Engineering Commons^™

Predictive Residual Neural Networks For Optical Trapping Of Small Particles, Nasim Mohammadi Estrakhri, Ponthea Zahraii, Saman Kashanchi, Nooshin M. Estakhri

Engineering Faculty Articles and Research

Optical tweezers provide a non-contact method to trap, move, and manipulate micro- and nano-sized objects. Using properly designed dielectric and plasmonic nanostructure configurations, optical tweezers have been tailored to create stable and precise trapping for nanoscale objects. Recent advances in numerical optimization techniques allow further enhancement in nanoscale optical traps through inverse optimization of such configurations. One of the main challenges in such optimization approaches is the time-consuming nature of full-wave simulation of nanostructures and postprocessing steps to extract optical forces. To address this challenge, we introduce a surrogate solver based on residual neural networks that can accurately predict the …

Data Provenance Via Differential Auditing, Xin Mu, Ming Pang, Feida Zhu

Research Collection School Of Computing and Information Systems

With the rising awareness of data assets, data governance, which is to understand where data comes from, how it is collected, and how it is used, has been assuming evergrowing importance. One critical component of data governance gaining increasing attention is auditing machine learning models to determine if specific data has been used for training. Existing auditing techniques, like shadow auditing methods, have shown feasibility under specific conditions such as having access to label information and knowledge of training protocols. However, these conditions are often not met in most real-world applications. In this paper, we introduce a practical framework for …

Data-Driven Roughness Estimation Of Additively Manufactured Samples Using Build Angles, Jose Galarza, Jose Barron Jr., Farid Ahmed, Jianzhi Li

Manufacturing & Industrial Engineering Faculty Publications

Achieving control of Laser Powder Bed Fusion (L-PBF) over the quality of the print is the main motivation for finding an optimum set of parameters in the process. Surface roughness is one of the characteristics of the print that impacts the performance of the desired functionality. This research focus is to relate the build angle with the surface roughness on the L-PBF printed specimens and utilize machine learning methods for roughness estimation of geometric features with varying build angles. The EOS M290 L-PBF printer was used to print Inconel-718 coupons using standard process parameters while varying build angles from 20 …

Etdsuite: A Toolkit To Mine Electronic Theses And Dissertations To Enrich Scholarly Big Data Using Natural Language Processing And Computer Vision, Muntabir Hasan Choudhury

Computer Science Theses & Dissertations

In the past decades, there has been a growing interest in mining scientific documents to obtain domain knowledge automatically. One of the understudied types of scientific documents is Electronic Theses and Dissertations (ETDs), as ETDs have distinct features compared with conference proceedings and journal articles. ETDs usually serve as partial requirements of academic degrees for students pursuing higher education. They are book-length documents (i.e., 100 – 400 pages long), and the topics may shift across chapters, exhibit the significant contribution of a student’s research over the entire degree pursuing period, and have unique metadata schema and page layouts. However, the …

Sub-Surface Geospatial Intelligence In Carbon Capture, Utilization And Storage: A Machine Learning Approach For Offshore Storage Site Selection, Mehdi Nassabeh, Zhenjiang You, Alireza Keshavarz, Stefan Iglauer

Research outputs 2022 to 2026

This study introduces an innovative data-driven and machine-learning framework designed to accurately predict site scores in the site screening study for specific offshore CO2 storage sites. The framework seamlessly integrates diverse sub-surface geospatial data sources with human aided expert-weighted criteria, thereby providing a high-resolution screening tool. Tailored to accommodate varying data accessibility and the significance of criteria, this approach considers both technical and non-technical factors. Its purpose is to facilitate the identification of priority locations for projects associated with Carbon Capture, Utilization, and Storage (CCUS). Through aggregating and analyzing geospatial datasets, the study employs machine learning algorithms and an expert-weighted …

Autonomous Real-Time Model Updating Within Digital Twin Frameworks For Thermal Systems, Braden Robert Priddy

Theses and Dissertations

As engineering systems increase in scale and complexity in the era of the Fourth Industrial Revolution, data-driven solutions will become essential in enabling the next generation of these systems. One of the trending tools that can aid in this transition is digital twins. As physical systems degrade throughout their life cycles, their behavior also changes. Digital twins use data assimilation to continuously update virtual models to represent the current state of their physical counterparts. A reliable digital twin can be leveraged by a system operator to perform diagnostics, optimize, and tests without ever needing the physical system. However, implementing effective …

