Master of Science, The Ohio State University, 2022, Industrial and Systems Engineering

Decades of cognitive systems engineering research has revealed that implementing human-machine teams into complex environments can consequently result in challenges that negatively impact human-machine teams. Such challenges and conflicts amongst team members can readily be observed in human-machine teams where agents are assigned heterogeneous tasks because the agents' individual goals may have a tendency to conflict and compete with one another in their shared environment. This conflict may also be magnified if the agents of our heterogeneously tasked human-machine team do not share a common goal and are not equipped with the resources to manage their differences. In our study, we set out to determine how the performance of our heterogeneously tasked agents in our simulated human-machine team was impacted in our full-motion video and intelligence analysis. By using joint-performance activity graphs, various statistical analyses, constant comparative analysis, and human-machine teaming heuristic analysis, we were able to determine that the performance of our human-machine team was not significantly different from the performance of our participants who worked alone. This led us to the conclusion that the machine agent insufficiently aided their human agent's decision making during the full motion video analysis and the design of the machine failed to adhere to known Human-Machine Teaming heuristics. Lastly, this holistic analysis revealed that the machine agent acted as if it did not have any knowledge of the ultimate goal of their human agent, and due to its limited capabilities, the machine was unable to contribute information in relation to the overarching goal. Even though the architecture of the human-machine team in this study failed to adhere to various human-machine teaming heuristics, failing to adhere to and implement the team so that both the agents' individual tasks meaningfully contributed the shared goal was determined to be the most criti (open full item for complete abstract)

Committee: Michael Rayo (Advisor); Samantha Krening (Committee Member); Michael Rayo (Committee Member) Subjects: Design; Engineering; Industrial Engineering; Systems Science

Doctor of Philosophy, University of Akron, 2021, Electrical Engineering

In this thesis, a numerical model is developed for estimating the reliability of a Switched Reluctance Machine (SRM), and is then used in its design optimization. SRM design optimization is performed using the multi-objective surrogate optimization method. The multi-objective surrogate optimization method is chosen because of its ability to find a global solution within a given number of function evaluations. In this study, optimization considers reliability as one of the objectives or constraints to obtain an optimal design with high reliability. The optimum design should operate for the target lifetime (20000 hours) before maintenance or complete replacement. Since optimization using 2D FEA is slow, a modified analytical model is developed to more quickly obtain a suitable design. This model predicts flux linkage and torque for a given phase current and rotor position with minimum error. Tapering in the stator and rotor poles is introduced in addition to traditional rectangular poles to reduce the error in predicting flux linkage and torque at unaligned and partially aligned positions. An improved BH curve model with a knee adjustment factor is used to predict the magnetic characteristics of steel used in the machine. Assumptions and detailed derivation of models to estimate flux linkage, co-energy, and torque are included. Reliability is defined as the probability of a component or system operating for a given time. The probability of a component to operate is calculated based on its failure rate. The failure rate of the machine is obtained as the combination of the machine's base failure rate and the failure rate of the windings, rotor, and shaft. The reliability of the machine is obtained using state transition diagrams associated with the Markov model under normal and fault conditions. Along with the machine's reliability, Mean Time to Failure (MTTF) is also predicted for all operating conditions. Using the developed analytical and reliability models f (open full item for complete abstract)

Committee: Yilmaz Sozer PhD (Advisor); Jose Alexis De Abreu-Garcia PhD (Committee Member); Malik Elbuluk PhD (Committee Member); Nao Mimoto PhD (Committee Member); Dane Quinn PhD (Committee Member) Subjects: Electrical Engineering; Electromagnetics

