Engineering Commons^™

A Hybrid Cnn-Lstm Surrogate Model For Hyper-Resolution Spatiotemporal Flood Forecasting In Norfolk, Virginia, Yidi Wang, Jonathan L. Goodall, Chetan Kumar, Diana Mcspadden, Sergio A. Barbosa, Binata Roy, Ali Shahabi, Navid Tahvildari

Data Science Faculty Publications

Study region

Norfolk, Virginia, United States

Study focus

Accurate and timely flood forecasting is essential for enhancing resilience in coastal urban areas in the context of increasing frequency and intensity of rainfall, sea level rise and urbanization. This study presents a hybrid deep learning-based surrogate model that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to enable real-time spatiotemporal flood forecasting. The model leverages CNN to capture spatial features from inputs such as elevation and Topographic Wetness Index (TWI), while LSTM processes time-series inputs of rainfall and tide data to capture temporal features.

New hydrologic insights for …

Fairness Without Demographic Attributes In Medical Vision–Language Models, Ahsan Habib Akash

Graduate Theses, Dissertations, and Problem Reports (ETD)

Vision–language models (VLMs) and the embedding models that underlie them have become the default representation layer for multimodal artificial intelligence. Trained by contrastive alignment over enormous, loosely curated image–text corpora, they inherit the demographic skew of that data, and they encode it in the geometry of the shared embedding space itself. The consequence is a model that performs unevenly across demographic groups even when no protected attribute is ever supplied as an input. In consumer applications this is an equity problem; in medicine it is a safety problem, because an embedding that is systematically less discriminative for one subpopulation translates …

Real-Time Lidar-Based Vehicle Detection Using Pointpillars For Connected And Automated Vehicle Applications, Jared Townsend Behr

Graduate Theses, Dissertations, and Problem Reports (ETD)

This work explores the implementation of a real-time LiDAR processing model for autonomous vehicle design. Automotive perception systems are becoming increasingly important as Advanced Driver Assistance Systems (ADAS) and Connected and Automated Vehicle (CAV) technologies continue to progress toward higher levels of autonomy. Reliable perception is paramount for autonomous features, such as adaptive cruise control (ACC), where identifying and tracking a lead vehicle is a safety critical task.

Many production ADAS perception systems primarily rely on radar and camera sensors for object detection and sensor fusion. These sensors work well to achieve classification and accurate distance detections; however, they are …

Analysis Of Battery Degradation Effects On Optimized Torque Splitting In Dual-Motor Electric Vehicles, Homer E. Butcher

Graduate Theses, Dissertations, and Problem Reports (ETD)

Battery electric vehicle performance depends on both powertrain efficiency and battery-health. In dual-motor all-wheel-drive battery electric vehicles, driver-requested torque can be distributed between front and rear electric drive units, creating an opportunity to reduce electrical energy demand through torque allocation. However, lithium-ion battery aging changes the electrical behavior of the energy storage system through capacity fade and internal resistance growth. Capacity fade reduces the usable energy and vehicle range, while resistance growth increases voltage drop, current-related losses, and battery electrical loading. This thesis evaluates how these degradation effects influence the energy consumption, simulated range, and battery electrical behavior of optimized …

Ai-Based Smart Proxy Models For Flow Assurance, Integrating Oil Rate Prediction, Pipeline Monitoring, And Leak Detection, Ali Sajedian

Graduate Theses, Dissertations, and Problem Reports (ETD)

Oil and gas pipeline networks operate under complex multiphase flow conditions and dynamic transient regimes that present significant challenges for traditional monitoring, anomaly detection, and operational decision-making methodologies. Many existing monitoring approaches rely on high-fidelity physics-based transient flow simulations to accurately represent system behavior. However, these models are computationally intensive and therefore difficult to implement in real-time operational environments.

This dissertation presents a physics-guided artificial intelligence framework for intelligent pipeline monitoring and integrity management based on the concept of Smart Proxy Modeling. By integrating first principles engineering knowledge with machine learning techniques, the proposed framework develops computationally efficient surrogate models …

Understanding And Preventing Air Problems In Irrigation Pipelines, Charles M. Burt

BioResource and Agricultural Engineering

The water in irrigation pipelines always contains some air. Air-related problems are especially challenging in hilly terrain, and create challenges for designers and those who attempt to troubleshoot problems after installation.

