An approach to fault diagnosis for non-linear system based on fuzzy cluster analysis

Author

Liu, Yiping ; Shen, Yi ; Liu, Zhiyan

Author_Institution

Dept. of Control Eng., Harbin Inst. of Technol., China

Volume

3

fYear

2000

fDate

2000

Firstpage

1469

Abstract

An approach to fault diagnosis based on fuzzy clustering is proposed. First, the fuzzy model representing each state of the system is built by extracting fuzzy rules from the sample data using fuzzy clustering algorithm. Then, the modified fuzzy models for fault diagnosis are obtained based on the original fuzzy models and constitute a whole rule-base. Furthermore, a strategy for fault diagnosis based on fuzzy clustering is presented to detect and locate faults in the system. Finally, some experimental results are shown to illustrate the effectiveness of the proposed approach

Keywords

fault diagnosis; fuzzy set theory; identification; knowledge based systems; nonlinear systems; pattern clustering; cost function; fault diagnosis; fault location; fault patterns; fuzzy cluster analysis; fuzzy clustering algorithm; fuzzy model; fuzzy rules; iterative optimisation; nonlinear system; simulation; whole rule-base; Clustering algorithms; Control engineering; Control systems; Diagnostic expert systems; Fault detection; Fault diagnosis; Fuzzy control; Fuzzy systems; Iterative algorithms; Telephony;

fLanguage

English

Publisher

ieee

Conference_Titel

Instrumentation and Measurement Technology Conference, 2000. IMTC 2000. Proceedings of the 17th IEEE

Conference_Location

Baltimore, MD

ISSN

1091-5281

Print_ISBN

0-7803-5890-2

Type

conf

DOI

10.1109/IMTC.2000.848718

Filename

848718