Multiscale fuzzy system identification

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

One of the biggest challenges in constructing empirical models is the presence of measurement errors in the data. These errors (or noise) can have a drastic effect on the accuracy and prediction of estimated models, and thus need to be removed for improved models accuracy. Multiscale representation of data has shown great noise-removal ability when used in data filtering. In this paper, this advantage of multiscale representation is exploited to improve the accuracy of the nonlinear Takagi-Sugeno (TS) fuzzy models by developing a multiscale fuzzy (MSF) system identification algorithm. The developed algorithm relies on constructing multiple TS fuzzy models at multiple scales using the scaled signal approximations of the input-output data, and then selecting the optimum multiscale model that maximizes the signal-to-noise ratio of the model prediction. The developed algorithm is shown to outperform the time domain fuzzy model, NARMAX model, and fuzzy model estimated from pre-filtered data using an Exponentially weighted Moving Average (EWMA) filter through a simulated shell and tube heat exchanger modeling example. The reason for this improvement is that the developed MSF modeling algorithm improves the model accuracy by integrating modeling and data filtering using a filter bank, from which the optimum filter (for modeling purposes) is selected.

Original languageEnglish
Pages (from-to)763-770
Number of pages8
JournalJournal of Process Control
Volume15
Issue number7
DOIs
Publication statusPublished - Oct 2005
Externally publishedYes

Fingerprint

Fuzzy Identification
Fuzzy systems
System Identification
Fuzzy Systems
Identification (control systems)
Takagi-Sugeno Fuzzy Model
Fuzzy Model
Filtering
Modeling
Filter
Exponentially Weighted Moving Average
Representation of data
Noise Removal
Multiscale Modeling
Multiscale Model
Fuzzy Modeling
Filter Banks
Domain Model
Heat Exchanger
Empirical Model

Keywords

  • Multiscale models
  • System identification
  • Takagi-Sugeno fuzzy models

ASJC Scopus subject areas

  • Process Chemistry and Technology
  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Cite this

Multiscale fuzzy system identification. / Nounou, Mohamed; Nounou, Hazem.

In: Journal of Process Control, Vol. 15, No. 7, 10.2005, p. 763-770.

Research output: Contribution to journalArticle

@article{2c883ced73c0480193c396f54bfb3c89,
title = "Multiscale fuzzy system identification",
abstract = "One of the biggest challenges in constructing empirical models is the presence of measurement errors in the data. These errors (or noise) can have a drastic effect on the accuracy and prediction of estimated models, and thus need to be removed for improved models accuracy. Multiscale representation of data has shown great noise-removal ability when used in data filtering. In this paper, this advantage of multiscale representation is exploited to improve the accuracy of the nonlinear Takagi-Sugeno (TS) fuzzy models by developing a multiscale fuzzy (MSF) system identification algorithm. The developed algorithm relies on constructing multiple TS fuzzy models at multiple scales using the scaled signal approximations of the input-output data, and then selecting the optimum multiscale model that maximizes the signal-to-noise ratio of the model prediction. The developed algorithm is shown to outperform the time domain fuzzy model, NARMAX model, and fuzzy model estimated from pre-filtered data using an Exponentially weighted Moving Average (EWMA) filter through a simulated shell and tube heat exchanger modeling example. The reason for this improvement is that the developed MSF modeling algorithm improves the model accuracy by integrating modeling and data filtering using a filter bank, from which the optimum filter (for modeling purposes) is selected.",
keywords = "Multiscale models, System identification, Takagi-Sugeno fuzzy models",
author = "Mohamed Nounou and Hazem Nounou",
year = "2005",
month = "10",
doi = "10.1016/j.jprocont.2005.03.005",
language = "English",
volume = "15",
pages = "763--770",
journal = "Journal of Process Control",
issn = "0959-1524",
publisher = "Elsevier Limited",
number = "7",

}

TY - JOUR

T1 - Multiscale fuzzy system identification

AU - Nounou, Mohamed

AU - Nounou, Hazem

PY - 2005/10

Y1 - 2005/10

N2 - One of the biggest challenges in constructing empirical models is the presence of measurement errors in the data. These errors (or noise) can have a drastic effect on the accuracy and prediction of estimated models, and thus need to be removed for improved models accuracy. Multiscale representation of data has shown great noise-removal ability when used in data filtering. In this paper, this advantage of multiscale representation is exploited to improve the accuracy of the nonlinear Takagi-Sugeno (TS) fuzzy models by developing a multiscale fuzzy (MSF) system identification algorithm. The developed algorithm relies on constructing multiple TS fuzzy models at multiple scales using the scaled signal approximations of the input-output data, and then selecting the optimum multiscale model that maximizes the signal-to-noise ratio of the model prediction. The developed algorithm is shown to outperform the time domain fuzzy model, NARMAX model, and fuzzy model estimated from pre-filtered data using an Exponentially weighted Moving Average (EWMA) filter through a simulated shell and tube heat exchanger modeling example. The reason for this improvement is that the developed MSF modeling algorithm improves the model accuracy by integrating modeling and data filtering using a filter bank, from which the optimum filter (for modeling purposes) is selected.

AB - One of the biggest challenges in constructing empirical models is the presence of measurement errors in the data. These errors (or noise) can have a drastic effect on the accuracy and prediction of estimated models, and thus need to be removed for improved models accuracy. Multiscale representation of data has shown great noise-removal ability when used in data filtering. In this paper, this advantage of multiscale representation is exploited to improve the accuracy of the nonlinear Takagi-Sugeno (TS) fuzzy models by developing a multiscale fuzzy (MSF) system identification algorithm. The developed algorithm relies on constructing multiple TS fuzzy models at multiple scales using the scaled signal approximations of the input-output data, and then selecting the optimum multiscale model that maximizes the signal-to-noise ratio of the model prediction. The developed algorithm is shown to outperform the time domain fuzzy model, NARMAX model, and fuzzy model estimated from pre-filtered data using an Exponentially weighted Moving Average (EWMA) filter through a simulated shell and tube heat exchanger modeling example. The reason for this improvement is that the developed MSF modeling algorithm improves the model accuracy by integrating modeling and data filtering using a filter bank, from which the optimum filter (for modeling purposes) is selected.

KW - Multiscale models

KW - System identification

KW - Takagi-Sugeno fuzzy models

UR - http://www.scopus.com/inward/record.url?scp=21844465654&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21844465654&partnerID=8YFLogxK

U2 - 10.1016/j.jprocont.2005.03.005

DO - 10.1016/j.jprocont.2005.03.005

M3 - Article

VL - 15

SP - 763

EP - 770

JO - Journal of Process Control

JF - Journal of Process Control

SN - 0959-1524

IS - 7

ER -