Issue in recreating plasma configuration with forward solver using coil currents from inverse solver.

[geekygiks]

Hi,

I am running freegsnke ver 2.0 and I am facing issues in recreating the plasma shape profile. The working schema is as follows,

1. Initially, I fix all the coil position and pass the following to inverse solver,

eq.tokamak.set_coil_current('CS1', 20e6)
eq.tokamak['CS1'].control = True
eq.tokamak.set_coil_current('CS2', 20e6)
eq.tokamak['CS2'].control = True
eq.tokamak.set_coil_current('CS3', 20e6)
eq.tokamak['CS3'].control = True

eq.tokamak.set_coil_current('PF1', 9.742e+06)
eq.tokamak['PF1'].control = True
eq.tokamak.set_coil_current('PF2', 10.0e+06)
eq.tokamak['PF2'].control = False
eq.tokamak.set_coil_current('PF3', -3.381e+06)
eq.tokamak['PF3'].control = True

I am using paxis module

from freegsnke.jtor_update import ConstrainPaxisIp
profiles = ConstrainPaxisIp(
eq=eq,
paxis=1.30E+06,
Ip=9.0e6,
fvac=24,
alpha_m=2.5,
alpha_n=1.2
)

with inverse_solver running as follows,

from freegsnke import GSstaticsolver
GSStaticSolver = GSstaticsolver.NKGSsolver(eq)
GSStaticSolver.inverse_solve(eq=eq,
profiles=profiles,
constrain=constrain,
target_relative_tolerance=1e-8,
verbose=False, # print output
)

I get a set of coil currents for my pre-defined constrain parameters.

{'CS1': np.float64(33775637.18984738),
'CS2': np.float64(8452064.547979241),
'CS3': np.float64(26347627.95361948),
'PF1': np.float64(6050308.471069523),
'PF2': np.float64(10000000.0),
'PF3': np.float64(-8388258.053080285)}

2. This I pass to my forward solver in different notebook keeping every other thing same.

eq.tokamak.set_coil_current('CS1', np.float64(33775637.18984738))
eq.tokamak['CS1'].control = False
eq.tokamak.set_coil_current('CS2', np.float64(8452064.547979241))
eq.tokamak['CS2'].control = False
eq.tokamak.set_coil_current('CS3', np.float64(26347627.95361948))
eq.tokamak['CS3'].control = False
eq.tokamak.set_coil_current('PF1', np.float64(6050308.471069523))
eq.tokamak['PF1'].control = False
eq.tokamak.set_coil_current('PF2', np.float64(10000000.0))
eq.tokamak['PF2'].control = False
eq.tokamak.set_coil_current('PF3', np.float64(-8388258.053080285))
eq.tokamak['PF3'].control = False

and solving with

#call the solver
GSStaticSolver.solve(eq=eq,
profiles=profiles,
constrain=None,
target_relative_tolerance=1e-8,
verbose=False, # print output
)
I am not getting exact plasma shape as I receive while inverse solver.