Researching Fault Analysis And Diagnosis Technology Of Vehicle Synthetic Transmission Device

As the main transmission component of the tracked armored vehicle,the comprehensive transmission device’s health status directly affects the maneuverability of the armored vehicle.The comprehensive transmission device has a harsh working environment and complex structure,which makes the device more prone to failure compared with other components;moreover,once a failure occurs,it is difficult to determine the fault location and difficult to maintain.Therefore,the systematic analysis and rapid diagnosis of the failure of the integrated transmission has become an urgent problem to be solved in the maintenance of armored vehicles.Aiming at this problem,this paper conducts an in-depth study on the fault analysis and diagnosis technology of vehicle integrated transmission.The specific work is described as follows:Firstly,based on the fuzzy fault tree theory,the system fault analysis of the vehicle integrated transmission is carried out.Through in-depth research on the composition,fault mechanism and propagation mechanism of the integrated transmission,the main fault types of the integrated transmission are analyzed and summarized,the top events of each fault tree are determined,and a fuzzy fault tree model is established;Analyze and solve the fuzzy importance of each bottom event.When a fault occurs,the corresponding fault tree is determined according to the fault phenomenon,and the fuzzy importance ranking of the bottom events in the fault tree provides an analysis basis for the subsequent component fault diagnosis.Then,based on the wavelet transform and convolutional neural network,the fault diagnosis algorithm of the key components of the integrated transmission is studied in depth.The noise reduction of the original vibration signal is carried out by using empirical mode decomposition（EEMD）and singular value difference spectrum method.On this basis,a fault diagnosis method combining continuous wavelet transform and two-dimensional convolutional neural network is proposed.In order to verify the fault diagnosis ability of the proposed method,the proposed method was compared with 1D＿CNN and Gray-2D＿CNN by taking the bearing public dataset of Western Reserve University as the verification object;the results show that the proposed fault diagnosis method has better performance.Troubleshooting capability.Finally,based on the open-source cross-platform Qt program development framework,the design and implementation of the fault diagnosis system software platform are studied.Through the overall design of the fault diagnosis system software platform and the design analysis of each functional module,the work flow of the fault diagnosis system is drawn up,and the Qt platform is used to develop the comprehensive transmission fault diagnosis system software platform.