Year: 2023
Author: Baoli Yin, Yang Liu, Hong Li, Zhimin Zhang
Journal of Computational Mathematics, Vol. 41 (2023), Iss. 5 : pp. 980–1002
Abstract
A simple criterion is studied for the first time for identifying the discrete energy dissipation of the Crank-Nicolson scheme for Maxwell’s equations in a Cole-Cole dispersive medium. Several numerical formulas that approximate the time fractional derivatives are investigated based on this criterion, including the L1 formula, the fractional BDF-2, and the shifted fractional trapezoidal rule (SFTR). Detailed error analysis is provided within the framework of time domain mixed finite element methods for smooth solutions. The convergence results and discrete energy dissipation law are confirmed by numerical tests. For nonsmooth solutions, the method SFTR can still maintain the optimal convergence order at a positive time on uniform meshes. Authors believe this is the first appearance that a second-order time-stepping method can restore the optimal convergence rate for Maxwell's equations in a Cole-Cole dispersive medium regardless of the initial singularity of the solution.
You do not have full access to this article.
Already a Subscriber? Sign in as an individual or via your institution
Journal Article Details
Publisher Name: Global Science Press
Language: English
DOI: https://doi.org/10.4208/jcm.2210-m2021-0257
Journal of Computational Mathematics, Vol. 41 (2023), Iss. 5 : pp. 980–1002
Published online: 2023-01
AMS Subject Headings:
Copyright: COPYRIGHT: © Global Science Press
Pages: 23
Keywords: Discrete energy dissipation Crank-Nicolson scheme Maxwell's equations Shifted fractional trapezoidal rule Mixed finite element methods.
Author Details
-
Multi-output physics-informed neural network for one- and two-dimensional nonlinear time distributed-order models
Liu, Wenkai
Liu, Yang
Li, Hong
Yang, Yining
Networks and Heterogeneous Media, Vol. 18 (2023), Iss. 4 P.1899
https://doi.org/10.3934/nhm.2023080 [Citations: 1]