Engineering Commons^™

Customer Churn Prediction, Deepshikha Wadikar

Dissertations

Churned customers identification plays an essential role for the functioning and growth of any business. Identification of churned customers can help the business to know the reasons for the churn and they can plan their market strategies accordingly to enhance the growth of a business. This research is aimed at developing a machine learning model that can precisely predict the churned customers from the total customers of a Credit Union financial institution. A quantitative and deductive research strategies are employed to build a supervised machine learning model that addresses the class imbalance problem handled feature selection and efficiently predict the …

Forecasting The Short-Term Value Of Wind Power For Risk-Aware Bidding Strategies In Single-Imbalance Price Electricity Markets, Gianni Goretti

Doctoral

The participation of wind energy in electricity markets and strategic bidding in the day-ahead market has been investigated with growing interest in recent years. However, markets adopting a single-price imbalance settlement where participants can increase their profits if they help to put the system back into balance have received very limited attention in the academic literature. In this thesis, new probabilistic models forecasting the short-term value of wind power are developed and their use in bidding in these types of markets is investigated. The proposed strategies are designed for participants who want to bid strategically in the day-ahead market to …

Investigating The Effects Of Flue Gas Injection, Hot Water Distribution, And Fill Distribution On Natural Draft Wet Cooling Tower Performance, John Stapleton

Senior Honors Theses

Natural draft wet cooling towers (NDWCT) are a common method of heat removal in powerplants. This study employs a numerical cooling tower model developed by Eldredge, Benton & Hodgson (1997) to examine whether utilizing nonuniform water, fill profiles, and flue gas injection can improve NDWCT efficiency. The results show that each of these variables can be optimized to lower outlet water temperature. Within the range tested for each parameter, the water profile had the most significant effect on outlet water temperature, followed by the flue gas temperature and then the fill profile. The optimum parameter combination reduced the predicted outlet …

Deep Learning Of Nonlinear Dynamical System, Aditya Wagh

Dissertations, Master's Theses and Master's Reports

A data-driven approach, such as neural networks, is an alternative to traditional parametric-model methods for nonlinear system identification. Recently, long Short- Term Memory (LSTM) neural networks have been studied to model nonlinear dynamical systems. However, many of these contributions are made considering that the input to the system is known or measurable, which often may not be the case. This thesis presents a method based on LSTM for output-only modeling, identification, and prediction of nonlinear systems. A numerical study is performed and discussed on Duffing systems with various cubic nonlinearity.

Magnetism In Γ-Fesi2 Nanostructures: A First Principles Study, Sahil Dhoka

Dissertations, Master's Theses and Master's Reports

First-principles calculations are performed on γ-FeSi₂ nanostructures grown on Si (111) and (001) substrate. An attempt to explain the origin of emergent magnetic properties of the metastable gamma phase of iron di-silicide (γ-FeSi₂) is made, which show ferromagnetic behavior on nanoscale, unlike its possible bulk form. Many papers try to explain this magnetism from factors like bulk, epitaxial strain, interface, surface, edges, and corners but doesn’t provide an analytical study for these explanations. Density functional theory is used to analyze the magnetic effects of these factors. The results for the epitaxial structures show no magnetic behavior for …

A Neural Network Approach To Estimate Buoy Mooring Line Sensor Deflection, Tom Price

Dissertations, Master's Theses and Master's Reports

Instrumented moorings are often used to measure characteristics, such as temperature and current, over the water column. However, the moorings deflect from the effects of currents and waves, which could lead to innacurate measurements. In this work, a computationally efficient method to compensate for mooring sensor position errors is developed. The two-step process first uses a hydrodynamic model of the buoy and mooring line system to create estimated mooring line deflections in a steady current. A neural network model is trained to approximate the hydrodynamic model’s mooring line displacement given the spatial location of the buoy and current profile measurements. …

Sensitivity And Uncertainty Quantification Of Continuum And Stochastic Bond Dissociation Kinetics Based Cohesive Laws, Rishabh Awasthi

