Research On Switch Rail Monitoring System Of Heavy Haul Railway Turnout Based On Ultrasonic Guided Wave

With the continuous increase of railway operation mileage in China,how to ensure the safety of railway operation has become an important research direction.Turnout switch rail is a necessary device for the train to realize line conversion.In order to ensure the normal service state of turnout switch rail,the detection and monitoring of turnout switch rail state is a necessary link to ensure the safe operation of the train.At present,the combination of rail flaw detection trolley and manual inspection is used to detect the switch rail at home and abroad,but this method can not realize the full section on-line monitoring.Ultrasonic guided wave technology can realize full section detection,but the propagation characteristics of guided wave are different for different structures.Due to the complex variable cross-section structure of turnout switch rail,the propagation of guided wave in switch rail is more chaotic.At the same time,the interference between signals makes the dispersion characteristics and multimodality of guided wave more significant.Aiming at the above problems,this thesis mainly studies the ultrasonic guided wave excitation method and crack identification and location algorithm of turnout switch rail,so as to provide theoretical support for the full section detection of turnout switch rail based on ultrasonic guided wave.Aiming at the selection of guided wave excitation mode in turnout switch rail,a variable section modal analysis method is proposed.Based on the semi analytical finite element method,nine sections of the switch rail are analyzed,and the section dispersion curve and vibration shape are obtained.The K-means clustering method is used to classify the modes according to the rail head,rail waist,rail bottom area and the overall average vibration energy.Through the analysis of the statistical characteristics of the classification results,the maximum vibration modes existing in all sections are obtained.Finally,the maximum vibration displacement node in all sections is selected as the excitation node to realize the guided wave excitation of the switch rail.In order to verify the effectiveness of the proposed excitation method,the finite element method is used to simulate and analyze the selected excitation nodes.Firstly,the crack models in different regions and positions are constructed.Through the wave number analysis and modal analysis of the simulation results,the existence of the selected mode is determined,and the feasibility of the excitation mode is verified.At the same time,the appropriate receiving nodes are selected based on the time-domain characteristics of the signal.Finally,the feasibility of the mass block crack simulation method is verified,and the simulation data set is constructed.Aiming at the problem that the guided wave signal is complex and difficult to decouple the crack information,a crack identification and location algorithm is proposed by analyzing the signal in time domain,time-frequency domain and signal separation.Based on this principle,a crack identification algorithm based on zero position difference is proposed to identify the crack by analyzing the zero crossing position of the signal;The method of independent component analysis is used to separate the crack signal.Finally,the envelope extraction algorithm based on Hilbert transform is used to locate the crack according to the peak position of the first wave packet of the crack signal.The results show that the crack location error is less than 5%,which meets the actual application requirements.Finally,the test experiment is carried out on the turnout switch rail outside the laboratory,and the group velocity measurement experiment and crack detection experiment are carried out respectively to verify the accuracy of the excitation mode and the crack identification and location method,which provides a theoretical basis and new ideas for solving the problem of switch rail crack detection.