The Research Of The Application Of Acoustic Emission Testing Technology In Faulting Diagnosis

As a kind of nondestructive testing technology developed in recent years,acoustic emission testing technology gets extensive attention due to its uniqueadvantages. This paper studied the application of acoustic emission testingtechnology in the field of fault diagnosis from different aspects, and the maincontents are as follows:This paper introduced the basic knowledge, concepts and characteristicsrelated to acoustic emission，acoustic emission technology and the processingmethods related to acoustic emission signal. The principles of localizationmethod of acoustic emission signal were expounded. The localization ofacoustic emission source in plane was realized in combine with MATLABprogram, which is based on the principle of localization method with foursensors in plane. The effectiveness of the localization program was verifiedthrough the experimental and the factors which affect the precision analyzed atthe same time.Different processing and analysis methods for acoustic emission signal ofbearings were researched, and the fault diagnosis of rolling bearing was realized. The symptom parameters related to the time-domain were used toanalyze the features of acoustic emission signal of bearing in different states.And it was processed and analyzed in frequency-domain through the Fouriertransform method. Otherwise, wavelet analysis also was used to analyzeacoustic emission signal of bearing, because of its advantages intime-frequency analysis. The intelligence diagnosis method based on the BPneural network was emphasis researched. This paper proposed an attributereduction method based on the combination of the principal componentanalysis and the rough set, and the realization process of this method wasintroduced specifically. At last, the parameters after reduction were inputted tothe BP neural network in order to recognize the fault type. At the same time,the analysis of the experiment data could show that this method can improvethe convergence speed of neural network and the recognition precisioneffectively.This paper expounded the principle, characteristics and methods of theapplication of acoustic emission technology, respectively in fatigue crackdetection, pressure vessel inspection and the partial discharge detection oftransformer. The testing processes of acoustic emission technology in threecases were detailed introduced combined with the instances, which could verify the effectiveness of the application of acoustic emission technology inthese aspects.