arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1306.0227math.NAcs.NA
收藏

高阶渐近保持DG-IMEX方案用于扩散尺度下的离散速度动能方程

High Order Asymptotic Preserving DG-IMEX Schemes for Discrete-Velocity Kinetic Equations in a Diffusive Scaling

University of California Riverside(加州大学河滨分校);Rensselaer Polytechnic Institute(伦斯勒理工学院);University of Houston(休斯顿大学)

Juhi Jang, Fengyan Li, Jing-Mei Qiu, Tao Xiong

AI总结:

本文提出高阶渐近保持方案,用于离散速度动能方程在扩散尺度下的求解,能稳定地过渡到宏观模型如热方程、多孔介质方程等,采用微-宏观重构和高阶DG-IMEX方法实现高精度与稳定性。

摘要:

本文开发了一类高阶渐近保持方案,用于某些离散速度动能方程在扩散尺度下的求解,其在渐近极限下可转化为宏观模型,如热方程、多孔介质方程、对流扩散方程和粘性伯格斯方程。我们的方法基于动能方程的微-宏观重构,涉及自然分解为平衡和非平衡部分。为实现高阶精度和均匀稳定性以及正确捕获渐近极限,所提出的方法采用了任意阶精度的不连续伽辽金空间离散化与合适的数值通量,以及高阶全局刚性准确的隐式-显式Runge-Kutta方案,配备适当的隐式-显式策略。形式渐近分析显示,在epsilon趋近于0的极限下,所提出的方案是一个显式、一致且高阶离散化方案。数值结果展示了所提方案的稳定性、高阶精度及其在极限下的性能。

英文摘要:

In this paper, we develop a family of high order asymptotic preserving schemes for some discrete-velocity kinetic equations under a diffusive scaling, that in the asymptotic limit lead to macroscopic models such as the heat equation, the porous media equation, the advection-diffusion equation, and the viscous Burgers equation. Our approach is based on the micro-macro reformulation of the kinetic equation which involves a natural decomposition of the equation to the equilibrium and non-equilibrium parts. To achieve high order accuracy and uniform stability as well as to capture the correct asymptotic limit, two new ingredients are employed in the proposed methods: discontinuous Galerkin spatial discretization of arbitrary order of accuracy with suitable numerical fluxes; high order globally stiffly accurate implicit-explicit Runge-Kutta scheme in time equipped with a properly chosen implicit-explicit strategy. Formal asymptotic analysis shows that the proposed scheme in the limit of epsilon -> 0 is an explicit, consistent and high order discretization for the limiting equation. Numerical results are presented to demonstrate the stability and high order accuracy of the proposed schemes together with their performance in the limit.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