An approach to detect broken rotor bars based on instantaneous frequency of the fault characteristic harmonic during the start-up transient

• Published in: International journal of advanced manufacturing technology Vol. 124; no. 11-12; pp. 4107 - 4119
• Main Authors: Zhu, Hongyu, Jia, Zhuzhi, Song, Xiangjin, Sun, Wenpeng
• Format: Journal Article
• Language: English
• Published: London Springer London 01-02-2023; Springer Nature B.V
• Subjects: CAE) and Design; Computer-Aided Engineering (CAD; Discrete Wavelet Transform; Engineering; Evolution; Fourier analysis; Harmonics; Hilbert transformation; Induction motors; Industrial and Production Engineering; Mechanical Engineering; Media Management; Original Article; Rotors; Signature analysis; Squirrel cage motors; Straight lines; Wavelet transforms; Constant frequency; Rotor; Squirrel cage induction motor; Diagnosis approach
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-022-08829-6

Broken rotor bar diagnosis in squirrel cage induction motors is a matter of increasing concern nowadays. The classical motor current signature analysis (MCSA), based on the Fourier analysis of the steady-state current, does not achieve good results when the motor runs in the start-up transient. In this paper, a novel approach is presented for detecting broken rotor bars (BRBs) in squirrel cage induction motors started at a constant frequency. It is shown that, in the case of motors with BRBs, the theoretical evolution of the instantaneous frequency
f
i
of the characteristic harmonic in squared Park vector modulus signal versus slip
s
is always manifests itself by a straight line with slope
k
equal to 100, which is physically justified using theoretical analysis. In the proposed approach, the discrete wavelet transform (DWT) is used to extract characteristic harmonics from the squared park vector modulus signal. Then, the instantaneous frequency (IF) of the resultant characteristic harmonics is calculated via the Hilbert transform (HT). The occurrence of failure is determined according to the correlation between the symbolized experimental
s
–
f
i
and its theoretical evolution in the
s
–
f
i
plane, and the fault severity is judged according to the energy of the fault characteristic harmonic signal. Experimental results of a 3 kW motor verify the validity of the theoretical analysis and the proposed diagnosis approach during the start-up transient.