER-like subset. There is also another scaling expressed in dimensionless variables, not intended for the purpose of prediction, denoted ITPA20- IL-dim, using only ITER-like data. Mikhail Maslov is one of the authors. Note that Mikhail also has very important data on the dependence of confinement on the separatrix density $`n_{e,sep}`$ Suggestions * Update PROCESS. * Investigate what the updated database says about low aspect ratio machines. * Investigate what the database can tell us about the dependence of confinement on the separatrix density. @stuartmuldrew @jmorris-uk @ajpearcey @cjwgriesel @dshort

[mkovari]

Definitions.

[mkovari]

The table of energy confinement times now looks like this:

****************** Energy confinement times, and required H-factors : ***********************

      scaling law              confinement time (s)     H-factor for
                                   for H = 1           power balance
   ITPA20               (H)        2.072                     1.165
   IPB98(y,2)           (H)        2.454                     0.984
   ISS95            (stell)        1.441                     1.675
   ISS04            (stell)        2.502                     0.965
   DS03                 (H)        3.803                     0.635
   NSTX (Spherical)     (H)        5.437                     0.444
   NSTX-Petty08 Hybrid  (H)        4.571                     0.528
   NSTX gyro-Bohm Buxton(H)        38.020                    0.063

You can see that the ITPA20 confinement time is less than the IPB98(y,2) value.

With the large-tokamak scenario (rmajor fixed, hfact variable), the outcome when ITPA20 is selected is a little unexpected: hfact hardly changes (and remains below the upper bound), but all the other variables change, including the plasma current, helium content and radiation power. The xenon content has doubled.

[mkovari]

Now that we have an up-to-date confinement scaling, I suggest people use it.

* Switch for energy confinement time scaling law
isc = 50

[mkovari]

A unit test can be derived from this check I did: