Identification Of Rail Material Damages By Acoustic Emission Technology

At present,the business mileage of High-speed rail in China has reached more than35,000 kilometers.With the gradual improvement of railway network consisted of 8 vertical and 8 horizontal trunk-lines,safe running is gradually replacing large-scale constructions and becoming the main focuses in future.Currently,major problems from railway transportation in China are large transportation volume and diversified locomotive varieties.In addition,rails always experience the harsh working conditions,such as complex loading conditions,service environment and running conditions.However,related safety operations are still under developed.For example,rail damage detection still adopts traditional technologies such as ultrasonic detection technology,eddy current detection technology,ray detection technology,penetration detection technology,magnetic detection technology.Nonetheless,it is difficult for these technologies to meet the requirements of the shortened window lead-time and complex conditions.Exploring new detection technologies has become the key to the current safe operation of railways.On the basis of in-depth analysis of switch damage types,this thesis aims to realize real-time monitor of rail switch damages,meanwhile compares among fiber Bragg grating detection,optical image detection and piezoelectric energy method based on acoustic emission technology,to select a desirable detection method.It is found based on experiments that the rail damage detection method based on acoustic emission technology is suitable for on-line monitoring the early failures of rail switch status,which have been received much more attention.In terms of signal processing,this thesis chose classical acoustic emission signal processing algorithms,energy spectrum ratios,and acoustic emission signal processing based on wavelet(packet)analysis for experiments to analyze the advantages and disadvantages of various methods.It founds that the signal processing method based on wavelet(packet)analysis and clustering of mega data mathematical tools,can give different responses to the dynamic crack,crack initiation and crack growth of the detection object.Based on a series of experiments,it is proved that acoustic emission technology to monitor rail crack damages in real time is feasible and has great advantages for real-time monitoring.However,the acoustic emission technology is still in the preliminary research stage in the detection of rail damage.Moreover,the feasibility verification of the technology in the detection of rail damage is still insufficient in theory and experiment.The field application still needs further research.