Doctor of Philosophy (PhD), Wright State University, 2022, Engineering PhD

Artificial Intelligence (AI) has seen a surge in popularity as increased computing power has made it more viable and useful. The increasing complexity of AI, however, leads to can lead to difficulty in understanding or interpreting the results of AI procedures, which can then lead to incorrect predictions, classifications, or analysis of outcomes. The result of these problems can be over-reliance on AI, under-reliance on AI, or simply confusion as to what the results mean. Additionally, the complexity of AI models can obscure the algorithmic, data and design biases to which all models are subject, which may exacerbate negative outcomes, particularly with respect to minority populations. Explainable AI (XAI) aims to mitigate these problems by providing information on the intent, performance, and reasoning process of the AI. Where time or cognitive resources are limited, the burden of additional information can negatively impact performance. Ensuring XAI information is intuitive and relevant allows the user to quickly calibrate their trust in the AI, in turn improving trust in suggested task alternatives, reducing workload and improving task performance. This study details a structured approach to the development of XAI in time-critical systems based on a design thinking framework that preserves the agile, fast-iterative approach characteristic of design thinking and augments it with practical tools and guides. The framework establishes a focus on shared situational perspective, and the deep understanding of both users and the AI in the empathy phase, provides a model with seven XAI levels and corresponding solution themes, and defines objective, physiological metrics for concurrent assessment of trust and workload.

Committee: Subhashini Ganapathy Ph.D. (Advisor); Sherif Elbasiouny Ph.D. (Committee Member); Asaf Harel Ph.D. (Committee Member); Victor Middleton Ph.D. (Committee Member) Subjects: Artificial Intelligence; Design; Engineering; Experiments; Industrial Engineering

Master of Science, The Ohio State University, 2022, Materials Science and Engineering

Recent increases in computational power have led to growing enthusiasm about the volume of data that can be collected and analyzed for many applications. However, the amount of data some physical/virtual systems generate is so great that an increased reliance on mathematical, statistical, and algorithmic based approaches to analyze and make decisions from the data is required. Application of these computational tools can lead to sharper decision making and vast amounts of knowledge discovered. The abstraction of the scientific decision-making process has led many researchers to consider observing systems with more tunable experimental parameters. This makes traditional experimentation, which is based on human researchers conducting the experiment and using their intuition to drive the next set of experiments, intractable for these applications. Autonomous experimentation (AE) systems, which are also a byproduct of the computational explosion, are able to address this issue and have found use across the fields of biology, chemistry, and materials science. AE systems are typically capable of conducting certain types of experiments with lower and more reliable turnaround times as opposed to their human counterparts. The automated execution of experiments naturally leads one to think about how those experiments can be parallelized and otherwise completed faster due to the lack of human presence in the experimentation environment. Therefore, AE systems are considered when designing many high-throughput experimentation (HTE) efforts. This thesis presents an overview of the current state-of-the-art for AE systems in Chapter 1, a framework developed to increase the independence of AE systems from human assistance in Chapter 2, and a machine-learning (ML) data processing pipeline that automates the image post-processing phase of the analysis of backscattered-electron scanning electron microscope images in Chapter 3.

Committee: Stephen Niezgoda (Advisor); Joerg Jinschek (Advisor); Sriram Vijayan (Other); Gopal Viswanathan (Committee Member); Oksana Chkrebtii (Committee Member) Subjects: Business Administration; Computer Science; Engineering; Experiments; Industrial Engineering; Information Science; Information Systems; Information Technology; Metallurgy; Operations Research; Robotics; Statistics

MDES, University of Cincinnati, 2021, Design, Architecture, Art and Planning: Design

Machine Learning (ML) plays an increasingly important role in modern user experience (UX) design practice. Being one of the fastest evolving tech phenomena within the `AI boom,' ML stands out as a field that could significantly benefit from more creative and critical thinking. Simultaneously, UX design students can benefit from increased literacy and competency in the already pervasive, highly relevant technology. However, contemporary UX design education fails to sufficiently empower young designers to work with cutting-edge technologies such as ML. A considerable amount of research has been conducted around designers' (lack of) comprehension of ML. These prior studies focus on identifying and discussing the challenges faced by UX design practitioners in designing for ML; not much research exists to propose and/or evaluate solutions for designers' lack of comprehension of ML. Further, none of the research in the UX design for ML space has focused on UX design students or education so far, though much of it identifies the lack of education of designers about ML as a major issue. Based on an analytical review of 88 introductory educational resources on UX design for ML, this thesis establishes a starting point for design educators to incorporate ML into undergraduate design curricula. It makes three primary contributions: 1) a set of guidelines for introductory education on UX design for ML, 2) a taxonomy of ML capabilities, use cases, and exemplars, and 3) a sample course proposal that demonstrates the application of (1) and (2) in undergraduate design education.

