Fault characteristics analysis for rotor systems with misalignment based on Wavelet Packet Decomposition and frequency-band energy ratio analysis

In this paper, combining with model experiment, fault characteristics of single-span and dual-disc rotor system with bearing misalignment are investigated by the combination of Wavelet Packet Decomposition (WDP) and frequency-band energy ratio analysis method. Firstly, resampling is applied to ensure the accurate and efficient decomposition of signal frequency-band. And then with the mother wavelet function ´db44´, the displacement signals were decomposed into 4 levels. Moreover, through a comparison with the time-domain waveforms and amplitude spectrums of decomposed signals in different decomposing levels, the effects of different misalignments to system vibration are shown. And corresponding frequency-band energy ratio values can have an excellent corresponding relationship with the misalignment degree and actual speed which can be applied to fault feature extraction to implement effective diagnosis of misalignment fault.