Research On Fault Diagnosis Of Rolling Bearing Based On Adaptive Local Iterative Filtering

As an indispensable and important part of rotating machinery,the failure of rolling bearings will seriously affect the safe and stable operation of the equipment.Therefore,it is of great significance to deeply carry out the research on the rolling bearing fault diagnosis technology for ensuring the safe and reliable operation of the equipment and avoiding the occurrence of sudden accidents.In this paper,taking the rolling bearing as the research object,the adaptive local iterative filtering(ALIF)method is applied to the rolling bearings fault diagnosis by processing the vibration signals of rolling bearings.The main contents of the study are as follows:The decomposition performance of ALIF is compared with that of empirical mode decomposition(EMD)by simulation signal.The results show that compared with the EMD method,ALIF method has advantages in the aspects of signal decomposition capability,accuracy of transient characteristics,and suppression capability of the mode aliasing.It is further proved that the ALIF method can effectively suppress the mode aliasing effect through analysis the vibration signals of rolling bearing inner ring fault.In order to accurately extract rolling bearing fault feature information,a method of fault feature extraction of rolling bearing based on ALIF and frequency weighted energy operator is proposed.Firstly,the vibration signal of bearing fault is decomposed into a series of single-modal components by ALIF method.Secondly,the first two components are selected according to the Kurtosis-Correlation coefficient criterion for reconstruction.Finally,the reconstructed signal is demodulated and analyzed by using the frequency weighted energy operator,and the obvious frequency components are extracted from the energy spectrum to diagnose the fault type.The proposed method is applied to the rolling bearing fault simulation signal and the outer ring fault vibration signal.Compared with the fault feature extraction method of EMD-FWEO,the results show that the proposed method can accurately extract the characteristic frequency of fault vibration signal and identify the fault type.In order to realize intelligent fault diagnosis of rolling bearing,a fault diagnosis method of rolling bearing based on ALIF,partial mean of composite multiscale dispersion entropy and Gustafson-Kessel(GK)clustering is proposed.Firstly,the vibration signal of bearing is decomposed into several single-modal components byALIF.Secondly,the first three components with largest correlation coefficients are selected as the object of study by using the Cross-Correlation coefficient method,and the partial mean of composite multiscale dispersion entropy is calculated as the characteristic vector.Finally,the obtained characteristic vector is recognized and classified through the GK clustering.The proposed method is applied to the rolling bearing experiment data of different fault types and different damage level,and the classification coefficient and average fuzzy entropy are used to evaluate the classification performance.The results show that the proposed method has better classification performance compared with the fault diagnosis method based on EMD,partial mean of composite multiscale dispersion entropy and GK clustering.