Quantifying Uncertainties in Parameterizations of Strength Models of Rolled Homogeneous Armor: Part 1, Overview

Abstract

Guidance is provided on how to obtain uncertainties in parameters of material strength models of rolled homogeneous armor,along with point estimates for those parameters, using existing software tools to implement two different approaches: Bayesianregression and the interval predictor model (IPM) approach. This report shows how to mathematically describe a Bayesianmodel associated with a material strength model and related experimental data, and how to express this Bayesian model in formsthat the aforementioned tools can process. It also describes how the IPM approach can be implemented in Python. The reportshows how the model parameter uncertainties can be visualized and how they may be presented in a form suitable for input tosoftware tools for uncertainty propagation analysis, such as Dakota. Finally, the report shows how Bayesian analysis may beused to evaluate the quality of the fit of a strength model to experimental data.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 30, 2019
Accession Number
AD1081740

Entities

People

  • J. J. Ramsey

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • C4I

DTIC Thesaurus Topics

  • Armor
  • Bayesian Networks
  • Computer Programs
  • Crystal Structure
  • Data Science
  • Databases
  • Experimental Data
  • Information Science
  • Linear Programming
  • Military Research
  • Monte Carlo Method
  • Normal Distribution
  • Probability
  • Probability Distributions
  • Random Variables
  • Rolled Homogeneous Armor
  • Yield Strength

Readers

  • Computational Modeling and Simulation
  • Data Mining and Knowledge Discovery.
  • Metallurgy

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference