Development Of Safety Prediction System For Urban Rail Train Based On Online Condition Monitoring

With the rapid development of China's urban rail transit industry,the security of train operation has aroused wide public concern.In order to ensure the safety of train operation,finding and solving problems is very important while the train is in motion.This thesis works on the design and implementation of the security forecasting system which is the subsystem of the existing system,vehicle maintenance and production management system.This subsystem using state of the running trains data from the parent system to solve the safety problems in the train operation,based on data analysis and data mining technologyThis system is composed of four functional modules,fault statistics,fault correlation analysis,fault identification and fault prediction.The main work of this thesis includes:Using raw table in the database to establish the corresponding fault table on the basis of the system settings and user requirements.Make the fault data statistics in the table and generate fault charts based on different conditions.By the large data platform for the train operation of the fault data,do data analysis and mining,the fault data associated analysis,the fault of the attributes and the association between the rules.Generate association rules data table.Using the association rules for fault identification,finding the fault data in the fault table and fault association rules in the association rules database.Generate hidden fault list after confirming the hidden trouble.Based on the relevant data,predict the developing trend of fault by regression analysis,and establish the regression model by constructing the regression equation between attributes.The system introduces data mining technology into the vehicle maintenance and production management system,which ensures the security of train operation by the identification and prediction of the abnormal data,and promptly notifying the user to deal with the hidden trouble.The system test results show that the functional requirements of the system have been basically realized.The vehicle maintenance and production management system has realized parts of basic functions.Software's follow-up development work is in progress.