Degree Type

Dissertation

Date of Award

2015

Degree Name

Doctor of Philosophy

Department

Electrical and Computer Engineering

First Advisor

David C. Jiles

Abstract

In this work, a new, non-destructive method for obtaining stress-depth gradients in ferromagnetic structures was developed, using the information contained within magnetic Barkhausen emissions. A depth- and stress-dependent model for the frequency spectrum of Barkhausen emissions was derived and fitted to measured data obtained from steel samples with controlled stress-depth gradients. To achieve this, a library of signal processing and optimization algorithms was developed, which allowed the analysis of large datasets. To validate experimental procedures, a number of solid mechanics finite element simulations were carried out. Proof of concept is demonstrated by assuming linear stress-depth gradients and successfully calculating the slopes of those, using a fitting algorithm.

Copyright Owner

Orfeas Kypris

Language

en

File Format

application/pdf

File Size

193 pages

Share

COinS