High-speed trains have become an important means of transportation for people’s daily travel,and the track structure as the carrier of trains directly affects the safety,stability and comfort of train operation.With the growth of high-speed railway operation period and the continuous erosion of natural environment,various damages and diseases will inevitably appear in the track structure.At present,the economy and practicability of the commonly used track damage detection methods in our country can not meet the real-time monitoring and detection of track structure,Therefore,it is of great engineering significance to carry out the rapid detection,identification and diagnosis method of high-speed railway track damage.Previous studies have shown that when the train passes through the damaged section of ballastless track structure,the vibration signal of the car body itself will have abnormal mutation.Based on this,this paper attempts to realize the location,qualitative and quantitative identification and diagnosis of ballastless track structure damage,using the dynamic response of high-speed operating vehicles as a medium.The main research contents of this paper are as follows:(1)Based on the joint simulation method of finite element software and multi-body dynamics software,the rigid-flexible coupling model of train-track is established.The vibration response of vehicle body and ballastless track structure is obtained by applying the track random irregularity spectrum as external excitation to the track.The results show that based on the train track coupling model,the vibration responses of the vehicle and track are in line with the specification limits,and are close to the relevant literature values,so the coupling model is reasonable.(2)Through the numerical simulation of ballastless track fastener damage,track plate crack damage,CA mortar layer debonding damage and bearing layer crack damage,we compare the sensitivity of vehicle dynamic response to track damage under various damage conditions.The results show that based on the vertical acceleration of car body,the common track structure damage can be identified and detected,and the vertical acceleration of train wheelset is the optimal basic data for track damage identification.(3)A method of ballastless track damage diagnosis based on wavelet analysis of vehicle vibration response is used,and the influence of the number of wavelet decomposition layers and the number of transformations on damage identification effect is studied.The results show that the abnormal mutation of reconstructed signal in wavelet sensitive frequency layer can realize the identification and location of single track damage and the synchronous identification and location of multiple degrees of the same track damage and multiple types of damage at the same time.(4)A method based on neural network to identify the damage state and degree of ballastless track is used..The results show that the RBF neural network is established by taking the time domain characteristic parameters of high frequency signal reconstructed by quadratic wavelet transform as the input samples of RBF neural network can realize the identification and classification of the damage state and degree of ballastless track. |