Disease Detection And Prediction Of Ballastless Track Slab

With the comprehensive start-up of Chinese high-speed rail network construction and the increasing in operating mileage and operating time,track problems,especially the deformation of the separation gap between slabs layers and the deformation of wide and narrow seams between connected track plates slabs,gradually attracted more attention.This paper mainly focuses on the problems of wide and narrow seams,and studies the influence of temperature factors on its deformation.In order to achieve these goals,the paper based on the actual investigation and analysis the problems of the track slabs,set up an effective monitoring platform,and reasonably analyze the temperature of the track slabs and its influence on the deformation of the wide and narrow seams.Based on the temperature changes of the track slabs,to use the LSTM model in the neural network predicts the deformation of the wide and narrow seams,and compared with the prediction results of the time series ARIMA model to verify the effectiveness of the prediction effect and facilitate the timely maintenance and repair work.On-site investigation and establish the monitoring platform construction.On-site investigation for the problems of track slabs in an area in East China.To classify the surface problems of ballastless track slabs according to the survey results and find out the main problem-deformation of wide and narrow seams,to be the research object.Taking a(subgrade)section of the investigation area as a test site,through reasonable establish the construction of the platform,to monitor the temperature inside the track slab and the deformation size of the wide and narrow seams,and check the monitoring data.Study the regularity of the deformation of the narrow and wide seams from three aspects: interlayer temperature,vertical temperature gradient and longitudinal temperature gradient.The results show that the interlayer temperature and longitudinal temperature gradient are strongly related to the deformation trends of the wide and narrow seams,however,the direction of change of interlayer temperature is basically the same trend as the deformation of the wide and narrow seams,while the direction of change of longitudinal temperature is basically the opposite trend as the deformation of the wide and narrow seams.The consistency of vertical temperature gradient and deformation of narrow seams is more obvious than wide joints.The correlation between the rate of the vertical temperature gradient and the deformation of the wide and narrow seams is very strong,especially with the deformation of the narrow seams.The extremity of track slabs is more sensitive to temperature changes than the middle of track slabs,and the shallow layer is more sensitive to temperature changes than the deep layer.Based on neural network,to establish LSTM model and predict the amount of deformation of wide and narrow seams.To verify the effectiveness of the LSTM model by comparison with the common ARIMA model.Preparation for the timely warning of track slab maintenance and repair by monitoring the temperature of the track slab and predict the changing size of the width and narrow seams of the track slab.