Theses and Dissertations
This collection includes both ASU Theses and Dissertations, submitted by graduate students, and the Barrett, Honors College theses submitted by undergraduate students.
Displaying 1 - 2 of 2
Filtering by
- Creators: Paradise, Paul David
- Creators: Turaga, Pavan
Description
Applications such as heat exchangers, surface-based cellular structures, rotating blades, and waveguides rely on thin metal walls as crucial constituent elements of the structure. The design freedom enabled by laser powder bed fusion has led to an interest in exploiting this technology to further the performance of these components, many of which retain their as-built surface morphologies on account of their design complexity. However, there is limited understanding of how and why mechanical properties vary by wall thickness for specimens that are additively manufactured and maintain an as-printed surface finish. Critically, the contributions of microstructure and morphology to the mechanical behavior of thin wall laser powder bed fusion structures have yet to be systematically identified and decoupled. This work focuses on elucidating the room temperature quasi-static tensile and high cycle fatigue properties of as-printed, thin-wall Inconel 718 fabricated using laser powder bed fusion, with the aim of addressing this critical gap in the literature. Wall thicknesses studied range from 0.3 - 2.0 mm, and the effects of Hot Isostatic Pressing are also examined, with sheet metal specimens used as a baseline for comparison. Statistical analyses are conducted to identify the significance of the dependence of properties on wall thickness and Hot Isostatic Pressing, as well as to examine correlations of these properties to section area, porosity, and surface roughness. A thorough microstructural study is complemented with a first-of-its-kind study of surface morphology to decouple their contributions and identify underlying causes for observed changes in mechanical properties. This thesis finds that mechanical properties in the quasi-static and fatigue framework do not see appreciable declines until specimen thickness is under 0.75 mm in thickness. The added Hot Isostatic Pressing heat treatment effectively closed pores, recrystallized the grain structure, and provided a more homogenous microstructure that benefits the modulus, tensile strength, elongation, and fatigue performance at higher stresses. Stress heterogeneities, primarily caused by surface defects, negatively affected the thinner specimens disproportionately. Without the use of the Hot Isostatic Pressing, the grain structure remained much more refined and benefitted the yield strength and fatigue endurance limit.
ContributorsParadise, Paul David (Author) / Bhate, Dhruv (Thesis advisor) / Chawla, Nikhilesh (Committee member) / Azeredo, Bruno (Committee member) / Jiao, Yang (Committee member) / Arizona State University (Publisher)
Created2022
Description
Semantic image segmentation has been a key topic in applications involving image processing and computer vision. Owing to the success and continuous research in the field of deep learning, there have been plenty of deep learning-based segmentation architectures that have been designed for various tasks. In this thesis, deep-learning architectures for a specific application in material science; namely the segmentation process for the non-destructive study of the microstructure of Aluminum Alloy AA 7075 have been developed. This process requires the use of various imaging tools and methodologies to obtain the ground-truth information. The image dataset obtained using Transmission X-ray microscopy (TXM) consists of raw 2D image specimens captured from the projections at every beam scan. The segmented 2D ground-truth images are obtained by applying reconstruction and filtering algorithms before using a scientific visualization tool for segmentation. These images represent the corrosive behavior caused by the precipitates and inclusions particles on the Aluminum AA 7075 alloy. The study of the tools that work best for X-ray microscopy-based imaging is still in its early stages.
In this thesis, the underlying concepts behind Convolutional Neural Networks (CNNs) and state-of-the-art Semantic Segmentation architectures have been discussed in detail. The data generation and pre-processing process applied to the AA 7075 Data have also been described, along with the experimentation methodologies performed on the baseline and four other state-of-the-art Segmentation architectures that predict the segmented boundaries from the raw 2D images. A performance analysis based on various factors to decide the best techniques and tools to apply Semantic image segmentation for X-ray microscopy-based imaging was also conducted.
In this thesis, the underlying concepts behind Convolutional Neural Networks (CNNs) and state-of-the-art Semantic Segmentation architectures have been discussed in detail. The data generation and pre-processing process applied to the AA 7075 Data have also been described, along with the experimentation methodologies performed on the baseline and four other state-of-the-art Segmentation architectures that predict the segmented boundaries from the raw 2D images. A performance analysis based on various factors to decide the best techniques and tools to apply Semantic image segmentation for X-ray microscopy-based imaging was also conducted.
ContributorsBarboza, Daniel (Author) / Turaga, Pavan (Thesis advisor) / Chawla, Nikhilesh (Committee member) / Jayasuriya, Suren (Committee member) / Arizona State University (Publisher)
Created2020