Study of 2.5D microstructural modeling techniques used for material property identification

Date of Award

2009

Degree Name

M.S. in Mechanical Engineering

Department

Department of Mechanical and Aerospace Engineering

Advisor/Chair

Advisor: Robert A. Brockman

Abstract

Advances in digital image correlation (DIC) techniques allow for the study of full-field surface deformations on a microscopic scale in metal alloys. Finite element-based microstructural models can be used for property identification using numerical optimization techniques. However, the exact microstructure in the interior of the specimen is not known without destroying the material, and DIC data is available only from the visible surface for comparison. An analytical methodology has been investigated that allows for the in-situ identification of orthotropic materials properties on a microstructural level in metallic materials using finite element analysis. A study has been conducted to determine the effects that various subsurface microstructural finite element (FE) modeling techniques have on the reduction of displacement errors on the surface of microstructural models. Multiple FE models with varying geometries and properties through the thickness (2.5D Models) are developed using surface geometry from a FE model created using known three-dimensional geometry through the thickness. Displacement errors are compared and an optimization procedure has been utilized to identify the linear elastic orthotropic material properties for various 2.5D models to determine which techniques are best suited for accurate material property identification. Using the recommended modeling techniques, relative errors in the surface displacements are on the order of three to six percent, and typical errors in the estimated material properties for a linear elastic orthotropic material are approximately one to five percent.

Keywords

Image processing Digital techniques, Surfaces, Deformation of, Nanostructured materials

Rights Statement

Copyright © 2009, author

Share

COinS