Committee: Matthew Wizinsky M.F.A. (Committee Chair); Renee Seward (Committee Member); Craig Vogel M.I.D. (Committee Member) Subjects: Design

Master of Science, The Ohio State University, 2020, Mechanical Engineering

Body structure design of automobiles is critical to achieve light weight and crash worthiness based on engineers' experience. To enhance experiential design, three computational methods are being developed: DOE based sampling, topology optimization and machine learning. This thesis produced data sets that can be used in DOE and machine learning. It also looked at different ways of using topology optimization for comparative studies. For all these tasks we need to extract feature attributes and patterns in association with performance parameters objectively. In the case of automotive hood models, the linear static analyses performance of hood lift and twist deflection, von Mises stress and geometry mass are the main performance factors of the comparative study. In this research, several pre-requisites, like geometry idealization, feature identification and parametrization, boundary condition standardization, were first investigated so that a standard way of Finite Element Modeling can be established. Then, an approach of generating large amount hood performance data, using ANSYS Workbench, was introduced to figure out the main effects parameter sets based on the design of experiments. By importing idealized model and design table, the hood performance data of 100 idealized models are generated in ANSYS DOE tool, accordingly. The hood performance data of idealized hood models generated in DOE are then exported to a statistic analysis software – MiniTab, to obtain the reduced set of parameters and parameter interactions through sensitivity study, based on factorial design analysis. Meanwhile, the topology optimization method for deriving the optimal structure of pocket feature pattern that can minimize deflection and improve the hood performance, was developed as a comparative study of structural optimization methods for automotive hood frames. Idealized solid model was created, with common imprinted surfaces, as well as boundary conditions with hood lift and hood twis (open full item for complete abstract)

Committee: Jami Shah (Advisor); Sandra Metzler (Committee Member) Subjects: Mechanical Engineering

Doctor of Philosophy, The Ohio State University, 2019, Pharmaceutical Sciences

The use of data-driven methods and machine learning has become increasingly pervasive in many industries, including drug discovery and design, as computing power and large amounts of data become increasingly available. In an effort to efficiently leverage this data, cheminformatics has emerged as a data-driven, interdisciplinary field that focuses on storing, accessing, and applying chemical information. Cheminformatics methods and tools facilitate the management and analysis of large annotated chemical datasets that would be difficult or impossible to do manually. A famous application of leveraging large amounts of chemical data was performed by Christopher A. Lipinski in 1997. Lipinski analyzed a large set of bioavailable synthetic drug molecules and identified trends in their molecular properties, which has since been referred to as the “Lipinski's Rule of 5”. While these rules are far from absolute, Lipinski's analysis demonstrates the utility of leveraging large amounts of chemical data to gain important insights. This thesis describes the application of cheminformatics methods to tackle two very different research problems: 1) the analysis and binding of a class of protein degraders called proteolysis targeting chimeras (PROTACs) and 2) the development of a target fishing application for the prediction of mechanism of action of natural products. PROTACs are a novel class of small molecule therapeutics that are garnering significant interest. Unlike traditional small molecule therapeutics, PROTACs simultaneously bind to both their protein target and an E3 ligase to induce degradation. The requirement to simultaneously bind two proteins necessitates a high molecular weight as PROTACs must contain two unique binding moieties that are connected by a linker. As a result, PROTAC molecules are expected to lie outside of the traditional drug-like chemical space described by Lipinski. To gain a better understanding of the physicochemical properties of PROTACs curre (open full item for complete abstract)

Committee: James Fuchs (Advisor); Xioalin Cheng (Advisor); Karl Werbovetz (Committee Member); Lara Sucheston-Campbell (Committee Member) Subjects: Chemistry; Computer Science; Molecular Chemistry; Molecules; Pharmacy Sciences

MS, University of Cincinnati, 2014, Engineering and Applied Science: Computer Engineering

Today embedded systems are found in all areas of our lives and have many different applications. They differ in their uses and properties as well as employing both software and hardware components in their implementations. This has made the design and development process for them much more complicated. Learning to use such a process is especially difficult for electrical engineering students, who have not been introduced to the systematic design and testing methodologies familiar to students trained in computer science and computer engineering. In this thesis, we illustrate the similarities and differences in the design and development design processes in for software systems and for software/hardware embedded systems. We give details for every stage for both types of systems and we develop detailed examples for example embedded systems, using a design process which extends the standard UML-based process used for software. In addition, we include details about project management. The examples and additional exercises and questions provide a set of tutorials which will assist students unfamiliar with complex design procedures in mastering the necessary skills to become well-trained embedded system developers.

