A meshless local Petrov-Galerkin method for the time-dependent Maxwell equations

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS

卷:268

Article

Dehghan, Mehdi¹  Salehi, Rezvan¹ 

[1] Amirkabir Univ Technol, Dept Appl Math, Fac Math & Comp Sci, Tehran 15914, Iran

关键词: Local polynomial approximation;    Moving least square reproducing kernel (MLSRK) method;    Meshless local Petrov-Galerkin (MLPG) method;    Maxwell equations;    Error estimation;    Electromagnetic equations;   

DOI  :  10.1016/j.cam.2014.02.013

来源: Elsevier

【 摘 要 】

In this paper, the meshless local Petrov-Galerkin (MLPG) method is employed to solve the 2-D time-dependent Maxwell equations. The MLPG method is a truly meshless method in which the trial and test functions are chosen from totally different functional spaces. In the current work, the moving least square reproducing kernel (MLSRK) scheme is chosen to be the trial function. The method is applied for the unsteady Maxwell equations in different media. In the local weak form, by employing the difference operator for evolution in time and simultaneously in time and space, the semi-discrete and fully discrete schemes are obtained respectively. The error estimation is discussed for both the semi-discrete and fully-discrete numerical schemes for modelling the time-dependent Maxwell equations. We show that provided that the time step size T is sufficiently small, the proposed scheme yields an error of O(p(2(m+1))+tau(2)) in the L-2 norm for the square of error. The new scheme is implemented and the numerical results are provided to justify our theoretical analysis. (C) 2014 Elsevier B.V. All rights reserved.

【 授权许可】

Free