Fault Feature Extraction Of Rolling Element Bearing Based On Encoder Signal

As an important part of the mechanical drive system,bearings play a vital role in mechanical equipment.However,due to factors such as working environment,load variation,improper manufacturing and assembly,bearing failures occur from time to time,and these failures often lead to equipment downtime and maintenance,which in turn have a large impact on productivity and production quality.Therefore,bearing fault diagnosis has been one of the important research contents in the field of rotating machinery fault diagnosis.To address the problem that the installation error in the encoder signal makes it difficult to extract the bearing outer race fault features,an error suppression scheme for encoder installation is proposed.The method firstly obtains the envelope on the original instantaneous angular speed signal based on the encoder mounting error characteristics combined with the root mean square envelope technique.Secondly,it improves the accuracy of fitting the encoder mounting error components based on the segmented stepwise approximation technique and the absolute mean difference index combined with the inverse slope correction method.Furthermore,the optimized envelope window length is determined adaptively by improving the energy ratio index,and the corresponding residual signal is obtained.Finally,the envelope spectrum analysis of the residual signal reveals the characteristics of the bearing outer race fault.For the problem of difficult rolling bearing fault feature extraction under low speed conditions,this paper proposes a method of rolling bearing fault feature extraction under low speed conditions based on improved smoothing a priori signal reorganization.The method firstly calculates the first 2 orders of the theoretical fault feature order of rolling bearing,on the basis of considering the slip,finds the maximum and minimum fault feature orders corresponding to the first 2 orders,and uses the maximum and minimum feature orders as the maximum and minimum cutoff frequencies of the filtering by the a priori smoothing method respectively;secondly,combines the maximum and minimum cutoff frequencies to obtain the regularization parameters corresponding to the improved smoothing a priori method,further divides the regularization parameters into and filter the instantaneous angular velocity signal with the regularization parameter and step size corresponding to the first order,determine the optimized regularization parameter adaptively with the improved energy ratio index,and obtain the corresponding filtered signal;then,take the regularization parameter and step size corresponding to the second order to filter the filtered signal obtained from the first order,determine the optimized regularization parameter adaptively with the improved energy ratio index,and obtain the corresponding filtered signal,combine the first order Finally,the envelope spectrum analysis is performed on the superimposed signals to reveal the bearing fault characteristics.Through simulation and experimental analysis,it is verified that the proposed method can effectively extract the fault characteristics of rolling bearings,which provides a reference for condition detection and fault diagnosis of rolling bearings.