Comparison of Prediction-Error-Modelling Criteria

John Bagterp Jørgensen; Sten Bay Jørgensen

doi:10.1109/ACC.2007.4283020

Comparison of Prediction-Error-Modelling Criteria

John Bagterp Jørgensen
, Sten Bay Jørgensen

Department of Chemical and Biochemical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

1454 Downloads (Orbit)

Abstract

Single and multi-step prediction-error-methods based on the maximum likelihood and least squares criteria are compared. The prediction-error methods studied are based on predictions using the Kalman filter and Kalman predictors for a linear discrete-time stochastic state space model, which is a realization of a continuous-discrete multivariate stochastic transfer function model. The proposed prediction error-methods are demonstrated for a SISO system parameterized by the transfer functions with time delays of a continuous-discrete-time linear stochastic system. The simulations for this case suggest to use the one-step-ahead prediction-error maximum-likelihood (or maximum a posteriori) estimator. It gives consistent estimates of all parameters and the parameter estimates are almost identical to the estimates obtained for long prediction horizons but with consumption of significantly less computational resources. The identification method is suitable for predictive control.

Original language	English
Title of host publication	American Control Conference, 2007. ACC '07
Publisher	IEEE
Publication date	2007
ISBN (Print)	1-4244-0988-8
DOIs	https://doi.org/10.1109/ACC.2007.4283020
Publication status	Published - 2007
Event	American Control Conference 2007 - New York City, United States Duration: 11 Jul 2007 → 13 Jul 2007 http://a2c2.org/conferences/acc2007/

Conference

Conference	American Control Conference 2007
Country/Territory	United States
City	New York City
Period	11/07/2007 → 13/07/2007
Internet address	http://a2c2.org/conferences/acc2007/

Bibliographical note

Copyright: 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/ACC.2007.4283020

J rgensenFinal published version, 210 KB

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{d7adf2646e094bfba896fca4f7e6523b,

title = "Comparison of Prediction-Error-Modelling Criteria",

abstract = "Single and multi-step prediction-error-methods based on the maximum likelihood and least squares criteria are compared. The prediction-error methods studied are based on predictions using the Kalman filter and Kalman predictors for a linear discrete-time stochastic state space model, which is a realization of a continuous-discrete multivariate stochastic transfer function model. The proposed prediction error-methods are demonstrated for a SISO system parameterized by the transfer functions with time delays of a continuous-discrete-time linear stochastic system. The simulations for this case suggest to use the one-step-ahead prediction-error maximum-likelihood (or maximum a posteriori) estimator. It gives consistent estimates of all parameters and the parameter estimates are almost identical to the estimates obtained for long prediction horizons but with consumption of significantly less computational resources. The identification method is suitable for predictive control.",

author = "J{\o}rgensen, \{John Bagterp\} and J{\o}rgensen, \{Sten Bay\}",

note = "Copyright: 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; American Control Conference 2007, ACC 2007 ; Conference date: 11-07-2007 Through 13-07-2007",

year = "2007",

doi = "10.1109/ACC.2007.4283020",

language = "English",

isbn = "1-4244-0988-8",

booktitle = "American Control Conference, 2007. ACC '07",

publisher = "IEEE",

address = "United States",

url = "http://a2c2.org/conferences/acc2007/",

}

TY - GEN

T1 - Comparison of Prediction-Error-Modelling Criteria

AU - Jørgensen, John Bagterp

AU - Jørgensen, Sten Bay

N1 - Copyright: 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

PY - 2007

Y1 - 2007

N2 - Single and multi-step prediction-error-methods based on the maximum likelihood and least squares criteria are compared. The prediction-error methods studied are based on predictions using the Kalman filter and Kalman predictors for a linear discrete-time stochastic state space model, which is a realization of a continuous-discrete multivariate stochastic transfer function model. The proposed prediction error-methods are demonstrated for a SISO system parameterized by the transfer functions with time delays of a continuous-discrete-time linear stochastic system. The simulations for this case suggest to use the one-step-ahead prediction-error maximum-likelihood (or maximum a posteriori) estimator. It gives consistent estimates of all parameters and the parameter estimates are almost identical to the estimates obtained for long prediction horizons but with consumption of significantly less computational resources. The identification method is suitable for predictive control.

AB - Single and multi-step prediction-error-methods based on the maximum likelihood and least squares criteria are compared. The prediction-error methods studied are based on predictions using the Kalman filter and Kalman predictors for a linear discrete-time stochastic state space model, which is a realization of a continuous-discrete multivariate stochastic transfer function model. The proposed prediction error-methods are demonstrated for a SISO system parameterized by the transfer functions with time delays of a continuous-discrete-time linear stochastic system. The simulations for this case suggest to use the one-step-ahead prediction-error maximum-likelihood (or maximum a posteriori) estimator. It gives consistent estimates of all parameters and the parameter estimates are almost identical to the estimates obtained for long prediction horizons but with consumption of significantly less computational resources. The identification method is suitable for predictive control.

U2 - 10.1109/ACC.2007.4283020

DO - 10.1109/ACC.2007.4283020

M3 - Article in proceedings

SN - 1-4244-0988-8

BT - American Control Conference, 2007. ACC '07

PB - IEEE

T2 - American Control Conference 2007

Y2 - 11 July 2007 through 13 July 2007

ER -

Comparison of Prediction-Error-Modelling Criteria

Abstract

Conference

Bibliographical note

Access to Document

OpenUrl availability

Fingerprint

Cite this