A physics-based reduced order basis method (RBM) for rapid and accurate computations of large-scale EM simulations

Author

Shankar, Vijaya ; Ota, Dale ; Sahely, Touraj ; Hall, William

fYear

2015

fDate

22-26 March 2015

Firstpage

1

Lastpage

2

Abstract

In response to the need for fast and accurate radar cross section (RCS) predictive capabilities for DoD warfighter mission, HyPerComp has been developing a state-of-the-art time-domain computational electromagnetics (CEM) technology known as HDphysics (High Definition physics). The HDphysics computational kernel is highly scalable on distributed memory systems, and can handle body-conforming unstructured grids, complex materials, and wideband excitations.

Keywords

computational electromagnetics; distributed memory systems; radar computing; radar cross-sections; reduced order systems; time-domain analysis; DoD warfighter mission; EM simulation; HDphysics computational kernel; HyPerComp; body conforming unstructured grid; complex materials; computational electromagnetics; distributed memory systems; high definition physics; physics-based reduced order basis method; radar cross section; time-domain CEM technology; Computational electromagnetics; Computational modeling; Geometry; Radar cross-sections; Scattering; Time-domain analysis; US Department of Defense; Broad band time-domain; CEM; Discontinuous Galerkin (DG) high order; Large-scale EM simulations; Parallel CPU/GPU processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Applied Computational Electromagnetics (ACES), 2015 31st International Review of Progress in

Conference_Location

Williamsburg, VA

Type

conf

Filename

7109664