Application And Research Of Acoustic Emission Technology In The Fault Diagnosis Of Ultra-Low Speed Bearing

Bearing is one of the most commonly used key parts in rotating machinery,its running state is often directly related to the safe and stable operation of the whole equipment,Once the fault occurs,it will greatly affect the safety and efficiency of the production of machinery and equipment..Therefore,it is very important to diagnose the damage state of rolling bearing.Due to the complexity of the operation and the structure of the low speed heavy load bearing,it is very difficult to monitor the damage state of the bearing.With the rapid development of machinery manufacturing industry,the application scope of low-speed rolling bearing is more and more widely,and this kind of bearings are generally installed in the rotating machinery equipment of large and medium-sized,once the damage caused by the shutdown,maintenance and replacement requires a lot of manpo wer and material resources,Therefore,monitoring damage state of the bearings in advance can avoid s the accident of shutdown and obtain a large economic benefit.Acoustic emission technique(AE)is a new technology of dynamic real-time monitoring,compared with the traditional detection methods,the acoustic emission signal is sensitive to dynamic defects,wide frequency band,high detection efficiency,It can find the early damage of low speed and heavy load rolling bearing in time,which has an important engineering application value for the maintenance and repair of bearing in the rotating machinery.This dissertation by means of acoustic emission technique,the test platform is built to simulate the running state of low-speed heavy-load bearing,artificial defects in different position and size of bearings,the acoustic emission signals of different damage states are collected,and the feasibility of acoustic emission technique in the monitoring of low speed bearing damage is studied theoretically and experimentally.The main work and achievements:With the help of bearing test bench to collect different damage state acoustic emission signal,wavelet analysis and wavelet packet analysis are used to process the signal respectively,by extracting the band for energy percentage,compared with that of wavelet analysis,wavelet packet analysis can extract the main frequency band of the bearing fault acoustic emission,higher proportion of energy bands are consistent with the spectrogram.And compared to the wavelet scale spectrum and STFT spectrum extraction performance characteristics of low speed bearing in AE signal,the results show that the wavelet scale spectrum for higher temporal resolution of non-stationary acoustic emission signal,And compared with the STFT spectrum,the frequency resolution of the wavelet scale spectrum is higher for the non-stationary signal,so the wavelet scale spectrum and the STFT spectrum can be used to extract the fault characteristics of the low-speed bearings.According to the low bearing weak fault features,it is easy to be submerged by noise.the dissertation present a new method for fault diagnosis of low speed bearings based on the combination of energy entropy and empirical mode decomposition(EEMD),and the effective screening of the IMFs by using the correlation coefficient method and the variance contribution rate method.The experimental results show that the cross correlation coefficient and the variance contribution rate can be used to screen the effective IMF components,and the extracted effective IMF energy entropy can be used to characterize the damage defects of the low speed bearings.And compare the classification results of low-speed bearing fault types of support vector machine and BP neural network,it is concluded that the recognition accuracy of SVM is better than of BP neural network.