Smartbot V3: A Scalable Educational And Research Robotics Platform And Decentralized Construction Framework, Robert Tyler Cook

Graduate Theses, Dissertations, and Problem Reports (ETD)

Autonomous robotic systems are increasingly used in both engineering education and robotics research, creating a need for systems capable of supporting both within a common platform. This thesis presents SMARTbot V3, a scalable robotics platform developed to support undergraduate robotics education and multi-robot research.

SMARTbot V3 is a differential-drive mobile manipulator incorporating a 3-degree-of-freedom manipulator, LiDAR, RGB cameras, onboard computation, and Robot Operating System 2 (ROS2) software. The platform was designed to support scalable deployment through standardized hardware, modular construction, and common software. A fleet of twenty robots was manufactured and deployed for both educational and research activities. Educational validation …

Sustainable Nutrient Management Opportunities For Small Communities In The U.S. With Lagoon Wastewater Treatment Systems, Denis Sigei Ruto

Graduate Theses, Dissertations, and Problem Reports (ETD)

Lagoon wastewater treatment systems are among the most widely used secondary treatment technologies serving small and rural communities in the United States due to their low capital and operating costs, minimal energy requirements, and operational simplicity. While effective for removal of organic matter and suspended solids, most lagoon systems were not designed to achieve advanced nutrient removal. As regulatory priorities increasingly emphasize control of total nitrogen and total phosphorus to protect receiving waters and public health, lagoon-dependent communities face growing compliance challenges. These challenges are exacerbated by aging infrastructure, limited technical capacity, financial constraints, and insufficient infrastructure data, highlighting the …

Computational Methods For Identification Of Molecular Signatures, Weijun Yi

Graduate Theses, Dissertations, and Problem Reports (ETD)

This work develops computational methods for identifying molecular signatures from high-throughput genomic data and for modeling long non-coding RNA (lncRNA) sub-cellular localization. The response of multiple myeloma to CB-6644, a selective RUVBL1/2 complex inhibitor with potential anti-tumor activity, is analyzed to identify drug-responsive pathways and molecular signatures. Conventional gene set enrichment analysis (GSEA) often excludes low-expression genes. Here, phenotype comparison is reformulated as a supervised machine learning problem: genes most informative for discrimination are first selected using a machine learning approach, and GSEA is then applied to these machine-learning derived gene sets. This framework improves detection of CB-6644-associated pathways. For …

Bridging Physics-Based Modeling And Machine Learning To Predict Material Behavior: Applications In Fatigue Crack Growth And Dielectric Property Characterization, Ansan Pokharel

Graduate Theses, Dissertations, and Problem Reports (ETD)

This dissertation integrates physics-based modeling with machine learning (ML) to predict how materials behave under complex thermal and mechanical conditions. A key innovation of this work is the use of finite element analysis (FEA) to supplement experimental data. This approach creates more diverse and representative synthetic datasets, helping to reduce the limitations and biases that arise when training ML models solely on experimental measurements. The research focuses on two applications: improving the prediction of fatigue properties in superalloys and estimating temperature-dependent, high-frequency dielectric properties relevant to microwave-based chemical processing.

In the first study, low-cycle fatigue experiments were performed on the …

Development Of Hydrogen Emissions Quantification Systems To Expand Understanding Of Leaks And Losses Associated With The Burgeoning Hydrogen Transportation Sector, Christopher Loomis

Graduate Theses, Dissertations, and Problem Reports (ETD)

To combat global warming, alternative fuels are being investigated. Hydrogen is at the forefront of this movement, advertised as having zero emissions due to the products of the energy reaction being only water. However, studies over recent decades indicate that hydrogen in the atmosphere acts as a form of greenhouse gas (GHG) by increasing the lifespan of methane in the atmosphere. Models are attempting to estimate the possible effects of a transition to a hydrogen-fueled economy but lack empirical data to provide confidence to these estimated values. There is little data that quantifies hydrogen from anthropogenic sources.

To address this …

Development And Additive Manufacturing Of Uv And Dual-Curable Elastomers For Soft Robotic Actuators And Embedded Sensors Via Direct Ink Writing, Emrah Demirkal

Graduate Theses, Dissertations, and Problem Reports (ETD)

The advancement of soft robotics requires materials and manufacturing methods that combine large deformation, structural stability, and integrated sensing within a single soft system. Conventional fabrication techniques often limit design flexibility and the incorporation of functional materials into soft structures. This dissertation addresses these challenges through the development of UV-curable and dual-curable silicone elastomers for the direct ink writing (DIW) of soft robotic actuators and embedded sensors.

