Digital Commons Network^™

Jointly-Learnt Networks For Future Action Anticipation Via Self-Knowledge Distillation And Cycle Consistency, Md Moniruzzaman, Zhaozheng Yin, Zhihai He, Ming-Chuan Leu, Ruwen Qin

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Future action anticipation aims to infer future actions from the observation of a small set of past video frames. In this paper, we propose a novel Jointly learnt Action Anticipation Network (J-AAN) via Self-Knowledge Distillation (Self-KD) and cycle consistency for future action anticipation. In contrast to the current state-of-the-art methods which anticipate the future actions either directly or recursively, our proposed J-AAN anticipates the future actions jointly in both direct and recursive ways. However, when dealing with future action anticipation, one important challenge to address is the future's uncertainty since multiple action sequences may come from or be followed by …

Additive Manufacturing Of Continuous Carbon Fiber-Reinforced Sic Ceramic Composite With Multiple Fiber Bundles By An Extrusion-Based Technique, Ruoyu Chen, Adam Bratten, Joshua Rittenhouse, Ming-Chuan Leu, Haiming Wen

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Due to the high cost, complex preparation process and difficulty in structural design, the traditional methods for carbon fiber reinforced SiC ceramic composite preparation have great limitations. This paper presents a technique for the additive manufacturing multiple continuous carbon fiber bundle-reinforced SiC ceramic composite with core-shell structure using an extrusion-based technique. A conventional nozzle system was modified to print simultaneously a water based SiC paste with continuous carbon fibers. Different levels of binder contents were investigated to optimize the stickiness, viscosity, thixotropy and viscoelasticity of the paste. After sintering, SiC whiskers were generated on the surface of fiber, which is …

Two-Level Targeter Convergence Study For Collinear Libration Point Spacecraft Formations, Donna Jennings, Henry J. Pernicka

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Spacecraft formation flying has attracted an increasing amount attention over the past several decades including a growing interest in relative motion at the collinear lib ration points. This current work builds on previous efforts by further developing and characterizing a two-level targeting (TLT) algorithm used to determine relative motion at lib ration points by performing a convergence study. A parametric study was conducted to identify the sensitivity of the algorithm to a variety of inputs and constraints. The parametric study revealed that the two-level targeting algorithm convergence properties significantly improve when using a higher order initialization estimate. The size of …

Direct Numerical Simulation Of Acoustic Disturbances In A Hypersonic Two-Dimensional Nozzle Configuration, Nathaniel Hildebrand, Meelan M. Choudhari, Cole P. Deegan, Junji Huang, Lian Duan

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Direct numerical simulations (DNS) are performed to study acoustic radiation in a quasi-two-dimensional nozzle with two independents spatially evolving turbulent boundary layers with an edge Mach number of 6. The emphasis of this work is to compare the radiated pressure fluctuations in a geometrically confined environment with those radiated from a single wall in an unconfined setting. The boundary-layer profile of the rms pressure fluctuation scaled by the mean shear stress at the wall is found to be in good agreement with prior flat-plate calculations at similar conditions. However, the normalized rms pressure fluctuation within the freestream p region is …

3d-Printed Gelatin-Alginate Hydrogel Dressings For Burn Wound Healing: A Comprehensive Study, Fateme Fayyazbakhsh, Michael J. Khayat, Ming-Chuan Leu

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Burn Wound Treatment is Still a Clinical Challenge Due to the Severity of Tissue Damage and Dehydration. among Various Wound Dressings, Hydrogel Materials Have Gained Significant Attention for Burn Wound Treatment in Clinical Practice Due to their Soothing and Moisturizing Activity. in This Study, 3D-Printed Dressings Were Fabricated using Clinically Relevant Hydrogels for Deep Partial-Thickness Burn (PTB) Wounds. Different Ratios of Gelatin and Alginate Mixture Were 3D-Printed and Examined in Terms of Rheological Behavior, Shear Thinning Behavior, Mechanical Properties, Degradation Rate, and Hydration Activity to Tune the Hydrogel Composition for Best Functionality. the Cell-Laden Dressings Were Bioprinted to Evaluate the …

