Analysis And Optimization Of High-Speed Railway Pantograph Based On Multi Sensor Layout

High-speed trains use the pantograph method to obtain power.The health and reliability of the pantograph are important guarantees to ensure the smooth and safe operation of high-speed trains.However,the working environment of the pantograph is changeable.If the faults or abnormal conditions are not eliminated in time,it will seriously affect the service life of the pantograph and even cause various accidents.This paper optimizes the DSA250 pantograph model used by CRH2 EMUs,and designs an effective sensor layout to detect the health status of key pantograph components in high-speed railways.Firstly,derive the theoretical formula of the ternary mass reduction model of the pantograph and the dynamic model.The solid model of the DSA250 pantograph system is established by the 3D modeling software Soild Works,and the equivalent simplified model of the pantograph structure is obtained.Secondly,the established pantograph model is imported into Ansys for modal analysis,and the natural frequency of the system is solved,perform model optimization.For the bow slide part of the pantograph structure,different materials are used for modal analysis.According to the comparison between the calculation results and the original model results,the natural vibration frequency is analyzed.It is concluded that the slide material is made of C/C composite material,and the vibration frequency can be reduced up to 0.74%;According to the improvement of the underframe structure of the pantograph and the modal analysis,according to the comparison of the results,the natural frequency of the underframe is reduced by 14.79%,and it is concluded that the optimal structure of the underframe is to set insulating sections in the four corners;improve the structure of the bow slide of the pantograph and carry out modal analysis.According to the comparison of the amplitude changes of the slide,the vibration frequency after the transformation was reduced by 7.75%,which was much better than.It is concluded that the optimal skateboard structure is four skateboards with equal symmetry.Thirdly,the static strength analysis of the pantograph model with optimized structure is carried out.Synthesize the modal analysis and static strength analysis of the pantograph.According to the mode shape,dynamic stress cloud diagram and deformation diagram,find out the most violent vibration and deformation,and the most stressful part,and get the weak link in the pantograph structure.Put the acceleration sensor on these positions to realize the multi-sensor effective monitoring of the key parts of the pantograph of high-speed trains.Finally,a Deep Confidence Network(DBN)based pantograph health status detection method is proposed.The vibration signals collected by the acceleration sensors are divided into processing and feature extraction,and the pantograph DBN network model constructed by training.Select the appropriate number of normal samples by comparing and analyzing the8 prediction results of different numbers of normal samples in the data set,and improving the diagnostic accuracy of the model to 97.87%,indicating that the fully trained DBN model can effectively identify and classify the fatigue of different types of pantographs,the accuracy of the result is high and the classification ability is strong.This article provides ideas for the optimal design of the pantograph model and the use of multi-sensor to detect the pantograph’s health status.