A method for identifying continuous delayed systems using rectangular-pulse testing

Shyh-Hong Hwang, Yu Chang Huang

Research output: Contribution to journalConference article

Abstract

An identification method is presented to deal with practical identification difficulties often encountered in plant testing, such as unsteady initial states and unknown load disturbances. A linear regression equation for step testing is first derived via integrating the differential system equation multiple times. A rectangular-pulse input is then considered as a combination of two step inputs. Based on the pulse response, the so-called two-stage least-squares algorithm is developed to overcome those identification difficulties and yield accurate estimates of the model parameters in a sequential manner. The method provides a convenient way to estimate the delay along with other model parameters. Furthermore, it is robust with respect to noise and model structure mismatch.

Original languageEnglish
Article number412-116
Pages (from-to)79-84
Number of pages6
JournalProceedings of the IASTED International Conference on Modelling, Identification and Control
Volume23
Publication statusPublished - 2004 Dec 1
EventProceedings of the 23rd IASTED International Conference on Modelling, Identification, and Control - Grindelwald, Switzerland
Duration: 2004 Feb 232004 Feb 25

Fingerprint

Testing
Model structures
Linear regression
Two-stage Least Squares
Identification (control systems)
Least Square Algorithm
Differential System
Estimate
Disturbance
Model
Unknown

All Science Journal Classification (ASJC) codes

  • Software
  • Modelling and Simulation
  • Computer Science Applications

Cite this

@article{ac25644e18744980a12bcecaece7bec3,
title = "A method for identifying continuous delayed systems using rectangular-pulse testing",
abstract = "An identification method is presented to deal with practical identification difficulties often encountered in plant testing, such as unsteady initial states and unknown load disturbances. A linear regression equation for step testing is first derived via integrating the differential system equation multiple times. A rectangular-pulse input is then considered as a combination of two step inputs. Based on the pulse response, the so-called two-stage least-squares algorithm is developed to overcome those identification difficulties and yield accurate estimates of the model parameters in a sequential manner. The method provides a convenient way to estimate the delay along with other model parameters. Furthermore, it is robust with respect to noise and model structure mismatch.",
author = "Shyh-Hong Hwang and Huang, {Yu Chang}",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "23",
pages = "79--84",
journal = "Proceedings of the IASTED International Conference on Intelligent Systems and Control",
issn = "1025-8973",
publisher = "ACTA Press",

}

A method for identifying continuous delayed systems using rectangular-pulse testing. / Hwang, Shyh-Hong; Huang, Yu Chang.

In: Proceedings of the IASTED International Conference on Modelling, Identification and Control, Vol. 23, 412-116, 01.12.2004, p. 79-84.

Research output: Contribution to journalConference article

TY - JOUR

T1 - A method for identifying continuous delayed systems using rectangular-pulse testing

AU - Hwang, Shyh-Hong

AU - Huang, Yu Chang

PY - 2004/12/1

Y1 - 2004/12/1

N2 - An identification method is presented to deal with practical identification difficulties often encountered in plant testing, such as unsteady initial states and unknown load disturbances. A linear regression equation for step testing is first derived via integrating the differential system equation multiple times. A rectangular-pulse input is then considered as a combination of two step inputs. Based on the pulse response, the so-called two-stage least-squares algorithm is developed to overcome those identification difficulties and yield accurate estimates of the model parameters in a sequential manner. The method provides a convenient way to estimate the delay along with other model parameters. Furthermore, it is robust with respect to noise and model structure mismatch.

AB - An identification method is presented to deal with practical identification difficulties often encountered in plant testing, such as unsteady initial states and unknown load disturbances. A linear regression equation for step testing is first derived via integrating the differential system equation multiple times. A rectangular-pulse input is then considered as a combination of two step inputs. Based on the pulse response, the so-called two-stage least-squares algorithm is developed to overcome those identification difficulties and yield accurate estimates of the model parameters in a sequential manner. The method provides a convenient way to estimate the delay along with other model parameters. Furthermore, it is robust with respect to noise and model structure mismatch.

UR - http://www.scopus.com/inward/record.url?scp=11844305597&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=11844305597&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:11844305597

VL - 23

SP - 79

EP - 84

JO - Proceedings of the IASTED International Conference on Intelligent Systems and Control

JF - Proceedings of the IASTED International Conference on Intelligent Systems and Control

SN - 1025-8973

M1 - 412-116

ER -