Multiscale ARX process modeling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of the popular linear auto-regressive with exogenous variable (ARX) model by developing a multiscale ARX (MSARX) modeling algorithm. The idea is to decompose the input-output data, construct multiple ARX models at multiple scales using the scaled signal approximations of the data, and then using cross validation, select among all MSARX models the one which best describes the process. Also, the MSARX modeling algorithm is shown to improve the parsimony of the estimated models, as ARX models with a fewer number of coefficients are needed at coarser scales. This advantage is attributed to the down-sampling used in multiscale decomposition of data. The main advantage of the MSARX algorithm is that it inherently accounts for the presence of noise in the data by the application of low pass filters used in the decomposition of the input-output data, which in turn improves the model robustness to measurement noise in the data and thus enhances its prediction. These prediction and parsimony advantages of the developed MSARX modeling algorithm are demonstrated using a simulated second order process.

Original languageEnglish
Title of host publicationProceedings of the 45th IEEE Conference on Decision and Control 2006, CDC
Pages823-828
Number of pages6
Publication statusPublished - 1 Dec 2006
Event45th IEEE Conference on Decision and Control 2006, CDC - San Diego, CA, United States
Duration: 13 Dec 200615 Dec 2006

Other

Other45th IEEE Conference on Decision and Control 2006, CDC
CountryUnited States
CitySan Diego, CA
Period13/12/0615/12/06

Fingerprint

Multiscale Modeling
Process Modeling
Parsimony
Decompose
Prediction
Decomposition
Model Robustness
Low pass filters
Representation of data
Multiscale Model
Low-pass Filter
Multiple Scales
Output
Multiple Models
Feature extraction
Cross-validation
Feature Extraction
Wavelets
Sampling
Model

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modelling and Simulation
  • Control and Optimization

Cite this

Nounou, M. (2006). Multiscale ARX process modeling. In Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC (pp. 823-828). [4177670]

Multiscale ARX process modeling. / Nounou, Mohamed.

Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC. 2006. p. 823-828 4177670.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nounou, M 2006, Multiscale ARX process modeling. in Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC., 4177670, pp. 823-828, 45th IEEE Conference on Decision and Control 2006, CDC, San Diego, CA, United States, 13/12/06.
Nounou M. Multiscale ARX process modeling. In Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC. 2006. p. 823-828. 4177670
Nounou, Mohamed. / Multiscale ARX process modeling. Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC. 2006. pp. 823-828
@inproceedings{9883f9cd2b2f41619ec48a70b67a698f,
title = "Multiscale ARX process modeling",
abstract = "Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of the popular linear auto-regressive with exogenous variable (ARX) model by developing a multiscale ARX (MSARX) modeling algorithm. The idea is to decompose the input-output data, construct multiple ARX models at multiple scales using the scaled signal approximations of the data, and then using cross validation, select among all MSARX models the one which best describes the process. Also, the MSARX modeling algorithm is shown to improve the parsimony of the estimated models, as ARX models with a fewer number of coefficients are needed at coarser scales. This advantage is attributed to the down-sampling used in multiscale decomposition of data. The main advantage of the MSARX algorithm is that it inherently accounts for the presence of noise in the data by the application of low pass filters used in the decomposition of the input-output data, which in turn improves the model robustness to measurement noise in the data and thus enhances its prediction. These prediction and parsimony advantages of the developed MSARX modeling algorithm are demonstrated using a simulated second order process.",
author = "Mohamed Nounou",
year = "2006",
month = "12",
day = "1",
language = "English",
isbn = "1424401712",
pages = "823--828",
booktitle = "Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC",

}

TY - GEN

T1 - Multiscale ARX process modeling

AU - Nounou, Mohamed

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of the popular linear auto-regressive with exogenous variable (ARX) model by developing a multiscale ARX (MSARX) modeling algorithm. The idea is to decompose the input-output data, construct multiple ARX models at multiple scales using the scaled signal approximations of the data, and then using cross validation, select among all MSARX models the one which best describes the process. Also, the MSARX modeling algorithm is shown to improve the parsimony of the estimated models, as ARX models with a fewer number of coefficients are needed at coarser scales. This advantage is attributed to the down-sampling used in multiscale decomposition of data. The main advantage of the MSARX algorithm is that it inherently accounts for the presence of noise in the data by the application of low pass filters used in the decomposition of the input-output data, which in turn improves the model robustness to measurement noise in the data and thus enhances its prediction. These prediction and parsimony advantages of the developed MSARX modeling algorithm are demonstrated using a simulated second order process.

AB - Multiscale wavelet-based representation of data has been shown to be a powerful tool in feature extraction from practical process data. In this paper, this characteristic of multiscale representation is utilized to improve the prediction accuracy of the popular linear auto-regressive with exogenous variable (ARX) model by developing a multiscale ARX (MSARX) modeling algorithm. The idea is to decompose the input-output data, construct multiple ARX models at multiple scales using the scaled signal approximations of the data, and then using cross validation, select among all MSARX models the one which best describes the process. Also, the MSARX modeling algorithm is shown to improve the parsimony of the estimated models, as ARX models with a fewer number of coefficients are needed at coarser scales. This advantage is attributed to the down-sampling used in multiscale decomposition of data. The main advantage of the MSARX algorithm is that it inherently accounts for the presence of noise in the data by the application of low pass filters used in the decomposition of the input-output data, which in turn improves the model robustness to measurement noise in the data and thus enhances its prediction. These prediction and parsimony advantages of the developed MSARX modeling algorithm are demonstrated using a simulated second order process.

UR - http://www.scopus.com/inward/record.url?scp=39649102919&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=39649102919&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1424401712

SN - 9781424401710

SP - 823

EP - 828

BT - Proceedings of the 45th IEEE Conference on Decision and Control 2006, CDC

ER -