UV-curable elastomers were first synthesized through thiol-ene click chemistry using poly(mercaptopropylmethylsiloxane) (MMPS) and vinyl-terminated polysiloxane (VPS). Their curing behavior, rheological response, printability, and mechanical properties were evaluated for DIW. These materials enabled rapid …

Using Aqueous Geochemistry Of Acid Mine Drainage To Spatially Investigate Concentrations Of Rare Earth Elements Throughout West Virginia, Pennsylvania, And Maryland, Elyssa Lynn Allen

Graduate Theses, Dissertations, and Problem Reports (ETD)

Rare earth elements (REEs) are an important commodity utilized in many industries such as electronics, defense, manufacturing, medicine, and green energy production. A lack of economical domestic sources forces the United States to import a majority of its supply, and incentivizes novel methods of acquisition. Mining regions of the United States, such as Appalachia or the West, are afflicted with acid mine drainage (AMD) impacted surface waters. REEs occur in varying concentrations within coal seams, indicating that coal and its waste are potential avenues of recovery that might also incentivize the remediation of coal impacted areas. Understanding spatial and geochemical …

Ai-Driven Prediction And Reconstruction Of Missing Cased-Hole Logs For Improved Well System Understanding, Samuel Avilez Martinez

Graduate Theses, Dissertations, and Problem Reports (ETD)

Well logging is a fundamental technique in formation evaluation providing continuous, real-time measurements of geological and petrophysical properties within a well. Through the systematic analysis of well logs, engineers and geoscientists can accurately determine critical formation characteristics, including porosity, permeability, lithology, and fluid composition. Well logging is fundamental for making informed decisions and reducing uncertainties in the exploration and development of oil and gas reservoirs.

Despite its significance, the acquisition of reliable well log data in oil and gas wells is often compromised by various operational, mechanical, and formation-related challenges, as well as pressure, fluid, and equipment constraints. In high-risk …

Ambulatory Monitoring Of Biomarkers For Enhanced Recovery, Lucas Abrahamian, Gurman Dhami

Bioengineering Senior Theses

Oxidative stress, an imbalance of reactive oxygen species and antioxidants, serves as both a marker and progressor of several chronic and degenerative diseases. These conditions create a major global health burden, accounting for significant cost and deaths. The prevalence of oxidative stress–driven diseases highlights the need for improved strategies to detect the imbalance before irreversible damage occurs. However, despite its major role in disease progression, oxidative stress is typically assessed using static, invasive, or laboratory methods. These approaches are poorly suited for continuous monitoring and fail to serve high-risk groups, such as patients with a family history of disease or …

Development Of In Vitro Hydrogel Blood Clots, Xiomara Garcia, Jasmine Urbina, Sapna Wolf

Bioengineering Senior Theses

Deep Vein Thrombosis (DVT) affects up to almost one million individuals in the United States annually and can lead to life-threatening complications, such as a pulmonary embolism. DVT occurs when a thrombus composed of platelets and fibrin forms within deep veins, obstructing blood flow. Current treatments include anticoagulants, thrombolytics, inferior vena cava filters, and balloon catheter procedures. However, the development and testing of these medical devices heavily rely on animal blood models due to physiological similarities in coagulation, raising ethical concerns and limitations related to cost, scalability, and sustainability. This project proposes a synthetic, in vitro alternative using agarose and …

Hvac Design And Calculations, Sarah Walters, Sarah Beyea

Williams Honors College, Honors Research Projects

This project studies Heating, Ventilation, and Air Conditioning (HVAC) systems for a car dealership located in Seguin, Texas. The building construction and architectural drawings were provided by EPIC Engineering Group. Comprehensive load calculations were done on the building in order to calculate necessary airflows throughout the building. Three possible HVAC design solutions were compared: Rooftop Units (RTUs), rooftop units paired with Variable Air Volume (VAV) Terminals with electric reheat, and Variable Refrigerant Flow (VRF) paired with a Dedicated Outdoor Air Unit (DOAS). Energy modeling was completed to compare the utility costs and energy consumption of each system. A comprehensive weighted …

Made On Mars: Design And Manufacturing Of Structural Polymer-Regolith Composites, Pailey M. Vitale

