DocumentCode :
3731004
Title :
Analysis and comparison of axis orbit characteristics in fault pattern recognition
Author :
Lu Yang; Jian Wang; Guigang Zhang; Zhaoping Ding
Author_Institution :
Institute of Automation, Chinese Academy of Sciences, Beijing, China
fYear :
2015
Firstpage :
859
Lastpage :
863
Abstract :
Axis orbit is very important in fault monitoring and diagnosis of rotary machines. However, characteristics of axis orbit have lots of varieties. To support information fusion of characteristics, this paper elaborates the analysis and comparison of 22 characteristics of axis orbit for the on-line fault pattern recognition. The stability, sensibility and validity of different characteristics are analyzed and compared. The results show that axis orbit faults can hardly be totally discriminated by using any single characteristic Many characteristics are useful to recognize one or several patterns, so that the method using the optimized characteristic combination and the optimized characteristic matching sequence is proposed in this paper that can be used to accurately and rapidly recognize axis orbit fault patterns. This method supports information fusion diagnosis and gives a reference to axis orbit recognition of different kinds of rotary machines.
Keywords :
"Orbits","Character recognition","Fault diagnosis","Standards","Monitoring","Stability analysis"
Publisher :
ieee
Conference_Titel :
Chinese Automation Congress (CAC), 2015
Type :
conf
DOI :
10.1109/CAC.2015.7382618
Filename :
7382618
Link To Document :
بازگشت