Pulsed-Active Microwave Thermography, Logan M. Wilcox, Mathias Bonmarin, Kristen M. Donnell

Electrical and Computer Engineering Faculty Research & Creative Works

Active microwave thermography (AMT) is a thermographic nondestructive testing and evaluation technique that utilizes an electromagnetic-based excitation with a subsequent infrared measurement of the surface thermal profile of the material or structure of interest. AMT has been successfully applied to several aerospace and civil infrastructure applications. This work seeks to expand the performance of AMT by incorporating a signal processing technique common to traditional (flash-lamp) thermography, referred to as pulsed thermography (PT). PT operates on the premise of a pulsed excitation, as opposed to a constant or step excitation (ST) over a given time-period that is typical to traditional active …

New Hybrid Model For Evaluating The Frequency-Dependent Leakage Inductance Of A Variable Inductance Transformer (Vit), Angshuman Sharma, Jonathan W. Kimball

Electrical and Computer Engineering Faculty Research & Creative Works

Skin and proximity effects can cause a significant drop in the effective leakage inductance of a transformer when the operating frequency is increased. Although the magnetic image method-based double-2-D model can calculate the low-frequency leakage inductance with sufficient accuracy, it is inherently a frequency-independent model. While Dowell's 1-D model uses frequency-dependent relations to account for both skin and proximity effects, its accuracy is severely affected by the assumed winding geometry. In this paper, a hybrid model is proposed that uses superposition to combine a modified Dowell's model with the double-2-D model. The proposed model is investigated on a variable inductance …

Improving Virtual Synchronous Generator Control In Microgrids Using Fuzzy Logic Control, Oroghene Oboreh-Snapps, Rui Bo, Buxin She, Fangxing Fran Li, Hantao Cui

Electrical and Computer Engineering Faculty Research & Creative Works

Virtual synchronous generators (VSG) are designed to mimic the inertia and damping characteristics of synchronous generators (SG), which can improve the frequency response of a microgrid. Unlike synchronous generators whose inertia and damping are restricted by the physical characteristics of the SG, VSG parameters can be more flexibly controlled to adapt to different disturbances. This paper therefore proposes a fuzzy logic controller designed to adaptively set the parameters of the VSG during a frequency event to ensure an improved frequency nadir and rate of change of frequency (ROCOF) response. The proposed control method is implemented and tested on the power …

A Dnn-Ensemble Method For Error Reduction And Training Data Selection In Dnn Based Modeling, Ling Zhang, Da Li, Jiayi He, Bhyrav Mutnury, Bo Pu, Xiao Ding Cai, Chulsoon Hwang, Jun Fan, James L. Drewniak, Er Ping Li

Electrical and Computer Engineering Faculty Research & Creative Works

Deep neural networks (DNNs) have been widely adopted in modeling electromagnetic compatibility (EMC) problems, but the training data acquisition is usually time-consuming through various simulators. This paper presents a powerful approach using an ensemble of DNN s to effectively reduce the training data size in DNN-based modeling problems. A batch of training data with the largest uncertainties is selected using active learning through the variance among the ensemble of DNNs. Subsequently, a greedy sampling algorithm is applied to select a data subset using diversity. Thus, the proposed method can achieve both uncertainty and diversity in data selection. By averaging the …

A Segmentation Strategy For Structures With Common Mode Coupling, James Hunter, Shengxuan Xia, Aaron Harmon, Mohamed Z.M. Hamdalla, Ahmed M. Hassan, Victor Khilkevich, Daryl G. Beetner