Dissertations, Master's Theses and Master's Reports

Polymeric adhesives are being extensively used to develop light weight and durable Composite Materials which are have been employed in automotive, aeronautical, medical industry. Polymeric adhesives undergo failure due to relative opening sliding of bonded interfaces which requires a need to characterize the fracture mechanism. Numerous experiments are conducted to predict the fracture response for Mode-I, II,III and numerical simulations are employed to inversely calibrate the material parameters governing the constitutive laws. Cohesive zone models are largely employed in fracture mechanics to predict failure of interfaces undergoing large plastic deformations. The objective of this report is to employ the cohesive …

Analysis Of A Circular Resonant Plate For Shock Testing, Jon Markl

Dissertations, Master's Theses and Master's Reports

Designing a resonant plate for shock test specifications involves an understanding of how a system will respond with a given set of input and output locations in all three directions. To investigate how to effectively change the off-axis responses to meet test specifications, a finite element model (FEM) was created and examinations of input location/angle of impact, placement of the test fixture, and placement of the impact pad were conducted. The primary tool used to determine how a structure will respond to a given input at a given output on the test structure was done using the shock response spectra …

Effect Of Cobalt In Thin Wall Ductile Iron And Solid Solution Strengthened Ferritic Ductile Iron, Alejandra I. Almanza

Dissertations, Master's Theses and Master's Reports

Ductile Iron is a material that is constantly evolving. Consequently, the ferrous industry is not only focusing on lightweighting but also on improving the impact strength and fracture toughness of typical ferritic-pearlitic ductile iron grades and solid solution strengthened ferritic ductile irons. Recently, the demand for thin-wall ductile iron and solid solution strengthened ferritic ductile iron grades has increased. The challenges behind the fabrication of these two ductile iron materials are the presence of carbides and the embrittlement of ferrite. In response, research has been focused on looking at alternative methods that can mitigate carbide formation in thin sections and …

Experimental Characterization Of Spool Valve Flow Noise, Zachary Cohen

Dissertations, Master's Theses and Master's Reports

The purpose of this research was to experimentally measure the flow noise produced by a spool valve and compare with measurements of orifice flow noise to reduce flow noise modelling complexity. The similarities and differences are compared over a range of volume flow rates, back pressures, and cross section areas. The impact of down stream boundary conditions on the flow noise are also examined. The results are generalized and plotted against Froude number. The valve results are curve fit to generate empirical equations that can be used to predict flow noise under typical operating conditions. The valve and orifice data …

Estimation Of Vehicle Trailer Rotation Using Calibrated Radars With Multiple Viewpoints, Kunle Olutomilayo

Dissertations, Master's Theses and Master's Reports

Multiple sensors are increasingly being deployed on systems for perception applications. In particular, vehicles are becoming equipped with a suite of sensors for advanced driver assistance features and autonomous driving. This dissertation considers automotive radar sensors mounted at the rear of a vehicle with the main objective of using their point cloud detections to estimate the rotation of a trailer which is attached to the vehicle's hitch ball. A simulation-based study of the problem is presented first. Thereafter, the problem is considered with respect to experimental radar data collected in both indoor and outdoor environments; the environmental difference is in …

Design, Build, And Analysis Of A Compressed Natural Gas Direct Injection Compression Ignition Single Cylinder Research Engine, Tyler Miller

Dissertations, Master's Theses and Master's Reports

Both experimental and simulation efforts have been employed to further advance the design and build of a single cylinder research engine. The engine will aid in advancing technology and understanding the operation of compression-ignition (CI) engines using natural gas within the heavy-duty engine industry. The basis for the engine is a Cummins 15L ISX engine that has been modified, retrofitted, and instrumented to allow for late-cycle direct-injection of high-pressure compressed natural gas. Along with the engine build, a one-dimensional GT-Power simulation model has been created and used to analyze the engine operation and specify components including the engine compression ratio …

A Study Of Model-Based Control Strategy For A Gasoline Turbocharged Direct Injection Spark Ignited Engine, Xin Wang

