Research On Fault Diagnosis Of Rolling Element Bearings Base On Full Vector MOMEDA

As an important significant part of rotating machinery,which the status of rolling element bearings is closely related to the operating efficiency of machinery.Therefore,it has important meaning to carry out the research on the fault diagnosis of rolling element bearings.Usually,the working environment of rolling bearings is complex and changeable.When a fault occurs,due to the influence of noise and other component vibration interference,the fault characteristics are often submerged.Meanwhile,the characteristics of the early faults are very weak,it is difficult to effectively extract fault features when using traditional signal processing methods.In this paper,the method of combining Full Vector(FV)spectrum and deconvolution technology is used to extract the fault features of rolling bearings,and at the same time,the Variable Predictive Model Based Class Discriminate(VPMCD)is used to realize intelligent fault classification of rolling bearings.The contents of this research are as follows:(1)Introduce in detail the basic principles and numerical calculation process of Full Vector Spectrum theory and Multipoint Optimal Minimum Entropy Deconvolution Adjusted(MOMEDA).Using MOMEDA for noise reduction filtering,extracting periodic pulse components in the signal,and giving the FV-MOMEDA rolling bearing fault feature extraction process,and through simulation analysis and experimental verification,it is proved that compared with the traditional single-channel signal processing method,the effect of FV-MOMEDA of extracting the fault features of the rolling bearing is more obvious.(2)Since MOMEDA is greatly affected by the length of its parameter filter in the filtering process,Particle Swarm Optimization(PSO)is introduced to optimize and improve the MOMEDA method.Meanwhile,in view of the non-stationary and non-linear characteristics of the vibration signal,the Local Mean Decomposition(LMD)method is used for preprocessing,and a rolling bearing fault feature extraction method based on LMD-MOMEDA is proposed,which can effectively extract the fault periodic pulse components and improve signal-noise ratio.On this basis,the method is combined with the FV spectrum and LMD-MOMEDA rolling element bearings fault diagnosis method is proposed,and the specific process of the algorithm is introduced in detail.Through simulation analysis and experimental verification,the method is effective of compound faults for the inner and outer rings of the rolling bearing,and have better diagnostic effect and accuracy.(3)Focused on the problem which the rolling element bearings fault classification.Variable Predictive Model Based Class Discriminate(VPMCD)is applied to the method.Firstly,extract the main vibration vector characteristics of the rolling bearing signal under different conditions.Secondly,form a feature vector group and finally input into the VPM prediction model for training and learning,and then verify through test samples.Finally,the intelligent fault classification of rolling bearings is realized.