Electrical and Computer Engineering Faculty Research & Creative Works

The level of electromagnetic coupling to electronic devices can vary widely from one device to another. When considering the induced voltage from an incoming plane wave on printed circuit boards (PCBs) and their attached cable harnesses, there is significant variety in the configuration of the devices that could be seen. This encourages the use of segmentation, so that the components of these devices (PCBs, connectors, and harnesses) can be modeled separately to alleviate simulation burden. This allows for a more flexible model and a 'toolbox' to construct devices with. The goal of this work is to use segmentation to model …

Practical Fixture Design For Passive Intermodulation Tests For Flexible Metallic Contacts, Shengxuan Xia, Yuchu He, Yansheng Wang, Hanfeng Wang, George Mankaruse, Danny Chan, Haicheng Zhou, Warren Lee, Nicholas Mcdonnell, Chulsoon Hwang

Electrical and Computer Engineering Faculty Research & Creative Works

Passive intermodulation (PIM) commonly exists in non-ideal metallic contacts. Since PIM typically represents an extremely low level of nonlinearity, it has not drawn enough attention over the years except for extremely high-power applications such as base stations. However, in recent years, the study on PIM has become essential in universally used consumer electronics design because of the higher requirement on the radio frequency (RF) sensitivity of wireless communications. The metal contacts caused PIM can create the sideband spectrum to interfere with the receiving band in the frequency divide duplex (FDD) mode. Therefore, the study on PIM for the frequently used …

Peec-Based On-Chip Pdn Impedance Modeling Using Layered Green's Function, Chaofeng Li, Biyao Zhao, Bo Pu, Xu Wang, Donghyun Kim, Jun Fan

Electrical and Computer Engineering Faculty Research & Creative Works

This paper presents an impedance model of on-chip power distribution network (PDN), which is an efficient criterion for estimating simultaneous switching noises (SSNs) on 3-D integrated circuit (IC). The impedance of on-chip PDN, including the effect of silicon substrate, is accurately modeled based on partial element equivalent circuit (PEEC) and layered Green's function (LGF). The equivalent circuit model of PDN is extracted based on the physical dimensions and electrical material characteristic of PDN at first. And then the LGF is used to consider the effect of silicon substrate for improving the accuracy of on-chip PDN impedance model. The effectiveness of …

Optimizing The Placement Of Non-Functional Pads On Signal Vias Using Multiple Reflection Analysis, Muqi Ouyang, Kevin Cai, Chaofeng Li, Anna Gao, Felen Fu, Hannah Bian, Bidyut Sen, Donghyun Kim

Electrical and Computer Engineering Faculty Research & Creative Works

In this study, the effects of using non-functional pads to optimize the performance of high-speed signal vias are investigated based on multiple reflection analysis. The non-functional pads on signal vias introduce more capacitive coupling and are possible to improve the response of the via structure if the original via has relatively larger impedance compared to the system reference impedance. The effectiveness of the non-functional pad optimization is validated through a numerical example, and the eye diagram of the via structure without and with non-functional pads are compared. The eye opening becomes 5.4 times larger after the via optimization using non-functional …

Incremental Cluster Validity Index-Guided Online Learning For Performance And Robustness To Presentation Order, Leonardo Enzo Brito Da Silva, Nagasharath Rayapati, Donald C. Wunsch

Electrical and Computer Engineering Faculty Research & Creative Works

In streaming data applications, the incoming samples are processed and discarded, and therefore, intelligent decision-making is crucial for the performance of lifelong learning systems. In addition, the order in which the samples arrive may heavily affect the performance of incremental learners. The recently introduced incremental cluster validity indices (iCVIs) provide valuable aid in addressing such class of problems. Their primary use case has been cluster quality monitoring; nonetheless, they have been recently integrated in a streaming clustering method. In this context, the work presented, here, introduces the first adaptive resonance theory (ART)-based model that uses iCVIs for unsupervised and semi-supervised …

Two-Dimensional In-Plane Strain Fss-Based Sensor, Mahboobeh Mahmoodi, Kristen M. Donnell

