Research On Incipient Fault Diagnosis Of Rolling Bearing Based On MCKD And Higher Order Spectrum

As an important transmission component of rotating machinery,it is important to diagnose the early faults of rolling bearings.When early faults occur in rolling bearings,their fault impact is small and noise is large.Traditional signal processing methods cannot effectively extract and identify fault features.In response to the above problems,this paper proposes an early fault diagnosis method under strong noise background.The main research contents are as follows:(1)Aiming at the small impact and high noise of bearing early fault signals,a method of rolling bearing early fault diagnosis based on MCKD and HOS is proposed.Firstly,GA is used to adaptively optimize the parameters of MCKD.After optimization,MCKD is used to enhance the impact components in the fault signal;Then,the shock coupling characteristics of the signal are extracted using HOS.Simulation and fault cases show that this method can effectively extract the impact characteristics of early fault of rolling bearings.(2)Aiming at the difference between the clustering features of MCKD and high order spectral processed signals on two-dimensional contour maps,a feature extraction method based on HOS and Tamura texture is proposed,and combined with support vector machines to achieve fault diagnosis of rolling bearings.Using MCKD and high order spectrum to process early fault signals,a two-dimensional contour map of the high order spectrum is obtained;Secondly,Tamura texture method is used to extract texture feature parameters from two-dimensional contour maps and construct fault feature vectors;Finally,using feature vectors as input to support vector machines,a bearing fault identification model based on support vector machines is established.The experimental results show that this method can effectively identify rolling bearing fault types with the same bearing fault size and different size and rotational speed with fewer characteristic parameters.