Unsupervised clustering for fault diagnosis

Author

Baraldi, Piero ; Maio, Francesco Di ; Zio, Enrico

Author_Institution

Dept. of Energy, Politec. di Milano, Milano, Italy

fYear

2012

fDate

23-25 May 2012

Firstpage

1

Lastpage

9

Abstract

We develop an unsupervised clustering method for the classification of transient data. A fuzzy-based technique is employed to measure the similarity among the transients; a spectral clustering technique, embedding the unsupervised Fuzzy C-Means (FMC) algorithm, is applied to the matrix of similarity values so that the clusters are formed by patterns most similar to each other. The performance of the proposed technique is tested with respect to a case study with data artificially generated.

Keywords

fault diagnosis; feature extraction; fuzzy set theory; matrix algebra; pattern classification; signal classification; spectral analysis; fault diagnosis; feature extraction; fuzzy-based technique; pattern clustering; plant operation; similarity value matrix; spectral clustering technique; transient data classification; transient similarity measure; unsupervised clustering method; unsupervised fuzzy C-means algorithm; Degradation; fault diagnosis; fuzzy c-means; fuzzy similarity; spectral analysis; transient data; unsupervised clustering;

fLanguage

English

Publisher

ieee

Conference_Titel

Prognostics and System Health Management (PHM), 2012 IEEE Conference on

Conference_Location

Beijing

ISSN

2166-563X

Print_ISBN

978-1-4577-1909-7

Electronic_ISBN

2166-563X

Type

conf

DOI

10.1109/PHM.2012.6228844

Filename

6228844