Engineering Commons^™

Building Energy Management: A Data-Driven Approach Using Clustering And Load Forecasting, Aviral Kandel

LSU New Orleans Theses and Dissertations

The increase of smart meters in the grid has led to the generation of a vast amount of high dimensional energy data with improving temporal resolution. During analysis, relying on short samples like a day or week of data, could lead to wrong conclusion due to seasonal dynamics and customer behavior variations. To effectively utilize the vast amount of information, it must be compressed into a low-dimensional representation. This thesis explores the state-of-the-art dimensionality reduction techniques for a high-dimensional, non-linear energy dataset and proposes a novel deep learning based method to address the limitation of existing approaches. The proposed method …

Manifold Learning And Undersampling Approaches For Imbalanced Class Sentiment Classification, L.M. Risman Dwi Jumansyah, Agus Mohamad Soleh, Utami Dyah Syafitri

Knowledge Engineering and Data Science

Movie reviews are crucial in determining a film's success by influencing audience decisions. Automating sentiment classification is essential for efficient public opinion analysis. However, it faces challenges such as high-dimensional data and imbalanced class distributions. This study addresses these issues by applying manifold learning techniques, Principal Component Analysis (PCA) and Laplacian Eigenmaps (LE) to reduce data complexity and undersampling strategies (Random Undersampling (RUS) and EasyEnsemble) to balance data and improve predictions for both sentiment classes. On reviews of The Raid 2: Berandal, EasyEnsemble achieved the highest average G-Mean of 0.694 using Term Frequency-Inverse Document Frequency (TF IDF) features with a …

Constructing Qur’An Recitation Classification Using Alexnet Algorithm, Harits Ar Rosyid, Dzulkifli Abdullah, Mohammed S. Alqahtani

Knowledge Engineering and Data Science

The growing demands for accurate and efficient methods in the Qur'an recitation classification highlight the limitations of existing models, particularly in assisting the memorization process. This study aims to address these challenges by implementing the AlexNet Convolutional Neural Network architecture, widely recognized for its effectiveness in image classification, to classify the Qur'an recitations using the Mel Frequency Cepstral Coefficient (MFCC) as the feature extraction method. The research involves several stages, including data collection, preprocessing (audio segmentation by verse), data augmentation, feature extraction, and classification using the AlexNet architecture, followed by performance evaluation. Key results demonstrate that the combination of MFCC …

Deep Learning Approach For Dental Anomalies X-Ray Imaging Using Yolov8, Amelia Ritahani Ismail, Md Salim Sadman Taseen

Knowledge Engineering and Data Science

Dental X-ray imaging is a critical diagnostic tool for identifying various dental anomalies. However, manual interpretation is time-consuming, prone to human error, and requires specialized expertise. Deep learning models, particularly object detection frameworks like YOLO, have demonstrated promising results in automating medical image analysis. This study aims to develop and evaluate a YOLOv8-based deep learning model for automated detection and classification of 14 dental anomaly categories, including Caries, Crowns, Fillings, Implants, and Periapical lesions. The proposed approach addresses limitations in previous YOLO versions by leveraging anchor-free detection and enhanced feature extraction for improved accuracy. The model was trained on a …

Classification Of Anxiety Levels Entering The World Of Work In Final Year Students Using The Neighbor Weighted K-Nearest Neighbor Method, Awang Hendrianto Pratomo, Muhammad Fahmi Adam, Dessyanto Boedi Prasetyo

Knowledge Engineering and Data Science

This study evaluates the accuracy of the Neighbor Weighted K-Nearest Neighbor (NWKNN) method in classifying the anxiety levels of final-year students as they prepare to enter the workforce, particularly in cases of unbalanced data distribution. The system was developed using the prototype method, and NWKNN was applied to classify anxiety levels into low, medium, and high categories. Testing using the Confusion Matrix demonstrated strong performance, achieving an accuracy of 94% based on a dataset of 1009 students, with a 90:10 ratio of training to test data. The results indicate that NWKNN effectively provides classification input values, making it a reliable …

Comparative Analysis Of Bpnn And Lvq For Sundanese Character Recognition, Haviluddin Haviluddin, Herman Santoso Pakpahan, Dinda Izmya Nurpadillah, Hario Jati Setyadi, Medi Taruk, Rayner Alfred