Dissertations, Master's Theses and Master's Reports

To meet increasingly stringent fuel economy and emissions legislation, more advanced technologies have been added to spark-ignition (SI) engines, thus exponentially increase the complexity and calibration work of traditional map-based engine control. To achieve better engine performance without introducing significant calibration efforts and make the developed control system easily adapt to future engines upgrades and designs, this research proposes a model-based optimal control system for cycle-by-cycle Gasoline Turbocharged Direct Injection (GTDI) SI engine control, which aims to deliver the requested torque output and operate the engine to achieve the best achievable fuel economy and minimum emission under wide range of …

Real-Time Predictive Control Of Connected Vehicle Powertrains For Improved Energy Efficiency, Joseph Oncken

Dissertations, Master's Theses and Master's Reports

The continued push for the reduction of energy consumption across the automotive vehicle fleet has led to widespread adoption of hybrid and plug-in hybrid electric vehicles (PHEV) by auto manufacturers. In addition, connected and automated vehicle (CAV) technologies have seen rapid development in recent years and bring with them the potential to significantly impact vehicle energy consumption. This dissertation studies predictive control methods for PHEV powertrains that are enabled by CAV technologies with the goal of reducing vehicle energy consumption.

First, a real-time predictive powertrain controller for PHEV energy management is developed. This controller utilizes predictions of future vehicle velocity …

Phase Field Fracture Modeling Of Chemically Strengthened Glass, Parag Nikam

Dissertations, Master's Theses and Master's Reports

The objective of the report is to implement the phase-field model in the Abaqus standard to compute the fracture properties of a glass strengthened by an ion-exchange process. Implemented the phase-field model which is based on the rate-independent variational principle of diffuse fracture in Abaqus standard software using the UEL and UMAT subroutines. SDVINI subroutine is used to give the residual stress or prestress conditions to simulate the stress profile of strengthened glass. Studied the effect of parameters such as length scale parameter and step size and the optimum parameters are selected. The experimental model of chemically strengthened glass is …

Numerical Investigation Of Spray Characterization Of Heater-Gdi System, Ashwin Karthik Purushothaman

Dissertations, Master's Theses and Master's Reports

Gasoline engines require fuel enrichment at low temperature cranking (cold start) conditions for efficient engine operation. Since the amount of fuel injected is high at cold start to compensate low fuel evaporation, fuel sprays tend to impinge on the cold surfaces of the piston and cylinder walls leading to the formation of excessive unburned hydrocarbons. One of the ways to ensure reliable cold start performance and to reduce UHC emissions is to have the fuel subjected to preheating. The objective of this study is to investigate the effects of fuel preheating on Gasoline Direct Injection sprays to improve the mixture …

Metalimnetic Oxygen Minimum In Green Lake, Wisconsin, Mahta Naziri Saeed

Dissertations, Master's Theses and Master's Reports

Green Lake, located in Green Lake County, Wisconsin, has been experiencing summer Metalimnetic Oxygen Minima (MOM). The severity of the MOM has increased in recent years and the lake was listed as impaired in 2014. To investigate the MOM, the lake was monitored during 2017 and 2018 at two sites using moored temperature and oxygen sensors recording at 1-hour intervals. Using these data, a hydrodynamic model (Simstrat) was configured for the lake. A new model was then developed to simulate the oxygen changes in the lake. Productivity, respiration, sediment oxygen demand (SOD) and fluxes were included, and hydrodynamics were driven …

Optimization Of Shape And Control Of Linear And Nonlinear Wave Energy Converters, Jiajun Song

Dissertations, Master's Theses and Master's Reports

In this dissertation, we address the optimal control and shape optimization of Wave Energy Converters. The wave energy converters considered in this study are the single-body heaving wave energy converters, and the two-body heaving wave energy converters. Different types of wave energy converters are modeled mathematically, and different optimal controls are developed for them. The concept of shape optimization is introduced in this dissertation; the goal is to leverage nonlinear hydrodynamic forces which are dependent on the buoy shape. In this dissertation, shape optimization is carried out and its impact on energy extraction is investigated. In all the studies conducted …

