An anisotropic nonlinear stabilization for finite element approximation of Vlasov-Poisson equations Journal: Journal of Computational Physics

Junjie Wen, Murtazo Nazarov · 2025

Read An anisotropic nonlinear stabilization for finite element approximation of Vlasov-Poisson equations Journal: Journal of Computational Physics by Junjie Wen, Murtazo Nazarov (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-07785.txt.

Introduction

Junjie Wen and Murtazo Nazarov steady a fast solver for the equations of charged gas in an electric field of its own making, in a 2025 paper. The Vlasov-Poisson system tracks how the gas spreads in space and speed while its charge bends its own paths. Their method lays smooth polynomial pieces over space and steps time with a classic explicit scheme. Raw, that recipe rings with fake waves.

Their cure is a crafted fake thickness, an artificial viscosity that leans by direction and bends with the solution itself. The damping kills the ringing while sparing the true waves. Tests show the best hoped-for accuracy in both the space pieces and the time steps. Three classic trials, Landau damping and two forms of stream instability, run clean in two-dimensional phase space. The authors offer the scheme for plasma runs where sharp fronts and true calm must coexist.

Catalog record

CREATORJunjie Wen, Murtazo Nazarov
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE61 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-07785

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

An anisotropic nonlinear stabilization for finite element approximation of Vlasov-Poisson equations Journal: Journal of Computational Physics Junjie Wen, Murtazo Nazarov Abstract We introduce a high-order finite element method for approximating the Vlasov-Poisson equations. This approach employs continuous Lagrange polynomials in space and explicit Runge-Kutta schemes for time discretization. To stabilize the numerical oscillations inherent in the scheme, a new anisotropic nonlinear artificial viscosity method is introduced. Numerical results demonstrate that this method achieves optimal convergence order with respect to both the polynomial space and time integration. The method is validated using classic benchmark problems for the Vlasov-Poisson equations, including Landau damping, two-stream instability, and bump-on-tail instability in a two-dimensional phase space. 1 Introduction Plasma, one of the fundamental states of matter, is a crucial element in many academic and industrial fields. However, it is usually difficult to predict the motion of the plasma in practical research; this difficulty is mainly caused by the fluid nature of plasma and its charged state. The simulation of fluid plasma can be considered as a computational fluid dynamics (CFD) problem, which is known to be complicated. In addition, modeling plasma becomes even more challenging when the effects of electromagnetic fields are taken into consideration. To simulate plasma, the kinetic model is widely applied, where the distribution of charged particles in the phase space is quantified by a distribution function f(\boldsymbol{x},\boldsymbol{v},t) . This distribution function is determined by the position \boldsymbol{x} , the velocity \boldsymbol{v} , and the time t . The time evolution of f(\boldsymbol{x},\boldsymbol{v},t) under self-consistent electromagnetic fields is described by the Vlasov equation. To estimate the electromagnetic fields in the system, one approach is disregarding the magnetic field \boldsymbol{B} and computing the electric field \boldsymbol{E} by solving the Poisson equation, which gives the name Vlasov-Poisson to the system. The history of numerical analysis for Vlasov-Poisson equations dates back to the 1960s, see [ 1 ] . In general, the numerical schemes for PDEs that describe particle propagation such as Vlasov-Poisson equations can be classified into particle-based and grid-based methods. The Particle-in-Cell (PIC) method is one of the most popular particle-based methods. Many attempts have been made to apply the PIC method to Vlasov-Poisson equations, and some of them have achieved satisfactory results, see [ 2 , 3 , 4 , 5 , 6 , 7 ] for instance. The PIC method makes use of so-called super nodes, and these super nodes are used to approximate the distribution of particles. The trajectories of these super nodes in the phase space are governed by the Vlasov equation, and the electric field is computed on a grid of physical space. The advantage that drives the application of the PIC method is its relatively low computational cost compared to other numerical schemes, especially when it is applied to high-dimensional problems including Vlasov-Poisson. Nevertheless, the