Gyrokinetic Simulation of ASDEX SOL with Implicit Moments-Conserving BGK Collision Operator
POSTER
Abstract
It is important to understand heat and particle transport in the scrape-off layer (SOL) of tokamaks to better understand power handling options in the divertor. Experiments on ASDEX have reported a significant enhancement of perpendicular transport as the effective collisionality increases in this region, due to a transition between the sheath-limited and the inertial filamentary regimes. It is therefore of interest to study the underlying mechanism with gyrokinetic simulation using Gkeyll, an electromagnetic continuum full-f code. High collisionality leads to tiny explicit time step limits and hence long simulation time. To achieve this regime with reasonable computational cost, we want to make the BGK collision operator implicit taking advantage of its simple form. The major challenge is the conservation of moments when projecting the Maxwellian onto the numerical basis functions. We matched the moments of the projected Maxwellian exactly to the moments of the numerical particle distribution function at each step by solving a nonlinear system of equations with an iterative method. While the collisionless term is still treated explicitly with our original SSP RK3 method, the BGK collision operator is treated implicitly after this explicit step using the backward Euler method. Applying the implicit moments-conserving BGK collision operator, we ran gyrokinetic simulation for realistic geometry of ASDEX SOL. Some preliminary results are shown in the poster.
Presenters
-
Dingyun Liu
Princeton University
Authors
-
Dingyun Liu
Princeton University
-
Gregory W Hammett
Princeton Plasma Physics Laboratory, Princeton Plasma Physics Laboratory (PPPL)
-
Ammar Hakim
Princeton Plasma Physics Laboratory (PPPL), Princeton Plasma Physics Laboratory
-
Manaure Francisquez
Princeton Plasma Physics Laboratory, Princeton Plasma Physics Laboratory (PPPL)
-
James L. Juno
Princeton Plasma Physics Laboratory, Princeton Plasma Physics Laboratory (PPPL)
-
Grant Johnson
PPPL
-
Akash Shukla
The University of Texas at Austin