Direct stellarator coil design using global optimization: application to a comprehensive exploration of quasi-axisymmetric devices

The examples directory contain scripts for executing the globalization then phase I of the workflow in

Direct stellarator coil design using global optimization: application to a comprehensive exploration of quasi-axisymmetric devices, A. Giuliani, Arxiv

Background

There are two options for globalization: a naive approach where an ensemble of initial guesses obtained by perturbing initially flat coils, or a less ad-hoc approach based on TuRBO. We search for coils with near-axis quasisymmetry using the optimization problem described in:

Single-stage gradient-based stellarator coil design: Optimization for near-axis quasi-symmetry, A Giuliani, F Wechsung, A Cerfon, G Stadler, M Landreman, Journal of Computational Physics 459, 111147

The goal of the scripts in this work is now to properly globalize the direct coil design algorithm.

Installation

To use this code, first clone the repository including all its submodules, via

git clone --recursive 

Next, best practice is to generate a virtual environment and install PyPlasmaOpt there:

cd PyPlasmaOpt
python -m venv venv
source venv/bin/activate
cd LinkingNumber; mkdir build; cd build; cmake ..; make; cd ../../
cd TuRBO; pip install -e .; cd ..
pip install -e .

Running the scripts

To run the near-axis optimization with TuRBO globalization:

./ex_TuRBO.py arguments.txt

with naive globalization (perturbing the initial guess with Gaussian noise):

./ex_naive.py arguments.txt

Typically, the practitioner will have to run the optimization multiple times for a fair comparison of the the two techniques.