Knowledge Engineering and Data Science

The Sundanese script (Aksara Sunda), an essential part of Sundanese cultural heritage, has been used since the 14th century AD. However, recognizing handwritten Sundanese characters remains challenging due to variations in individual writing styles. This study compares the performance of Backpropagation Neural Network (BPNN) and Learning Vector Quantization (LVQ) for recognizing handwritten Sundanese vowel (Swara) characters. A dataset was collected from 15 individuals, each writing seven Sundanese vowel characters, which were then used for training and testing the recognition models. Experimental results show that BPNN outperforms LVQ, achieving a higher classification accuracy (95.23%), lower Mean Squared Error (MSE), and faster …

Real-Time Predictive Analytics For Healthcare Monitoring And Terrain Classification Using Data Streaming And Online Learning, Ayemon Baraka

UNLV Theses, Dissertations, Professional Papers, and Capstones

This thesis develops a scalable, real-time predictive framework aimed at improving outcomes in healthcare monitoring and gait classification. Specifically, it addresses two primary applications: early detection of cardiovascular events, myocardial infarction (MI) and arrhythmia, through continuous analysis of ECG data streams; and adaptive terrain classification from gait data. Utilizing data streaming technologies such as Apache Kafka and Apache Spark, coupled with machine learning models, this framework enables near real-time and precise predictions from high-throughput, real-time data. Furthermore, the integration of online learning frameworks, including the River API and TensorFlow I/O, allows for continuous model adaptation, enhancing the system’s responsiveness and …

Tracking Joint Movement Using Optical Flow, Isabella Paperno

UNLV Theses, Dissertations, Professional Papers, and Capstones

We developed an algorithm that aims to move us closer to detecting early signs of arthritis. The program processes and analyzes X-ray videos using coyote and dog cadavers as models to examine the range of motion around the hip and connecting joints using optical flow techniques that track motion and velocity. We focus on how optical flow techniques track embedded metal markers and verify accuracy through comparisons with XMALab (X-ray motion analysis lab). Once proven as an accurate alternative, the focus will switch to markerless tracking and become a proof-of-concept for optical flow to be used in place of XMALab, …

Alteration Of Peripheral Cytokines In Alzheimer’S Disease, Zhengshi Yang

UNLV Theses, Dissertations, Professional Papers, and Capstones

Alzheimer’s disease (AD) is a progressive neurodegenerative disease with the clinical symptoms of gradual cognitive decline and memory loss. Biologically, abnormal accumulation of Aβ (Aβ) plaque and tau tangles in the brain are the core pathological hallmarks in AD. Although inflammatory response is not specific to AD, emerging evidence suggests that the disrupted innate immune system within the central nervous system, particularly microglial activation, might act as the bridge linking Aβ aggregation to pathological phosphorylation and aggregation of tau. The inflammatory response was suggested to initially have a neuroprotective role of attempting to clear Aβ plaques, and then become neurotoxic …

Integrating Deep Traffic Prediction And Environmental Impact Assessment Using Noisy Real-World Data In Las Vegas, Tarek Bin Zahid

UNLV Theses, Dissertations, Professional Papers, and Capstones

This thesis introduces an integrated framework for advanced traffic prediction and real-time emission estimation, designed to aid urban planning and environmental monitoring. Utilizing a graph-based transformer model, it predicts traffic conditions across the Las Vegas road network, drawing on spatial and temporal data from a large-scale sensor network. The study significantly expands the dataset from 26 to approximately 900 sensors, enhancing predictive accuracy and regional coverage. Inspired by masking techniques and strategies tailored to incomplete datasets, the model effectively handles real-world, noisy data without relying on resource-intensive imputation. Innovative training approaches enable robust traffic flow predictions despite missing or imperfect …

Mechanically Cost-Effective Approach For Bipedal Walking In Robots Using Instantaneous Collision Angle, Smit R. Patel

UNLV Theses, Dissertations, Professional Papers, and Capstones

Humans, as bipedal locomotors, are effective at reducing the mechanical cost of transport (CoTmech) by adopting movement strategies and gaits that minimize energy expenditure for a given distance. By using different gaits at different speeds, leveraging their long spring-like tendons and muscle elasticity which store and release energy during movement, humans reduce the mechanical effort required for locomotion. Current locomotion solutions offered in bipedal robots, based on legacy walking and running gait models, are not great at energy efficiency unless walking at very low speeds. Additionally, the control system of robots, designed to ensure stability and adaptability, requires substantial resources, …

Delineating Genetic Influences On Neurodegenerative Disorders And Infectious Diseases Through Advanced Computational Methods, Xiaowei Zhuang