Dynamic Testing: An Experimental Approach To Defect Identification In Additive Manufactured Parts, Aimee Allen

Dissertations, Master's Theses and Master's Reports

Today’s additive manufacturing (AM) industry produces specialized parts at low volume or with complex geometries. Traditional testing methods are effective, but costly and time consuming to perform. The AM industry lacks an optimized testing method for identifying internal defects that occur in parts. The evaluation of multiple parts printed on the same build plate for internal defects using various nondestructive (dynamic) testing techniques is presented. From these experiments, perspective was gained on when and how dynamic testing can be used to find defects. Insight gained from these experiments can help the industry in future testing for internal defects.

Multiscale Modeling Of Carbon Fibers/Graphene Nanoplatelets/Epoxy Hybrid Composites For Aerospace Applications, Hashim Al Mahmud

Dissertations, Master's Theses and Master's Reports

Significant research effort has been dedicated for decades to improve the mechanical properties of aerospace polymer-based composite materials. Lightweight epoxy-based composite materials have increasingly replaced the comparatively heavy and expensive metal alloys used in aeronautical and aerospace structural components. In particular, carbon fibers (CF)/graphene nanoplatelets (GNP)/epoxy hybrid composites can be used for this purpose owing to their high specific stiffness and strength. Therefore, this work has been completed to design, predict, and optimize the effective mechanical properties of CF/GNP/epoxy composite materials at different length scales using a multiscale modeling approach. The work-flow of modeling involves a first step of using …

Increasing Access To Pv Technology Through Sustainable Racking: A Review Of Existing Literature And Ground Mounted Fixed Tilt Designs, And What Can Be Done Next, Parijata Prabhakara

Dissertations, Master's Theses and Master's Reports

There is a lack of literature on the structural balance of systems (BOS), also called racking, for ground mount, fixed-tilt solar PV systems. Literature that exists discusses mostly rooftop racking and installations, additional wind loading, and weight considerations imposed on roofs, and little guidance is provided in building codes. The lack of peer-reviewed guidance on design requirements for domestic and large-scale application solar PV racking systems leaves most consumers relying on expensive, patented, off-the-shelf hardware. As PV cell technology and module costs have improved, we can start to focus on PV BOS improvements (particularly racking) and move towards sustainable designs …

Generative Design In Energy Efficient Buildings, Johanna Thill

Dissertations, Master's Theses and Master's Reports

This is an investigative report on the use of generative design and genetic algorithms in the design of energy-efficient buildings. It focuses on a workflow using Project Refinery© and Dynamo© in Revit® for multi-objective optimization and visualization, and Honeybee© for daylighting analysis. The workflow was not about to run completely due to instability in Dynamo, however daylighting analysis and visualizations were produced separately from demonstrations of Refinery. It concludes that while genetic algorithms have potential to be useful in energy-efficient building design, these programs are not yet fully developed and difficult to use without extensive background knowledge and fluency in …

Towards Efficient Water Treatment: Mechanism Of Colloidal Fouling Of Ultrafiltration Membranes, Ikenna Henry Ozofor

Dissertations, Master's Theses and Master's Reports

This research study first reviewed challenges of conventional and membrane separation systems for water treatment. Though membrane separation systems appeared superior to the conventional counterparts for water and wastewater treatment, wider applications of membrane systems have been limited by some factors, most notably is membrane fouling. Experimental studies were therefore conducted to achieve the goal of this research, which is to investigate how feed properties affect fouling of ultrafiltration (UF) membranes by colloids.

