Bearing faults are always observed as cyclical impulses in the vibration signal. Detecting and extracting the impulse response signal is the primary feature extraction means of bearing faults diagnose. However, the impulse response is mostly immersed in strong noise, which makes it difficult to diagnose the bearing faults. In order to effectively remove this noise and detect the impulses, a hybrid method combining ensemble empirical mode decomposition (EEMD) method and an improved morphological filtering based on kurtosis criterion was proposed in this paper. In this method, a new decision strategy of intrinsic mode function (IMF) and morphological structure element (SE) is based on kurtosis criterion. The signal reconstructed by the selected IMFs is processed by improved morphological filtering based on kurtosis criterion. Meanwhile, it also avoids the selection of center frequency and filter band in resonance demodulation method and has good adaptability. When analyzed with the inner and outer ring faults of rolling bearing, the results show that this method can distinctly and accurately extract the fault information and the noise is well suppressed. So, it can be used to diagnose the bearing faults accurately.
ＷＵＸｉａｏｔａｏ;ＹＡＮＧＭｅｎｇ;ＹＵＡＮＸｉａｏｈｕｉ;ＧＯＮＧＴｉｎｇｋａｉ.
Bearing fault diagnosis using EEMD and improved morphological filtering method based on kurtosis criterion[J]. Journal of Vibration and Shock, 2015, 34(2): 38-44
PDF(2562 KB)
