U.S. Department of EnergyOffice of Scientific and Technical Information
Parallel computation of electromagnetic fields
Abstract
The DSI3D code is designed to numerically solve electromagnetics problems involving complex objects by solving Maxwell`s curl equations in the time-domain and in three space dimensions. The code has been designed to run on the new parallel processing computers as well as on conventional serial computers. The DSI3D code is unique for the following reasons: It runs efficiently on a variety of parallel computers, Allows the use of unstructured non-orthogonal grids, Allows a variety of cell or element types, Reduces to be the Finite Difference Time Domain (FDID) method when orthogonal grids are used, Preserves charge or divergence locally (and globally), Is non- dissipative, and Is accurate for non-orthogonal grids. This method is derived using a Discrete Surface Integration (DSI) technique. As formulated, the DSI technique can be used with essentially arbitrary unstructured grids composed of convex polyhedral cells. This implementation of the DSI algorithm allows the use of unstructured grids that are composed of combinations of non-orthogonal hexahedrons, tetrahedrons, triangular prisms and pyramids. This algorithm reduces to the conventional FDTD method when applied on a structured orthogonal hexahedral grid.
- Authors:
- Publication Date:
- Research Org.:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Org.:
- USDOE, Washington, DC (United States)
- OSTI Identifier:
- 620998
- Report Number(s):
- UCRL-ID-127450
ON: DE98051059
- DOE Contract Number:
- W-7405-ENG-48
- Resource Type:
- Technical Report
- Resource Relation:
- Other Information: PBD: 21 May 1997
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; 66 PHYSICS; PARALLEL PROCESSING; COMPUTER CODES; MAXWELL EQUATIONS; COMPUTER CALCULATIONS; ARRAY PROCESSORS; VECTOR PROCESSING
Citation Formats
Madsen, N K. Parallel computation of electromagnetic fields. United States: N. p., 1997.
Web. doi:10.2172/620998.
Madsen, N K. Parallel computation of electromagnetic fields. United States. https://doi.org/10.2172/620998
Madsen, N K. 1997.
"Parallel computation of electromagnetic fields". United States. https://doi.org/10.2172/620998. https://www.osti.gov/servlets/purl/620998.
@article{osti_620998,
title = {Parallel computation of electromagnetic fields},
author = {Madsen, N K},
abstractNote = {The DSI3D code is designed to numerically solve electromagnetics problems involving complex objects by solving Maxwell`s curl equations in the time-domain and in three space dimensions. The code has been designed to run on the new parallel processing computers as well as on conventional serial computers. The DSI3D code is unique for the following reasons: It runs efficiently on a variety of parallel computers, Allows the use of unstructured non-orthogonal grids, Allows a variety of cell or element types, Reduces to be the Finite Difference Time Domain (FDID) method when orthogonal grids are used, Preserves charge or divergence locally (and globally), Is non- dissipative, and Is accurate for non-orthogonal grids. This method is derived using a Discrete Surface Integration (DSI) technique. As formulated, the DSI technique can be used with essentially arbitrary unstructured grids composed of convex polyhedral cells. This implementation of the DSI algorithm allows the use of unstructured grids that are composed of combinations of non-orthogonal hexahedrons, tetrahedrons, triangular prisms and pyramids. This algorithm reduces to the conventional FDTD method when applied on a structured orthogonal hexahedral grid.},
doi = {10.2172/620998},
url = {https://www.osti.gov/biblio/620998},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed May 21 00:00:00 EDT 1997},
month = {Wed May 21 00:00:00 EDT 1997}
}