FOCUS updates on 07/18/2017

[zhucaoxiang]

Dear all,

I pushed several updates in both master and develop branches. They include

• Transplant coil diagnostic subroutine
  Now, if you use case_postproc = 1, the subroutine diagnos will be called. It evaluates the present coils set and show the results of

  • the cost functions;
  • average coil length (in debugging mode make dfocus, it shows all the coil length);
  • maximum coil curvature;
  • minimum coil-coil separation;
  • minimum coil-plasma separation.

  It could be used for running FOCUS without any optimizations (case_optimize = 0) and evaluating some the present coils set performance.

• Implement plasma and coil periodicity
  In previous versions, FOCUS was treating the plasma boundary as one period and all the coils independent with enforcing IsSymmetric = 0. Now, you have more freedoms.

  • IsSymmetric = 0

    the same as previous, no field periodicity, all coils independent, plasma boundary discretized in the whole toroidal angle with Nzeta.

  • IsSymmetric = 1

    apply the field periodicity of the plasma boundary, all coils independent, plasma boundary discretized in only one period with Nzeta.

    This could decrease the resolution required for discretizing plasma surface.

  • IsSymmetric = 2
    apply the field periodicity of both the plasma boundary and coils, Ncoils would be the number of coils per period and coils in other periods will be mapped automatically, plasma boundary discretized in only one period with Nzeta.

    This could decrease the number of degree of freedoms and accelerate the computation speed.

• Minor improvements
  There are also some internal improvments, such as,

  • change the code structure in bnormal to use less memory;
  • update documentations in github pages;
  • fix several hidden bugs

If you have any questions, pleas let me know (via email or github issues).

[lazersos]

It would be if there was an option that specified a coil as non-stellarator symmetric. For example in the CNT device the interlinked coils are stellarator symmetric, but the vertical field coils aren't. Maybe an option to specify coil type as -1.

[lazersos]

There seems to be a problem with the torflux routine and idof when issymmetric=2

[zhucaoxiang]

@lazersos Hi, Sam, do you have the case path which you have problem with? I can test it and try to fix the bugs. Yes, you are right. I haven't put stellarator symmetry in it yet. For CNT case, you can just use IsSymmetry = 0/1. I will let you know, if I put them in.

[zhucaoxiang]

I fixed the problem Sam mentioned. Please pull the updates. Thanks!