Electrical and Computer Engineering Faculty Research & Creative Works

Frequency selective surface (FSS) based sensors have found application as sensors in the last decade. In this paper, a new sensor design is proposed for two-dimensional in-plane strain sensing. The unit cell of the FSS-based sensor includes two slot dipoles, oriented normal to one another and each with different dimensions, to create two unique resonant frequencies when interrogated with an incident electric field normal to the direction of measured strain. In this way, 2D strain can be characterized concurrently and independently. The error due to strain orthogonal to the direction of interest, along with the error due to the presence …

A Meander Line-Based Frequency Selective Surface For Strain Sensing, Swathi Muthyala Ramesh, Kristen M. Donnell

Electrical and Computer Engineering Faculty Research & Creative Works

Frequency selective surfaces (FSSs) are periodic arrays of conductive elements that have distinct reflection and transmission responses. In this work, an FSS sensor designed to operate in Ka-band to measure a wide range of uni-directional strain using a meander line-based unit cell is presented. Specifically, the proposed unit cell of the sensor consists of a convoluted meander line geometry designed on a thin dielectric substrate. Strain sensing is achieved by monitoring the change in the resonant frequency of the FSS when under strain that is parallel to an interrogating signal linearly polarized and aligned with the convoluted dimension of the …

Locality-Aware Qubit Routing For The Grid Architecture, Avah Banerjee, Xin Liang, R. Tohid

Computer Science Faculty Research & Creative Works

Due to the short decohorence time of qubits available in the NISQ-era, it is essential to pack (minimize the size and or the depth of) a logical quantum circuit as efficiently as possible given a sparsely coupled physical architecture. In this work we introduce a locality-aware qubit routing algorithm based on a graph theoretic framework. Our algorithm is designed for the grid and certain 'grid-like' architectures. We experimentally show the competitiveness of algorithm by comparing it against the approximate token swapping algorithm, which is used as a primitive in many state-of-the-art quantum trans pilers. Our algorithm produces circuits of comparable …

Experimental Investigation Of Asphaltene Deposition And Its Impact On Oil Recovery In Eagle Ford Shale During Miscible And Immiscible Co2huff-N-Puff Gas Injection, Mukhtar Elturki, Abdulmohsin Imqam

Geosciences and Geological and Petroleum Engineering Faculty Research & Creative Works

One of the challenges in extracting oil from unconventional resources using hydraulic fracturing and horizontal drilling techniques is the low primary recovery rate, which is caused by the ultra-small permeability of these resources. Consequently, it is essential to investigate gas injection methods to produce the trapped oil in shale formations. However, the injection process can cause asphaltene depositions inside the reservoir, leading to plugging of pores and oil recovery (OR) reduction. There has been limited research on using gas injection techniques to improve oil production in tight/unconventional resources, although carbon dioxide (CO2) and gas-enhanced oil recovery methods have been used …

Comprehensive Evaluation Of Novel Medium-Temperature And High-Temperature Resistant Re-Crosslinkable Preformed Particle Gels For Conformance Control, Zhanmiao Zhai

Masters Theses

"Gel treatment has been widely used to control excessive water production and improve oil recovery in mature oilfields. In comparison to traditional preformed particle gel (PPG), re-cross linkable preformed particle gel (RPPG) has a significant advantage in treating abnormal features such as open fractures, fracture-like channels, and void space conduits (VSC). This research provides a comprehensive evaluation of two novel RPPG products developed in our lab: medium-temperature resistant RPPG (MT-RPPG) for a Middle East reservoir and a high-temperature resistant RPPG (HT-RPPG) for North Sea reservoirs (130 ^oC).

