Based On Blind Source Separation Of Railway Wagon Bearing Acoustic Emission Signal Feature Extraction And Intelligent Diagnosis Research

Railroad freight axle of the rolling bearing is to support, guide, and reduce fixed machinery and mechanical parts of the friction between the movement. It is the most important parts of rotating machine. The bearing fault of railroad freight bring on serious risk to safe operation of the wagon and cause serious traffic accident. Although the safety accidents of railway wagon for various reasons can’t avoid completely, accurate monitoring railroad freight axle bearing condition to take the appropriate and effective forecast method to its failure can effectively avoid the accident because of the bearing fault of train running. Railroad freight axle bearing state monitoring is the important parts of the future railroad freight safety information system.This paper discusses the acousticemission monitoring, analyzing acoustic emission signals (AE) whichdetected from the train bearing components.Use blindsource separation technology of signal processing and analysis, and then usepattern recognition method realize intelligent fault diagnosis.In order to test early bearing fault easily, through the contrast analysis, the paper finally choose acoustic emission signal as fault signal source that its anti-interference and sensitivity are strong to realize fault diagnosis through the blind source separation and support vector machine (SVM) algorithm. Through the simulation experiments the thesis shows that the algorithm is effective and practical and acoustic emission signal extraction based on blind source separation and support vector machine (SVM) calculation method can well realize on-line intelligent diagnosis of the axle bearing fault.