Greifswald 2024 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

P: Fachverband Plasmaphysik

P 5: Magnetic Confinement II/HEPP II

P 5.4: Talk

Monday, February 26, 2024, 17:45–18:00, ELP 6: HS 3

Stochastic Single-Stage Stellarator Optimization for EPOS and Analysis of Coil Perturbations — •Pedro F. Gil, Jason Smoniewski, Paul Huslage, and Eve V. Stenson — Max-Planck-Institute for Plasma Physics, Garching, Germany

The EPOS (Electrons and Positrons in an Optimised Stellarator) project, as part of the APEX (A Positron Electron eXperiment) Collaboration, aims to build a small-scale stellarator for the confinement of pair plasmas. The magnetic field in EPOS will be quasi-axisymmetric, meaning that the magnetic field amplitude is invariant along a toroidal coordinate. This symmetry ensures the good confinement of trapped particle orbits. The device will receive a limited amount of positrons setting a constraint on its size, leading to unmanageable coil manufacturing and assembly tolerances. An analysis of the perturbations that affect the induced magnetic field is performed in order to guide the optimization towards robust configurations.

Stellarator optimization is usually a two-step process: find a target equilibrium, and design coils to match that desired equilibrium. Stochastic optimization of the coils randomly perturbs the shape of the coils N times and averages the magnetic field. This both broadens the width of the minima allowing to find more robust configurations and reduces the likelihood of getting trapped in local minima. Combined with a single-stage approach it smooths the objective function while searching for both coils and plasma. This method is expected to relax the coil construction constraints for EPOS into the ± 0.3 mm range, making it buildable.

Keywords: Optimization; Stellarator; Coils; Pair Plasma; Stochastic