Research On Feature Extraction Of Early Acoustic Emission Signal Of Bearings

As a necessary part of rotating machinery,rolling bearings will affect the operation of the whole system.The importance of bearing monitoring,fault diagnosis and life prediction is self-evident.In this paper,the acoustic emission signal with more information than the vibration signal is taken as the research object.Firstly,the principle and mechanism of acoustic emission are introduced.The reason of the acoustic emission phenomenon and the analysis method of the signal are explained.The signal processing extracts the performance degradation index of each domain rolling bearing,and then seeks the index which best represents the bearing fatigue law among these indicators to measure the running state of the bearing.This is of great significance for ensuring the reliability of equipment operation and maximizing the actual production benefits.The main research contents of this thesis are as follows:(1)From the time domain parameters,acoustic emission parameters and information entropy change trend,study the degradation characteristics of the bearing life cycle,and obtain the change of the characteristic parameters in the whole life cycle,so that the characteristic parameters from different angles are within the whole life cycle of the rolling bearing.Comparative analysis of the performance of recession performance.The results show that although these parameters can relatively better reflect the degradation state of the bearing,the selection of a single characteristic parameter as the characterization of the decay performance often fails to balance sensitivity and stability,and a single characteristic parameter is usually only for one defect or The fault is more sensitive.In actual operation,due to the complexity and variability of bearing work,relying on a single characteristic parameter in a certain domain as a degradation performance index cannot be effectively characterized.(2)For the problem of diversification of bearing performance degradation indicators,Principal Component Analysis(PCA)is used to fuse the characteristic parameters in the time domain,frequency domain and time-frequency domain that can be relatively well characterized by degraded states.The first principal component is used as an evaluation criterion for bearing performance degradation.(3)A new method for determining the pulse frequency of the acoustic emission signal is approached.Starting from the time domain waveform,the interval between pulses is used to preliminarily determine the fault frequency,and the pulse frequency weighting and threshold correction method are used to obtain the precise pulse frequency.The calculation is simple and quick,and the application range of the wide multi-fault diagnosis is the biggest feature of the method.At the same time,it also provides a strong guarantee for the fault diagnosis and shutdown conditions of the bearing fatigue test.