Numerical methods for modeling of anisotropic plasmas in far-SOL

The environment of the far scrape-off layer is known to be extremely challenging for numerical modeling due to the anisotropy induced by the magnetic field and the gradient of plasma density. However, understanding of this region is crucial for simulations of radio frequency (RF) heating and optimization of this process for higher efficiency and stability. The traditional approaches for fluid modeling of the plasma transport rely on computational meshes aligned with the magnetic field lines, which become an unfeasible avenue for more geometrically accurate modeling of the near-wall plasma. The numerical methods developed within the MFEM (Modular Finite Element Methods) library offer an alternative pathway, relying on high-order finite elements on arbitrary (non-aligned) meshes fitting the detailed design of the fusion device. The numerical approach presented here, based on the Hybridizable Discontinuous Galerkin (HDG) discretization, is robust with respect to geometry and anisotropy. The favorable numerical properties of the method enable large-scale simulations on high performance computing infrastructure with highly scalable parallel iterative solvers. The properties of the method are showcased on multiple relevant problems and measured by benchmarks.