Committee: Carla Purdy Ph.D. (Committee Chair); Raj Bhatnagar Ph.D. (Committee Member); George Purdy Ph.D. (Committee Member) Subjects: Computer Engineering

Master of Science in Computer Science, Miami University, 2025, Computer Science and Software Engineering

Computer-aided design (CAD) is widely used for 3D modeling in many technical fields, yet the creation of 2D sketches remains a manual step in typical CAD modeling workflows. Automatically generating 2D sketches can help users in CAD modeling by reducing their workload and by streamlining the design process. While sketches inherently possess a graph structure, with geometric primitives as nodes and constraints as edges, the application of graph neural networks (GNNs) to this domain remains relatively unexplored. To address this gap, we introduce SketchGNN, a graph diffusion model designed to generate CAD sketches using a joint continuous-discrete diffusion process. Our approach includes a novel discrete diffusion technique, wherein Gaussian-perturbed logits are projected onto the probability simplex via a softmax transformation. This enables our model to express uncertainty in the discrete diffusion process unlike traditional methods. We demonstrate that SketchGNN achieves state-of-the-art performance, reducing the Frechet Inception Distance (FID) from 16.04 to 7.80 and the negative log-likelihood (NLL) from 84.8 to 81.33.

Committee: John Femiani (Advisor); Khodakhast Bibak (Committee Member); Karen Davis (Committee Member) Subjects: Artificial Intelligence; Computer Science; Information Science

PhD, University of Cincinnati, 2023, Engineering and Applied Science: Mechanical Engineering

In recent years, the demand for high-functioning industrial parts and components has increased significantly. In addition to the enhanced functional performance, the parts are also expected to be manufacturable at minimum cost and lead times. Additive manufacturing processes have the capability to meet these demands. The inherent nature of additive manufacturing enables the users to fabricate complicated parts with reduced cost and material requirements. Designers can apply scenario-specific topology optimization approaches, infill lattices, and multi-material structures to ensure the eventual design performs well under the applied loading. However, most of these design methods fail to consider the manufacturability of that particular part. Typical additive manufacturing post-processing operations, such as support structure removal and entrapped powder elimination, are not addressed during a part's design phase. This dissertation aims to bridge the gap between a high-performance part design and its additive manufacturability to achieve a cost and time efficient product design and manufacturing pipeline. To simultaneously optimize the part geometry while ensuring improved manufacturability, a novel support structure accessibility filter is embedded inside a compliance minimization topology optimization model. Computational geometry and graph-based algorithms are developed to detect and characterize part geometries for powder entrapment challenges. Also, a graph-search approach for optimum rotation sequence for powder removal is developed as a preliminary tool for designers to consider depowdering challenges during the design phase. Typical requirements for additively manufactured parts consist of high functional performance with appropriate weight reduction. Multi-material topology optimization methods combined with variable-density lattice structures are used to obtain lightweight parts that perform well under thermal and mechanical loading. Due to the prohibitive c (open full item for complete abstract)

Committee: Sam Anand Ph.D. (Committee Chair); Michael Alexander-Ramos Ph.D. (Committee Member); Gen Satoh Ph.D. (Committee Member); Manish Kumar Ph.D. (Committee Member); Prashant Khare Ph.D. (Committee Member) Subjects: Mechanical Engineering