The Evaluation Of Machine Learning Techniques For Isotope Identification Contextualized By Training And Testing Spectral Similarity, Aaron P. Fjelsted, Tyler J. Morrow, Clayton D. Scott, Yilun Zhu, Darren E. Holland, Azaree T. Lintereur, Douglas E. Wolfe

Faculty Publications

Precise gamma-ray spectral analysis is crucial in high-stakes applications, such as nuclear security. Research efforts toward implementing machine learning (ML) approaches for accurate analysis are limited by the resemblance of the training data to the testing scenarios. The underlying spectral shape of synthetic data may not perfectly reflect measured configurations, and measurement campaigns may be limited by resource constraints. Consequently, ML algorithms for isotope identification must maintain accurate classification performance under domain shifts between the training and testing data. To this end, four different classifiers (Ridge, Random Forest, Extreme Gradient Boosting, and Multilayer Perceptron) were trained on the same dataset …

Text-To-Model Transformation: Natural Language-Based Model Generation Framework, Aditya Akundi, Joshua Ontiveros, Sergio Luna

Mechanical Engineering Faculty Publications

System modeling language (SysML) diagrams generated manually by system modelers can sometimes be prone to errors, which are time-consuming and introduce subjectivity. Natural language processing (NLP) techniques and tools to create SysML diagrams can aid in improving software and systems design processes. Though NLP effectively extracts and analyzes raw text data, such as text-based requirement documents, to assist in design specification, natural language, inherent complexity, and variability pose challenges in accurately interpreting the data. In this paper, we explore the integration of NLP with SysML to automate the generation of system models from input textual requirements. We propose a model …

2024 Summer Proceedings Teuscher Lab, Teuscher Group, Christof Teuscher, Chelsea Ogbede, Lauren Sanday, Sofia Vargas, Artem Arefev

altREU Projects

How will computation evolve in the coming years? What problems can be tackled using artificial intelligence, in a world increasingly driven by data? And how can that data be used to better inform our decisions as a society? In this unique collection of research projects, each chapter represents a distinct work undertaken by a single individual or a group of students as part of the altREU program led by Christof Teuscher. The projects, rooted in applications of artificial intelligence and innovative computation techniques, examine impactful solutions to numerous pressing challenges affecting communities around the world.

Federated Learning Systems For Mobile Sensing Data, Xiaopeng Jiang

Dissertations

Federated Learning (FL) has emerged as a new distributed Deep Learning (DL) paradigm that enables privacy-aware training and inference on mobile devices with help from the cloud. This dissertation presents a comprehensive exploration of FL with mobile sensing data, covering systems, applications, and optimizations.

First, a mobile-cloud FL system, FLSys, is designed to balance model performance with resource consumption, tolerate communication failures, and achieve scalability. In FLSys, different DL models with different FL aggregation methods can be trained and accessed concurrently by different apps. In addition, FLSys provides advanced privacy-preserving mechanisms and a common API for third-party app developers to …

Collision Dynamics Of Compound Droplets In Microchannels: A Combined Numerical And Data-Driven Study, S M Abdullah Al Mamun

Dissertations

Understanding and predicting the hydrodynamic interactions of micron-scale droplets is crucial in a wide range of industrial and real-life applications, including microfluidics, pharmaceutics, drug delivery, food science, and enhanced oil recovery. These multi-phase and multi-scale phenomena are further complicated by the presence of core droplets of an immiscible fluid within shell droplets, known as compound droplets. The collisions and interactions of droplets in emulsions are influenced by various physical and geometric parameters, leading to distinct rheological and dynamic responses. This research employs numerical methods for a systematic parametric study of both simple and compound droplet pair collisions under confined shear …

Scla 521 Ai In Society, Bert Chapman

Libraries Faculty and Staff Presentations

Provides access to information resources on societal impacts of artificial intelligence from multiple libraries databases covering multiple disciplines including government information resources.

