Publication Date

2017

Document Type

Dissertation

Committee Members

Ha-Rok Bae (Committee Member), Ramana Grandhi (Advisor), Peter Hovey (Committee Member), Shamachary Sathish (Committee Member), Michael Saville (Committee Member)

Degree Name

Doctor of Philosophy (PhD)

Abstract

Recent research in extending nondestructive evaluation (NDE) methods toward damage and materials characterization has led to an emphasis on the development of physics-based forward models in NDE. Furthermore, a desire to quantify levels of uncertainty in NDE inspections, particularly in Air Force applications, has driven the need for quick solving surrogate models for the traditional forward models, which have largely focused on accurately capturing many aspects of the problem with little emphasis on computational effort. In the past 5 years, many different techniques for creating lower-fidelity, quickly solving models to be used in place of the expensive high fidelity simulations have been explored in the context of NDE simulations, however there is still much to be done. This dissertation is focused on the development of physics-based models which rely on approximations to simplify the calculations while maintaining accuracy. Models were developed for prediction of the signal from random, heterogeneous materials during eddy current testing (ECT) inspections with absolute and reflection differential coils. A model was also developed for a specific scanning acoustic microscopy (SAM) experiment that has been used previously to characterize the elastic properties of a material. Furthermore, improvements were made to the post-processing of the data from these experiments that enabled high fidelity analysis of the wave behavior in single crystals of titanium alloys. The models developed as part of this work were shown to be relatively accurate, but much quicker to solve than the commercially available software for performing the same simulations. The models were tested against considerable amounts of experimental data from four samples with known microstructure, and they were found to be in good qualitative agreement with the experiments in most cases. The models were also verified against other numerical techniques for performing the same computations.

Page Count

184

Department or Program

Ph.D. in Engineering

Year Degree Awarded

2017

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.


Included in

Engineering Commons

Share

COinS