Title :
Analysis of a novel method of autotuning a multivariable plant based on quantisation
Author :
Goodwin, Graham C. ; Welsh, James S.
Author_Institution :
Newcastle Univ., NSW, Australia
Abstract :
Surveys of industry reveal that an alarmingly high proportion of industrial controllers are operated far from their optimal settings. There is thus strong motivation to develop automated tuning methods. One class of tuning methods that has found industrial acceptance is based on the use of relays. A related idea, that we have previously explored, replaces the relays by quantisers placed at the controller output. This approach retains the core merits of the relay method whilst overcoming some of the disadvantages. The current paper contributes to this approach by analysing noise performance, by comparing relay/quantiser tuning with periodic signal injection and by extending the results to multivariable plants
Keywords :
multivariable control systems; noise; process control; quantisation (signal); relays; stochastic processes; transfer functions; tuning; automated tuning methods; autotuning; multivariable plant; noise performance; periodic signal injection; quantisation; relays; Automatic control; Industrial control; MIMO; Nonlinear control systems; Parametric statistics; Quantization; Relays; Signal analysis; Testing; Transfer functions;
Conference_Titel :
American Control Conference, 1999. Proceedings of the 1999
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4990-3
DOI :
10.1109/ACC.1999.782385