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 - 5 of 5
Filtering by
- Creators: Ren, Yi
Description
In this dissertation, three complex material systems including a novel class of hyperuniform composite materials, cellularized collagen gel and low melting point alloy (LMPA) composite are investigated, using statistical pattern characterization, stochastic microstructure reconstruction and micromechanical analysis. In Chapter 1, an introduction of this report is provided, in which a brief review is made about these three material systems. In Chapter 2, detailed discussion of the statistical morphological descriptors and a stochastic optimization approach for microstructure reconstruction is presented. In Chapter 3, the lattice particle method for micromechanical analysis of complex heterogeneous materials is introduced. In Chapter 4, a new class of hyperuniform heterogeneous material with superior mechanical properties is investigated. In Chapter 5, a bio-material system, i.e., cellularized collagen gel is modeled using correlation functions and stochastic reconstruction to study the collective dynamic behavior of the embed tumor cells. In chapter 6, LMPA soft robotic system is generated by generalizing the correlation functions and the rigidity tunability of this smart composite is discussed. In Chapter 7, a future work plan is presented.
ContributorsXu, Yaopengxiao (Author) / Jiao, Yang (Thesis advisor) / Liu, Yongming (Committee member) / Wang, Qing Hua (Committee member) / Ren, Yi (Committee member) / Arizona State University (Publisher)
Created2018
Description
Advanced material systems refer to materials that are comprised of multiple traditional constituents but complex microstructure morphologies, which lead to their superior properties over conventional materials. This dissertation is motivated by the grand challenge in accelerating the design of advanced material systems through systematic optimization with respect to material microstructures or processing settings. While optimization techniques have mature applications to a large range of engineering systems, their application to material design meets unique challenges due to the high dimensionality of microstructures and the high costs in computing process-structure-property (PSP) mappings. The key to addressing these challenges is the learning of material representations and predictive PSP mappings while managing a small data acquisition budget. This dissertation thus focuses on developing learning mechanisms that leverage context-specific meta-data and physics-based theories. Two research tasks will be conducted: In the first, we develop a statistical generative model that learns to characterize high-dimensional microstructure samples using low-dimensional features. We improve the data efficiency of a variational autoencoder by introducing a morphology loss to the training. We demonstrate that the resultant microstructure generator is morphology-aware when trained on a small set of material samples, and can effectively constrain the microstructure space during material design. In the second task, we investigate an active learning mechanism where new samples are acquired based on their violation to a theory-driven constraint on the physics-based model. We demonstrate using a topology optimization case that while data acquisition through the physics-based model is often expensive (e.g., obtaining microstructures through simulation or optimization processes), the evaluation of the constraint can be far more affordable (e.g., checking whether a solution is optimal or equilibrium). We show that this theory-driven learning algorithm can lead to much improved learning efficiency and generalization performance when such constraints can be derived. The outcomes of this research is a better understanding of how physics knowledge about material systems can be integrated into machine learning frameworks, in order to achieve more cost-effective and reliable learning of material representations and predictive models, which are essential to accelerate computational material design.
ContributorsCang, Ruijin (Author) / Ren, Yi (Thesis advisor) / Liu, Yongming (Committee member) / Jiao, Yang (Committee member) / Nian, Qiong (Committee member) / Zhuang, Houlong (Committee member) / Arizona State University (Publisher)
Created2018
Description
An accurate knowledge of the complex microstructure of a heterogeneous material is crucial for quantitative structure-property relations establishment and its performance prediction and optimization. X-ray tomography has provided a non-destructive means for microstructure characterization in both 3D and 4D (i.e., structural evolution over time). Traditional reconstruction algorithms like filtered-back-projection (FBP) method or algebraic reconstruction techniques (ART) require huge number of tomographic projections and segmentation process before conducting microstructural quantification. This can be quite time consuming and computationally intensive.
In this thesis, a novel procedure is first presented that allows one to directly extract key structural information in forms of spatial correlation functions from limited x-ray tomography data. The key component of the procedure is the computation of a “probability map”, which provides the probability of an arbitrary point in the material system belonging to specific phase. The correlation functions of interest are then readily computed from the probability map. Using effective medium theory, accurate predictions of physical properties (e.g., elastic moduli) can be obtained.
