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: Rajan, Subramanian
- Creators: Turaga, Pavan
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
Description
There has been a renewed interest to understand the degradation mechanism of concrete under radiation as many nuclear reactors are reaching their expiration date. Much of the information on the degradation mechanism of concrete under radiation comes from the experiments, which are carried out on very small specimens. With the advent of finite element analysis, a numerical predictive tool is desired that can predict the extent of damage in the nuclear concrete structure.
A mesoscale micro-structural framework is proposed in Multiphysics Object-Oriented Simulation Environment (MOOSE) finite element framework which represents the first step in this direction. As part of the framework, a coupled creep damage algorithm was developed and implemented in MOOSE. The algorithm considers creep through rheological models, while damage evolves exponentially as a function of elastic strain and creep strain. A characteristic length is introduced in the formulation such that the energy release rate associated with each element remains the same to avoid vanishing energy dissipation with mesh refinement. A creep damage parameter quantifies the effect of creep strain on the damage that was calibrated using three-point bending experiments with varying rates of loading.
The creep damage model was also validated with restrained ring shrinkage tests on cementitious materials containing compliant/stiff inclusions subjected to variable drying conditions. The simulation approach explicitly considers: (i) moisture diffusion driven differential shrinkage along the depth of the specimen (ii) viscoelastic response of aging cementitious materials (iii) isotropic damage model with Rankine′s failure initiation criterion, and (iv) random distribution of tensile strengths of individual finite elements.
The model was finally validated with experimental results on neutron-irradiated concrete. The simulation approach considers: (i) coupled hygro-thermal model to predict the temperature and humidity profile inside the specimen (ii) radiation-induced volumetric expansion of aggregates (RIVE) (iii) thermal, shrinkage and creep effects based on the temperature and humidity profile and (iv) isotropic damage model with Rankine’s criterion to determine failure initiation.
A mesoscale micro-structural framework is proposed in Multiphysics Object-Oriented Simulation Environment (MOOSE) finite element framework which represents the first step in this direction. As part of the framework, a coupled creep damage algorithm was developed and implemented in MOOSE. The algorithm considers creep through rheological models, while damage evolves exponentially as a function of elastic strain and creep strain. A characteristic length is introduced in the formulation such that the energy release rate associated with each element remains the same to avoid vanishing energy dissipation with mesh refinement. A creep damage parameter quantifies the effect of creep strain on the damage that was calibrated using three-point bending experiments with varying rates of loading.
The creep damage model was also validated with restrained ring shrinkage tests on cementitious materials containing compliant/stiff inclusions subjected to variable drying conditions. The simulation approach explicitly considers: (i) moisture diffusion driven differential shrinkage along the depth of the specimen (ii) viscoelastic response of aging cementitious materials (iii) isotropic damage model with Rankine′s failure initiation criterion, and (iv) random distribution of tensile strengths of individual finite elements.
The model was finally validated with experimental results on neutron-irradiated concrete. The simulation approach considers: (i) coupled hygro-thermal model to predict the temperature and humidity profile inside the specimen (ii) radiation-induced volumetric expansion of aggregates (RIVE) (iii) thermal, shrinkage and creep effects based on the temperature and humidity profile and (iv) isotropic damage model with Rankine’s criterion to determine failure initiation.
ContributorsSaklani, Naman (Author) / Neithalath, Narayanan (Thesis advisor) / Rajan, Subramanian (Committee member) / Mobasher, Barzin (Committee member) / Hoover, Christian (Committee member) / Chawla, Nikhilesh (Committee member) / Arizona State University (Publisher)
Created2020