SearcharxivSearch

arXiv subjects

Jan van Dijk

Publications and source records attributed to Jan van Dijk.

6 recordsLinked to original sources

Implementation and verification of the avalanche source in a 3D full-f particle-in-cell model of relativistic electrons for studies of tokamak disruptions

Disruptions threaten tokamak operation not only because of large in-vessel forces and thermal heat loads, but also because some electrons may be accelerated to relativistic energies. These so-called runaway electrons (REs) can multiply exponentially via knock-on collisions with thermal electrons. As the resulting RE avalanche is exponentially sensitive to the pre-disruption plasma current, multi-MA RE beams may form in large future devices, risking severe localized wall damage. Detailed understanding of RE beam formation and the particle phase-space distribution requires a self-consistent treatment of the RE avalanche and competing losses in the stochastic fields of MHD-active plasmas. Such simulations including the RE sources in 3D fields are needed to develop viable mitigation scenarios. For this, the 3D nonlinear MHD code JOREK includes a hybrid fluid-kinetic model, describing REs with a full-f relativistic particle-in-cell (PiC) approach using full-orbit or drift-kinetic descriptions. In this work, an energy and momentum conserving knock-on collision operator is implemented to enable accurate modeling of the RE phase-space dynamics in 3D electromagnetic fields. To make such novel high-fidelity simulations computationally viable, a resampling technique was also implemented to restrict the number of markers. The avalanche model is verified using analytical expressions from literature and applied to a JET-like termination scenario, demonstrating its applicability to realistic 3D MHD active scenarios. Future work on porting to accelerated high-performance computing systems will be needed to cross the long time scales involved, e.g., in periodic termination and re-avalanching that could occur in large devices like ITER.

physics.plasm-ph

MCPlas, a MATLAB toolbox for reproducible plasma modelling with COMSOL

The MCPlas toolbox represents a collection of MATLAB functions for the automated generation of an equation-based fluid-Poisson model for non-thermal plasmas in the multiphysics simulation software COMSOL. Following the development of the new generation of the LXCat platform, all input data are prepared in a structured and interoperable JSON format and can be supplied and validated using existing JSON schemas. The toolbox includes fully transparent, editable MATLAB source code and offers an advanced description of electron transport in addition to commonly used approaches in the plasma modelling community. It supports one-dimensional and two-dimensional modelling geometries employing Cartesian, polar and cylindrical coordinate systems. MCPlas is tested on two reference cases: DC- and RF-driven low-pressure glow discharges in argon. Comparison of MCPlas results with results obtained by employing COMSOL's Plasma Module verifies the reliability of the plasma model implemented by MCPlas and demonstrates the significance of electron transport treatment and boundary conditions applied in the toolbox. Using the same examples, the easy handling of complex reaction kinetic models in MCPlas and the reusability of its JSON input data across different modelling platforms are illustrated. This demonstrates that MCPlas provides a transparent and reproducible workflow for the simulation of non-thermal plasmas using COMSOL.

physics.plasm-ph

MHD simulations of formation, sustainment and loss of Quiescent H-mode in the all-tungsten ASDEX Upgrade

Periodic edge localized modes (ELMs) are the non-linear consequences of pressure-gradient-driven ballooning modes and current-driven peeling modes becoming unstable in the pedestal region of high confinement fusion plasmas. In future tokamaks like ITER, large ELMs are foreseen to severely affect the lifetime of wall components as they transiently deposit large amounts of heat onto a narrow region at the divertor targets. Several strategies exist for avoidance, suppression, or mitigation of these instabilities, such as the naturally ELM-free quiescent H-mode (QH-mode). In the present article, an ASDEX Upgrade equilibrium that features a QH-mode is investigated through non-linear extended MHD simulations covering the dynamics over tens of milliseconds. The equilibrium is close to the ideal peeling limit and non-linearly develops saturated modes at the edge of the plasma. A dominant toroidal mode number of $n=1$ is found, for which the characteristic features of the edge harmonic oscillation are recovered. The saturated modes contribute to heat and particle transport preventing pedestal build-up to the ELM triggering threshold. The non-linear dynamics of the mode, in particular its interaction with the evolution of the edge safety factor is studied, which suggest a possible new saturation mechanism for the QH-mode. The simulations show good qualitative and quantitative agreement to experiments in AUG. In particular, the processes leading to the termination of QH-mode above a density threshold is studied, which results in the transition into an ELM regime. In the vicinity of this threshold, limit cycle oscillations are observed.

physics.plasm-ph

Combining the hybrid mimetic mixed method with the Scharfetter-Gummel scheme for magnetised transport in plasmas

In this paper, we propose a numerical scheme for fluid models of magnetised plasmas. One important feature of the numerical scheme is that it should be able to handle the anisotropy induced by the magnetic field. In order to do so, we propose the use of the hybrid mimetic mixed (HMM) scheme for diffusion. This is combined with a hybridised variant of the Scharfetter-Gummel (SG) scheme for advection. The proposed hybrid scheme can be implemented very efficiently via static condensation. Numerical tests are then performed to show the applicability of the combined HMM-SG scheme, even for highly anisotropic magnetic fields.

physics.plasm-ph

2022 Review of Data-Driven Plasma Science

Data science and technology offer transformative tools and methods to science. This review article highlights latest development and progress in the interdisciplinary field of data-driven plasma science (DDPS). A large amount of data and machine learning algorithms go hand in hand. Most plasma data, whether experimental, observational or computational, are generated or collected by machines today. It is now becoming impractical for humans to analyze all the data manually. Therefore, it is imperative to train machines to analyze and interpret (eventually) such data as intelligently as humans but far more efficiently in quantity. Despite the recent impressive progress in applications of data science to plasma science and technology, the emerging field of DDPS is still in its infancy. Fueled by some of the most challenging problems such as fusion energy, plasma processing of materials, and fundamental understanding of the universe through observable plasma phenomena, it is expected that DDPS continues to benefit significantly from the interdisciplinary marriage between plasma science and data science into the foreseeable future.

physics.plasm-ph

Evaluation of a measure on the quasi-steady state assumption of Collisional Radiative Models via Intrinsic Low Dimensional Manifold Technique

Collisional and radiative dynamics of a plasma is exposed by so-called Collisional Radiative Models [1] that simplify the chemical kinetics by quasi-steady state assignment on certain types of particles. The assignment is conventionally based on the classification of the plasma species by the ratio of the transport to the local destruction frequencies. We show that the classification is not exact due to the role of the time-dependent local production, and a measure is necessary to confirm the validity of the assignment. The main goal of this study is to evaluate a measure on the quasi-steady state assumptions of these models. Inspired by a chemical reduction technique called Intrinsic Low Dimensional Manifolds [2, 3], an estimate local source is provided at the transport time-scale. This source is a deviation from the quasi-steady state for the particle and its value is assigned as an error of the quasi-steady state assumption. The propagation of this error on the derived quantities is formulated in the Collisional Radiative Models. Based on the error a novel technique is proposed to discriminate the quasi-steady states. The developed analysis is applied to mercury and argon fluorescent lamps separately and the corresponding errors are presented. We observe that the novel and conventional technique agrees for most of the excited levels but disagrees for a few low energy excited states.

physics.plasm-ph