Feed salinity, pH, and nanoparticle concentration were the variables studied to unravel how different UF membranes are fouled by model silica colloids (with average diameter of 25nm). …

Virus Purification Framework And Enhancement In Aqueous Two-Phase System, Pratik Umesh Joshi

Dissertations, Master's Theses and Master's Reports

Viral infections regularly pose detrimental health risks to humans. Preventing viral infections through global immunization requires the production of large doses of vaccines. The increasing demand for vaccines, especially during pandemics such as COVID-19, has challenged current manufacturing strategy to develop advanced unit operations with high throughput capability. Over the decade, the upstream processing responsible for synthesizing viral products in cell cultures has shown significant success in yielding high titers of viruses and virus-like particles. The progress in the upstream stage has now shifted the bottleneck to the downstream processing (DSP). Overlooked for decades, the DSP responsible for viral product …

Effect Of Cerium And Magnesium Additions On The Strength And Morphology Of The Intermetallic Compound Layers In Aluminum-Steel Dissimilar Castings, Matt Thomas

Dissertations, Master's Theses and Master's Reports

Dissimilar castings are under investigation due to their light-weighting opportunities within the automotive industry. In this case, dissimilar castings consist of a solid steel core and a poured aluminum component. The strength and integrity of the casting relies on the coherency of the bond between the two components; by optimizing the metallurgical component of the bonding phenomena, the weight and complexity of the insert can be reduced, further increasing the light-weighting opportunities. An interface comprised of intermetallics, generally Fe₄Al₁₃ and Fe₂Al₅, forms between the two components with the latter forming nearest the steel. …

Lagrangian Relaxation Method For Integrated Vehicles And Drones Routing Model, Perawit Charoenwut

Chulalongkorn University Theses and Dissertations (Chula ETD)

Nowadays, e-commerce increases parcel transportation demand year by year. Many companies are seeking the way to improve performance of delivery service and cost-efficiency. One among many solutions is using a drone attached with a truck for the last mile delivery. A drone can be launched from either depot or truck and returned to the depot or wait for the truck to take them back from the docking station around the delivery area. The vehicle routing problem with drones considers truck- and drone-route simultaneously. It is an enormous problem that cannot be solved by a normal linear programming method. This research …

Effects Of Oxygen Modification On The Structural And Magnetic Properties Of Highly Epitaxial La0.7sr0.3mno3 (Lsmo) Thin Films, Shalini Kumari, Navid Mottaghi, Chih-Yeh Huang, Robbyn Trappen, Ghadendra Bhandari, Saeed Yousefi, Guerau Cabrera, Mohindar S. Seehra, Mikel B. Holcomb

Faculty & Staff Scholarship

La0.7Sr0.3MnO3, a strong semi-metallic ferromagnet having robust spin polarization and magnetic transition temperature (TC) well above 300 K, has attracted significant attention as a possible candidate for a wide range of memory, spintronic, and multifunctional devices. Since varying the oxygen partial pressure during growth is likely to change the structural and other physical functionalities of La0.7Sr0.3MnO3 (LSMO) films, here we report detailed investigations on structure, along with magnetic behavior of LSMO films with same thickness (~30 nm) but synthesized at various oxygen partial pressures: 10, 30, 50, 100, 150, 200 and 250 mTorr. The observation of only ( …

Large-Scale Inflatable Structures For Tunnel Protection: A Review Of The Resilient Tunnel Plug Project, Eduardo M. Sosa, Gregory J. Thompson, Gregory M. Holter, John M. Fortune

Faculty & Staff Scholarship

The protection of underground civil infrastructure continues to be a high priority for transportation and transit security agencies. In particular, rail transit tunnels running under bodies of water are susceptible to disruptions due to flooding caused by extraordinary climatic events such as hurricanes or other events resulting from human activities. Several events have taken place in the past decades that have demonstrated the need to mitigate vulnerabilities or, at least, minimize the consequences of catastrophic events. Although it is impossible to prevent all situations that can lead to flooding, damage can be substantially decreased by reducing the area affected by …

