Self-Tuning Control for Two-Dimensional Processes - UMIST Control Systems Centre Series

In this book, least squares optimal prediction, minimum variance control and generalised mimimum variance control algorithms for two-dimensional CARMA (controlled auto-regressive moving average) processes are developed. The algorithms are suitable for combination with recursive parameter estimation routines to form self-tuning predictors and controllers.;A forgetting strategy that forgets in two dimensions is introduced, and the problems of setpoint tracking and offset handling for the self-tuning controllers are considered. For postgraduate students interested in two-dimensional systems of self-tuning systems, and for researchers in industry involved in image processing, papermaking or production of plastics.;Despite the complexity of two-dimensional systems theory, the book shows how simple algorithms can be achieved via a solution of a specific Diophantine equation. The combination of these control algorithms with proven estimation routines developed for image processing paves the way for application to industrial processes.