Compact TS-fuzzy models through clustering and OLS plus FIS model reduction

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

24 Citations (Scopus)

Abstract

Identification of uncertain and nonlinear systems is an important and challenging problem. Fuzzy models of the Takagi-Sugeno (TS) type may be a good choice to describe such systems, however in many cases these become soon complex. We noticed that generally too less effort is put into variable selection and the creation of suitable local rules. Moreover, in most cases no model reduction is applied, while this may simplify the model by removing redundant information. To handle these issues, we propose a three-step method to obtain compact TS-models that can be effectively used to represent complex systems. First, a new fuzzy clustering method is proposed for identification of compact TS-models. Second, the most relevant consequent variables of the TS-model are selected by an orthogonal least squares method based on the obtain clusters. Third, for selection or relevant antecedent variables, a new method is proposed based on Fisher's interclass separability criterion. The overall approach is demonstrated by means of the MPG (miles per gallon) nonlinear regression benchmark. Results are compared with those obtained by standard linear, neuro-fuzzy and advanced fuzzy clustering-based identification tools.

Original languageEnglish
Title of host publicationIEEE International Conference on Fuzzy Systems
Pages1420-1423
Number of pages4
Volume3
Publication statusPublished - 2001
Event10th IEEE International Conference on Fuzzy Systems - Melbourne, Australia
Duration: Dec 2 2001Dec 5 2001

Other

Other10th IEEE International Conference on Fuzzy Systems
CountryAustralia
CityMelbourne
Period12/2/0112/5/01

Fingerprint

Fuzzy clustering
Identification (control systems)
Uncertain systems
Large scale systems
Nonlinear systems

Keywords

  • Compact TS-fuzzy model
  • Fischer's interclass separability
  • Gath-Geva fuzzy clustering
  • Model reduction
  • OLS

ASJC Scopus subject areas

  • Software
  • Safety, Risk, Reliability and Quality
  • Chemical Health and Safety

Cite this

Abonyi, J., Roubos, J., Oosterom, M., & Szeifert, F. (2001). Compact TS-fuzzy models through clustering and OLS plus FIS model reduction. In IEEE International Conference on Fuzzy Systems (Vol. 3, pp. 1420-1423)

Compact TS-fuzzy models through clustering and OLS plus FIS model reduction. / Abonyi, J.; Roubos, J.; Oosterom, M.; Szeifert, F.

IEEE International Conference on Fuzzy Systems. Vol. 3 2001. p. 1420-1423.

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

Abonyi, J, Roubos, J, Oosterom, M & Szeifert, F 2001, Compact TS-fuzzy models through clustering and OLS plus FIS model reduction. in IEEE International Conference on Fuzzy Systems. vol. 3, pp. 1420-1423, 10th IEEE International Conference on Fuzzy Systems, Melbourne, Australia, 12/2/01.
Abonyi J, Roubos J, Oosterom M, Szeifert F. Compact TS-fuzzy models through clustering and OLS plus FIS model reduction. In IEEE International Conference on Fuzzy Systems. Vol. 3. 2001. p. 1420-1423
Abonyi, J. ; Roubos, J. ; Oosterom, M. ; Szeifert, F. / Compact TS-fuzzy models through clustering and OLS plus FIS model reduction. IEEE International Conference on Fuzzy Systems. Vol. 3 2001. pp. 1420-1423
@inproceedings{260ba812fa034b7a9ec705a617bb10ba,
title = "Compact TS-fuzzy models through clustering and OLS plus FIS model reduction",
abstract = "Identification of uncertain and nonlinear systems is an important and challenging problem. Fuzzy models of the Takagi-Sugeno (TS) type may be a good choice to describe such systems, however in many cases these become soon complex. We noticed that generally too less effort is put into variable selection and the creation of suitable local rules. Moreover, in most cases no model reduction is applied, while this may simplify the model by removing redundant information. To handle these issues, we propose a three-step method to obtain compact TS-models that can be effectively used to represent complex systems. First, a new fuzzy clustering method is proposed for identification of compact TS-models. Second, the most relevant consequent variables of the TS-model are selected by an orthogonal least squares method based on the obtain clusters. Third, for selection or relevant antecedent variables, a new method is proposed based on Fisher's interclass separability criterion. The overall approach is demonstrated by means of the MPG (miles per gallon) nonlinear regression benchmark. Results are compared with those obtained by standard linear, neuro-fuzzy and advanced fuzzy clustering-based identification tools.",
keywords = "Compact TS-fuzzy model, Fischer's interclass separability, Gath-Geva fuzzy clustering, Model reduction, OLS",
author = "J. Abonyi and J. Roubos and M. Oosterom and F. Szeifert",
year = "2001",
language = "English",
volume = "3",
pages = "1420--1423",
booktitle = "IEEE International Conference on Fuzzy Systems",

}

TY - GEN

T1 - Compact TS-fuzzy models through clustering and OLS plus FIS model reduction

AU - Abonyi, J.

AU - Roubos, J.

AU - Oosterom, M.

AU - Szeifert, F.

PY - 2001

Y1 - 2001

N2 - Identification of uncertain and nonlinear systems is an important and challenging problem. Fuzzy models of the Takagi-Sugeno (TS) type may be a good choice to describe such systems, however in many cases these become soon complex. We noticed that generally too less effort is put into variable selection and the creation of suitable local rules. Moreover, in most cases no model reduction is applied, while this may simplify the model by removing redundant information. To handle these issues, we propose a three-step method to obtain compact TS-models that can be effectively used to represent complex systems. First, a new fuzzy clustering method is proposed for identification of compact TS-models. Second, the most relevant consequent variables of the TS-model are selected by an orthogonal least squares method based on the obtain clusters. Third, for selection or relevant antecedent variables, a new method is proposed based on Fisher's interclass separability criterion. The overall approach is demonstrated by means of the MPG (miles per gallon) nonlinear regression benchmark. Results are compared with those obtained by standard linear, neuro-fuzzy and advanced fuzzy clustering-based identification tools.

AB - Identification of uncertain and nonlinear systems is an important and challenging problem. Fuzzy models of the Takagi-Sugeno (TS) type may be a good choice to describe such systems, however in many cases these become soon complex. We noticed that generally too less effort is put into variable selection and the creation of suitable local rules. Moreover, in most cases no model reduction is applied, while this may simplify the model by removing redundant information. To handle these issues, we propose a three-step method to obtain compact TS-models that can be effectively used to represent complex systems. First, a new fuzzy clustering method is proposed for identification of compact TS-models. Second, the most relevant consequent variables of the TS-model are selected by an orthogonal least squares method based on the obtain clusters. Third, for selection or relevant antecedent variables, a new method is proposed based on Fisher's interclass separability criterion. The overall approach is demonstrated by means of the MPG (miles per gallon) nonlinear regression benchmark. Results are compared with those obtained by standard linear, neuro-fuzzy and advanced fuzzy clustering-based identification tools.

KW - Compact TS-fuzzy model

KW - Fischer's interclass separability

KW - Gath-Geva fuzzy clustering

KW - Model reduction

KW - OLS

UR - http://www.scopus.com/inward/record.url?scp=0036349115&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036349115&partnerID=8YFLogxK

M3 - Conference contribution

VL - 3

SP - 1420

EP - 1423

BT - IEEE International Conference on Fuzzy Systems

ER -