Williams Honors College, Honors Research Projects

Human exploration of Mars is constrained by harsh environmental conditions and the prohibitive cost of transporting materials from Earth. Long-term sustainability requires the local production of mechanical and structural components using resources available on Mars, thereby minimizing payload mass. This project investigates polymer–regolith composites derived from atmospheric CO₂ and mineral-rich regolith as a pathway toward in-situ manufacturing. These composites, when compatible with additive manufacturing, could replace imported plastics, enable on-demand fabrication, and support closed-loop recycling systems. The objective is to design and evaluate polymer–regolith composites that maintain mechanical integrity and environmental resistance under Martian conditions, including extreme temperature swings, radiation …

Active Altitude Control For High Powered Rockets, Henry Allen, Jason Secora, Charles Williams, Donavon Sanchez, Caleb Nedoma

Williams Honors College, Honors Research Projects

The International Rocket Engineering Competition (IREC) is the premier collegiate high power rocketry competition in the world hosted by the Experimental Sounding Rocket Association (ESRA). The Akronauts Rocket Design team has been competing in IREC since 2015 and have won several awards. The main goal of IREC is to launch a rocket to a specified altitude of 10k, 30k, or 45k ft while carrying a payload and successfully recover with little to no damage.

The control of a typical high-powered rocket is purely passive; the center of pressure of the rocket must be behind the center of gravity with respect …

Energy Efficient Windows Using Pvme, James S. Bronstrup

Williams Honors College, Honors Research Projects

This Honors Project aims to improve window energy efficiency using the thermoplastic polymer PVME (polyvinyl methyl ether), which displays temperature-dependent transparency that can be adjusted by mixing it with water and salt to reduce heat transfer through a window. The objective is to determine the optimal PVME water-salt ratios for placement inside a windowpane, evaluate energy savings compared to a traditional argon-filled window, and compare estimated costs. A small-scale model will test different concentrations for various climates, with heat transfer measured by temperature indicators via a LabQuest. Building on research by Yang Zhou at Nanyang Technological University, who used a …

Microscopic Insights On Shear Rheology Of Polyampholyte Ionomers Via Molecular Simulations, Kyle Vonscio

Williams Honors College, Honors Research Projects

Ionomers are polymers that contain a small fraction of charged groups. The ionic attraction of charged groups allows ionomers to act similarly to crosslinked polymers. Depending on the strength of electrostatic interactions and structure of the ionomers, different dynamics and rheological responses can be observed. With moderate attraction between ions, these noncovalent crosslinks can be broken when heated, allowing ionomers to flow and show the strength of a crosslinked network at low temperatures. Nonequilibrium molecular dynamics (NEMD) simulations will be performed under shear in the xy-plane and the macroscopic rheology and conformation of the polyampholyte ionomer will be determined. The …

A 1d Symmetric Interior Penalty Discontinuous Galerkin Solver In Rust, William Aey

Williams Honors College, Honors Research Projects

This honors project will build a 1D Symmetric Interior Discontinuous Galerkin (SIPDG) solver in Rust for Stum-Liouville type problems such as the Poisson equation, with Robin, Dirichlet, and Neumann boundary conditions. The work will cover the full pipeline: starting from the strong form of the PDE, deriving the DG weak form, implementing element and interface operators, and assembling or apply the discrete operator. Rust's safety and concurrency (e.g, via Rayon) will be used to explore serial and parallel performance. A test-driven development approach will be used to maintain a strong suite of tests. The project will result in a documented …

Ai-Based Porosity Detection In Babbitt Bore Turning, Kaylee Leggett, Hannah Adams Gemmell, Kaisa Shingleton

Williams Honors College, Honors Research Projects

For this project, an external company reached out to the University of Akron requesting assistance with defect detection during their vertical turning operations. As babbitt is removed in a vertical turning process, it occasionally reveals defects, mainly porosity, which can lead to costly downstream failures of the part. Current inspection techniques involve use of dye penetrant, which is time consuming, labor intensive, unergonomic, and a source of human error. The goal of the project is to create an alternative inspection method using an AI-based machine-learning model. After the turning operation, a camera is deployed to perform an in-place inspection, taking …