Doctor of Philosophy, University of Toledo, 2023, Mechanical Engineering

Shape Memory Effect (SME) and Superelasticity (SE) are key characteristics of shape memory alloy (SMA) materials. SME allows the material to return to its original shape after heating, while superelasticity enables recovery from significant inelastic deformation. NiTiHf is a highly promising SMA, known for its elevated SME and SE performances. Designing and controlling NiTiHf SMA properties as desired poses challenges due to its dependence on many factors. Three core characteristics define SMA materials: transformation temperatures (TTs), thermal hysteresis (TH), and actuation strain (AS). TTs are crucial design properties that determine the activation threshold for SME and SE effects. TH, resulting from TTs differences, reflects the energy loss during each SME action. AS represents the amount of recoverable strain during each SME actuation. Traditional approaches to designing NiTiHf TTs, TH, and AS have relied solely on experimental studies, which have not yielded comprehensive results and can be impractical due to high costs and time requirements. Cost-effective modeling approaches, including physics-based and data-driven methods, expedite material design and process optimization. Machine learning (ML) modeling, equipped with strong regression analyses, significantly reduces the need for experimental trials to optimize alloy design. Physics-based modeling, considering underlying physical principles, plays a critical role as error compensation tools. In this study, both data-driven and physics-based modeling were utilized to overcome the high-dimensional dependency of NiTiHf TTs and AS on various factors and the limited understanding of governing physics. The input parameters for the machine learning models included elemental composition, thermal treatments, and common post-processing steps used in NiTiHf fabrication. This feature selection incorporated a majority of accessible information from the literature on NiTiHf TTs and AS, making use of all essential proce (open full item for complete abstract)

Committee: Mohammad Elahinia (Committee Chair); Ala Qattawi (Committee Chair); Othmane Benafan (Committee Member); Behrang Poorganji (Committee Member); Meysam Haghshenas (Committee Member) Subjects: Mechanical Engineering

Master of Science, Miami University, 2023, Computer Science and Software Engineering

Software models are used to analyze and understand the properties of the system, providing stakeholders with an overview of how the system should work before actually implementing it. Such models are usually created informally, such as drawing sketches on a whiteboard or paper, especially during the early design phase, because these methods foster communication and collaboration among stakeholders. However, these informal sketches must be formalized to be useful in later applications, such as analysis, code generation, and documentation. This formalization process is often tedious, error-prone, and time-consuming. In an effort to avoid recreating formal models from scratch, this thesis presents SkeMo, a sketch-based software modeling tool. SkeMo is built on a CNN-based image classifier using 3000 input sketches of class diagram components and integrated into the functionality of an existing web-based model editor, the Instructional Modeling Language (IML), with a newly implemented touch interface. SkeMo was evaluated using a ten-fold cross-validation to assess the image classifier and through a user study involving 20 participants to collect metrics and feedback. The results demonstrate the promising potential of sketch-based modeling as an intuitive and efficient modeling practice, allowing users to quickly and easily create models to design complex software systems.

Committee: Eric Rapos (Advisor); Christopher Vendome (Committee Member); Xianglong Feng (Committee Member); Douglas Troy (Committee Member) Subjects: Computer Science; Engineering

Master of Science, The Ohio State University, 2023, Electrical and Computer Engineering

Over the past decade, there has been a growing movement towards reducing the carbon footprint which involves striving for net-zero emissions and developing an infrastructure that can sustain it. Among the end-use sectors, transportation accounts for nearly a third of overall greenhouse gas (GHG) emissions, with commercial vehicles as a huge contributor, making it imperative for this industry to adapt to emerging technologies to accommodate the expectations of a green and sustainable mobility vision. Battery electric and fuel cell vehicle technologies are suitable candidates to replace the existing conventional fossil fuel powered vehicle architectures but a complete transformation is nigh realizable due to various impediments. A soft impact is vital to ease the transformation process to foster growth and acceptance among industry partners as well as prepare for any unseen hurdles along the way. The work presented in this thesis focuses on designing and operating commercial vehicle fleets, introducing a novel fleet management system (FMS) framework capable of providing energy efficient mobility solutions. The FMS uses a recommender system comprised of a Design Space Filter (DSF) module to provide a feasible set of powertrains from the vehicle configuration database and uses machine learning algorithms to estimate the energy consumption for a given drive cycle, ranking them on the basis of their freight energy efficiency metric and ultimately aiding in the fleet composition design process. Due to the usage of highly confidential data pertaining to vehicle behavior, operators and OEMs are not keen on sharing this data unless there are agreements and secure data-sharing procedures established. Aware of this data bottleneck, the FMS leverages federated learning technique to estimate vehicular performance attributes and provides inferences which can be utilized for analyzing fleet behavior and enhancing fleet operations. This is extended to learn the mobility dynamics of (open full item for complete abstract)

