An evaluation of robust gains for linear systems
Date Issued
December 1, 1980
Author(s)
Suthiboon, Smoothara
Advisor(s)
J. D. Birdwell
Additional Advisor(s)
J. M. Bailey
J. M. Googe
Abstract
Given a linear system with an actuator structure which takes on a finite number of discrete values, it is possible to determine when a linear feedback regulator exists which stabilizes all system structures. Such a feedback regulator is called robust in this thesis. When such a regulator exists, the available methodology calculates a robust feedback gain which is the solution to a specific optimization problem, and which is unique up to the choice of a probability distribution vector over the structural set.
This thesis describes the initial research on two methods of evaluating the selection of the distribution vector, and, therefore, of the robust gain matrix. The first method calculates the expected cost of using a specific gain, given the system structure. This cost is a function of the distribution vector and can be used to define an optimization problem over that vector. The second method evaluates the stability margins as a function of the distribution and attempts to define a similar class of optimization problems using these margins.
Degree
Master of Science
Major
Electrical Engineering
File(s)![Thumbnail Image]()
Name
Thesis80S983.pdf
Size
2.62 MB
Format
Unknown
Checksum (MD5)
57ba3249164e1fb66fc300c99e0db04c