Defect Size Estimation Of Rolling Bearings Based On Double Impulse Characteristic Method

Rolling bearings are the key components of rotating machinery,and their health determines the safe operation of the entire system.The vibration signals collected by the sensors usually exhibit non-stationary characteristics affected by the working environment,and it is difficult to accurately identify weak fault features.Therefore,the in-depth study of signal processing methods for the implementation of condition monitoring and fault diagnosis of rolling bearings can effectively avoid the occurrence of accidents.In this paper,starting from the vibration signal characteristics of fault bearing,the double pulse feature extraction technology is studied when rolling element passes through the defect of raceway.The relationship between the signal waveform and the defect size is established to determine the degree of the defect of the rolling bearing.The specific research contents of the thesis are as follows:(1)A fault diagnosis method for rolling bearing based on AR-MCKD is proposed.Aiming at the problem of order determination of the AR model in fault diagnosis,a method of determining the order of the maximum kurtosis of the residual signal is proposed to suppress the intrinsic components in the signal and strengthen the impact component.In order to make the periodic component of the vibration signal more prominent,the pre-whitened signal is processed by the method of maximum correlation kurtosis deconvolution to achieve effective enhancement of weak faults.The envelope spectrum is analyzed to achieve accurate judgment of fault location.(2)The influence of decomposition modes on the results of variational mode decomposition is studied.A method for ensuring the decomposition accuracy by the correlation between each modal component and the signal is proposed,and the optimal modulus is determined by setting the threshold.The influence of the penalty factor on the bandwidth of each modal component is analyzed,and the sampling frequency is used as the penalty factor in the application to improve the accuracy of the decomposition.(3)The mechanism of the double-pulse feature of the rolling element through the raceway defect is studied.A double-pulse feature extraction method based on VMD is proposed.The original signal is pre-whitened by the AR model,and the pre-whitened signal is subjected to VMD decomposition.The separation and extraction of double-pulse features are realized by the indexes of mutual information and kurtosisaccording to the different characteristics of step and impact components,and the signal characteristics are enhanced by envelope analysis.(4)A method for estimating the size of defects of rolling bearing based on double-pulse characteristics is proposed.The corresponding defect size estimation model is proposed by studying the relationship between the characteristics of the double-pulse and the position of the rolling element passing through the defect.The influence of rotational speed and defect size on the accuracy of the estimation method are studied,and the causes of the errors are analyzed.