UNLV Theses, Dissertations, Professional Papers, and Capstones

Genetics plays a critical role in understanding the molecular mechanisms underlying neurodegenerative disorders and pathogen evolution in infectious diseases. For example, identifying genetic variants associated with a disease phenotype uncovers functional pathways that could lead to potential drug targets and therapeutic interventions. In addition, tracking the genetic evolution of pathogens enables early detection and warning of infectious disease outbreaks. In both applications, given the large amount of genetic data, advanced computational methods, including longitudinal and multivariate models, could significantly boost the statistical power and capture interrelationships among traits, environmental factors and genetic influences. This dissertation focuses on four applications of …

Development Of A Subsurface Water Velocity And Quality Measurement Payload For Suas, Mitchell Bailey

All Graduate Theses and Dissertations, Fall 2023 to Present

Aquaculture farms, like oyster farms, often use outdated techniques to measure the quality of their waters. These techniques are often stationary or are deployed by an operator who may struggle to reach a location depending on conditions like the presence of a storm or low tide. This creates data sets with gaps over time and across the area of the farm, meaning the data cannot be used to model the standard farm conditions. An Uncrewed Aerial System (UAS) payload is developed in this paper that is capable of measuring depth-velocity profiles through the use of a tethered velocity sensor. The …

Dual-Channel Side Channel Attack: Improved Aes Key Decryption By Combining Power And Electromagnetic Side Channels With Convolutional Neural Networks, Sean P. O'Neill

Theses and Dissertations

This research introduces a novel DL approach for SCA that combines power consumption and EM signals to enhance encryption key deduction by leveraging a dual-channel CNN architecture. A new dataset, consisting of simultaneous power and EM signal collections during 128-bitAES encryption, was developed to train and evaluate the model’s effectiveness. The combined approach achieved an 88% reduction in traces needed, from 50 traces to 6, for encryption key classification, outperforming traditional methods such as random forest, DPA, DEMA,and individual side channel CNN models. These findings highlight the potential of integrating multiple side channels in SCA to improve performance without the …

Clusteredlog: Optimizing Log Structures For Efficient Data Recovery And Integrity Management In Database Systems, Mariha Siddika Ahmad, Brajendra Panda

Electrical Engineering and Computer Science Faculty Publications and Presentations

In modern database systems, efficient log management is crucial for ensuring data integrity and facilitating swift recovery from potential data corruption or system failures. Traditional log structures, which store operations sequentially as they occur, often lead to significant delays in accessing and recovering specific data objects due to their scattered nature across the log. ClusteredLog addresses the limitations of traditional logging methods by implementing a novel logical organization of log entries. Instead of simply storing operations sequentially, it groups related operations for each data item into clusters. As a result, ClusteredLog enables faster identification and recovery of damaged data items …

Investigation Of Conductive Die-Top Thermal Capacitors During Short Circuit Operation Of Silicon Carbide Devices, Youssef Abotaleb

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

Silicon carbide (SiC) power devices have garnered significant attention in recent years due to their superior thermal and electrical properties compared to traditional silicon devices. However, SiC power devices suffer from severe reliability issues arising from their mechanical properties. This is because the Young's modulus of SiC is about three times larger than that of silicon, which correspondingly raises the mechanical stresses acting on the bonding structure, often leading to device failure under power shock events.

Traditional packaging materials and designs tend to constrain SiC performance, as hot-spot temperature resulting from a power shock is one of the key factors …

Extracting Permittivity And Permeability Using The Position-Insensitive And Calibration-Independent Method On A Rectangular Waveguide, James Conrad Denemark

Theses and Dissertations

Classic methods for extracting material characteristics require known measurements to accurately calibrate the network analyzer. Previous work demonstrated a position-insensitive and calibration-independent (PiCi) transmission/reflection method to extract a material’s permittivity. This thesis proposes a method with the same function, manipulated to use one empty measurement and then two samples of different thicknesses. The PiCi method is first adopted for rectangular waveguide which resulted in inaccurate permittivity data when compared to the calibrated solution. Once detector mismatch corrections were applied, the PiCi method produced accurate results. Using a 2-D numerical root search, permittivity and permeability material characteristics are now successfully extracted …

Gate Driver Design In High-Temperature Cmos Process For Heterogeneous Integration Inside Sic Power Module, Khandoker Asif Faruque

Graduate Theses and Dissertations

The shift toward electrification in transportation, including electric and hybrid vehicles, presents challenges for power electronic converters. Silicon Carbide (SiC) power devices are promising due to their high efficiency, temperature tolerance, and reduced losses compared to traditional silicon devices. However, their use is hindered by issues such as parasitic capacitances and gate inductances, which can lead to voltage spikes and stress on the device gate oxide, hence impacting converter reliability. Increasing gate resistance can help manage these issues, but it reduces switching speed and increases losses. Snubber circuits can mitigate switching stress but add extra components, reducing efficiency. Active gate …