We assessed the effect of the monomer types and concentration, pH, salinity, …

Molten Steel Interactions With Tundish Linings And Refractory Glazes, Tyler Michael Richards

Doctoral Dissertations

"In the final stages of casting, molten steel is in contact with two main types of refractories: the working lining of the tundish used to serve as an intermediate vessel between the ladle and copper mold, and the various glazed carbon-bearing refractories used to control the flow of steel from the ladle to the tundish and from the tundish to the mold. If steel interacts with these refractories in such a way as to result in the formation of inclusions, it may be too late to remove them, leading to degradation of the steel’s mechanical properties. Understanding these interactions and …

Creation Of A Neural Network For The American Sign Language To Russian Translation App, John T. Simmons

Capstone Projects

1. A large population of people utilize American Sign Language for their primary method of communication.
2. No commercially available product is available for these people for when they need to communicate with speakers of a foreign language.
3. We must investigate methods to make communication between these two parties easier and more accessible.
4. By using a neural network to classify images of American Sign Language letters, we can build a service to make translation of American Sign Language into foreign languages possible.

Applying Cfd Model Studies To Determine Zones At Risk Of Methane Explosion And Spontaneous Combustion Of Coal In Goaves, Magdalena Tutak, Jarosław Brodny, Greg Galecki

Mining Engineering Faculty Research & Creative Works

Underground mining operations are subject to a number of natural hazards. Events resulting from these hazards are difficult to predict, and if they occur, they disrupt the entire mining process and pose a great danger to the crew. Some of the most dangerous include ventilation hazards involving methane explosions and fires caused by the spontaneous combustion of coal. The complex state of the underground environment means that these hazards oftentimes occur simultaneously, making mining conditions even worse. The following paper addresses this issue by developing the methodology for determining areas endangered by methane explosions and spontaneous coal combustion in goaves. …

Spatial Transformation Of A Layer-To-Layer Control Model For Selective Laser Melting, Xin Wang, Robert G. Landers, Douglas A. Bristow

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Selective Laser Melting (SLM) is an Additive Manufacturing (AM) technique with challenges in its complexity of process parameters and lack of control schemes. Traditionally, people tried time-domain or frequency-domain control methods, but the complexity of the process goes beyond these methods. In this paper, a novel spatial transformation of SLM models is proposed, which transforms the time-domain process into a spatial domain model and, thus, allows for state-space layer-to-layer control methods. In a space domain, this also provides the convenience of modelling laser path changes. Finally, a layer-to-layer Iterative Learning Control (ILC) method is designed and demonstrates the methodology of …

High-Pressure Kinetic Interactions Between Co And H2 During Syngas Catalytic Combustion On Pdo, Ran Sui, John Mantzaras, Rolf Bombach, Meysam Khatoonabadi

Mechanical and Aerospace Engineering Faculty Research & Creative Works

The catalytic combustion of H2/CO/O2/N2 mixtures over PdO was investigated at pressures 3 to 10 bar, H2:CO volumetric ratios 1:5 to 3:1, and global equivalence ratios ψ = 0.13 and 0.23. The catalyst surface temperatures were controlled to 540-690 K, a range especially important for hybrid hetero-/homogeneous combustion approaches with large gas turbines at idle or part-load operation and for microreactors with recuperative small-scale turbines. In situ Raman measurements determined the major gas-phase species concentrations over the catalyst boundary layers in a channel-flow reactor, thermocouples monitored the surface temperatures, and surface characterization identified the catalyst oxidation state (PdO) and surface …

Parenting Pre-Teens During Covid-19 In A Rural Midwestern Community: An Interpretive Phenomenological Study, Sarah Oerther, Daniel B. Oerther

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

To uncover the experiences of parenting Generation Z pre-teen children in rural communities impacted by the Stay Home Missouri order from April through May 2020. Researchers have focused on urban parents, leading to gaps in understanding the impact of the COVID-19 quarantine on rural parents and children. A qualitative study employing interpretive phenomenology. 14 white cis-male-sexed fathers and cis-female-sexed mothers living in midwestern rural communities participated in this study. Semi-structured interviews with 14 participants parenting pre-teen children were conducted. The interviews were analyzed using interpretive phenomenology. The COREQ checklist was followed. One theme that emerged from the narratives was the …

