Quantitative Evaluation for Two-Dimensional Electromagnetic Propagation in a Dielectric Medium using Markov Chain Monte Carlo Method (Preprint)

Abstract

The motivation of this work is to quantify the degradation of aging electrical cables. The dielectric material parameter of insulation can be correlated with degradation. In this paper, the forward problem is posed as a microwave nondestructive evaluation (NDE) problem. A 2D finite difference time domain (FDTD) was used to evaluate the forward problem. The inverse problem is solved via a Bayesian approach. The Bayesian formulation describes the solution as a posterior distribution over the model parameters such as complex permittivity. Since there is no analytical solution for the posterior distribution, the Markov Chain Monte Carlo (MCMC) method is employed to numerically solve for it. The Metropolis-Hasting algorithm is used in particular. Results for computational experiments are demonstrated to show feasibility of this approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 01, 2012
Accession Number
ADA557040

Entities

People

  • Fumio Kojima
  • Jeremy S. Knopp

Organizations

  • Air Force Research Laboratory

Tags

Communities of Interest

  • Advanced Electronics
  • Air Platforms

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Computational Science
  • Finite Difference Time Domain
  • Insulation
  • Markov Chains
  • Materials
  • Mathematical Models
  • Microwaves
  • Models
  • Monte Carlo Method
  • Plane Waves
  • Probability
  • Sampling
  • Test And Evaluation
  • Two Dimensional

Readers

  • Electromagnetic Wave Scattering and Antenna Radiation Engineering
  • Statistical inference.
  • Structural Health Monitoring of Composite Structures.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • Microelectronics