ISME Theses and Dissertations

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 189
  • 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
    A study of stress and temperature distribution on tool face while cutting materials prone to shear banding
    (Wichita State University, 2023-12) Pandey, Aditya; Moscoso-Kingsley, Wilfredo
    Shear banded chip formation typically occurs with a shear banding frequency between 1 kHz and 100 kHz. Large cyclic changes in cutting force, contact length and stress distribution along the tool tip are predicted by finite element analysis. However, these output variables are sensitive to the mechanical behavior of the material under the large strain, strain rate and temperature conditions that exist in and around these shear bands. This study is aimed at experimental measurement of the stress distribution within cutting tools at very high speeds in order to measure the cyclic variation in cutting forces and contact stresses along the tool rake face. Orthogonal machining of Al-7075-T6 and Ti6Al4V tubular specimens is carried out with a transparent sapphire plate as the cutting tool. Machining is done at high feed and relatively low cutting speed to result in shear banded chips exhibiting a shear banding frequency of 3 to 5 kHz, as inferred from measured cutting forces and chip morphology. A Photron Crysta high-speed polarization measuring camera system is used to measure the principal stresses as well as the principal stress direction at 60,000 Hz. This high frame rate has helped resolve cyclic variations in stresses within individual shear band cycles. Quantitative analysis of the stress images using the shear difference method is carried out to yield the distribution of shear stress and normal stress over the rake face contact. The thermography study has been carried out to obtain the thermal behavior of Ti6Al4V when machined at high velocities and in order to compliment previously reported data on effect of high-speed machining of Ti6Al4V on the tool wear. A controlled effort was made to anneal and bevel the cutting edges of the tools and towards removal of radial and lateral runout from the workpiece material in an attempt to reduce tool chipping.
  • Item
    Assessing and predicting ambulance crew workload in emergency medical services operations
    (Wichita State University, 2023-12) Paul, Kevin Alex; Cure, Laila
    This thesis investigates the relationship between ambulance crew workload and operational metrics in Emergency Medical Systems (EMS) and the applicability of machine learning models for predicting crew workload during a shift based on dispatch data. Utilizing call dispatch data from a collaborating EMS organization, the study incorporates observational task analysis, employing cumulative work time and a VACP (Visual, Auditory, Cognitive, and Psycho-motoric) scale to quantify task durations and estimate workload scores. A trace-based simulation approach is developed which integrates dispatch data and task analysis results to quantify workload. This simulation is used to study and reveal statistically significant correlations between crew workload and commonly used operational metrics, including unit hour utilization, number of calls, and scene time interval. Response time utilization, a not commonly used metric but one that can be easily estimated using dispatch data, was found to correlate better with workload than traditional metrics. Multiple machine learning models are identified from the literature review for the problem of predicting shift end workload. These models which include linear regression, decision tree regression, random forest regression, recurrent neural network (RNN), and long short-term memory (LSTM) are tuned, trained, and then evaluated in terms of performance, adaptability to input, and ease of use /applicability in EMS. A sequentially trained linear regression model is identified as the best solution for predicting ambulance crew workload at specified posts. This study lays the groundwork for future research, providing a framework to quantify medics' workload and offering data-driven insights to advance the understanding of workload dynamics for informed management decisions, potentially mitigating burnout.
  • Item
    Data analytics solutions for asset management of overhead transmission lines
    (Wichita State University, 2023-12) Meier, Kylie; Tamimi, Al
    The energy grid relies on a dependable transmission system to traverse electricity across miles through plains, mountains, and cities. The grid is in a continuous state of expansion and older lines are expected to operate for decades longer than their originally anticipated lifespan when they were installed. Each line is exposed to many different environmental modalities that cause the integrity of the line to erode. Animals, fungi, moisture and air, storms, hurricanes, car crashes, and the strain of the loaded lines all affect the transmission line’s states of degradation. Knowing how and when to replace these lines is paramount to keeping a reliable energy system functional. While the transmission system has many components that are required to perform well to keep energy flowing across the lines, there are two parts that have the most detrimental consequences, given their failure. The pole or structure and the conductor are the foundation of the transmission line and require the most attention at their demise. Repairs and replacements are heavily dependent on the budget of the utility company. The goal of this thesis is to develop analytical models that support maintenance and replacement decisions for overhead transmission lines. In collaboration with a local utility company, we develop data-driven and physics-informed predictive models for degradation of overhead transmission conductors due to environmental and operational conditions. Additionally, we develop time-based and condition-based maintenance strategies for treatment and replacement of utility poles under budget limitation using integer programming and Markov decision processes, respectively.
  • Item
    Quantum machine intelligence in smart transportation
    (Wichita State University, 2023-12) Harikrishnakumar, Ramkumar; Nannapaneni, Saideep; Krishnan, Krishna K.
    Smart Mobility is the key component of the Smart City initiative that is being explored throughout the world. In recent years, bike-sharing systems (BSS) are being widely established in urban cities to provide a sustainable mode of transport, by fulfilling the mobility requirements of public residents. The application of BSS in highly congested urban cities reduces the effect of overcrowding, pollution, and traffic congestion problems. The crucial role behind incorporating BSS depends on the prediction and rebalancing of bike demand across all the bike stations. The bike demand prediction involves real-time analysis for identifying the discrepancy between the bike pick-up and drop-off throughout all the bike stations in a given time period. The critical part of a BSS operation is the effective management of rebalancing vehicle carrier operations that ensure bikes are restored in each station to their target value during every pick-up and drop-off operation. To enhance the prediction and rebalancing analysis of bike demand we propose quantum computing algorithms to provide computational speedup in comparison with classical algorithms. In my thesis, we focused on developing algorithms for solving prediction and combinatorial optimization problems with applications in Shared Mobility. We extensively used three methods of approach (a) Quantum Bayesian network which is quantum equivalent to classical Bayesian network for bike sharing demand prediction problems, (b) Optimization models and Quadratic unconstrained binary optimization models for solving combinatorial optimization problems such as rebalancing bike sharing systems, (c) Ensemble of prediction models with Deep learning models to measure the accuracy and computational performance of both (Quantum & Classical) computing platforms.
All items in SOAR are protected by copyright, with all rights reserved, unless otherwise indicated.