Application of artificial intelligence techniques to the study of machine signatures

Author

Chen, W.-Y. ; Xu, J.-X. ; Panda, S.K.

Author_Institution

Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore, Singapore

fYear

2012

fDate

2-5 Sept. 2012

Firstpage

2390

Lastpage

2396

Abstract

This paper presents demonstration on the application of artificial intelligence techniques to the study of machine vibration signatures. First, a Self-Organizing Map (SOM) is used to discover cluster information from frequency-domain vibration signatures for the detection and diagnosis of unbalanced rotor and bearing faults. In the next, with further feature extraction in frequency-domain, a 2-dimensional multi-class Support Vector Machine (SVM) is used to predict these fault modes with an error rate of 1.48% over a wide machine operation speed.

Keywords

artificial intelligence; electric machine analysis computing; fault diagnosis; feature extraction; frequency-domain analysis; rotors; support vector machines; 2D multiclass SVM; SOM; artificial intelligence techniques; bearing fault detection; bearing fault diagnosis; cluster information; fault modes; feature extraction; frequency-domain vibration signatures; machine vibration signatures; self-organizing map; two-dimensional multiclass support vector machine; unbalanced rotor detection; unbalanced rotor diagnosis; Feature extraction; Frequency domain analysis; Rotors; Support vector machines; Vectors; Vibrations; Wiener filters;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Machines (ICEM), 2012 XXth International Conference on

Conference_Location

Marseille

Print_ISBN

978-1-4673-0143-5

Electronic_ISBN

978-1-4673-0141-1

Type

conf

DOI

10.1109/ICElMach.2012.6350218

Filename

6350218