Digital Application of High-Speed Railway Maintenance Equipment Inspection and Repair Based on Deep Learning
High-speed railway engineering equipment is the basis of railway transportation,and its use status is directly related to the safety of transportation.In order to improve the inspection efficiency and decision-making accuracy of inspectors and managers of the engineering departments,solve the problems such as lack of intuitiveness of inspection early warning status,unreasonable cycle days setting and lack of electronization of spring and autumn special inspection,and realize intelligent and digital management,this paper studies two key technologies to build a digital technology system.Firstly,deep learning was used to establish an automatic disease recognition model based on convolutional neural network.Through data collection and verification analysis of workshops along several high-speed railways,it is proved that the model is able to automatically identify diseases after uploading disease images.The training accuracy is up to 97%,and the verification accuracy is up to 76%.Secondly,based on big data technology,this paper combines convolutional neural network and long-short term memory to establish a equipment state judgment model,and builds an inspection frequency prediction algorithm.By analyzing the data of two high-speed railway work sections,the results show that the equipment status judgment model is able to accurately judge the equipment operating status by capturing the key information in the inspection records,and the average relative error of predicting the inspection times is 14.3%,and the accuracy is 85.7%.Finally,through the case of digital realization of inspection and repair of high-speed railway engineering equipment,it is fully proved that the design and fusion of these two applications can provide sufficient technical support for the inspection and repair of high-speed railway engineering equipment,and can provide comprehensive and reliable data basis for real-time and efficient inspection and repair decisions.
NETL
NSTL
万方数据