Committee: Qadeer Ahmed (Advisor); Parinaz Naghizadeh (Committee Member); Manfredi Villani (Other) Subjects: Artificial Intelligence; Automotive Engineering; Computer Engineering; Electrical Engineering; Sustainability; Transportation

Doctor of Philosophy, The Ohio State University, 2023, Mechanical Engineering

Traditional design tasks and manufacturing systems often require a multitude of manual efforts to fabricate sophisticated artifacts with desired performance characteristics. Engineers typically iterate on a first principles-based model to make design decisions and then iterate once more by manufacturing the artifacts to take manufacturability into design consideration. Such manual decision-making is inefficient because it is prone to errors, labor intensive and often fails to discover process-structure-property relationships for novel materials. As robotics is an integral part of modern manufacturing, building autonomous robots with decision-making capabilities is of crucial importance for this application domain. Unfortunately, most of the robots and manufacturing systems in the industries lack such cognitive abilities. We argue that this whole process can be made more efficient by utilizing machine learning (ML) approaches, more specifically by leveraging sequential decision-making, and thus making these robots, design processes and manufacturing systems autonomous. Such data-driven decision-making has multiple benefits over traditional approaches; 1) machine learning approaches may discover interesting correlations in the data or process-structure-property relationship, 2) ML algorithms are scalable, can work with high dimensional unstructured problems and learn in highly nonlinear systems where a model is not available or feasible, 3) thousands of man-hour and extensive manual labor can be saved by building autonomous data-driven methods. Due to the sequential nature of the problem, we consider reinforcement learning (RL), a type of machine learning algorithm that can take sequential decisions under uncertainty by interacting with the environment and observing the feedback, to build autonomous manufacturing systems (AMS). Unfortunately, traditional RL is not suitable for such hardware implementation because (a) data collection for AMS is expensive and (b) tradit (open full item for complete abstract)

Committee: David Hoelzle (Advisor); Parinaz Naghizadeh (Committee Member); Jieliang Luo (Committee Member); Kira Barton (Committee Member); Michael Groeber (Committee Member) Subjects: Artificial Intelligence; Design; Mechanical Engineering; Robotics

Doctor of Philosophy, The Ohio State University, 2023, Mechanical Engineering

Metamaterials are deliberately architected artificial materials that can achieve unconventional properties not observed in nature, showing potential for various applications. Mechanical metamaterials are a new branch of metamaterials using geometry designs to control mechanical properties such as stiffness, deformation, and energy absorption. To date, most of the research on mechanical metamaterials considers an array of unit cells distributed in a uniform pattern, and the properties of those mechanical metamaterials are restricted by the unit cell structure. By introducing multiple unit cells into the array with non-uniform patterns, a much wider variety of mechanical properties become possible. However, such non-uniform mechanical metamaterials with extensive design domains bring challenges to the design, especially when specific target properties are desired. Motivated by the development of deep learning, we develop a framework based on feedforward neural networks (FNN) to systematically explore a large design domain of non-uniform mechanical metamaterials with nonlinearity in material, geometry, and boundary condition, realizing the mechanical response curve predictions of non-uniform patterns and the inverse designs for given target response curves. But for conventional mechanical metamaterials, their properties are significantly confined by the original geometries and lack in-situ tunability. Thus by a direct ink writing (DIW) technique, we combine hard-magnetic soft materials (MSMs) and hard-magnetic shape memory polymers (M-SMPs), which demonstrate superior shape manipulation performance by realizing reprogrammable, untethered, fast, and reversible shape transformation and shape locking in one material system, to develop magneto-mechanical metamaterials that are capable of shifting between various macroscopic mechanical behaviors such as expansion, contraction, shear, and bending under cooperative thermal and magnetic actuation, enabling wide-range i (open full item for complete abstract)

Committee: Ruike Zhao (Advisor); Haijun Su (Committee Member); Carlos Castro (Committee Chair) Subjects: Materials Science; Mechanical Engineering; Mechanics

