Identifier

etd-06112009-075955

Degree

Doctor of Philosophy (PhD)

Department

Mechanical Engineering

Document Type

Dissertation

Abstract

In this thesis we start with an introduction to the theory of vibration control. We broadly classify the control methods into passive and active schemes. We introduce the problem of state feedback control and provide the classical solution in the form of Ackermann formula. We then identify the limitations of the classical approach and present the more elegant solution of partial pole assignment without spillover. We highlight the problem with model uncertainties and describe the method of pole assignment using data from measured receptances. This approach is extended for pole assignment for a linear vibrating system by using state feedback control delayed in time. This approach is significantly advantageous over various conventional state-space approaches which need to use information of , and matrices. Since the method relies solely on measured receptances, it negates the need to know , and matrices. It is shown that for a system with degrees of freedom, we may assign eigenvalues. Assigning eigenvalues in a time delayed system does not necessarily regulate the dynamics of the system or guarantee its stability. We separate the eigenvalues into two groups, primary and secondary, and propose method of a posteriori analysis to ensure that the primary eigenvalues have been assigned. The method is demonstrated by various examples. For state feedback control, the control is achieved by measuring the states of the system and feeding them back into the system after multiplying them with appropriate control gain. This makes it imperative to measure all the states of the system. In practical control applications, all states are not accessible for measurement. We address the problem of inaccessibility of states making it difficult to implement the state feedback control. We introduce the theory of linear state estimation also called observer design. We identify the limitations of this approach and introduce the concept of state reconstruction by delayed action. We develop a method to reconstruct the inaccessible states by introducing delay in the system and using information from accessible states. The results are demonstrated by examples.

Date

2009

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Yitshak M. Ram

Share

COinS