Speaker
Description
The controlled and reliable generation of a high-power burning plasma in ITER requires an
understanding of energetic particle (EP) dynamics and subsequent wall heat loads [1]. Detailed
studies must however include realistic 3D magnetic equilibria, which are widely thought to
worsen EP confinement; this is particularly true for the interaction between edge-localised NBI
ions and RMP fields [2][3]. In ITER plasmas, EP slowing-down timescales are long (? 1s) and
plasma-facing component (PFC) geometry is complex (e.g. pipework under the ITER divertor
dome). Therefore generating smooth, high-fidelity EP distribution functions and wall heat loads
require a high-performance computing approach.
LOCUST-GPU has been designed in response to this challenge. The Monte Carlo algorithm
solves the Lorentz equation of motion for millions of EP markers in a static or rotating 3D
magnetic field together with a collision operator and high-precision PFC model. This is achieved
in hours by utilising GPGPU cards, each controlled by OpenMP threads, to track markers in
parallel.
Current efforts to validate LOCUST-GPU against similar fast ion codes are presented here,
and a close match with TRANSP predictions is shown. Additionally, preliminary studies of the
effects of DIII-D RMP fields on EP confinement are also presented.
References
[1] N. N. Gorelenkov, S. D. Pinches, K. Toi, Nuclear Fusion 54, 125001 (2014)
[2] K. Särkimäki et al., Nuclear Fusion 58, 076021 (2018).
[3] Van Zeeland, M. A., et al. Plasma Physics and Controlled Fusion 56.1, 015009 (2013)
This material is based upon work supported in part by the USA Department of Energy under
Award Number DE-FC02- 04ER54698. This work was supported by the Engineering and Physical
Sciences Research Council [EP/L01663X/1]. The views and opinions expressed herein do
not necessarily reflect those of the ITER Organization or the European Commission.
| Country or International Organization | United Kingdom |
|---|
