Bearing rolling body discrete failure method based on vibration signals

Abstract

The invention relates to the technical field of retaining-frame-free ball bearings, in particular to a bearing rolling body discrete failure method based on vibration signals, which comprises the following steps: firstly, calculating permutation entropy by using vibration signal data of a retaining-frame-free ball bearing discrete groove, and then selecting at least three optimal intrinsic mode functions of each damage; signal features in the optimal intrinsic mode function are substituted into a sparrow search algorithm for iterative optimization, a support vector machine model is trained, and finally the trained support vector machine model performs classification prediction on vibration signal data of the discrete groove. The result shows that the bearing rolling body discrete failure method based on the vibration signal can effectively classify complex vibration characteristics of discrete groove damage evolution.

Description

technical field [0001] The invention relates to the technical field of cageless ball bearings, in particular to a method for discrete failure of bearing rolling bodies based on vibration signals. Background technique [0002] Bearings are one of the most widely used rotating mechanical parts in the mechanical field. They are widely used in aerospace, automobiles, home appliances and industrial equipment. The quality of the bearings directly determines whether the equipment can operate normally. It plays the role of bearing force and transmitting load. Due to the different working environment of the bearing, it is easy to wear and relatively easy to wear and tear, which has a direct impact on the production process and product quality. Moreover, with the development of modern science and technology, equipment is developing in the direction of large-scale, complex, and automated. If one of the firmware fails, it will have a very serious impact on the entire production. Bearin...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is: in order to solve the permutation entropy of the vibration time-series signals of the existing discrete groove bearings without cages in the four damage states of the discrete grooves are all greater than 0.9, and in mild damage, moderate damage and severe damage Although the permutation entropy has increased to a certain extent, the permutation entropy data overlap phenomenon is serious, and it cannot be directly used for the classification and prediction of the damage degree of non-cage bearings. The present invention provides a method for discrete failure of bearing rolling elements based on vibration signals. The vibration testing machine is used to test the discrete groove bearing without cage to measure the original time-series vibration signal of the bearing under different damage degrees of the discrete groove, obtain the data set, and through the integrated empirical mode decomposition with adaptive addition of white noise, the The original vibration signal is decomposed into multiple eigenmode function components, and the permutation entropy PE of the components is obtained as the eigenvalue of the complexity of the vibration signal with different damage degrees of the bearing, and it is input into the support vector machine model optimized by the sparrow search algorithm. The neural network can effectively classify the complex vibration characteristics of discrete slot damage evolution

Images