Gear fault diagnosis based on SVM

Author

Ma, Shang-jun ; Liu, Geng ; Xu, Yongqiang

Author_Institution

Sch. of Mechatronical Eng., Northwestern Polytech. Univ., Xi´´an, China

fYear

2010

fDate

11-14 July 2010

Firstpage

140

Lastpage

143

Abstract

Elements of Support Vector Machine was applied to the fault diagnosis of gear system, and the two-class algorithms for 3 individual fault modes, which are No Fault Gear Mode, Crack of Dedendum Mode and Tooth Surface Abrasion Mode respectively, are well developed and set up. Through the training and testing simulation data samples and the signal samples from gear oscillation, these 3 different types of gear fault modes are finally identified and distinguished from each other at the rotating speed of 300r/min and 900r/min. The result validates that the Support Vector Machine is with excellent diagnostic ability in the fault diagnosis system of gear and with favorable prospect in this filed of application.

Keywords

abrasion; cracks; fault diagnosis; gears; maintenance engineering; mechanical engineering computing; support vector machines; SVM; crack; dedendum mode; gear fault diagnosis; gear fault modes; support vector machine; tooth surface abrasion mode; Classification algorithms; Fault diagnosis; Gears; Kernel; Support vector machines; Testing; Training; Fault diagnosis; Feature extraction; Gear system; Support Vector Machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Wavelet Analysis and Pattern Recognition (ICWAPR), 2010 International Conference on

Conference_Location

Qingdao

Print_ISBN

978-1-4244-6530-9

Type

conf

DOI

10.1109/ICWAPR.2010.5576299

Filename

5576299