Master of Science in Electrical Engineering (MSEE)
Electrical and Computer Engineering
The Minimum-degree Pole Placement algorithm for Self-tuning Regulator design and the Recursive Least-square method and the projection algorithm for plant estimation are studied first in this thesis. Simulation studies for the estimator and controller algorithms are mainly undertaken after describing how to use MATLAB S-function in detail. Not only do Simulation experiments illustrates how and how well the MDPP and RLS algorithms work, but also show how to write and debug MATLAB codes for S-function programs. The robustness of the adaptive control system is intensively discussed subsequently. By using an estimator resistant to the noise contamination, the adaptive control system can not be destablized by the introduced noise at the input of the plant or the estimator. However, the adaptive control system is subject to the unmodeled dynamics that have a magnitude response like an impulse at the crossover frequency of the system. Simulation results also show that a classic feedback controller has a better performance, compared with the adaptive controller.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Wu, Zhongshan, "Simulation study and instability of adaptive control" (2001). LSU Master's Theses. 3603.