drawback of the PIC method is that applying super nodes can introduce a numerical error, and this error is decreased by 1/\sqrt{N} when the number of super nodes increases by N , according to the study [ 7 ] and references therein. Unlike particle-based methods, which track the trajectories of the super nodes, grid-based methods operate on a grid of the phase space and discretize the PDE by evaluating the function values at grid nodes. Popular grid-based methods for Vlasov-Poisson equations include finite volume (FV) and semi-Lagrangian schemes. The FV scheme was proven feasible for Vlasov-Poisson equations in [ 8 ] , where a first-order Vlasov-Poisson approximation is introduced. Subsequently, researchers have attempted to achieve higher-order and stabilized FV approximations with proper reconstruction algorithms and stabilization techniques; relevant studies can be found in [ 9 , 10 , 11 ] . The semi-Lagrangian method moves function data on the grid of phase space based on the characteristics of the Vlasov equation. This approach is unconditionally stable and allows large time steps, as introduced in [ 12 ] . The semi-Lagrangian method is often used in combination with other numerical frameworks, such as discontinuous Galerkin (DG), for instance [ 13 , 14 ] . In addition to the approaches mentioned above, some Vlasov-Poisson solvers employing other numerical schemes such as finite difference (FD) and spectral methods have also been proposed, see [ 15 , 16 , 17 ] and references therein for instance. The finite element (FE) scheme is a robust numerical method that divides the computation domain into a finite number of subdomains and approximates unknown functions within each subdomain using a set of basis functions. This method is known to be able to handle complex geometries and boundary conditions, and can achieve high accuracy through mesh refinement. The application of the continuous FE method to Vlasov-Poisson equations has not been extensively studied yet. A FE approximation of the Vlasov-Poisson equation was published in 1988; however, there has been a notable lack of subsequent research building on this foundation in recent decades. The authors of the 1988 paper verified some properties of the FE approximations without exploring the high accuracy potential of the FE method, see more details in [ 18 , 19 ] . Additionally, it is well-known that the FE method introduces numerical oscillations when it is applied to conservation laws, including Vlasov-Poisson equations. Notably, this issue was not addressed in that paper. The residual-based artificial viscosity method serves as an effective stabilization technique for FE approximations. This method enhances stability by incorporating a viscosity term based on the PDE residual. It was proved in [ 20 ] that the FE approximation of scalar conservation laws converges to the unique entropy solution when using the residual-based artificial viscosity. The FE method with the residual-based artificial viscosity has been applied to many models of CFD and computational plasma, such as Navier–Stokes equations and Magnetohydrodynamics, e.g., [ 21 , 22 ] . In this article, we introduce a continuous FE approximation of the Vlasov-Poisson equation; this approximation utilizes high-order Lagrange polynomials in space and is explicit in time. The oscillations in the approximation are addressed by the residual-based artificial viscosity scheme and the numerical results prove that this approach can stabilize the FE approximation without degrading its accuracy. The Poisson equation is solved on a grid of the physical space using the continuous FE method. We implement our methods on the DOLFINx platform [ 23 ] , an open-source FE library that includes parallelization packages to speed up the calculations. We would like to clarify that the implementation of the methods to high-dimensional Vlasov-Poisson is not included in the article. This article is organized as follows: In Section 2 , all the equations in the system are introduced together with some conservation properties of Vlasov-Poisson equations. In Section 3 , important settings for the FE schemes such as the meshes, the elements, and the function spaces are defined. In Section 3.1 , we introduce a nonlinear finite element framework for Vlasov-Poisson equations using a residual-based artificial viscosity method. The results of numerical experiments are shown in Section 4 . In addition, we apply our methods to the guiding-center model, which serves as a variant of Vlasov-Poisson, and the results are presented in A . 2 Preliminaries In this section, we introduce the equations that model the distribution and propagation of plasma. We

Read the complete text