Research On Fault Diagnosis Of Rolling Bearing Based On Improved EWT And MCKD

With the improvement of modern industrial level,the condition monitoring of mechanical equipment has become an important link to ensure the normal operation of industrial production.Rolling bearings are widely used in the industrial field and play an important role,and most of the rolling bearings operate in the complex and harsh working environment,its parts are easy to damage,leading to the overall equipment failure,which will bring immeasurable accidents and losses in production and life.In order to reduce the loss caused by the failure of rolling bearing,it is of great practical significance to study the fault diagnosis method of rolling bearing,so as to improve the reliability of rolling bearing.In this paper,the rolling bearing is taken as the research target,and the improved empirical wavelet transform(EWT)and maximum correlation kurtosis deconvolution(MCKD)are combined to apply to the rolling bearing fault diagnosis.The theory and principle of empirical wavelet transform are analyzed in depth.The defects of old EMD and EEMD are analyzed.Because it can not realize the adaptive segmentation of Fourier spectrum accurately,an improved iewt method is proposed and applied to fault diagnosis of rolling bearings.The method takes the Fourier spectrum as the appropriate envelope,regards the envelope as "new Fourier spectrum",then divides the spectrum of "new Fourier spectrum" to get the modal components,and then performs feature extraction and fault analysis.The analysis of rolling bearing fault data shows that the method can adaptively and reasonably divide the fault signal,and accurately analyze the fault characteristic frequency of rolling bearing.Compared with the original EWT method,it shows that the method is more accurate for the fault detection of rolling bearing.Rolling bearing is easy to be damaged in the process of mechanical operation.In the case of weak impact characteristics in the early stage of fault,MCKD method can effectively reduce the noise of fault signal,and its unique kurtosis index can be more effective for feature extraction of periodic impact.After a deep study of MCKD,an improved maximum correlation kurtosis deconvolution method(IMCKD)is proposed to solve the problem that MCKD needs to rely on artificial experience to select parameters.That is to say,particle swarm optimization(PSO)is used to determine the optimal filter length and fault period of MCKD algorithm.The imckd algorithm is compared with the minimum entropy deconvolution(MED)algorithm,and the effectiveness and correctness of imckd are verified by analyzing the rolling bearing fault data.A fault diagnosis method combining IEWT and IMCKD based on margin factor is proposed.IEWT can effectively separate the bearing fault signals hidden in strong noise.The dimensionless parameter margin factor is introduced into the selection of modal components to further select the modal components with the most fault components.IMCKD can well de noise the fault signals and express the fault signals The periodicity of the signal.By analyzing the simulation signals and experimental signals,it is proved that the proposed method has high diagnostic accuracy and robustness in multiple signal decomposition,and has good reliability and practicability in practical engineering applications.It also provides a new idea and method for the fusion algorithm.