Design New Stage For Vacuum Pin On Disk, Miranda Boyd

Williams Honors College, Honors Research Projects

For my senior design project, I am redesigning the moving stage for the vacuum pin-on-disk test used in the Akron Engineering Tribology Lab. This system is used to evaluate bearings and lubricants for aerospace applications under controlled vacuum conditions. The current stage design limits flexibility when adjusting for different testing parameters, so my goal is to develop a new movable stage that allows for varying weights and loads to be applied accurately and safely. The redesigned stage will improve test efficiency, adaptability, and precision, ensuring the system can accommodate a wider range of experimental conditions and materials while maintaining compatibility …

Printer For Music Box, Chad Lewis, Caleb Murawski, Zion Smith, Bryan Tibbs

Williams Honors College, Honors Research Projects

The traditional method of creating music box sheet music involves manually punching holes into a paper strip using a hand-operated hole punch. This process involves precise knowledge of each note’s location and the ability to achieve perfect accuracy for hours.

The goal of this system is to automate this process, significantly reducing the time required while greatly improving the accuracy of the resulting music box playback. The user simply uploads a MIDI file of their choice into a user-friendly application. Here, the file is modified based on the user’s needs and sent to an automated hole-punching system to punch the …

Security Onion Ids Case Study: Detecting And Investigating Threat Traffic Using Suricata, Zeek, And Pcap Evidence, Daryna Myroniuk

Williams Honors College, Honors Research Projects

A key component of cybersecurity is network intrusion detection, which is used to examine network traffic for any malicious activity. Many organizations deploy intrusion detection systems (IDS) but still face challenges such as validating detections, investigating alerts in a timely manner, and producing clear and repeatable evidence of what has occurred, especially when live traffic capture may be limited to risks, permissions, or privacy concerns. The goal of this project was to use Security Onion, which includes Suricata and Zeek to create and demonstrate a controllable and manageable evidence-based IDS investigation workflow. This project is focused on deploying and validating …

Analyzing Network Traffic And Data Exfiltration Via Smb In Post-Vm Escape Scenarios, Noah M. Disanza

Williams Honors College, Honors Research Projects

Virtual machines (VMs) play a crucial role in modern IT infrastructure environments by providing isolation and enhanced security, among other things, for both personal and corporate systems. VMs are heavily rely upon to safely test malware, manage infrastructure, and reduce risk to host systems. This reliance is so substantial that the idea of reducing risk to the host system is believed to be erasing risk entirely. However, this mindset has shown to be challenged time and time again by the emergence of exploits known as virtual machine escapes. These exploits allow malicious actors to break out of the virtualized environment …

Lumacurve Control System Update, Jacob Kustra, Caleb Wengerd

Williams Honors College, Honors Research Projects

The existing track heater control system used by Lumacurve has become outdated and requires frequent maintenance due to inconsistent operation and component failures. This project aims to modernize the system through the implementation of a programmable logic controller (PLC), infrared non-contact temperature sensing, and an automated pneumatic ejection mechanism.

The redesigned system improves temperature monitoring accuracy and reduces operator intervention by automatically ejecting heated plastic tracks from the heater once the desired temperature is reached. An emergency ejection subsystem is also incorporated to safely clear tracks from the heater in the event of improper operation or system faults. Additional emphasis …

Rapid Skin Stapler, Emma J. Patterson, Allison M. Milligan, Morgan E. Dunlap

Williams Honors College, Honors Research Projects

Severe burn injuries frequently require split-thickness skin grafts, with metal surgical staples serving as the current standard for fixation. While staples provide rapid and secure placement of graft material, they also introduce significant challenges. Removal is painful, time-consuming, and often requires sedation in pediatric patients, leading to added risk, resource strain, and poor patient experience. These drawbacks highlight the need for improved fixation methods that maintain graft stability while reducing procedural burden. This project applies the engineering design process to address these challenges through identification of user needs, translation into design inputs, risk assessment, and development of a prototype aimed …

Lunar Regolith Densometer, Liam Lee, Vincent Innocenzi

Williams Honors College, Honors Research Projects

Testing of extraterrestrial rovers is a growing concern as lunar exploration as an industry is becoming increasingly commercialized. Currently, the way rover testing is conducted is on a soil regolith which is already available. The structure of the soil, however, is harmful to humans and requires a full bunny suit and respirator to be worn during the conduction of the experiment. This limitation, along with the expensive price tag, prevents college university students and companies from establishing a rover test area in labs. The goal of the overall project is to create a safer testing environment by mimicking the soil-to-tire …