🐛 Incorrect calculation of line and Bremsstrahlung plasma radiation components

[chris-ashe]

Describe the bug

Currently the total, core and edge radiation are calculated as follows in impurity_radiation.py for calculate_radiation_loss_profiles():

pden_impurity_rad_total = self.pimp_profile * self.rho
        pden_impurity_core_rad_total = self.pimp_profile * (
            self.rho
            * fradcore(
                self.rho,
                impurity_radiation_module.radius_plasma_core_norm,
                impurity_radiation_module.f_p_plasma_core_rad_reduction,
            )
        )

        self.pden_impurity_rad_profile = np.add(
            self.pden_impurity_rad_profile, pden_impurity_rad_total
        )
        self.pden_impurity_core_rad_profile = np.add(
            self.pden_impurity_core_rad_profile, pden_impurity_core_rad_total
        )

The total raidaition is calculated first, followed by the core (which may have a reduction factor applied). The edge is then calculated as the difference between the core and the total as in calculate_radiation_powers() :

pden_plasma_outer_rad_mw = (
        imp_rad.pden_impurity_rad_total_mw - imp_rad.pden_impurity_core_rad_total_mw
    )

This means a deficit of core radiation is instantly transferred to the edge even if the profiles are not consitent with this.

This can be seen in the large_tokamak file which shows that the majority of the radiation is in the edge when the radiation profile shows the contrary

Steps to reproduce

Present in all runs

Expected behaviour

The core and edge radiation components should be calculated first and then added together to produce the total. It is perfectly fine to add a core radiation change via f_p_plasma_core_rad_reduction to take into account the radiation data assumptions.

Environment

• OS:
• Version: 3.2.0

Additional context

[jonmaddock]

This has been fixed on a branch of mine, but I want to investigate the consequences combined with other changes more fully first before making a more isolated PR here.

[jonmaddock]

There are 4 related bugs going on here:

• Multiplication by rho
• Edge integral not being performed
• Core radiation used in the plasma power balance has been reduced artificially
• Edge radiation is not included in the plasma power balance

The reduction in radiation should only be used in the confinement time calculation to correct for highly-radiative regimes, and should not affect other calculations, especially the plasma power balance, where it will have a major effect on the location of solution vectors.