Rolling Bearing Fault Diagnosis Based On Acoustic Emission Technique

Rolling bearing plays significant role in rotating machine, and it is also one of the easiest parts which will casue problem. Once the fault happened, one the way of high speed and heavy haul railway, the problem will be expanded, and it will cause hotbox axle,ignition, cutting axle, and lead to train overturnning in the end,ext.To ensure the safety operation of train transportation, research on failure diagnosis of bearing is practically significant. This essay focus on several common failures of train rolling bearing, uses Acoustic Emission technology, chooses wavelet packet analysis as signal primary application method, and builds a method system which based on the combination of wavelet packet analysis and neural network identification. Research on related information, systematically introduce failure modes and failure reasons of rolling bearing, and also the common failure diagnosis mothods and technologies. Through compare and contrast analysis, propose rolling bearing diagnosis method base on acoustic emission technology. The method introduces acoustic emisson diagnosis techonology mechanism as well as common acoustic emission application methods.Meticulously illustrates basic theories and algorithms of wavelet, as well as applications in acoustic emission process. Propose wavelet packet decomposition frequency-band monitoring method, diagnose typical failures of rolling bearing, and extract eigenvector to prepare follow-up Neural network identification.Adopt nonlinear mapping function of neural network and build neural network, realize mapping relationship of rolling bearing between running status and acoustic emission eigenvector, network trainning for extracted rolling bearing eigenvenctor, and get ideal identification result.