Linear-wavelet models for system identification

R. K. H. Galvão; V. M. Becerra

Linear-wavelet models for system identification

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

A model structure comprising a wavelet network and a linear term is proposed for nonlinear system identification. It is shown that under certain conditions wavelets are orthogonal to linear functions and, as a result, the two parts of the model can be identified separately. The linear-wavelet model is compared to a standard wavelet network using data from a simulated fermentation process. The results show that the linear-wavelet model yields a smaller modelling error when compared to a wavelet network using the same number of regressors.

Original language	English
Title of host publication	15th Triennial World Congress of the International Federation of Automatic Control (IFAC)
Publisher	International Federation of Automatic Control (IFAC)
Publication status	Published - 2002

Keywords

neural-network models, system identification, nonlinear models, function approximation, non-parametric identification, fermentation processes

Access to Document

http://www.nt.ntnu.no/users/skoge/prost/proceedings/ifac2002/data/content/01280/abstract.htm

Cite this

@inproceedings{c2789ca83b0c4536849dfa9f67f49f8d,

title = "Linear-wavelet models for system identification",

abstract = "A model structure comprising a wavelet network and a linear term is proposed for nonlinear system identification. It is shown that under certain conditions wavelets are orthogonal to linear functions and, as a result, the two parts of the model can be identified separately. The linear-wavelet model is compared to a standard wavelet network using data from a simulated fermentation process. The results show that the linear-wavelet model yields a smaller modelling error when compared to a wavelet network using the same number of regressors.",

keywords = "neural-network models, system identification, nonlinear models, function approximation, non-parametric identification, fermentation processes",

author = "Galv{\~a}o, \{R. K. H.\} and Becerra, \{V. M.\}",

year = "2002",

language = "English",

booktitle = "15th Triennial World Congress of the International Federation of Automatic Control (IFAC)",

publisher = "International Federation of Automatic Control (IFAC)",

}

TY - GEN

T1 - Linear-wavelet models for system identification

AU - Galvão, R. K. H.

AU - Becerra, V. M.

PY - 2002

Y1 - 2002

N2 - A model structure comprising a wavelet network and a linear term is proposed for nonlinear system identification. It is shown that under certain conditions wavelets are orthogonal to linear functions and, as a result, the two parts of the model can be identified separately. The linear-wavelet model is compared to a standard wavelet network using data from a simulated fermentation process. The results show that the linear-wavelet model yields a smaller modelling error when compared to a wavelet network using the same number of regressors.

AB - A model structure comprising a wavelet network and a linear term is proposed for nonlinear system identification. It is shown that under certain conditions wavelets are orthogonal to linear functions and, as a result, the two parts of the model can be identified separately. The linear-wavelet model is compared to a standard wavelet network using data from a simulated fermentation process. The results show that the linear-wavelet model yields a smaller modelling error when compared to a wavelet network using the same number of regressors.

KW - neural-network models, system identification, nonlinear models, function approximation, non-parametric identification, fermentation processes

M3 - Conference contribution

BT - 15th Triennial World Congress of the International Federation of Automatic Control (IFAC)

PB - International Federation of Automatic Control (IFAC)

ER -

Linear-wavelet models for system identification

Abstract

Keywords

Access to Document

Fingerprint

Cite this