Research On Fault Feature Extraction And Fault Diagnosis Of Rolling Bearing Based On Multi-stable Stochastic Resonance

As a commonly used rotating machinery component,rolling bearing is often accompanied by a lot of noise interference signals during its failure vibration signal acquisition.Aiming at the difficult problem of extracting the characteristic of weak fault signal of rolling bearing under strong background noise,this paper mainly proposed two kinds of rolling bearing weak fault feature extraction and diagnosis method which combines the theory of Multi-stable stochastic resonance(MSSR),the variational mode decomposition(VMD)and analytical modal decomposition(AMD)method.The research shows that the method proposed in this paper is conducive to the extraction and diagnosis of rolling bearing's weak fault characteristics under noise background.By considering the deep groove ball bearing as the research object,and using Hertzian contact theory and elastic mechanics and geometry,a nonlinear dynamic model is established herein for the study of vibration response of deep groove ball bearings having single defect on surfaces of inner and outer races.The outer race and inner race defect size parameters are introduced into this nonlinear dynamic model,and dynamic models of localized fault on outer race,inner race of rolling element bearing are simulated and analyzed by using Runge-Kutta method.Meanwhile,the simulation results have been compared with the fault bearing case.The MSSR induced by Dichotomous noise and Lévy noise have been investigated respectively.The mean of signal-to-noise ratio gain(SNR-GM)of the MSSR induced by Dichotomous noise and Lévy noise are plotted and summarized,and the theory is applied in the rolling element experiment.The experiments prove that it can realize the extraction of weak fault signal.Finally,the comparison of multi-stable SR simulation and experiment of rolling bearings induced by Gaussian noise,Dichotomous noise and Lévy noise has been made.To realize the feature extraction of rotating machinery in the strong noise environment,a feature extraction method of weak fault signal based on VMD and re-scaling MSSR is proposed.The first application of parameter optimization of VMD algorithm for fault signal is decomposed into several intrinsic mode functions(IMFs),andthen through the kurtosis criterion and find the maximum kurtosis of IMF component,finally the characteristic frequency of the IMF component through the re-scaling MSSR system will be enhanced,which is easily and clearly detected.Aimed at solving detection problems in rolling bear multi-frequency signals in noisy backgrounds,a novel method is proposed,based on re-scaling the frequency-shifted MSSR with AMD-VMD.In this method,different signal frequency bands are processed by re-scaling sub-sampling compression,to make each frequency band meet the conditions of stochastic resonance.Before the enhanced signal components are synthesized,they are processed to achieve the enhanced signal by means of AMD,leaving only the enhanced sections of the signal.The processed signal is decomposed into intrinsic mode functions by VMD,achieving the detection and extraction of characteristic frequency of multi-frequency signals.