A discontinuous galerkin time-domain method with dynamically adaptive cartesian mesh for computational electromagnetics

Su Yan, Chao Ping Lin, Robert R. Arslanbekov, Vladimir I. Kolobov, Jian Ming Jin

Research output: Contribution to journalArticlepeer-review

Abstract

A discontinuous Galerkin time-domain (DGTD) method based on dynamically adaptive Cartesian mesh (ACM) is developed for a full-wave analysis of electromagnetic (EM) fields. The benefits of hierarchical Cartesian grids and adaptive mesh refinement are demonstrated for linear EM propagation problems. The developed DGTD-ACM achieves a desired accuracy by refining nonconformal meshes near material interfaces to reduce stair-casing errors without sacrificing the high efficiency afforded with uniform Cartesian meshes. More importantly, DGTD-ACM can dynamically refine the mesh to resolve the local variation of the fields during propagation of EM pulses. A local time-stepping scheme is adopted to alleviate the constraint on the time-step size due to the stability condition of the explicit time integration. It is shown by numerical examples that the proposed method can achieve a good numerical accuracy and reduce the computational time effectively for linear problems of EM propagation in dispersive media. With further development, the method is expected to provide a powerful tool for solving nonlinear EM problems in plasma physics and electronics.

Original languageEnglish (US)
Article number7888928
Pages (from-to)3122-3133
Number of pages12
JournalIEEE Transactions on Antennas and Propagation
Volume65
Issue number6
DOIs
StatePublished - 2017

Keywords

  • Adaptive Cartesian mesh (ACM)
  • Discontinuous Galerkin time-domain (DGTD) method
  • Dynamic mesh adaptation
  • Electromagnetic (EM) simulation
  • Local time stepping (LTS)
  • Runge-Kutta method

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A discontinuous galerkin time-domain method with dynamically adaptive cartesian mesh for computational electromagnetics'. Together they form a unique fingerprint.

Cite this