Spectral Stochastic FEM for Uncertainty Quantification Due to Multiple Dielectric Variabilities

Narendranath, Abhijith B and Vinoy, Kalarickaparambil Joseph (2019) Spectral Stochastic FEM for Uncertainty Quantification Due to Multiple Dielectric Variabilities. In: IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, 18 (10). pp. 1961-1965.

PDF Iee_Ant_Wir_Pro_Let_18-10_1961.pdf - Published Version Restricted to Registered users only Download (886kB) | Request a copy

Abstract

The impact of fabrication uncertainties such as random variations in material properties of the media on the electromagnetic system response is usually ignored due to computational cost. This is evaluated in this study using a spectral representation of these random parameters. The finite-element formulation for electromagnetics with edge elements is extended to include these stochastic variations. The numerical results are found to match the Monte Carlo simulation with a very large number of samples. The system matrix is found to be sparse, and hence, the method discussed is shown to be computationally efficient. Since this approach can handle uncertainty in multiple subregions, such as dielectric inserts in a waveguide, it can be used to analyze the sensitivity of practical electromagnetic systems over a range of frequencies.

Item Type:	Journal Article
Publication:	IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS
Publisher:	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Additional Information:	Copyright of this article belongs to IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Keywords:	Stochastic processes; Mathematical model; Standards; Transmission line matrix methods; Dielectrics; Uncertainty; Permittivity; Electromagnetics; edge element; material uncertainty; polynomial chaos; spectral stochastic; uncertainty quantification
Department/Centre:	Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited:	12 Dec 2019 09:44
Last Modified:	12 Dec 2019 09:44
URI:	http://eprints.iisc.ac.in/id/eprint/63978

Actions (login required)

View Item