Event-triggered deadtime estimation from closed-loop data

Author

Isaksson, Alf J. ; Horch, Alexander ; Dumont, Guy A.

Author_Institution

R. Inst. of Technol., Stockholm, Sweden

Volume

4

fYear

2001

fDate

2001

Firstpage

3280

Abstract

Our objective is to estimate process deadtime, based on normal operating data only. The first issue is to examine data to find intervals with sufficient excitation for identification. We conclude that the only useful event to hope for is a significant enough change in setpoint. A method is then proposed for finding the transient parts of data after setpoint changes. A deadtime estimation method based on Laguerre models and a rational approximation for the deadtime is then applied to the selected data. The method has been tested extensively via simulation as well as on real industrial data, and found to give very satisfactory results

Keywords

Monte Carlo methods; closed loop systems; decision theory; discrete time systems; identification; process control; Laguerre models; closed-loop data; event-triggered deadtime estimation; industrial process; normal operating data; process deadtime; rational approximation; setpoint change; sufficient excitation; Additives; Automatic control; Control systems; Milling machines; Monitoring; Signal processing; Steady-state; Testing; User-generated content;

fLanguage

English

Publisher

ieee

Conference_Titel

American Control Conference, 2001. Proceedings of the 2001

Conference_Location

Arlington, VA

ISSN

0743-1619

Print_ISBN

0-7803-6495-3

Type

conf

DOI

10.1109/ACC.2001.946428

Filename

946428