A fuzzy clustering method for multidimensional parameter selection in system with uncertain parameters

Author

Kamei, Katsuari ; Auslander, David M. ; Inoue, Kazuo

Author_Institution

Dept. of Comput. Sci. & Syst. Eng., Ritsumeikan Univ., Kyoto, Japan

fYear

1992

fDate

8-12 Mar 1992

Firstpage

355

Lastpage

362

Abstract

The authors present a new multivariate analysis method for multidimensional parameter selection in such problems as controller design for nonlinear systems or other systems for which analytical solutions are not available. The method is based on the fuzzy clustering technique. This new multivariate analysis method is based on the most popular fuzzy clustering algorithm, the fuzzy c-means algorithm (FCM). To apply the FCM method to multivariate binary analysis an effective distance replaces the distance as originally defined in FCM, and a distance weight is defined. Together, they allow the finding of pass regions in the parameter space, and the estimation of the number of pass regions

Keywords

fuzzy set theory; pattern recognition; statistical analysis; binary analysis; controller design; fuzzy c-means algorithm; fuzzy clustering method; fuzzy set theory; multidimensional parameter selection; multivariate analysis; nonlinear systems; pass regions; pattern recognition; statistical analysis; uncertain parameters; Algorithm design and analysis; Clustering algorithms; Clustering methods; Control systems; Fuzzy systems; Multidimensional systems; Nonlinear control systems; Nonlinear systems; Performance analysis; System performance;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems, 1992., IEEE International Conference on

Conference_Location

San Diego, CA

Print_ISBN

0-7803-0236-2

Type

conf

DOI

10.1109/FUZZY.1992.258641

Filename

258641