Secondly, a stochastic optimization procedure that enables one to accurately reconstruct material microstructure from a small number of x-ray tomographic projections (e.g., 20 - 40) is presented. Moreover, a stochastic procedure for multi-modal data fusion is proposed, where both X-ray projections and correlation functions computed from limited 2D optical images are fused to accurately reconstruct complex heterogeneous materials in 3D. This multi-modal reconstruction algorithm is proved to be able to integrate the complementary data to perform an excellent optimization procedure, which indicates its high efficiency in using limited structural information.
Finally, the accuracy of the stochastic reconstruction procedure using limited X-ray projection data is ascertained by analyzing the microstructural degeneracy and the roughness of energy landscape associated with different number of projections. Ground-state degeneracy of a microstructure is found to decrease with increasing number of projections, which indicates a higher probability that the reconstructed configurations match the actual microstructure. The roughness of energy landscape can also provide information about the complexity and convergence behavior of the reconstruction for given microstructures and projection number.
In this thesis, a novel procedure is first presented that allows one to directly extract key structural information in forms of spatial correlation functions from limited x-ray tomography data. The key component of the procedure is the computation of a “probability map”, which provides the probability of an arbitrary point in the material system belonging to specific phase. The correlation functions of interest are then readily computed from the probability map. Using effective medium theory, accurate predictions of physical properties (e.g., elastic moduli) can be obtained.
Secondly, a stochastic optimization procedure that enables one to accurately reconstruct material microstructure from a small number of x-ray tomographic projections (e.g., 20 - 40) is presented. Moreover, a stochastic procedure for multi-modal data fusion is proposed, where both X-ray projections and correlation functions computed from limited 2D optical images are fused to accurately reconstruct complex heterogeneous materials in 3D. This multi-modal reconstruction algorithm is proved to be able to integrate the complementary data to perform an excellent optimization procedure, which indicates its high efficiency in using limited structural information.
Finally, the accuracy of the stochastic reconstruction procedure using limited X-ray projection data is ascertained by analyzing the microstructural degeneracy and the roughness of energy landscape associated with different number of projections. Ground-state degeneracy of a microstructure is found to decrease with increasing number of projections, which indicates a higher probability that the reconstructed configurations match the actual microstructure. The roughness of energy landscape can also provide information about the complexity and convergence behavior of the reconstruction for given microstructures and projection number.
ContributorsLi, Hechao (Author) / Jiao, Yang (Thesis advisor) / Chawla, Nikhilesh (Committee member) / Liu, Yongming (Committee member) / Ren, Yi (Committee member) / Mu, Bin (Committee member) / Arizona State University (Publisher)
Created2017
Description
Machine learning has demonstrated great potential across a wide range of applications such as computer vision, robotics, speech recognition, drug discovery, material science, and physics simulation. Despite its current success, however, there are still two major challenges for machine learning algorithms: limited robustness and generalizability.
The robustness of a neural network is defined as the stability of the network output under small input perturbations. It has been shown that neural networks are very sensitive to input perturbations, and the prediction from convolutional neural networks can be totally different for input images that are visually indistinguishable to human eyes. Based on such property, hackers can reversely engineer the input to trick machine learning systems in targeted ways. These adversarial attacks have shown to be surprisingly effective, which has raised serious concerns over safety-critical applications like autonomous driving. In the meantime, many established defense mechanisms have shown to be vulnerable under more advanced attacks proposed later, and how to improve the robustness of neural networks is still an open question.
The generalizability of neural networks refers to the ability of networks to perform well on unseen data rather than just the data that they were trained on. Neural networks often fail to carry out reliable generalizations when the testing data is of different distribution compared with the training one, which will make autonomous driving systems risky under new environment. The generalizability of neural networks can also be limited whenever there is a scarcity of training data, while it can be expensive to acquire large datasets either experimentally or numerically for engineering applications, such as material and chemical design.
In this dissertation, we are thus motivated to improve the robustness and generalizability of neural networks. Firstly, unlike traditional bottom-up classifiers, we use a pre-trained generative model to perform top-down reasoning and infer the label information. The proposed generative classifier has shown to be promising in handling input distribution shifts. Secondly, we focus on improving the network robustness and propose an extension to adversarial training by considering the transformation invariance. Proposed method improves the robustness over state-of-the-art methods by 2.5% on MNIST and 3.7% on CIFAR-10. Thirdly, we focus on designing networks that generalize well at predicting physics response. Our physics prior knowledge is used to guide the designing of the network architecture, which enables efficient learning and inference. Proposed network is able to generalize well even when it is trained with a single image pair.
