Subway Train Bogie Condition Monitoring System

The bogie is the traction and load-bearing device of the train.If the bogie is malfunctioning,it will definitely pose a hidden danger to the safety of train operation.The timely detection and treatment of bogie faults is an effective way to ensure the safety of train operation.In order to monitor the bogie status timely and accurately,this paper develops a non-contact infrared thermal imaging camera-based bogie status monitoring system for subway trains,which mainly includes:(1)Aiming at the problems of serious data distortion,numerous sensors and single monitoring point in traditional monitoring methods,a bogie monitoring system scheme applicable to the line mainline is designed and a field hardware platform is built.The system is divided into a field inspection system and a trackside service system.In the field inspection system,a temperature detection scheme combining visible image positioning and infrared image temperature measurement and a basic data acquisition scheme are designed;in the trackside service system,a data management scheme,a data display scheme and a data reporting scheme are designed.Compared with the traditional scheme,this scheme has the advantages of more monitoring points,real data and good real-time.(2)Based on the design scheme and hardware platform,the software of field detection system and trackside service system is developed.The field inspection system software includes two parts:host and slave,which can complete sensor data acquisition,dual-light image processing,walking part fault diagnosis,data tran smission and system management tasks;the trackside service system is developed based on B/S architecture,divided into front-end and back-end software,which can complete data parsing,data management,Web interface display and data reporting tasks to the intelligent operation and maintenance platform.(3)In order to accurately extract the component temperature and improve the accuracy of fault diagnosis,a dual-light image processing strategy and a comprehensive fault diagnosisanalysis strategy are proposed and implemented.For component temperature extraction,a dual-light imagc processing strategy based on YOLOv4 target recognition algorithm combined with affine transformation is proposed;for component fault diagnosis,a bogie component fault diagnosis strategy based on data correlation and multi-dimensional data analysis is proposed.After the field application,the dual-light image processing strategy can complete the target recognition and dual-light image alignment tasks;the fault diagnosis strategy can accurately identify and classify the component faults.(4)In order to verify the functional integrity and reliability of the system,actual testing and blackbody verification tests were conducted.After long-term testing and data analysis,the system data acquisition,processing,transmission,storage,display and reporting functions are normal and can complete the task of bogie status monitoring;the system monitoring loss rate is low,the anti-interference ability is good,and it has the ability of long-term stable operation in the line main line;the system temperature measurement accuracy and temperature measurement accuracy is high,which ensures the data is real and reliable and effectively reduces the system misdiagnosis rate.