Efficient solution on solving 3D Maxwell equations using stable semi-implicit splitting method
Author(s) -
Wei Cen,
Ning Gu
Publication year - 2016
Publication title -
aip advances
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.421
H-Index - 58
ISSN - 2158-3226
DOI - 10.1063/1.4948771
Subject(s) - discretization , alternating direction implicit method , finite difference time domain method , maxwell's equations , stability (learning theory) , domain (mathematical analysis) , mathematics , crank–nicolson method , finite difference method , computer science , scheme (mathematics) , mathematical analysis , physics , quantum mechanics , machine learning
In this paper, we propose an efficient solution on solving 3-dimensional (3D) time-domain Maxwell equations using the semi-implicit Crank-Nicholson (CN) method for time domain discretization with advantage of unconditional time stability. By applying the idea of fractional steps method (FSM) to the CN scheme, the proposed method provides a much simpler and efficient implementation than a direct implementation of the CN scheme. Compared with the alternating-direction implicit (ADI) method and explicit finite-difference time-domain approach (FDTD), it significantly saves the computational resource like memory and CPU time while remains similar numerical accuracy
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom