This is a repository of project scripts for the ITCR Plasma Laboratory. It considers some magnetic field calculation routines for the MEDUSA tokamak, particle trajectories using the SCR-1 magnetic field, and diagnostics for SCR-1 magnetic field errors.

The jupyter notebooks in this project fit the following function to BS-SOLCTRA's data output,

$$R(\theta, \phi) = \sum_{i=0}^{N}\sum_{j=0}^{M} A_{ij}\cos(j\theta-iN_{fp}\phi)$$

$$Z(\theta, \phi) = \sum_{i=0}^{N}\sum_{j=0}^{M} B_{ij}\sin(j\theta-iN_{fp}\phi)$$

This allows for the analytical calculation of the area and volume of the magnetic field surfaces, which is used in the magnetic field error minimization routine.

In this module, a magnetic field coil optimization was carried out for the SCR-1 stellarator's modular coil design. The code was designed to optimize the fourier coefficients of the 12 modular coils using a genetic algorithm based on the article "A Continuous Genetic Algorithm Designed for the Global Optimization of Multimodal Functions" and concepts from chapter three of "Practical genetic algorithms" by Randy and Sue Haupt. The error function used was defined as the plasma surface average of the local magnetic field error function given by,

$$e_l = \frac{\textbf{B}·\textbf{n}}{|\textbf{B}|}$$

So the average magnetic field error which is to be optimized is,

$$e_{avg} = \frac{1}{A}\int_\Omega e_l dA$$

Test runs for this module haven't been carried out, but the genetic algorithms programmed were tested in jupyter notebooks.

These projects are still on-course and updated periodically.

Developer:

• Johansell Villalobos Cubillo -- IF