Ground And Aerial Robots For Agricultural Production: Opportunities And Challenges, Santosh Pitla, Sreekala Bajwa, Santosh Bhusal, Tom Brumm, Tami M. Brown-Brandl, Dennis R. Buckmaster, Isabella Condotta, John Fulton, Todd J. Janzen, Manoj Karkee, Mario Lopez, Robert Moorhead, Michael Sama, Leon Schumacher, Scott Shearer, Alex Thomasson

Department of Agricultural and Biological Systems Engineering: Faculty Publications

Crop and animal production techniques have changed significantly over the last century. In the early 1900s, animal power was replaced by tractor power that resulted in tremendous improvements in field productivity, which subsequently laid foundation for mechanized agriculture. While precision agriculture has enabled site-specific management of crop inputs for improved yields and quality, precision livestock farming has boosted efficiencies in animal and dairy industries. By 2020, highly automated systems are employed in crop and animal agriculture to increase input efficiency and agricultural output with reduced adverse impact on the environment. Ground and aerial robots combined with artificial intelligence (AI) techniques …

การลดของเสียในกระบวนการประกอบสายส่งสัญญาณทางแสง (Active Optical Cable), อาภานวล แก้วกาหลง

Chulalongkorn University Theses and Dissertations (Chula ETD)

งานวิจัยนี้มีวัตถุประสงค์เพื่อปรับปรุงกระบวนการประกอบสายส่งสัญญาณทางแสง ที่ระดับความเร็ว 100Gb/s ซึ่งเป็นเทคโนโลยีการส่งสัญญาณสำหรับดาต้าเซ็นเตอร์ที่เติบโตอย่างต่อเนื่อง โดยประยุกต์ใช้เครื่องมือปรับปรุงคุณภาพตามแนวทางซิกซ์ ซิกมา หรือ DMAIC ประกอบด้วย (1) ระยะนิยามปัญหา พิจารณาอัตราผลิตผลดีในกระบวนการประกอบระดับออพติคอลเอนจิน พบว่ากระบวนการวัดกำลังแสงขาออกมีสัดส่วนของเสียเฉลี่ยต่ำที่สุด ประมาณ 10.29% ซึ่งเป็นตัวชี้วัดว่ากระบวนการวางเลนส์ด้วยวิธีการแบบแพสซิพ (Passive Alignment technique) ซึ่งเป็นวิธีการวางตำแหน่งเลนส์โดยใช้ภาพเอกลักษณ์และการคำนวณระยะสัมพัทธ์ระหว่างกึ่งกลางของเลนส์ และกึ่งกลางของเลเซอร์ VCSEL ยังมีคุณภาพดีไม่เพียงพอ (2) ระยะการวัด การวิเคราะห์ความพ้องกันของระบบวัดกำลังแสงขาออก โดยใช้ข้อมูลแบบนับ (Attribute gage R&R) แสดงให้เห็นว่าระบบวัดมีความสามารถในการวัดซ้ำ, วัดได้อย่างถูกต้อง, และระบบมีความพ้องกันดี (Kappa > 0.9) โดยค่า Cpk และ Ppk จากการวิเคราะห์ความสามารถของกระบวนการ เป็น 0.923 และ 0.423 ตามลำดับ (3) ระยะวิเคราะห์หาสาเหตุของปัญหาด้วยการทดสอบสมมติฐานสัดส่วนประชากร 2 กลุ่ม พบว่ามี 4 ปัจจัยที่ส่งผลต่อสัดส่วนของเสียได้แก่ อัลกอริทึมการหาอไลน์เมนท์ออฟเซ็ต, แผ่นวงจรรวมที่ผลิตจากแหล่งที่ต่างกัน, ระนาบแกน Z หรือบอนด์ไลน์, และอุณหภูมิการอบกาวให้แห้งหลังวางเลนส์ (4) การดำเนินการปรับปรุง และ (5) การติดตามผล ซึ่งหลังการปรับปรุงพบว่าว่าสัดส่วนของเสียลดลงจาก 10.29% เหลือ 1.98% หรือลดลง 80% กระบวนการมี Cpk และ Ppk ดีขึ้นเป็น 1.187 และ 0.687 ตามลำดับ