Research On Rail Weld Damage Monitoring Technology Based On Ultrasonic Guided Waves

The operating mileage of China’s high-speed railway has reached 38000 kilometers by 2020.As an important equipment for high-speed railway operation,seamless rail plays a key role in the operation of high-speed railway trains.During the service period of seamless rail,if there are defects in the welding process,the position of rail weld is more likely to become a weak link.The continuous load of the train and the change of external environment will make its fatigue loss faster,resulting in damage or even rail breakage,which seriously threatens the safety of high-speed railway operation.Due to the complex structure of the weld,the existing regular detection schemes and technologies have low detection efficiency for the rail weld,and the detection accuracy is insufficient,so it is impossible to monitor on-line.Therefore,taking the rail weld as the object,this paper uses the ultrasonic guided wave,which is sensitive to damage and has high accuracy,to study the structural health monitoring and damage identification of the rail weld.The specific research work is as follows:(1)The basic theories of ultrasonic guided wave,such as characteristic equation,group velocity,phase velocity and dispersion,are described.Using the numerical analysis method to solve the dispersion equation in the rail,the theoretical dispersion of the rail is analyzed.Combined with the dispersion results,the relationship between wavelength and recognition accuracy,the ultrasonic guided wave excitation signal suitable for the rail is selected.Aiming at the special structure of rail weld,a dispersion back analysis method based on actual guided wave signal is proposed,which makes up for the problems existing in the dispersion analysis method of numerical model.(2)The complete integrated empirical mode decomposition(CEEMDAN)of adaptive noise is used to preprocess the guided wave signal.The mode decomposition of ultrasonic guided wave signal is realized by CEEMDAN method,which solves the problem that the received signal has mode aliasing due to the dispersion characteristics of guided wave and cannot be analyzed and processed directly when using ultrasonic guided wave for structural health monitoring,so that more effective features can be extracted from ultrasonic guided wave signal in the future.(3)The damage features in ultrasonic guided wave signals are extracted,quantified and located.The effectiveness of the method is verified by experiments.Time delay,normalized residual signal energy,time coherence peak and correlation coefficient are used as the quantitative characteristics of the difference between ultrasonic guided wave health reference signal and damage signal.U75 V steel plate of the same material as rail is used for simulated damage experiment.The above quantitative characteristics are used to compare the results of linear path location,elliptical damage location method and damage factor probability location method,It is concluded that the probability of damage factor is the best identification and location method,and it is improved.(4)The finite element model of rail and rail weld is established as the basis of sensor array layout.With reference to the construction standards of high-speed railway,a monorail model with sleeper is established to simulate the process of EMU trains with axle load of 17 tons and speeds of 250 km / h,300 km / h,350 km / h and 380 km / h passing through the rail,explore the rail deformation and stress concentration area,and simulate the ultrasonic guided wave propagation process around the above areas to visually display the guided wave propagation process at each position of the rail.(5)Ultrasonic guided wave health monitoring experiment of rail weld.Chn60 rail with weld is selected for ultrasonic guided wave health monitoring research.The sensor array is reasonably set according to the finite element simulation results,and the simulated damage test of rail weld is carried out.The improved damage factor probability location method is applied to rail weld.The location method is effective and the results have high accuracy.At the same time,the ultrasonic guided wave characteristics of the rail under load are explored,and the influence of the load on the ultrasonic guided wave health monitoring is determined,which is helpful to the further study of the structural health monitoring framework of the in-service railway track.