The pseudo-inverse mapping as a noise reduction technique in global gyrokinetic magnetic fusion plasma simulations with the XGC code
ORAL
Abstract
We study a velocity-space coarse-graining technique based on an arbitrary/flexible-order pseudo-inverse mapping to control particle noise in gyrokinetic particle-in-cell (PIC) simulations with the XGC code (the X-point Gyrokinetic Code). XGC is a 5D total-f code used to study multiscale kinetic turbulence in magnetic fusion plasmas. The code evaluates dissipative operations such as Fokker-Planck Coulomb collisions on a 5D (3D configuration and 2D velocity space) grid. XGC was recently equipped with a novel mapping between the marker particles and the uniform 2D velocity-space grid based on the calculation of a pseudo-inverse to enable exact conservation properties of particle number, momentum and energy [A. Mollen et al. J. Plasma Phys. 87 905870229 (2021)]. The pseudo-inverse mapping carries several interesting mathematical properties and has subsequently been generalized to use a discretization space of any order and to allow for the usage of unstructured 2D grids in the coarse-graining operation (mapping back-and-forth between the particles and the grid).
*This work was supported by the U.S. Department of Energy under contract number DE-AC02-09CH11466, the ECP program via funding to WDMApp (17-SC-20-SC), and by the U.S. Department of Energy Office of Science ASCR and FES through the SciDAC-4 Partnership Center for High-fidelity Boundary Plasma Simulation (HBPS).
–
Presenters
-
Albert V Mollen
- Princeton Plasma Physics Laboratory