Enhancing Predictive Accuracy Of Compressive Strength In Recycled Concrete Using Advanced Machine Learning Techniques With K-Means Clustering, Ziaul Haq Doost, Leonardo Goliatt, Mohammed Suleman Aldlemy, Mumtaz Ali, Bruno Da S. Macêdo

AUIQ Technical Engineering Science

The urgent need to mitigate environmental impacts in the construction industry drives the exploration of sustainable practices, such as the use of recycled materials in concrete production. The primary objective of this study was to enhance the predictability of compressive strength in the concrete through the application of advanced machine learning (ML) techniques, specifically Gradient Boosting Regression (GBR) and Random Forest Regression (RFR). Using a comprehensive dataset of 353 eco-friendly concrete samples, the study carefully developed and validated these models to evaluate their performance. The findings exposed that the GBR model outperformed the RFR model, obtained an R² of 0.97 …

Facial Expression Recognition And Classification For Autism Spectrum Disorder Based On Sdft Images Using Deep Learning, Sameer Hameed Abdulshahed, Ahmad Taha Abdulsadda

AUIQ Technical Engineering Science

Autism spectrum disorder (ASD) is a prevalent condition in childhood, affecting around 1 in 44 individuals. Approximately 53% of children with ASD exhibit one or more challenging behaviors (CBs), which include aggression, self-injury, property destruction, elopement, and more. This percentage is significantly higher compared to their typically developing peers or those with other developmental disorders. Cognitive-behavioral therapy (CB) has numerous detrimental effects on the individual, all of which are linked to an unfavorable long-term result. When it comes to caregivers of children with ASD, the presence of Challenging Behaviors (CB) is a more dependable indicator of stress than the intensity …

Approximate Computing And In-Memory Computing: The Best Of The Two Worlds!, Mohammed Essa Fawzy Essa

Theses and Dissertations

Machine learning (ML) has become ubiquitous, integrating into numerous real-life applications. However, meeting the computational demands of ML systems is challenging, as existing computing platforms are constrained by memory bandwidth, and technology scaling no longer yields substantial improvements in system performance. This work introduces novel hardware architectures to accelerate ML workloads, addressing both compute and memory challenges. In the compute domain, we explore various approximate computing techniques to assess their efficacy in accelerating ML computations. Subsequently, we propose the Approximate Tensor Processing Unit (APTPU), a hardware accelerator that utilizes approximate processing elements to replace direct quantization of inputs and weights …

Simultaneous Crack & Wave Propagation And Acoustic Emission Signal Modelling Using Peri-Elastodynamic, Md Mushfiqur Rahman Fahim

Theses and Dissertations

This work presents the use of Peri-Elastodynamic, a guided wave simulation method based on meshfree non-local Peridynamics theory. To model simultaneous crack and wave propagation simulation and its application on Acoustic Emission (AE) signal modelling, a new formulation is presented. The field of nondestructive evaluation (NDE) and structural health monitoring (SHM) is slowly embracing the advancement in Machine Learning (ML) / Artificial Intelligence (AI) for better and cost-effective assessment of structures and damage prediction. AI/ML can revolutionize the NDE/SHM field by automating the data collection and analyzing processes. However, the big challenge is that the model needs a sheer amount …

Application Of Machine Learning Techniques And The Unscented Kalman Filter To Real-Time Gas Turbine Clearance Prediction, Donald Earl Floyd

Theses and Dissertations

The growth in renewable energy sources and retirement of large baseload coal-fired power stations has led to an accompanying decrease in reliability and security of the electrical grid. Since renewable energy sources are typically non-dispatchable, this can lead to blackouts and/or brownouts for customers. Heavy duty gas turbine power plants (HDGT) offer a solution to this problem. HDGTs are dispatchable, clean, and offer flexibility in the fuel they consume, but operational limitations must be well understood to fully exploit their benefits.

One of the main operational limitations is the tip clearances in the gas turbine. In many cases, the gas …

Implementation Of Machine Learning Using Deep Neural Networks To Estimate The Failure Risk Caused By Leakage In Pressure Relief Devices, Adi Yudho Wijayanto, Yossi Andreano, M. Ali Yafi Rizky, Dedi Priadi

Journal of Materials Exploration and Findings

The primary objective of deploying Pressure Relief Device (PRD) equipment is to ensure the safety of pressure vessels within a pressurized system. Over time, PRD equipment may degrade and fail to perform its intended function, which must be identified as a failure mode. To mitigate potential risks associated with this, it is recommended that an approach such as risk-based inspection (RBI) be implemented. Despite the widespread adoption of RBI, the method relies on qualitative techniques, leading to significant variations in equipment risk assessments. This study proposes a novel risk analysis method that uses deep learning-based machine learning to develop a …

Identifying Waterway Traffic Flow Patterns Using Modified Clustering, Shihao Pang

