Date of Award
Summer 2004
Document Type
Thesis - Restricted
Degree Name
Master of Science (MS)
Department
Electrical and Computer Engineering
First Advisor
Yaz, Edwin E.
Second Advisor
Johnson, Michael T.
Third Advisor
Josse, Fabien
Abstract
Research in the field of industrial control centers around methods that quickly and effectively identify and eliminate external disturbances that drive processes out of control. Much work has been done in recent years to merge the concepts of Statistical Process Control (SPC) and Automatic Process Control (APC) to develop Statistical Automatic Process Control (SAPC). There are limitations to SPC, APC, and SAPC because the coefficients of the disturbance model need to be known in order to be effective. The advantage of Disturbance Accommodation Control (DAC), a method first introduced by C.D. Johnson (1967), is that DAC is an active control scheme that I can greatly reduce or eliminate the effects of an unmeasurable deterministic external disturbance. Unlike SAPC, one does not need to know the associated coefficients in the disturbance model. This thesis has five major contributions: l) Determining that the external input must have a finite bound to guarantee an output with finite bound. 2) Introducing an N-step receding horizon extension to the Euclidean-Norn solution. 3) Presenting a procedure that calculates the steady-state output for a system based on the location and magnitude of parameter perturbations. 4) Determining limitations of the N -step solution based on the system parameters. 5) Concluding that using an extended Luenberger observer to yield extra error correction terms does not provide good results.
Recommended Citation
Black, Barry D., "Robustness Study and Further Extensions of Disturbance Accommodation Control" (2004). Master's Theses (1922-2009) Access restricted to Marquette Campus. 4033.
https://epublications.marquette.edu/theses/4033