Master's Theses

Permanent URI for this collection

This collection consists of digital copies of master's theses submitted for degree at the colleges and departments of Wichita State University. The collection includes theses beginning of fall 2005 -- summer 2022 as well as selected historical theses.

The complete set of all WSU theses may be found in the WSU Library Catalog. University Libraries has two paper copies of each theses submitted before 2006 and archival microfilm copies in the Libraries Special Collections for theses after 2006.

News

The latest addition to this collection is theses defended in summer 2023

Browse

Recent Submissions

Now showing 1 - 5 of 2030
  • Item
    The application of machine learning algorithms for flood susceptibility assessment for the state of Kansas
    (Wichita State University, 2023-12) Rimal, Prashant; Demissie, Zelalem S.
    Flooding has been a significant problem in the United States (US) over the past century. Since 1996, more than 1,500 flood events have been recorded in Kansas, resulting in more than billions in losses. This project explored the use of machine learning and publicly available data to assess factors affecting flooding and develop a flood susceptibility map for Kansas at multiple resolutions. It aims to explore the major predictor variables or flood-controlling factors and the response of the Stack Generalization across multiple resolutions and scenarios. Six machine learning (ML) algorithms: Logistic Regression (LR); Random Forest (RF); Support Vector Machine (SVM); K-nearest neighbor (KNN); Adaptive Boosting (Ada Boost); Extreme Gradient Boosting (XG Boost) were employed to determine the most important factors influencing the susceptibility of an area to flooding. The learning set for the ML algorithms comprised geospatial datasets of thirteen flood-controlling factors: rainfall, elevation, slope, aspect, flow direction, flow accumulation, Topographic Wetness Index (TWI), distance from the nearest stream, evapotranspiration, land cover, impervious surface, land surface temperature, and hydrologic soil type. A total of 1,528 non-flood inventories were created for two different scenarios, with the only difference being the inclusion of stream buffers for overall analysis. The ML algorithms were compared and used to estimate flood susceptibility for each location in the geodatabase resulting in a flood-susceptibility map for both cases. Overall, testing results showed that the tree-based ensemble algorithms; XGB and RF ML models performed relatively well in both cases over multiple resolutions compared to other models in predicting flooding with an accuracy ranging from 0.82 to 0.97, respectively. Also, variable importance analysis depicted that predictor variables such as distance from the streams, hydrologic soil type, rainfall, elevation, and impervious surfaces significantly affect flood prediction.
  • Item
    Fastener size metrology with machine vision
    (Wichita State University, 2023-12) Sekar, Nirmal Kumar; Boldsaikhan, Enkhsaikhan
    This study aims to establish a new manufacturing systems integration method that enhances the fastener size metrology using a machine vision sensor that is mounted on the wrist of an industrial robot. This metrology method is applicable to inspection of any parts with varying sizes in advanced manufacturing applications, particularly in automotive and aircraft manufacturing. The proposed method offers a new way of integrating algorithmic principles into existing manufacturing systems. It consists of data processing and analysis steps that involve image acquisition via machine vision followed by image processing for feature extraction and metrology. Firstly, a machine vision camera is mounted on the end of a robotic arm and then calibrated. The robot arm is used to automatically move the camera to different perspective view poses for capturing images. Images from two different perspective views are used for disparity mapping that produces a depth map generated from two stereo images. Secondly, the disparity map edges are identified by using edge detection and metrology tools for fastener size metrology. The experimentation used ideal simulation images instead of actual camera images for analysis and validation. The results with simulation images indicate that the proposed methodology can detect ±0.005 cm variations in the fastener length. The accuracy of fastener size metrology depends on the accuracy of edge detection as the edge detection tool may make mistakes due to sporadic variations in the image quality. The hit/miss data of edge detection with the intensity difference threshold of 64 is statistically evaluated by the Probability of detection (POD) analysis. According to the POD analysis, an intensity difference greater than 192 can guarantee the 1.0 (100%) mean probability of detection with the 95% lower confidence interval curve that is greater than 0.8 (80%). Keywords: Stereo Images, Disparity Map, Probability of Detection, Machine Vision, Metrology.
  • Item
    Exploring the synthesis of boron nitride at low temperature
    (Wichita State University, 2023-12) Yara, Nikhil Kumar; Wei, Wei
