Feedback control of non-minimum phase systems with applications to piezo-driven active noise control
Abstract
An active adaptive feedback method capable of achieving feedback active noise control attenuation through the use of a non-minimum phase piezoelectric activated plate actuator was developed and implemented. The piezo/plate actuator was investigated by using vibro-acoustic closed form modeling, system identification and experimental testing. It was found that feedback active noise control performance using the piezo/plate actuators is limited by plant dead time, time varying system characteristics and non-minimum phase zeros. Various other implementation issues were also investigated and resolved in order to successfully implement a FANC system with a piezo/plate actuator. Several adaptive controllers capable of performing under these limitations were analytically investigated, and their respective limitations were presented. Based on the adaptive general minimum variance algorithm, a new feedback active noise controller was developed for the non-minimum phase piezo/plate actuator. Utilizing the waterbed effect and a user defined performance cost function, a systematic method of selecting an optimal set of feedback control design parameters was presented. The waterbed effect states that a controller which attempts to push down the magnitude of the sensitivity function (:S(jω):) in a certain frequency range will necessarily push up the magnitude of the sensitivity function at another frequency range. Also, the ratio of attenuation to amplification is frequency dependent. Experimental results illustrate that the proposed feedback active noise control algorithm provides better attenuation results for the piezo/plate system than the minimum variance algorithm. Lastly, simulations were presented to illustrated that the proposed active adaptive feedback control scheme could be performed in real time if suitable electronic equipment were available.
Degree
Ph.D.
Advisors
Krousgrill, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.