Method Research Of Early Fault Feature Extraction Of Rolling Bearing Based On EEMD And Cyclostationary

Rolling bearing is one of the most important and widely used mechanical parts.It is very necessary to analyze the early fault bearing.Cyclostationary signal is a kind of non-stationary signal with periodic characteristic,and the vibration signal of the bearing is not only periodic,but also has the characteristics of non-stationary.It is very necessary to study the cyclostationary theory,which is helpful to improve the accuracy and efficiency of the diagnosis.The relationship between the number and the kurtosis criterion based on the proposed IMF optimization principle of a cross-correlation,kurtosis and wavelet soft threshold combination.Verified by experiment,found that the IMF optimization principle of this correlation-kurtosis and wavelet soft threshold combination is superior to the single use of cross-correlation coefficient and kurtosis criterion,which laid the foundation for the weak bearing fault feature extraction.On the basis of the above research,a fault diagnosis algorithm based on EEMD and cyclostationary is proposed,which can effectively improve the signal to noise ratio(SNR)and suppress the aliasing.The validity of the method is verified by the test data of bearing failure.Compared with the traditional resonance demodulation method,the superiority of the new method in suppressing noise,aliasing and fault feature extraction is proved.To build the experimental system for fault diagnosis of rolling bearing rolling bearing early,36 sets of different fault types and fault severity of processing,and large scale tests were carried out,finishing the 900 group fault bearing test data.The new algorithm is applied to the early fault diagnosis of bearing,and the validity of the new algorithm is further verified by comparing with the traditional resonance demodulation algorithm.The proposed new algorithm and carry out related experimental research,to further promote the cyclostationary theory in fault diagnosis of rolling bearing application,and technical support to the safety condition of rolling bearing detection.