CRACK FAULT DIAGNOSIS OF GEAR BASED ON MORPHOLOGICAL WAVELET DE-NOISING

The gear fault signal is often submerged by a lot of background noise,so it is hard for signal-noise separation and for further fault diagnosis. A novel diagnosis method is proposed to performance the pretreatment of gear fault signal based on morphological wavelet de-noising. This method combines the characteristics of identification from mathematical morphology and the characteristics of multi-resolution analysis from wavelet decomposition. Firstly the gear vibration signal is de-noised by using the morphological wavelet de-noising method. Then the fault features is extracted by calculating the power spectrum of de-noised signal. The principle and steps of method are given and its de-noising effect is evaluated by using some parameters. The simulated signal and actual signal is analyzed. The results show that the proposed method can suppress the noise interference greatly. After de-noising,the SNR is improved and the fault characteristic of signal is highlighted better. The accuracy of fault diagnosis of bearing is improved effectively. And the proposed method has good engineer practicability