Theoretical work in robust control has lead to the development of comprehensive tools for the analysis and synthesis of feedback controllers for uncertain systems. In this context, the uncertainty is modeled by perturbations about which only the maximum size is known. This framework leads to models which describe sets of behaviour.
The weak link is the application of this theory is the modeling of physical systems within this framework; how does one select a nominal model and perturbations of the appropriate structure and size? This talk will approach this problem from a complementary viewpoint: given a perturbation model, to what extent does it describe the observed behaviours of a system. We extend the modeling and validation methods to a class of nonlinear uncertain systems in feedback configurations, and look at an example which exhibits hysteresis.
The talk will be aimed at a general technical audience and will discuss the modeling and analysis framework and how it might be applied to general engineering systems. To illustrate some of the ideas we will use the approach to study a model of combustion instability in jet engines.
Brief bio of speaker: Roy Smith received the BE and ME (Hons) degrees from the University of Canterbury in 1980 and 1981 respectively. He has worked for a variety of government laboratories, industrial companies and consultancies in New Zealand and California on: automotive engine control systems, instrumentation and automation of mass spectrometers and linear accelerators; and industrial boiler systems.
After completing a PhD at Caltech in 1989, he spent a year teaching at Caltech and doing research at JPL on the design of vibration supression systems in large space telescopes. In 1990 he joined the Electrical and Computer Engineering Dept. at the University of California, Santa Barbara, where he now holds the rank of Professor.
Roy Smith's research interests include the identification and control of uncertain (and possibly nonlinear) systems; process control problems, including rapid thermal processing and semiconductor deposition; automotive control systems; flexible space structures; aeromanoeuvring spacecraft; formation flying of space vehicles, and magnetically levitated bearings.
Back to Control Seminars Page