Damage Identification Of Rail Welds Based On Ultrasonic Lamb Compact Array

As a pillar industry of national economy,rail transit undertakes the important responsibility of promoting the stable development of social economy.However,with the long-term operation of the in-service rail,as the weak part of the rail structure,the weld is prone to fatigue fracture and other problems under the action of the train and the external environment,which will also lead to vehicle overturning,derailment and other phenomena,seriously threatening the safety of railway transportation.At present,the common rail detection methods have the problems of insufficient detection accuracy,unable to online monitoring and low labor efficiency.Therefore,it is of great significance to study an efficient detection method for the weld,to find the damage in time at the initial stage of crack initiation,and to prevent the further expansion of the damage that may lead to major accidents,so as to ensure the safety of railway transportation.In this paper,aiming at the rail weld structure,the weld damage identification technology based on ultrasonic Lamb wave is studied to accurately detect the weld structure.The research methods and main results are as follows:(1)The dispersion equation of Lamb wave in rail is solved by numerical method,and the dispersion curve of rail is drawn by MATLAB.Combined with the dispersion curve and experimental experience,the key parameters of excitation signal,such as center frequency,peak number,window function,amplitude and sensor distance,are optimized and analyzed,which provides theoretical basis for compact array design and weld detection method.(2)Using ANSYS finite element simulation method,the static simulation and dynamic simulation of rail are analyzed and studied.Firstly,through the static simulation of rail stress,the key parts of stress concentration are selected as the key parts of damage simulation and detection,and are used as the design basis of array detection method;secondly,the dynamic simulation of Lamb wave propagation in rail is carried out,and the propagation characteristic curve of Lamb wave is drawn,which is compared with the numerical solution to verify the correctness of the model;finally,the damage is simulated and studied The results show that the traditional Lamb wave detection method is difficult to be used in the complex structure of rail weld,and it is urgent to propose a new array detection method.(3)The synthetic directivity model of two isotropic sound fields is studied,and the damage identification technology of compact array based on dual source excitation is proposed.The simulated damage experiment is designed,and the window function filtering and cross-correlation algorithm for damage identification are proposed.The effectiveness of compact array is verified,and the advantages and disadvantages of the two array detection methods are analyzed by comparing with sparse array.(4)This paper proposes to extract the scattering characteristics of Lamb wave by comparing the Lamb wave in the area without weld and with weld,and designs a damage identification algorithm to compare the imf1 power spectrum by db5 wavelet denoising and empirical mode analysis.The results show that the proposed compact array detection method is more suitable for rail weld damage detection.