Surface Damage Location Method Of Low Speed Large Bearing Based On Acoustic Emission

The application of low-speed & large rolling bearings in rotating machinery is becoming more and more common.For the fault diagnosis of rolling bearings,vibration signal detection is usually the most effective method,but the vibration detection technology relies on accurate speed information.For equipment with stable speed operation,traditional vibration detection technology can be used to detect bearing failure,however,low-speed heavy-load bearings are usually accompanied by relatively large speed fluctuations,and the frequency of failure is low,it is difficult to diagnose failures with traditional spectrum analysis.Therefore,based on the advantages of acoustic emission spatial localization to development the new technologies and means.Find the exact location of low-speed large bearing damage is a great significance for confirming the type of failure,avoiding the misdiagnosis of bearing failure,and improving efficiency during maintenance.For the structure type of the bearing ring,based on the principle of linear localization of acoustic emission,a model that only solves the time difference of the acoustic emission signal and does not depend on solving the wave speed is proposed.For the problem of the actual bearing fault acoustic emission signal frequency bandwidth,the composition is complex,and it is difficult to accurately calculate the signal arrival time difference,an optimal time difference calculation method based on wavelet packet decomposition is proposed.On the thrust ball bearing with a pitch diameter of 150 mm,the accuracy of the model and the proposed algorithm were verified by carrying out the simulation experiment of lead-break signal and the experiment of bearing surface damage.The localization of the fault source depends on the accuracy of the signal arrival time difference calculation among sensors.To compensate for the problem of calculating the signal arrival time difference band by band based on the wavelet packet decomposition time difference calculation method,a more reliable method AIC(Akiake Information Criterion)is used to calculate the signal arrival time difference.By introducing the characteristic function of the original signal,the AIC function is made more sensitive,and the accuracy of the time difference calculation is improved by calculating the AIC of the signal characteristic function.In order to verify the universality of the research method,a static test of leadbearing signal localization was carried out on the large bearing model HRB811-500,and accurate localization results were obtained by the two proposed algorithms.In order to carry out the damage location experiment of large bearings in the laboratory environment,we independently designed and implemented a large-scale bearing experiment equipment.For the problem of poor signal quality of large bearing faults,the methods of median filtering and high-pass filtering are used to preprocess the AE signals of bearing faults,and experimental verification is carried out through two algorithms.The results show that the method studied can estimate the location of largescale bearing surface damage sources.