Research On Fault Identification Of Gearbox And Quantification Method Of Gear Crack For Railway Vehicle

With the continuous development of China’s railways towards high speed and heavy haul,the vibration characteristics of railway vehicle system components and the vibration transmission laws between various parts have undergone great changes,and the safety of vehicle operation suffers huge challenges.As a key component of vehicle power transmission,the operation status of the gearbox directly affects the safety of vehicle operation.In addition,no matter whether it is a high-speed train or a heavy haul locomotive in China,there are problems of over-repair,backlog of inventory parts and low utilization rate of maintenance resources.Therefore,it is imperative to optimize the maintenance standard,and it will become a major trend for railway vehicles to change from regular repair to state repair.While avoiding over-repair,it is also necessary to prevent disrepair,which requires real-time monitoring of vehicle operating status.The development of fault identification and quantification technology plays an important role in condition monitoring and subsequent full life cycle optimization.The main research contents of this article are as follows:(1)According to the on-site investigation and the statistical data of domestic researchers,the common failure modes of the railway vehicle gearbox are sorted out.On this basis,the vibration characteristics of gears and bearings are analyzed,and a signal simulation model of gearbox for fault diagnosis is established.(2)In view of the difficulties of various types and large size differences of gearbox components of railway vehicles,the overlapping of the characteristic frequencies and their multiple frequencies of various components,difficult to extract weak faults or easily covered by strong faults and background noise etc,a multi-channel scheme for fault identification of railway vehicles gearbox is designed and a gearbox fault diagnosis method based on optimized morphological component analysis and envelope demodulation is applied.The simulation of gearbox fault diagnosis is completed and the method is applied to the experiment.(3)The energy method is used to calculate the time-varying meshing stiffness of gear with different crack depths,which provides data for the finite element model of traction transmission system established later.(4)In order to further verify the signal simulation model and quantify the fault,so as to determine the degree of fault,and then carry out preliminary research for the subsequent prediction of remaining useful life,a dynamic model of the vehicle traction transmission system is established based on the finite element method,and the dynamic response of gear under different crack depths is obtained by Newmark-? method.Finally,according to the principle of simplicity and quickness,a quantization method for railway vehicle gears based on time-domain characteristic parameters and back propagation neural network is proposed.