The robustness of a neural network is defined as the stability of the network output under small input perturbations. It has been shown that neural networks are very sensitive to input perturbations, and the prediction from convolutional neural networks can be totally different for input images that are visually indistinguishable to human eyes. Based on such property, hackers can reversely engineer the input to trick machine learning systems in targeted ways. These adversarial attacks have shown to be surprisingly effective, which has raised serious concerns over safety-critical applications like autonomous driving. In the meantime, many established defense mechanisms have shown to be vulnerable under more advanced attacks proposed later, and how to improve the robustness of neural networks is still an open question.
The generalizability of neural networks refers to the ability of networks to perform well on unseen data rather than just the data that they were trained on. Neural networks often fail to carry out reliable generalizations when the testing data is of different distribution compared with the training one, which will make autonomous driving systems risky under new environment. The generalizability of neural networks can also be limited whenever there is a scarcity of training data, while it can be expensive to acquire large datasets either experimentally or numerically for engineering applications, such as material and chemical design.
In this dissertation, we are thus motivated to improve the robustness and generalizability of neural networks. Firstly, unlike traditional bottom-up classifiers, we use a pre-trained generative model to perform top-down reasoning and infer the label information. The proposed generative classifier has shown to be promising in handling input distribution shifts. Secondly, we focus on improving the network robustness and propose an extension to adversarial training by considering the transformation invariance. Proposed method improves the robustness over state-of-the-art methods by 2.5% on MNIST and 3.7% on CIFAR-10. Thirdly, we focus on designing networks that generalize well at predicting physics response. Our physics prior knowledge is used to guide the designing of the network architecture, which enables efficient learning and inference. Proposed network is able to generalize well even when it is trained with a single image pair.
ContributorsYao, Houpu (Author) / Ren, Yi (Thesis advisor) / Liu, Yongming (Committee member) / Li, Baoxin (Committee member) / Yang, Yezhou (Committee member) / Marvi, Hamidreza (Committee member) / Arizona State University (Publisher)
Created2019
Description
How to effectively and accurately describe, character and quantify the microstructure of the heterogeneous material and its 4D evolution process with time suffered from external stimuli or provocations is very difficult and challenging, but it’s significant and crucial for its performance prediction, processing, optimization and design. The goal of this research is to overcome these challenges by developing a series of novel hierarchical statistical microstructure descriptors called “n-point polytope functions” which is as known as Pn functions to quantify heterogeneous material’s microstructure and creating Pn functions related quantification methods which are Omega Metric and Differential Omega Metric to analyze its 4D processing.In this dissertation, a series of powerful programming tools are used to demonstrate that Pn functions can be used up to n=8 for chaotically scattered images which can hardly be distinguished by our naked eyes in chapter 3 to find or compare the potential configuration feature of structure such as symmetry or polygon geometry relation between the different targets when target’s multi-modal imaging is provided. These n-point statistic results calculated from Pn functions for features of interest in the microstructure can efficiently decompose the structural hidden features into a set of “polytope basis” to provide a concise, explainable, expressive, universal and efficient quantifying manner.
In Chapter 4, the Pn functions can also be incorporated into material reconstruction algorithms readily for fast virtualizing 3D microstructure regeneration and also allowing instant material property prediction via analytical structure-property mappings for material design.
In Chapter 5, Omega Metric and Differential Omega Metric are further created and used to provide a time-dependent reduced-dimension metric to analyze the 4D evaluation processing instead of using Pn functions directly because these 2 simplified methods can provide undistorted results to be easily compared. The real case of vapor-deposition alloy films analysis are implemented in this dissertation to demonstrate that One can use these methods to predict or optimize the design for 4D evolution of heterogeneous material.
The advantages of the all quantification methods in this dissertation can let us economically and efficiently quantify, design, predict the microstructure and 4D evolution of the heterogeneous material in various fields.
ContributorsCHEN, PEI-EN (Author) / Jiao, Yang (Thesis advisor) / Ren, Yi (Thesis advisor) / Liu, Yongming (Committee member) / Zhuang, Houlong (Committee member) / Nian, Qiong (Committee member) / Arizona State University (Publisher)
Created2021