Marian K. Kazimierczuk (Committee Member), Kuldip S. Rattan (Advisor), Xiaodong Zhang (Committee Member)
Master of Science in Engineering (MSEgr)
High temperature thermal testing is used to validate the effectiveness of thermal protection systems (TPS) and to simulate aerodynamic heating in high speed flight vehicle structures. Silicon controlled rectifiers (SCR) are utilized to drive resistive heating elements and producing temperatures up to 5000 degrees Fahrenheit. Thermocouples used to provide feedback to the control system have a high probability of failure at extreme temperatures. Traditional solutions include switching to redundant thermocouples, changing to an open-loop control scheme, or using power as the control signal. The first methods is limited in that redundant thermocouples are also not completely reliable. Both open-loop control and power control do not provide temperature data and require one complete test to be accomplished with temperature data available. In this thesis, fuzzy logic learning is used on previously recorded power data to estimate the temperature of the test article. The results of the learning algorithm fit the recorded data closely. Additionally, a fuzzy logic prediction algorithm is developed to estimate the temperature from the power data in the event of thermocouple failure. The algorithm is shown to be effective in predicting temperature from power data if sufficient data is collected prior to thermocouple failure.
Department or Program
Department of Electrical Engineering
Year Degree Awarded
Copyright 2010, all rights reserved. This open access ETD is published by Wright State University and OhioLINK.