Graduate Theses and Dissertations

Efficient management of inland waterways is essential for the economic and operational efficiency of transportation networks. Characterization and prediction of waterway vessel traffic flow patterns by time of day are critical for optimizing planned disruptive events like maintenance activities. This study identifies and predicts inland waterway traffic flow patterns along the Lower Mississippi River (LMR) using a modified clustering approach. A five-year period of Automatic Identification System (AIS) data, which tracks vessel movements in real-time, is used for model development and evaluation. The model first segments the river into approximately one-mile-long traffic message channels (TMCs) to estimate vessel counts and …

Real-Time Barge Detection Using Traffic Cameras And Deep Learning On Inland Waterways, Geoffery Eyram Agorku

Graduate Theses and Dissertations

Inland waterways are critical for freight movement, but limited means exist for monitoring their performance and usage by freight-carrying vessels, e.g., barges. While methods to track vessels, e.g., tug and tow boats, are publicly available through Automatic Identification Systems (AIS), ways to track freight tonnages and commodity flows carried on barges along these critical marine highways are non-existent, especially in real-time settings. This paper develops a method to detect barge traffic on inland waterways using existing traffic cameras with opportune viewing angles. Deep learning models, specifically, You Only Look Once (YOLO), Single Shot MultiBox Detector (SSD), and EfficientDet are employed. …

Reconfigurable Over-The-Air Chamber: Measuring Radio Frequency Device Performance, Benjamin T. Arnold

Theses and Dissertations

Over-the-air (OTA) testing is particularly useful in determining the performance of multi-antenna communication devices in real-world environments. Traditional OTA testing technologies include the reverberation chamber (RC) and the multi-probe anechoic chamber. More recently, the reconfigurable OTA chamber (ROTAC), which is a RC that has probes lining its chamber walls, has been proposed. These probes are either driven with a source, possibly combined with a channel emulator, or are terminated with a tunable impedance. Controlling the excitations and terminations on the probes can alter the fields within the chamber and thereby synthesize an antenna response at the device under test (DUT). …

Exploring Telehealth Utilization Through Data Analytics, Statistical Analyses, And Machine Learning Techniques, Aysenur Betul Cengil

Graduate Theses and Dissertations

This dissertation investigates the utilization of telehealth services, initially focusing on the Arkansas healthcare system and then extending the analysis nationwide. It aims to understand the factors influencing telehealth adoption and its impact on healthcare delivery. After examining telehealth utilization in Arkansas from 2018 to 2022, the research utilizes a comprehensive dataset from Epic Cosmos, which includes a wide range of patient and visit data from multiple healthcare facilities across the United States from 2018 to 2023. This timeframe allows for a detailed analysis of telehealth trends before, during, and after the COVID-19 pandemic. In Chapter 2, we analyze key …

Integration Of Matlab And Machine Learning To Accelerate Evaluation Of Biological Activity In Agricultural Soils And Promote Soil Health Improvement Goals, Andrew Stiven Ortiz Balsero

Department of Agricultural and Biological Systems Engineering: Dissertations, Theses, and Student Research

Traditionally, assessments of soil biological activity have been confined to laboratory settings, creating a disconnect with practical in-field methods. To bridge this gap, cotton fabric degradation has been used to illustrate soil microbial activity under different management practices. While effective, these demonstrations are subjective and labor-intensive.

Researchers have explored using image processing software like ImageJ and Adobe Photoshop to streamline this process. Although these tools accurately quantified fabric degradation under varying soil conditions, the methods remained labor-intensive and complex. Consequently, these methods were still not ideal for on-farm use by agricultural practitioners.

To further address labor and complexity limitations, the …

Development Of A Turning Movement Estimator, Somayeh Nazari Enjedani

Boise State University Theses and Dissertations

Turning Movement (TM) counts at intersections are crucial for several reasons. There has been extensive research on this topic throughout the history of traffic research, which reflects its importance in transportation engineering and urban planning. It can be said that one of the most significant applications of TMs is signal timing at intersections. Signal timing design requires data on the turning decisions of vehicles that travel through intersections to allow enough time for each movement at the intersection. Most of organizations around the USA are still applying traditional methods for collecting TM data, such as manual counts. These methods are …

Emerging Technologies And Advanced Analyses For Non-Invasive Near-Surface Site Characterization, Aser Abbas

