Research On Acoustic Emission Signal Recognition Of Axle Fatigue Crack Based On Parametric Analysis And EEMD Energy Entropy

The axle is an important part of the train running directly related to the safety of the train.It not only bears a huge static load and dynamic load,but it is also often affected by external factors such as friction,corrosion,cyclic stress,impact,and temperature.These external factors often cause the train axles to wear,crack or even break.If the axle fails during operation,it will cause extremely serious losses to the national property and passengers' lives.In order to ensure the smoothness and safety of the vehicle operation,it is of great significance to diagnose the fault of the axle.Acoustic emission technology is an important non-destructive testing technology and has extensive applications in machinery,construction,mining,and oceans.Therefore,applying acoustic emission technology to axle fault detection is a hot topic in the field of non-destructive testing.In this paper,according to the basic principles of acoustic emission detection technology and its feature extraction methods,a method for fault diagnosis of axle cracks based on parametric analysis method is first proposed.The process of designing the axle crack fault diagnosis experiment of parametric analysis method was written in detail,and the parameters of the acoustic emission signal data collected from the experiment were analyzed to obtain the experimental results,and feature extraction was performed on the fault signal.In order to apply a more comprehensive feature extraction method to the fault diagnosis of axle cracks,an axle crack diagnosis method based on Ensemble Empirical Mode Decomposition(EEMD)energy entropy and LVQ neural network is proposed.The collected acoustic emission signal is decomposed into several stationary intrinsic mode functions(IMF)by the EEMD method.When the axle fails,the energy value of the signal changes,so it can be determined by calculating the energies of the EEMD energy of different signals.malfunction.Finally,LVQ neural network is used to classify fault types.Experimental results show that the proposed method can be effectively applied to acoustic emission signal recognition.