Current drive fractions do not add to 1

[jonmaddock]

In GitLab by @mkovari on May 18, 2023, 11:53

This code falsifies the current drive fractions so they do not add to 1. No error message appears in the output file.
EDIT: the variable names have been changed. The code is now as follows.

        current_drive_variables.f_c_plasma_internal = (
            current_drive_variables.f_c_plasma_bootstrap
            + current_drive_variables.f_c_plasma_diamagnetic
            + current_drive_variables.f_c_plasma_pfirsch_schluter
        )

        # Plasma driven current fraction (Bootstrap + Diamagnetic
        # + Pfirsch-Schlüter) constrained to be less than
        # or equal to the total fraction of the plasma current
        # produced by non-inductive means (which also includes
        # the current drive proportion)
        physics_module.err243 = 0
        if (
            current_drive_variables.f_c_plasma_internal
            > physics_variables.f_c_plasma_non_inductive
        ):
            current_drive_variables.f_c_plasma_internal = min(
                current_drive_variables.f_c_plasma_internal,
                physics_variables.f_c_plasma_non_inductive,
            )

This variable is only used in one place:

        #  Fraction of plasma current produced by auxiliary current drive
        physics_variables.f_c_plasma_auxiliary = (
            physics_variables.f_c_plasma_non_inductive
            - current_drive_variables.f_c_plasma_internal
        )

This is needed because f_c_plasma_non_inductive can be an iteration variable, so we need to be able to calculate things from it.

Although negative auxiliary current drive is physically possible, we don't have any way of handling it in PROCESS.

As far as I can see, f_c_plasma_auxiliary is only used in one place:

          # Calculate the injected power for the primary heating method
            current_drive_variables.p_hcd_primary_injected_mw = (
                1.0e-6
                * (
                    physics_variables.f_c_plasma_auxiliary
                    - current_drive_variables.f_c_plasma_hcd_secondary
                )
                * physics_variables.plasma_current
                / current_drive_variables.eta_cd_hcd_primary
            )

@stuartmuldrew ?

[jonmaddock]

In GitLab by @kj5248 on May 18, 2023, 12:00

Bootstrap_fraction______________________________________________________ (bootipf)_____________________ 7.0000E-01
Diamagnetic_fraction____________________________________________________ (diaipf)______________________ 0.0000E+00
Pfirsch-Schlueter_fraction______________________________________________ (psipf)_______________________ 0.0000E+00
Auxiliary_current_drive_fraction________________________________________ (faccd)_______________________ 0.0000E+00
Inductive_fraction______________________________________________________ (facoh)_______________________ 3.0550E-01
Total___________________________________________________________________ (plasipf+faccd+facoh)_________ 1.0000E+00
Fraction_of_the_plasma_current_produced_by_non-inductive_means__________ (fvsbrnni)____________________ 6.9450E-01 ITV

[jonmaddock]

In GitLab by @stuartmuldrew on May 18, 2023, 16:53

I'm confused in the above example as there is no heating and current drive. I think this goes back to the odd way PROCESS has handled the bootstrap current limit, where it enforces the maximum by overwriting the value rather than modifying the plasma solution.

[chris-ashe]

Related issue #3273

[j-a-foster]

@chris-ashe and I have looked at this and it seems that the current fractions are decoupled in how they're calculated and relate to one another. The non-inductive current fraction is an iteration variable but I can't see what constrains its value? This then sets the plasma driven current but as that is bootipf + diaipf + psipf, it doesn't then report the fractions correctly summing to 1 as bootipf is the only value reported, not plasipf.

Also, the inductive fraction has no relation to the CS whatsoever, it's simply: 1 - non-inductive fraction

[mkovari]

The relevant section of a new OUT.DAT is now like this:

 Bootstrap fraction                                                        (f_c_plasma_bootstrap)          4.21908069934171792e-01 OP
 Diamagnetic fraction                                                      (f_c_plasma_diamagnetic)        0.00000000000000000e+00 OP
 Pfirsch-Schlueter fraction                                                (f_c_plasma_pfirsch_schluter)   0.00000000000000000e+00 OP
 Auxiliary current drive fraction                                          (f_c_plasma_auxiliary)          2.31037373389791911e-03 OP
 Inductive fraction                                                        (f_c_plasma_inductive)          5.75781556331930289e-01 OP
 Fraction of the plasma current produced by non-inductive means            (f_c_plasma_non_inductive)      4.24218443668069711e-01

The Total (plasipf+faccd+facoh) line has been removed.

[grmtrkngtn]

Has this issue been resolved @mkovari ?

As far as I can see, on the current Large Tokamak example the fractions sum to 1.

Closing this issue pending further comment, as I cannot replicate the problem.

[mkovari]

The questionable code is still present (now with an error code):

        # Plasma driven current fraction (Bootstrap + Diamagnetic
        # + Pfirsch-Schlüter) constrained to be less than
        # or equal to the total fraction of the plasma current
        # produced by non-inductive means (which also includes
        # the current drive proportion)
        physics_variables.err243 = 0
        if (
            current_drive_variables.f_c_plasma_internal
            > physics_variables.f_c_plasma_non_inductive
        ):
            current_drive_variables.f_c_plasma_internal = min(
                current_drive_variables.f_c_plasma_internal,
                physics_variables.f_c_plasma_non_inductive,
            )
            physics_variables.err243 = 1

        # Fraction of plasma current produced by inductive means
        physics_variables.f_c_plasma_inductive = max(
            1.0e-10, (1.0e0 - physics_variables.f_c_plasma_non_inductive)
        )

@ym1906 I suggest we leave this open but put it down as low priority.