DocumentCode :
2534278
Title :
Adaptive relay based identification and control of Hammerstein-type nonlinear processes
Author :
Mehta, Utkal ; Majhi, Somanath
Author_Institution :
Dept. of Electron. & Commun. Eng., Indian Inst. of Technol., Guwahati
Volume :
2
fYear :
2008
fDate :
11-13 Dec. 2008
Firstpage :
355
Lastpage :
360
Abstract :
This paper presents a very simple and effective identification and control method for Hammerstein type nonlinear processes. Generally, a nonlinear process possessing static nonlinearity experiences an asymmetrical limit cycle output. During the relay feedback test, the relay height is adjusted such that the output becomes symmetrical. The static nonlinearity is identified from a single test and then its inverse function is used to nullify static nonlinearity in the process. Thereafter, a linear process model is identified and subsequently a linear PID controller is designed. Finally, the PID controller along with the inverse nonlinear function yields a simple nonlinear PID controller that is capable of controlling the nonlinear process effectively. The usefulness of proposed identification and control strategy is illustrated by two simulation examples.
Keywords :
adaptive control; control nonlinearities; feedback; limit cycles; linear systems; nonlinear control systems; three-term control; Hammerstein-type nonlinear processes; adaptive relay based identification; asymmetrical limit cycle output; linear PID controller; linear process model; relay feedback test; static nonlinearity; Adaptive control; Communication system control; Limit-cycles; Nonlinear dynamical systems; Output feedback; Process control; Programmable control; Relays; Testing; Three-term control; Control; Hammerstein Nonlinear Process; Identification; Relay Feedback;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
India Conference, 2008. INDICON 2008. Annual IEEE
Conference_Location :
Kanpur
Print_ISBN :
978-1-4244-3825-9
Electronic_ISBN :
978-1-4244-2747-5
Type :
conf
DOI :
10.1109/INDCON.2008.4768749
Filename :
4768749
Link To Document :
بازگشت