Machine Learning for Predicting Properties of Silicon Carbide Grain Boundaries

Abstract

Statistical techniques are utilized to determine the efficacy of physics-based descriptors to predict the energetic properties of silicon carbide grain boundaries. These descriptors are utilized in kernel ridge regression models with a radial basis function kernel for the prediction of grain boundary energetics. Models derived from this approach have been implemented as a replacement to the insertion, removal, and replacement probability functions in a Monte Carlo based selection scheme for sampling the microscopic degrees of freedom in silicon carbide grain boundaries. Preliminary results show these models increase the overall computational efficiency of finding low-energy minimized states compared to current techniques.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2019
Accession Number
AD1087034

Entities

People

  • Dennis Trujillo
  • Matthew Guziewski
  • Shawn P Coleman

Organizations

  • "Idiohypophyseal" diabetes mellitus in two hypophysectomised women

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Boundaries
  • Carbides
  • Ceramic Materials
  • Ceramic Matrix Composites
  • Compound Semiconductors
  • Crystal Structure
  • Data Sets
  • Elements
  • Grain Boundaries
  • High Performance Computing
  • Machine Learning
  • Materials
  • Materials Science
  • Mechanical Properties
  • Molecular Dynamics
  • Sampling
  • Silicon Carbide

Fields of Study

  • Physics

Readers

  • Coastal Oceanography
  • Computational Modeling and Simulation
  • Powder metallurgy of Titanium alloys.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference