EECS Theses and Dissertations

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 363
  • Item
    Time sharing of control and interlock circuitry by one wire.
    (Wichita State University, 1962-08) Kelley, Rolland C.; Dunn, Colon H.
    SUMMARY The concept of "time-sharing" of control and interlock wiring is introduced in this paper. This concept calls for the use of one wire to transmit many functions each in its own distinct time period, thereby reducing the number of wires necessary. The type of functions that are adaptable and some aircraft systems that utilize the functions are listed. A comparison of rotating vs. solid state commutators is made, and considerable effort is directed into understanding solid state switching theory. The use and operation in switching circuits of three types of solid state components, (silicon controlled rectifiers, unijunction transistors, and transistors) is presented. The theory of operation of the circuits required to build the "time-sharing" units is discussed. Some circuits included in this discussion are an astable multivibrator, a ring counter, and a channel synchronizing circuit. Successful operation of the "time-sharing" concept was demonstrated by connecting control relays and interlock circuits through the solid state commutators. Some limitations are mentioned as well as possible future uses of "time-sharing". Twenty-six oscillograph traces and the necessary circuit diagrams are found in the appendix.
  • Item
    An electrolytic tank analog to the root locus plot
    (Wichita State University, 1962-12) Benton, Crayton D.; Ford, W. R.; Dunn, Colon H.
    The purpose of this thesis is two-fold; first, it gives the author a fundamental understanding of the root-locus concept or system analysis, and second it presents the development or an electrolytic tank to be used with an automatic plotter for the construction or root-locus diagrams.
  • Item
    Artificial intelligence -based distance relay behavior for future power systems with 100% clean electricity
    (Wichita State University, 2024-07) Oke, Kolade Oladimeji; Pang, Chengzong
    The production of electricity in any society is an essential tool and symbol of development for such a community. In the past, the generation of electrical power has primarily relied on fossil fuels, such as coal, which emit CO2 into the atmosphere, thereby depleting the ozone layer and leaving our planet at risk of destruction. In today's world, efforts are being made to ensure the generation of electrical power through carbon-free generators. Regional Transmission Operators (RTOs) and Independent System Operators (ISOs) have set targets to ensure carbon-free generation in the near future. Owing to this development, there is an increasing addition of renewable generators to the grid, and a future of a 100% renewable generation mix is envisaged. The integration of renewables into the existing grid is also subject to changes in power flow within the power system, which may lead to surges in voltage or current, potentially causing electrical faults in the transmission line. In this research, the behavior and operation of distance relays are simulated, and a protective dataset is created using artificial intelligence methods to predict faults in the line in a 100% clean electricity. This study explores the application of machine learning models in predicting fault conditions in transmission lines based on resistance, inductance, impedance, power, voltage, and current values of the electrical power supply. Three models—logistic regress (Korstanje, sept 2022)ion, support vector machine (SVM), and K-nearest neighbors (KNN)[6]—were trained and evaluated using a dataset emphasizing the power flow results of the transmission line model derived from the Matpower. Hyperparameter tuning was performed to improve predictive accuracy, with SVM [5] (W.Urooj, 2021)and logistic regression showing superior performance in using indices such as current, voltage, and impedance to determine any presumed line faults.
  • Item
    Minimization of expected energy not served by optimal use of electric vehicles in distribution network post extreme weather event via dynamic programming
    (Wichita State University, 2024-07) Ohaeri, Josephine Ezinne; Aravinthan, Visvakumar
    The occurrence of extreme weather events seems to be on the rise, causing system outages and leaving customers without power supply for long time periods. Considerable research has been done to explore the use of distributed energy resources in distribution system restoration post extreme weather event. Due to the global clamor for reduced carbon emissions, emphasis has been placed on the use of renewable stationary power sources such as solar photovoltaic systems and more affordable mobile power sources such as rented electric vehicles to supply as much load as possible till grid supply is restored. This thesis models the resilience level of a typical distribution system having rooftop solar PV system, electric buses and electric vans as distributed energy resources post extreme weather event. These electric vehicles are supplied by transportation and logistics companies that have little or no use of these vehicles post extreme weather event. The location of rooftop solar and the number of electric vehicles is randomized and applied to different network reconfiguration topologies to evaluate the Expected Energy Not Served (EENS) using Monte Carlo Simulation. A base case scenario without electric vehicle integration and a second scenario with electric vehicle integration was considered using this simulation. The appropriate network reconfiguration topology with minimum value of EENS was obtained from generated results. The EENS value of the second scenario was further optimized using dynamic programming as a third scenario. This optimization approach also provides the minimum number of electric vans and buses needed to give minimized EENS when certain constraints are applied. Comparative analysis of the three scenarios is then done to demonstrate the need for use of rented electric vehicles in the enhancement of system resilience post extreme weather event.
  • Item
    Electrically scannable antenna using ferrite radiating elements
    (Wichita State University, 1961-08) Brandshaw, Martin D.; Schrag, Robert L.; Murphy, A. T.
    This thesis is the result of an exploratory investigation concerning the use of magnetized ferrite radiating elements for the construction of electronically scannable antennas Included within the thesis is a brief survey of three electronic scanning techniques and a short summary of the microwave properties of ferrites. A derivation of the ferrite tensor permeability is given in Appendix I o A discussion of three techniques which could be used to analytically determine the radiation pattern is also given. The most promising of the three appears to be the Schelunkoff Equivalence Principle. An experimental device is described which can be used with one or two element arrays to obtain beam scanning. The azimuth radiation patterns produced by these two types of antennas are presented. Measurements of the input VOS. W o R. show that the ferrite sets up large reflections at certain v a lues of applied D.C o magnetic field.
All items in SOAR are protected by copyright, with all rights reserved, unless otherwise indicated.