Research On Pantograph State Maintenance Based On Data Mining

Urban rail transit vehicles are complex mechatronic systems.Pantograph is an important part of the urban rail transit power supply system and is the core equipment for power transmission in the bow network.If the pantograph breaks down during operation,it will not only interrupt the flow but also often cause the destruction of the catenary system,which will directly affect the safety and reliability of the train operation.In addition,the harsh working environment,complex mechanics and electrical Interaction make the pantograph more prone to failure.At present,the vast majority of urban rail transit vehicle companies using vehicle maintenance system planned,from the current frequent occurrence of pantograph line failure has gradually can’t meet the needs of subway vehicle maintenance,and it did not consider the different conditions caused by the different failure conditions of the factors.What’s more,vehicle maintenance accumulated a large number of fault data resources can’t be fully utilized.Therefore,by evaluating the reliability of the pantograph structure and the actual overhaul of the special line and the in-depth data mining,the status of the pantograph of the vehicle is designed for the overhaul and the potential hazards are detected in time and the corresponding overhaul and maintenance are carried out.Therefore,by evaluating the reliability of the pantograph structure and the actual overhaul of the special line and the in-depth data mining,the status of the pantograph of the vehicle is designed for the overhaul and the potential hazards are detected in time,which has important research significance.In this paper,the data mining algorithm as a tool for urban rail transit vehicle pantograph status maintenance research,the main contents are as follows:(1)Analyzed the basic structure of the pantograph and the working principle of the circuit,the structural functions and failure modes are analyzed emphatically,the key components of each subsystem are clarified,which provides the theoretical basis for the status maintenance of the following pantograph key subsystem.(2)Aiming at the limitation of fault tree analysis for pantograph reliability evaluation,the reliability analysis of pantograph is combined with fault tree analysis and Bayesian network to find out the weak points of pantograph.Quantitative assessment of the reliability of the pantograph of urban rail transit vehicles.(3)Based on the maintenance data accumulated in a subway,the data analysis is carried out based on the data mining algorithm.A fault diagnosis method based on principal component analysis and probabilistic neural network is proposed.The simulation results show that the diagnostic model is short in time and high in diagnosis,which proves the validity of the model.(4)Based on the analysis method and the diagnosis model proposed,based on the existing maintenance system,the weak link of the pantograph subsystem and the components with higher fault frequency are proposed,and the pantograph condition inspection strategy of the subway vehicle is proposed Suggest.