Reliability Analysis And Optimization Of EMU Bogie Frame

When the EMU is running,the bogie frame accepts the load from other parts in all directions and the impact load of the track irregularity.With the increase of the running time,the frame is more prone to strength problems.Therefore,considering the uncertainty factors to analyze the reliability of the frame has important engineering significance for the operation safety of the EMU.This thesis takes the bogie frame of EMU as the research object,improves BP neural network based on Atom Search Optimization(ASO)algorithm to carry out structural reliability analysis of the frame,calculates the material utilization rate of weld area to carry out fatigue reliability analysis of the frame,and adopts the Golden Eagle Optimization(GEO)algorithm to carry out structural optimization design of the frame.Specific research contents are as follows:In order to verify whether the structure satisfies the structural and fatigue design requirements,Hyper Mesh software is used to establish the finite element model of the structure.According to UIC 615-4 and EN 13749 standards,the constrained position and load condition of the structure are determined.The static strength of the structure is analyzed by ANSYS software.According to the allowable stress of the material,the static strength of each case is checked to evaluate whether the structure satisfies the requirements of structural strength.The modified Goodman-Smith fatigue limit diagram is used to check the fatigue strength of the welded joints of the frame to evaluate whether the welded joints satisfy the fatigue strength requirements.Through the above finite element analysis,the maximum stress case is determined,which provides the preparation for the reliability analysis of the frame.In order to solve the influence of machining errors and other factors on the structural strength of the frame,uncertainty factors are introduced to analyze the structural reliability of the frame.The design variables influencing Von Mises stress are determined by sensitivity analysis.The test sample points are obtained by Latin hypercube design,and the ASO-BP neural network is established to train and predict the test sample points,so as to obtain the response relationship between the design variables and the maximum Von Mises stress.Then10,000 test samples are randomly selected to calculate the reliability of the structure.In this thesis,the utilization rate of materials is used to calculate the fatigue reliability of the frame.The frame welds are divided into 96 sections,and the maximum material utilization rate of each section is calculated.According to the material utilization rate theory,whether the welds satisfy the fatigue design requirements is judged,which provide preparation for the structural optimization design of the frame.The Golden Eagle Optimization algorithm is used to optimize the structural design of the frame,so as to improve the anti-fatigue performance of the frame.The optimal design variables are selected and the sample points are obtained by the Latin hypercube experimental design.The maximum material utilization rate is taken as the output response,and the quadratic polynomial response surface is fitted based on the test sample points.The optimization mathematical model is established,and GEO algorithm is used for iterative optimization to obtain the optimal solution of the objective function.Then the finite element modeling is carried out again based on the optimized design variables,and the maximum utilization rate of weld material is calculated.The results show that the anti-fatigue performance of the framework is significantly improved.This thesis systematically studies the structural performance and fatigue resistance of the bogie frame,proposes to use ASO-BP neural network and material utilization rate to analyze the structural reliability and fatigue reliability of the frame,and uses GEO algorithm to optimize the structural design of the frame,which greatly improves the anti-fatigue performance of the frame.The results of simulation analysis can provide theoretical basis and reference for the design of other parts of the EMU.