DocumentCode
2950258
Title
Broken Rotor Bar Detection in Induction Motors via Wavelet Ridge
Author
Wei, Yunbing ; Shi, Baoshan ; Cui, Guangzhao ; Yin, Jun
Author_Institution
Coll. of Electr. & Inf. Eng., Zhengzhou Univ. of Light Ind., Zhengzhou, China
Volume
2
fYear
2009
fDate
11-12 April 2009
Firstpage
625
Lastpage
628
Abstract
On-line diagnostics of induction motor faults such as broken rotor bars can be accomplished by analyzing the anomalies of motor stator current. This paper presents a novel approach for the detection of the broken rotor bars in induction motors based on wavelet ridge. The characteristic frequency component (CFC) of the broken rotor bars is very close to power frequency in frequency domain but far less in amplitude in steady state, so it is very difficult to detect the broken rotor bars by Fourier transform. As the CFC of the broken rotor bars approaches power frequency gradually during the motorpsilas startup process, the wavelet ridge-based method is introduced to analyze this transient procedure and the CFC of the broken rotor bars is extracted effectively. The influence of power frequency can be well eliminated, and the detection accuracy is greatly improved. Experimental results also demonstrate this is truly an excellent approach for the detection of the broken rotor bars.
Keywords
fault diagnosis; frequency-domain analysis; induction motors; rotors; starting; wavelet transforms; broken rotor bar detection; fault online diagnostics; frequency domain; induction motor fault; motor startup process; wavelet ridge-based method; Bars; Circuit faults; Electrical fault detection; Frequency domain analysis; Hardware; Induction motors; Rotors; Stators; Transient analysis; Wavelet analysis; broken rotor bar; detection; induction motor; wavelet ridge;
fLanguage
English
Publisher
ieee
Conference_Titel
Measuring Technology and Mechatronics Automation, 2009. ICMTMA '09. International Conference on
Conference_Location
Zhangjiajie, Hunan
Print_ISBN
978-0-7695-3583-8
Type
conf
DOI
10.1109/ICMTMA.2009.510
Filename
5203511
Link To Document