Central Solenoid current constraints

[jonmaddock]

In GitLab by @mkovari on Aug 15, 2014, 14:00

There is an odd fudge when calculating the Central Solenoid currents.
cohbof = min(fcohbof, 1.0D0)
fcohbof = max(fcohbof, -1.0D0)
where
fcohbof = ratio of OH coil overall current density at BOF / coheof,
coheof = OH coil overall current density at EOF.
I assume that cohbof is the CS current density at BOF.

I believe the two constraints should be sufficient:
(26) OH coil EOF current density limit,
(27) OH coil BOP current density limit,
without the need for this fudge.
(Formerly change request 43b)

[jonmaddock]

In GitLab by @pknight on Oct 16, 2014, 12:55

Actually, the above should read:

fcohbof = min(fcohbof, 1.0D0)
fcohbof = max(fcohbof, -1.0D0)

that is, abs(fcohbof) <= 1.0D0. (I'm not sure if this restriction is necessary, though - working on it now.)

The two constraints you mention are unrelated to this.

[jonmaddock]

In GitLab by @mkovari on Oct 16, 2014, 15:07

Code is still in place. We think it isn't causing a problem.

[mkovari]

The code now reads:

PROCESS/process/pfcoil.py

Lines 482 to 483 in 127ff4a

	pfv.fcohbof = min(pfv.fcohbof, 1.0e0) # constrains abs(fcohbof) <= 1.0;
	pfv.fcohbof = max(pfv.fcohbof, -1.0e0) # probably un-necessary

  real(dp) :: fcohbof
  !! ratio of central solenoid overall current density at beginning of flat-top / end of flat-top

The optimised value will depend on what constraints apply to the CS current, but generally we expect that the peak allowable current density will be reached immediately after pre-charge and also at the end of flat-top. Since some flux is used up during ram-up, it makes sense that |fcohbof| < 1. However, I can't see the need to impose this.

This is another example of code that could cause discontinuties in the objective function or constraints, interfering with convergence. I suggest we replace this code by a warning that |fcohbof| > 1 and see what happens.

@jonmaddock @timothy-nunn @clmould