All Graduate Theses and Dissertations, Fall 2023 to Present

This dissertation introduces novel techniques for estimating the soil small-strain shear modulus (G_max) and damping ratio (D), crucial for modeling soil behavior in various geotechnical engineering problems. For G_max estimation, a machine learning approach is proposed, capable of generating two-dimensional (2D) images of the subsurface shear wave velocity, which is directly related to G_max. The dissertation also presents a method for estimating frequency dependent attenuation coefficients from ambient vibrations collected using 2D arrays of seismic sensors deployed across the ground surface. These attenuation coefficients can then be used in an inversion process …

Ensemble Machine Learning At The Edge Using The Codec Classifier Structure And Weak Learners Guided By Mutual Information, Aj Beckwith

All Graduate Theses and Dissertations, Fall 2023 to Present

The Codec Classifier is a low-computation, low-memory tree ensemble method that dramatically improves feasibility of image classification on resource-constrained edge devices. It achieves advantages over other tree ensemble methods due the separation of encoder and decoder tasks in the classifier. The encoder partitions feature space, and the decoder labels the regions in the partition. This functional separation of tasks enables the encoder design (partitioning) to be guided by maximizing the mutual information (MI) between class labels and the features (i.e. the encoded representation of the data) without regard to the error performance of the classifier. Experiments show maximizing MI leads …

Leveraging Machine Learning And Stochastic Programming To Address Vaccine Hesitancy In Public Health Resource Allocation, Hieu Trung Bui

Graduate Theses and Dissertations

Infectious disease outbreaks highlight the urgent need for effective strategies to distribute vaccines and allocate critical healthcare resources to contain the disease and reduce its negative impacts on the population. Managing these allocations is a significant challenge, especially in marginalized communities facing uncertainty in healthcare demand and logistical constraints. This dissertation addresses these challenges by investigating factors that influence dynamic changes in vaccine hesitancy (VH) and its implications for disease spread and healthcare resource demand. It develops optimization models for vaccine distribution and resource allocation under uncertainty, validated with data from the COVID-19 pandemic in the U.S. The first study …

Mixed Potential Sensor For Natural Gas Leak Detection Design, Manufacturing, And Simulation, Sleight Halley

Chemical and Biological Engineering ETDs

There is a need for a low-cost robust sensor accurately detecting leaks from oil and gas infrastructure. This work presents the development of an electrochemical sensor based on an yttria stabilized zirconia (YSZ) electrolyte with La_0.87Sr_0.13CrO₃, Indium Tin Oxide (In₂O₃ 90 wt%, SnO₂ 10 wt%), Au, Pt electrodes. Selectivity to target gasses of the three sensing electrodes in conjunction with machine learning allows for the accurate discrimination between possible methane sources as well as the quantification of methane concentration. Sensor sensitivity is improved through a low-ionic conductivity substrate of magnesia …

A New Approach: Ordinal Predictive Maintenance With Ensemble Binary Decomposition (Opmeb), Ozlem Ece Yurek, Derya Birant

Turkish Journal of Electrical Engineering and Computer Sciences

Predictive maintenance (PdM), a fundamental element of modern industrial systems, employs machine learning to monitor equipment conditions, estimate failure probabilities, and optimize maintenance schedules. Its core objective is to enhance equipment reliability, extend lifespan, and minimize costs through data-driven insights by enabling efficient maintenance scheduling, reducing downtime, and optimizing resource allocation. In this paper, we propose a novel ordinal predictive maintenance with ensemble binary decomposition (OPMEB) method for the PdM domain, considering the hierarchical nature of class labels reflecting the machine's health status, including categories like healthy, low risk, moderate risk, and high risk. The proposed OPMEB method was validated …

A Method For Predicting The Tar Yield Of Tar-Rich Coals Based On The Bp Neural Network Using Multiple Indicators Of Coal Petrography And Coal Quality, Qiao Junwei, Wang Changjian, Zhao Hongchao, Shi Qingmin, Zhang Yu, Fan Qi, Wang Duo, Yuan Dandan

Coal Geology & Exploration

Objective Tar yield, the most important coal quality parameter for coal utilization through low-temperature pyrolysis, determines the clean utilization of tar-rich coals. However, various constraints result in limited test data on tar yield in the geological exploration stage of coals, substantially restricting the fine-scale assessment and efficient utilization of tar-rich coals. Methods To achieve more scientific and accurate fine-scale tar-rich coal assessments, this study examined 1073 sets of lithotype and coal quality data obtained previously from a Jurassic coalfield in northern Shaanxi. From these data, 141 sets with 20 lithotype and coal quality parameters regarding macerals, proximate analysis, ultimate analysis, …