The On-Line Detection Of Rolling Contact Fatigue Of Bearing Based On Acoustic Emission

Rolling contact fatigue is one of the main failure form of mechanical parts. The happening of rolling contact fatigue can often be covert and the fatigue damage was paroxysmal, so the real-time detection of rolling contact fatigue is of great significance. Aiming at the application status of the acoustic emission technology used on rolling contact fatigue test at home and abroad, a kind of intelligent rolling contact fatigue on-line detecting system scheme based on acoustic emission technology is put forward in this paper, and the system is applied to the rolling contact fatigue test of bearings with the tester. The main research work in this paper is outlined as follows.Firstly, the acoustic emission signal processing method for rolling contact fatigue test is studied. Through filter denoising, feature extraction, feature selection and pattern recognition of the acoustic emission signal, the detection the rolling contact fatigue is completed. In feature extraction, modern signal processing methods, such as wavelet decomposition, empirical mode decomposition and local mean decomposition are used to extract three kinds of characteristics of the rolling contact fatigue acoustic emission signal respectively, those are, energy spectrum, information entropy and AR model parameters moment; Principal component analysis(PCA)method is used to select the extracted feature; Intelligent pattern recognition technologies such as artificial intelligence neural network and support vector machine(SVM) are used to identify the fatigue state of spacimen.Secondly, the rolling contact fatigue on-line detection system based on acoustic emission is designed. Combined with the existing equipment of the lab, the hardware part of the system consists of the lever load rolling contact fatigue tester and acoustic emission signal acquisition system; The software part is made of six modules which consists of data acquisition, denoising processing, feature extraction and selection, pattern recognition, display interface and file management. At the same time, the parameter optimal configuration of the acoustic emission capture card is also emphasized in this research.Finally, the experimental verification is conducted for the online rolling contact fatigue test system. Through the filter denoising, feature extraction and pattern recognition of the test data, and compared the test results with the specimen pattern in a stereoscopic microscope, the effectiveness of the system is verified.