Multi-fault detection and failure analysis of wind turbine gearbox using complex wavelet transform

Abstract

Multi-fault detection is a challengeable task for fault feature extraction because the weak faults are always buried in intensive vibration energy, especially in the wind turbine gearbox consisting of numerous gears and bearings under severe operation condition. The vibration signals originated from a real multi-fault wind turbine gearbox with catastrophic failure are analyzed in this paper. Conventional narrow-band filtering and Hilbert transform are used to detect distinct harmonics representing broken teeth faults of gears. The cepstrum method is adopted to distinguish the approached frequency components. However, the fault features representing bearing failure doesn't emerge in the demodulation and cepstrum analysis. Complex wavelet transform can provide a multi-scale enveloping spectrogram (MuSEnS) to decompose and demodulate signals simultaneously. Using this method, the weak bearing fault features buried in intensive energies can be detected readily through analyzing the sclies of the MuSEnS at different scales. The disassembled results of the wind turbine gearbox demonstrate the effectiveness of the applied methods. The failure mechanism of the multiple faults in the wind turbine gearbox is discussed, which reveals that the weak bearing faults can lead to catastrophic failure.