    Boron nitride (BN) is one of the most advanced ceramic materials that have appealing properties such as electrical insulation, mechanical strength, and high thermal conductivity. There are a lot of methods to synthesize BNs. In this work, the chemical vapor deposition (CVD) which requires temperatures around 1000°C was used. This thesis work aimed to design and develop a lower- temperature CVD method for the production of BN to improve efficiency and reduce costs. B, MgO and Fe2O3 were used as the precursors in the ratio of 2:1:1 respectively. Along with this, He and NH3 gases were used to carry out the reaction to produce the BN as the end product which has nanotubes, flakes, hair-like structures, and bubbles. Various temperatures in the range of 800 to 1000ºC with varying reactions, flow rates of gases, and pressure were investigated. There was an effect on BN production by varying the flow rate and reaction time. XRD and SEM were employed to characterize the obtained BN. It showed that BNNTs were obtained at 800ºC, 850ºC and 900ºC with a shorter reaction time between 30 to 45 mins and an NH3 flow rate of 1.00 - 1.25 L/min. At various higher temperatures, BN with flakes, hair-like structures and bubbles were obtained. Under less optimal parameters, amorphous boron nitride nanostructures were formed. This thesis demonstrates a promising energy-efficient CVD route for BN and also synthesized BN with some structures like nanotubes, flakes and bubbles. Usually, these BNNTs are synthesized at very high temperatures but this work was able to produce BNs at about 200°C below the conventional temperatures. The results provide new insight into the relationships between temperature, flow rates, duration, and BN yield. Further work will be needed to improve nanotube purity and density. Nonetheless, the technique developed represents progress toward greener, more cost-effective BNNT production.
  • Item
    Evaluating machine learning models using seismic data sets from different settings
    (Wichita State University, 2023-12) Taskinen, Eldon; Demissie, Zelalem S.
    In prior decades the concept of using mathematical methods to predict earthquakes was considered infeasible. Recent advances in machine learning and predictive modeling offer promising avenues to potentially realize earthquake prediction. In order to test the viability of machine learning methods, experiments were made with earthquake datasets from Kansas and Puerto Rico. The two datasets were chosen for the distinct differences in their tectonic settings. Kansas has few major faults, with a largely inactive subsurface, this produced a smaller dataset with a few large clusters. Puerto Rico is complexly faulted, with an extremely active tectonic setting, this produced a larger dataset with a large number of small clusters. In order to test the effectiveness of these two datasets for machine learning and prediction they were run through four different machine learning algorithms including an LSTM model, Bi-LSTM model, Bi-LSTM model with attention, as well as a transformer algorithm. Not only were the four different machine learning methods compared against each other for accuracy but also the datasets as well. Conclusive findings show that the two different data sets favor different processing methods. The Kansas data performs the best with the Bi-LSTM with attention model, while the Puerto Rico data performs the best with the LSTM model. This is likely due to the tectonic settings of the two regions, since the Kansas dataset has less overall data, and earthquakes are concentrated in a few large clusters, while the Puerto Rico data set has a more even distribution.
  • Item
    Butler county at a crossroads: What was lost is remembered
    (Wichita State University, 2023-12) Walenta, Suzanne; Price, Jay M., 1969-
    At some point in all communities, large or small, a challenge emerges of how to sustain and grow overtime. This crossroads can manifest itself when faced with a choice or challenge of honoring its past and identity with a perception of rural life that is ever changing. A crossroads can often apply to memory and the moments that we choose to reflect on. However, the nostalgia of memory doesn’t always reflect the complex reality. In Butler County, an observer can witness a microcosm of the national experience - similar events that were tucked away, glossed over or hidden from view. The objective for Kansas Crossroads of Butler County was to move beyond the founder story narrative and unveil the complexities and richness of a rural community. The Butler County Historical Society’s Rural Crossroads/Kansas Crossroads project started out as a “quick” history retelling of selected towns which grew into a longer episodic series named, Kansas Crossroads of Butler County. The foundation was to use the community representative perspective through recorded interviews; use historical photos and documents, newspapers, and books to explore each community and to share stories that people may not be aware of, negative and positive. The result of this series was a version of rural communities that nostalgia often forgets. There are commonalities in each of the fifteen communities explored – farming, ranching, and oil discovery. However, there are many stories that have been hidden away and forgotten. Nostalgia often forgets community struggles with the Klu Klux Klan; white mob violence trigged in the oil boom days; 19th century physical fights and guns at the ready as two towns fight over the county seat; lawlessness and horse thieves. There are also hidden triumphs such as the first all-women jury. Looking back there are lessons revealed not only in lived experience but in how stories are presented.