Feasibility Study On Microwave Glucose Level Monitoring, Ricardo Cepeda

Theses and Dissertations

Diabetes management heavily relies on regular blood glucose monitoring, which traditionally involves invasive techniques. In recent years, significant research has been directed towards developing non-invasive glucose monitoring methods, particularly using electromagnetic waves. This study explores the feasibility of detecting blood glucose levels non-invasively by measuring the resonant frequency shifts of an antenna, influenced by the dielectric properties of blood. Microwave measurement techniques have garnered attention due to their ability to safely penetrate human tissue and detect biochemical markers like glucose. While previous studies have established a link between glucose concentration and dielectric properties, many techniques suffer from poor sensitivity or …

Enhancing Webtas: From An Atam Simulator To A Developmental Environment, Elizabeth G. Schmidt

Electrical Engineering and Computer Science Undergraduate Honors Theses

As interest in a niche research area, like the abstract Tile Assembly Model (aTAM), grows it is crucial to ensure that simulation tools are accessible and user-friendly for a broad audience. Existing resources can pose challenges for less experienced users, highlighting the need for enhancements that improve usability and functionality. To address these issues, key upgrades to the current online simulation platform, WebTAS, were implemented, including breakpoint functionality and nondeterminism checks, alongside features such as XML file integration, image capturing, and simulation view resets. These enhancements collectively elevate the platform from a basic learning tool to a comprehensive environment for …

Leveraging P4 Programmable-Hardware Switches For In-Network Pmu Packet Recovery, Evan Michael Bonar

Electrical Engineering and Computer Science Undergraduate Honors Theses

Phasor Measurement Unit (PMU) systems are essential for real-time power grid monitor- ing but often face data loss due to network delays, equipment malfunctions, or transmis- sion errors. Traditional centralized recovery solutions introduce significant latency and scalability challenges. This thesis presents a P4-based in-network recovery mechanism that embeds detection and recovery directly into the data plane of P4-enabled programmable switches, significantly reducing recovery time and infrastructure complexity. Using the Aurora 610 switch, the system detects missing packets via sequence number analysis and recovers magnitudes with an efficient register-based algorithm.

Evaluation demonstrates high accuracy and low latency, achieving a mean absolute …

Dynamic Optimization Of Directed Energy Deposition Build Conditions Using Real-Time Monitoring Via Closed-Loop Control, Callan Herberger

Open Access Theses & Dissertations

Directed Energy Deposition (DED) is an additive manufacturing process that is being rapidly adopted by industry and is well suited for the fabrication of complex components in various metal alloys. DED provides unique benefits such as design flexibility, the potential for in-situ alloying, and an open environment that allows for unobstructed monitoring within the build chamber. Despite these benefits, fully exploiting additive manufacturing's (AM) potential remains a complex task for designers. This dissertation presents a framework for controlling Directed Energy Deposition process variables through in-situ monitoring. An exploration into modifying AM build conditions through the development and implementation of a …

Uncovering The Light Network Load Performance Penalty Of The Network Link Outlier Factor (Nlof), Michelle Lara

Open Access Theses & Dissertations

This thesis evaluates the effectiveness of the Network Link Outlier Factor with Most Likely Link (NLOF: MLL) algorithm under varying network load conditions. Repeated simulation experiments using Mininet were conducted for four different network-wide load levels: 100 Mbps, 500 Mbps, 1 Gbps, and 5 Gbps. Using statistical inference, our experimental results indicate that NLOF: MLL is ineffective under light load conditions (i.e., 100Mbps load) due to the limited network flow data available for its learning process. This limitation highlights a key challenge in applying the algorithm to lightly loaded networks. A preliminary algorithm was proposed to address this light-load performance …

Advancing Grid Modernization Through Data-Driven Resilience Modeling And Hosting Capacity Of Distributed Energy Resources, Oscar Samuel Acosta

Open Access Theses & Dissertations

This Ph.D. dissertation focuses on advancing the integration of distributed energy resources (DERs) through concepts surrounding their impacts on power system stability, resilience, and hosting capacity (HC). This dissertation addresses crucial topics in renewable energy deployment, transient fault response, and dynamic modeling. The work begins with the development of renewable energy source (RES) models tailored for offsetting residential heating, ventilation, and air conditioning~(HVAC) and commercial cooling loads. These models utilize solar photovoltaic (PV) and wind energy systems to produce scalable frameworks adapted across diverse climates and building types in application of a partial-load targeting methodology. The dissertation then transitions from …