Variational Data Assimilation For Two Interface Problems, Xuejian Li

Doctoral Dissertations

“Variational data assimilation (VDA) is a process that uses optimization techniques to determine an initial condition of a dynamical system such that its evolution best fits the observed data. In this dissertation, we develop and analyze the variational data assimilation method with finite element discretization for two interface problems, including the Parabolic Interface equation and the Stokes-Darcy equation with the Beavers-Joseph interface condition. By using Tikhonov regularization and formulating the VDA into an optimization problem, we establish the existence, uniqueness and stability of the optimal solution for each concerned case. Based on weak formulations of the Parabolic Interface equation and …

Dc Ir-Drop Analysis Of Power Distribution Networks By A Robin Transmission Condition-Enhanced Discontinuous Galerkin Method, An Fa Yang, Min Tang, Jun Fa Mao, Li (Lijun) Jun Jiang, Hakan Bagcii, Ping Li

Electrical and Computer Engineering Faculty Research & Creative Works

In this work, a novel Robin transmission condition (RTC)-enhanced discontinuous Galerkin (DG) method is proposed for the dc IR-drop analysis of power distribution networks with Joule heating effects included. Unlike the conventional DG method, the proposed DG method straightforwardly applied to discretize the second-order spatial partial differential governing equations for the electrostatic potential $\Phi $ and the steady-state temperature $T$. The numerical flux in DG used to facilitate the information exchange among neighboring subdomains introduces two additional variables: the current density $J$ for the electrical potential equation and the thermal flux $q$ for the thermal equation. To solve them, at …

Early Wildfire Detection Using Uavs Integrated With Air Quality And Lidar Sensors, Doaa Rjoub, Ahmad Alsharoa, Ala'eddin Masadeh

Electrical and Computer Engineering Faculty Research & Creative Works

Every year, wildfires burn out countless hectares of lands, resulting in ecological, environmental, and economic damage. This paper presents an energy management system that consists of an unmanned aerial vehicle (UAV) equipped with air quality and light detection and ranging (LiDAR) sensors for monitoring forests and recognizing flames early. We develop a novel approach for autonomous patrolling system. This approach has the advantage of effectively detecting wildfire incidents, while optimizing the energy consumption of the UAV's battery to cover large areas. When a wildfire is detected, the UAV is able to transmit real-time data, such as sensor readings and LiDAR …

An Interleaved High Step-Up Dc-Dc Converter With Coupled Inductor And Built-In Transformer For Renewable Energy Applications, Ramin Rahimi, Saeed Habibi, Mehdi Ferdowsi, Pourya Shamsi

Electrical and Computer Engineering Faculty Research & Creative Works

This paper introduces an interleaved high step-up DC-DC converter with high voltage gain, low voltage stresses on the switches, low current stresses on the components, and continuous input current with low ripple, all of which are beneficial for the renewable energy (RE) applications. The proposed converter is based on the integration of three voltage-boosting (VB) methods: coupled inductor (CI), built-in transformer (BIT), and switched-capacitor (SC) cells. The energies of the leakage inductances of the CIs and BIT are absorbed and recirculated to the output side, which further extends the voltage gain. In addition, the current-falling rates of the diodes are …

A Spline-Based Partial Element Equivalent Circuit Method For Electrostatics, Riccardo Torchio, Maximilian Nolte, Sebastian Schops, Albert E. Ruehli

Electrical and Computer Engineering Faculty Research & Creative Works

This contribution investigates the connection between Isogeometric Analysis (IgA) and the Partial Element Equivalent Circuit (PEEC) method for electrostatic problems. We demonstrate that using the spline-based geometry concepts from IgA allows for extracting circuit elements without an explicit meshing step. Moreover, the proposed IgA-PEEC method converges for complex geometries up to three times faster than the conventional PEEC approach and, in turn, it requires a significantly lower number of degrees of freedom to solve a problem with comparable accuracy. The resulting method is closely related to the isogeometric boundary element method. However, it uses lowest-order basis functions to allow for …