PHD, Kent State University, 2022, College of Arts and Sciences / Materials Science Graduate Program

In this dissertation, we use the Finite Element Method (FEM) to model surface deformations produced by liquid crystal elastomer (LCE) coatings with different director microstructures. We first study director fields containing arrays of topological defects of integer and half-integer charges. We demonstrate that deformations are driven by competition between bend and splay in the director field, which in turn depend on the LCE film's defect structure. For example, a +1 defect with azimuthal anchoring has a bend-dominated pattern, and therefore, results in an elevation on heating. By contrast, we find that an LCE coating imprinted with a +1/2 defect demonstrates a surface topography whose deformation is a combination of out-of-plane and in-plane displacements. By modifying the defect core structure, we design an LCE coating with switchable topography containing an array of "suction cups" that resemble octopus suckers. Next, we study LCE coatings prepared between photopatterned substrates with antagonistic surface anchoring patterns. We devise a two-step modeling process in which we first minimize the Frank-Oseen Free Energy via a Finite Difference Method (FDM) to obtain a relaxed director field. We import the director field into the FEM code by interpolating the director field from a cubic lattice to a structured tetrahedral FEM mesh. Combination of FDM and FEM opens a new era to study shape morphing material whose director field is of a 3D form and cannot be solved analytically. For example, we investigate samples constructed from substrates with different director field patterns that produce periodic in-plane disclinations. We show that existence of these disclinations can generate temperature-induced switchable topography with microchannels on the surface of LCE coatings. Each channel can have either a single or double valley shape, depending on the position of the disclination lines with respect to the two film surfaces. Finally, we address the challenging (open full item for complete abstract)

Committee: Robin Selinger (Committee Chair); Badel Mbanga (Committee Co-Chair); Yaorong Zheng (Committee Member); Oleg Lavrentovich (Committee Member); Xiaoyu Zheng (Committee Member); Hiroshi Yokoyama (Committee Member) Subjects: Materials Science

Master of Science (MS), Wright State University, 2022, Physics

A phonon is a discrete unit of vibrational motion that occurs in a crystal lattice. Phonons and the frequency at which they propagate play a significant role in the thermal, optical, and electronic properties of a material. A phononic material/device is similar to a photonic material/device, except that it is fabricated to manipulate certain bands of acoustic waves instead of electromagnetic waves. Phononic materials and devices have been studied much less than their photonic analogues and as such current materials exhibit control over a smaller range of frequencies. This study aims to test the viability of machine learning, specifically neural networks in aiding in phononic crystal design. Multiple combinations of training datasets, neural network configuration, and data formatting methods are attempted with performance metrics recorded. A novel inverse design scheme is proposed that utilizes phonon density of states to perform prediction of phononic crystal parameters given a desired band gap and center frequency.

Committee: Chandriker Dass Ph.D. (Committee Co-Chair); Amit Sharma Ph.D. (Committee Co-Chair); Ivan Medvedev Ph.D. (Committee Member) Subjects: Computer Science; Materials Science; Physics

MS, University of Cincinnati, 2022, Engineering and Applied Science: Computer Engineering

Software defined radio (SDR) provides significant advantages over traditional analog radio systems and are becoming increasingly relied on for ”mission critical” applications. This along with risk of trojans, single-event upsets and human error creates the necessity for fault tolerant systems. Redundancy has been traditionally used to implement fault tolerance but incurs a substantial area overhead which is undesirable in most applications. Advancements in field-programmable gate array and system on a chip technologies have made implementing machine learning (ML) algorithms within embedded systems feasible. In this thesis we explore the use of ML to implement fault tolerance in an SDR. Our approach, which we call adaptive component-level degeneracy (ACD), uses a ML model to learn the functionality of an SDR component. Once trained, the model can detect when the component is compromised and mitigate the issue with its own output. We demonstrate the ability of our model to learn multiple simulated SDR components. We compare the one-dimensional convolutional neural network and bidirectional recurrent neural network architectures at modeling time series components. We also implement ACD within a real-time SDR system using GNU Radio Companion. The results show great potential for the utilization of ML techniques for improving embedded system reliability.