Reference Dependence In Queue Design And Pricing Strategies, Jian Liu, Yongpin Zhou, Jian Chen, Peng Li

Electrical and Computer Engineering Faculty Research & Creative Works

This research investigates the effect of reference dependence on waiting times in service systems which formerly used a first-in-first-out (FIFO) service but have introduced a priority line with a fee. Our model combines reference-dependent gain-loss utility with standard customer utility, and we posit that customers are pleased with shorter-than-expected waiting times, whereas longer-than-expected times lead to dissatisfaction and an increased likelihood of balking. The study explores two scenarios: a captive customer system (CCS) and a noncaptive customer system (NCCS), with a focus on optimal pricing and segmentation strategies for revenue and social welfare maximization. The results reveal that, in a …

Impact Of Electrical Testing Strategies On The Performance Metrics Of Bio-Organic-Based Resistive Switching Memory, Muhammad Awais, Hao Zhe Leong, Nadras Othman, Mohamad Danial Shafiq, Feng Zhao, Kuan Yew Cheong

Electrical and Computer Engineering Faculty Research & Creative Works

Resistive Random-Access Memory (ReRAM) is considered as one of the most promising non-volatile memory technologies because of its high scalability, fast switching speed, and low power consumption. While many review papers are focused on investigating material types, material properties, device fabrication methods, and device structures, the influence of electrical testing strategies on ReRAM performance has yet been reviewed, particularly for bio-organic-based ReRAM. This review compiled, analyzed, and discussed how compliance current, voltage sweep rate, voltage sweep range, and voltage sweeping direction affect the ON/OFF ratio, read memory window, and both SET and RESET voltages of ReRAM.

Study Of Temperature Effects Of Glucose And Microwave Dielectric Dispersion Model, Ethan Grant Boone

Theses and Dissertations

In this study, the temperature effect of dielectric properties in various aqueous solutions containing glucose are examined. A design for a new testing chamber is introduced to provide improved conditions for testing to ensure accurate and precise measurements. A N-SMA Adaptor is used when obtaining dielectric characteristics of glucose solutions ranging 100-300mg/dl. Dielectric parameters are obtained adopting a modified Debye dispersion model. MATLAB is used to confirm coaxial dimensions are suitable fabrication designs using fixed dielectric parameters of air and water. Simulations and design models were implemented utilizing ANSYS High Frequency Structure Simulator and 3-D Modeler software to develop the …

Hardware Applications Of High-Speed Fault Detection Algorithms, Daniel Zintsmaster

All Theses

The growing need for a sustainable electric energy infrastructure has driven research into control, protection, and optimization of power systems. A key challenge is that the changing grid must operate at increasingly higher speeds, but many current hardware devices cannot meet these demands. This thesis focuses on power system protection, aiming to design a hardware solution that can detect and isolate faults in microseconds, ensuring faster, more reliable grid operations. The first hardware developed was for a low-voltage direct current (LVdc) microgrid, which faces challenges due to a lack of protection schemes and novel speed requirements. This thesis presents protection …

Higher Order Bessel Beams Integrated In Time (Hobbit) With Engineered Light Frequencies And Applications In Computational Imaging, Tyler Cramer

All Theses

This thesis presents an adaptive optical system which leverages the frequency shifts from an acousto-optic deflector (AOD) form an array of uniformly spaced frequency diverse beamlets. These beamlets are then spatially transformed (wrapped) into a circular array forming a non-diffracting ring with an unprecedented orbital angular momentum (OAM) mode switching rate of 50Mhz with an OAM range of ±64. Along with frequency diversity, the non-diffracting range of the beams are analyzed 6 Rayleigh ranges of an equivalently sized gaussian beam. Inherent in the frequencies superposed on the AOD are associated relative phases which can be controlled to form sets of …

Techno-Economic Factors Impacting The Intrinsic Value Of Behind-The-Meter Distributed Storage, Ingrid Hopley, Mehrdad Ghahramani, Asma Aziz

Research outputs 2022 to 2026

With the increasing adoption of renewable energy, there is a growing need for efficient storage solutions. Battery storage is becoming an essential tool for maintaining grid reliability and handling the variable nature of renewable energy sources. This research focuses on behind-the-meter, grid-connected household systems in Western Australia, adopting a consumer perspective to evaluate the financial viability of residential batteries. Using the HOMER Grid for techno-economic modeling, eight factors influencing financial viability were analyzed, with results validated through two external case studies. The findings suggest that photovoltaic (PV) systems paired with batteries can be cost-effective at current prices, depending on load …