Matthew Clark (Committee Co-chair), Kuldip Rattan (Committee Co-chair), Zhang Xiadong (Committee Member)
Master of Science in Engineering (MSEgr)
Complex cyber-physical systems are difficult to model and control. However, humans are capable of accomplishing these tasks by constantly adapting and redefining the rules to control these complex systems. Fuzzy logic provides a means of encoding human inference into a control methodology. However, the fuzzy logic controllers are nonlinear and their stability is difficult to verify. Therefore, the widespread usefulness of fuzzy logic controllers is limited. It has been proven that fuzzy logic controllers can be implemented as piecewise linear switched controllers. It has also been shown that the piecewise linear system can be implemented as a hybrid system. Piecewise linear hybrid system stability can be verified by extending the Lyapunov proof for one linear system to multiple decreasing Lyapunov functions. The objective of this thesis is to implement fuzzy logic control systems as a piecewise linear hybrid system and examine their stability. A proportional fuzzy logic controller with constant derivative gain is implemented as a piecewise linear hybrid system using Matlab Simulink Stateflow. Stability of the system is examined by obtaining the Lyapunov function of each subsystem and stitching them according to the fuzzy rules. It is shown that the stitching of Lyapunov functions must successively decrease for the system to be stable. Further implications of robustness are examined by varying the fuzzy logic rules and observing the effect on the corresponding stitched Lyapunov functions.
Department or Program
Department of Electrical Engineering
Year Degree Awarded
Copyright 2013, all rights reserved. This open access ETD is published by Wright State University and OhioLINK.