Committee: Rashmi Jha Ph.D. (Committee Member); Temesguen Messay Kebede Ph.D. (Committee Member); David Kapp PhD (Committee Member); John Emmert Ph.D. (Committee Member) Subjects: Computer Engineering

PhD, University of Cincinnati, 2022, Engineering and Applied Science: Computer Science and Engineering

The constantly increasing global elderly population is associated with a wide range of needs due to the varying degrees of decline in behavioral, social, emotional, mental, psychological, and motor abilities. Falls, highly common in the elderly, on account of such declining abilities and environmental factors, can limit their abilities to perform Activities of Daily Living (ADLs), as well as cause multiple health related complications including death. Therefore, it is essential that the future of intelligent living spaces, such as Smart Homes, are equipped with adaptable, pervasive, and ubiquitous systems that can anticipate and respond to these diverse needs of the elderly, while being able to track their dynamic indoor location, to provide solutions as and where such needs arise. The interdisciplinary work presented in this dissertation, aims to address these challenges and makes ten scientific contributions to these fields. First, it presents a methodology that investigates the many modalities of user interactions to deduce a user's indoor location in a particular "activity-based zone" during ADLs. Next, it presents a context-independent solution to determine the "zone-based" indoor position of a user in any indoor environment. These two approaches achieved performance accuracies of 81.36% and 81.13%, respectively, when tested on a dataset. Third, it presents a methodology for detecting a user's location in an indoor environment in terms of the X and Y coordinate information. This methodology outperformed all prior works in this field when evaluated using the Root Mean Squared Error (RMSE)-based performance evaluation metrics as per ISO/IEC18305:2016—an international standard for testing Localization and Tracking Systems. Fourth, it presents the findings from a comparison study of multiple learning approaches that were developed, implemented, and evaluated to address the challenge of determining the best machine learning method for Indoor Localization. The findin (open full item for complete abstract)

Committee: Chia Han Ph.D. (Committee Member); Juan E. Gilbert Ph.D. (Committee Member); Xuefu Zhou Ph.D. (Committee Member); Anca Ralescu Ph.D. (Committee Member); Wen-Ben Jone Ph.D. (Committee Member) Subjects: Computer Science

Doctor of Philosophy, University of Akron, 2021, Electrical Engineering

Acoustic noise and vibration prediction, mitigation, and performance optimization, in electric machines, are studied in this dissertation. First, vibration prediction enhancement in electric machines through frequency-dependent damping characterization is proposed in this dissertation. Different methods of mass and stiffness-dependent Rayleigh damping coefficient calculation are studied to identify the best damping estimation strategy. The proposed damping estimation strategy is used to predict the vibration spectrums of two 12-slot 10-pole (12s10p) permanent magnet synchronous machine (PMSM) designs and predicted vibration spectrums are experimentally validated through run-up tests of two prototypes. Moreover, to eliminate the dependency of the damping estimation strategy on the availability of a prototype, a damping coefficient prediction methodology is proposed. The proposed prediction methodology is experimentally validated using a third 12s10p PMSM prototype. Secondly, a lumped unit response-based sensitivity analysis procedure is introduced, which isolates electromagnetic and structural impacts brought by variation of different design parameters in an electric machine. The lumped unit response strategy utilizes the frequency-dependent damping estimation method developed early in the dissertation. The impact of different generic design parameters and a structural feature on a range of output quantities are studied in detail for a 12s10p PMSM. Analysis reveals that on a 12s10p PMSM, slot opening has a very high impact on the dominant airgap force component. A multi-level non-linear regression model-based optimization strategy is introduced considering electromagnetic and structural design objectives and constraints following the sensitivity analysis. A 12s10p PMSM prototype is tested to validate the FEA simulations used during the optimization process. Finally, the lumped unit response-based vibration prediction methodology devel (open full item for complete abstract)

Committee: Dr. Yilmaz Sozer (Advisor); Dr. Malik E. Elbuluk (Committee Member); Dr. J. Alexis De Abreu Garcia (Committee Member); Dr. D. Dane Quinn (Committee Member); Dr. Kevin Kreider (Committee Member) Subjects: Automotive Engineering; Electrical Engineering; Electromagnetics; Electromagnetism; Engineering; Mechanical Engineering; Technology