On Rotary Machine´s Multi-Class Fault Recognition Based on SVM

Author

Xiaojun, Gu ; Shixi, Yang ; Suxiang, Qian

Author_Institution

Zhejiang Univ., Hangzhou

fYear

2007

Firstpage

460

Lastpage

463

Abstract

In response to the lack of rotary mechanical diagnostic samples, this paper takes the advantages of support vector machine (SVM) in small sample classification for rotary machine multi-class fault pattern recognition, and introduces three methods based on binary classifications: "one-against-all", "one-against-one", and "directed acyclic graph" SVM (DAGSVM) and then compare their performance. The experiments indicate that the SVM has high adaptability for rotary machine fault diagnosis in the case of small number of samples.

Keywords

directed graphs; fault diagnosis; machinery; mechanical engineering computing; pattern recognition; support vector machines; binary classification; directed acyclic graph SVM; fault pattern recognition; multiclass fault recognition; rotary machine; rotary mechanical diagnostic samples; small sample classification; support vector machine; Educational institutions; Fault diagnosis; Hydrogen; Mechanical engineering; Pattern recognition; Power engineering and energy; Support vector machine classification; Support vector machines; Virtual colonoscopy; Fault Diagnosis; Pattern Recognition; Rotary Machine; Support Vector Machine (SVM);

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference, 2007. CCC 2007. Chinese

Conference_Location

Hunan

Print_ISBN

978-7-81124-055-9

Electronic_ISBN

978-7-900719-22-5

Type

conf

DOI

10.1109/CHICC.2006.4347003

Filename

4347003