Velocity Synchrosqueezing Windowed Fourier Transform for Fault Diagnosis of Fixed-Shaft Gearbox Under Nonstationary Conditions

The Velocity synchrosqueezing transform (VST) is a good method to analyze the vibration signal for condition monitoring of planetary gearbox under non-stationary conditions. The VST is realized by jointly applying domain mapping, synchrosqueezing transform (SST) and time-freqeuncy representation (TFR) restoration. It features a smear-free result for time-frequency analysis. However the VST has lower frequency resolution in the higher frequency region, this drawback limits its effectiveness in analyzing the vibration signal of fixed-shaft gearbox. To resolve this problem, this paper proposes the velocity synchrosqueezing windowed Fourier transform (VSWFT) method. Compared with the VST, this method employs the synchrosqueezing windowed Fourier transform (SWFT), instead of the SST, to process the angle-domain signal. As the SWFT uses window with fixed window length, it has fixed time-frequency resolution, which is more suitable for analyzing vibration signal of fixed-shaft gearbox. Finally the TFR is restored from the SWFT. The fault, if any, can be diagnosed by identifying the revealed sidebands of meshing frequency in the TFR. The effectiveness of the proposed method is validated using experimental vibration signal collected from a faulty gearbox under a non-stationary condition.

[1]  K. R. Fyfe,et al.  ANALYSIS OF COMPUTED ORDER TRACKING , 1997 .

[2]  Robert B. Randall,et al.  A New Method of Modeling Gear Faults , 1982 .

[3]  Guoyu Meng,et al.  Vibration signal analysis using parameterized time–frequency method for features extraction of varying-speed rotary machinery , 2015 .

[4]  M. Plancherel,et al.  Contribution À ĽÉtude de la reprÉsentation D’une fonction arbitraire par des intÉgrales dÉfinies , 1910 .

[5]  Franz Hlawatsch,et al.  Duality and classification of bilinear time-frequency signal representations , 1991, IEEE Trans. Signal Process..

[6]  Fulei Chu,et al.  Recent advances in time–frequency analysis methods for machinery fault diagnosis: A review with application examples , 2013 .

[7]  Giorgio Dalpiaz,et al.  Effectiveness and Sensitivity of Vibration Processing Techniques for Local Fault Detection in Gears , 2000 .

[8]  Wenyi Wang,et al.  EARLY DETECTION OF GEAR TOOTH CRACKING USING THE RESONANCE DEMODULATION TECHNIQUE , 2001 .

[9]  Ming Liang,et al.  A velocity synchrosqueezing transform for fault diagnosis of planetary gearboxes under nonstationary conditions , 2017 .

[10]  Ming Liang,et al.  Time–frequency analysis based on Vold-Kalman filter and higher order energy separation for fault diagnosis of wind turbine planetary gearbox under nonstationary conditions , 2016 .

[11]  P. D. McFadden,et al.  Detecting Fatigue Cracks in Gears by Amplitude and Phase Demodulation of the Meshing Vibration , 1986 .

[12]  I. Daubechies,et al.  Synchrosqueezed wavelet transforms: An empirical mode decomposition-like tool , 2011 .

[13]  Peter V. E. McClintock,et al.  Linear and synchrosqueezed time-frequency representations revisited: Overview, standards of use, resolution, reconstruction, concentration, and algorithms , 2015, Digit. Signal Process..

[14]  Zhipeng Feng,et al.  Iterative generalized synchrosqueezing transform for fault diagnosis of wind turbine planetary gearbox under nonstationary conditions , 2015 .

[15]  Hau-Tieng Wu,et al.  Synchrosqueezing-Based Recovery of Instantaneous Frequency from Nonuniform Samples , 2010, SIAM J. Math. Anal..

[16]  Gang Yu,et al.  General linear chirplet transform , 2016 .

[17]  L. Cohen,et al.  Time-frequency distributions-a review , 1989, Proc. IEEE.