Structural Performance Analysis And Multi-objective Optimization Of Subway Head Car Body

In recent years,subway vehicles have developed rapidly in our country and other countries in the world.Because of its convenience and comfort,subway vehicles have gradually become one of the transportation means preferred by people during urban commuting.In this paper,a certain iron head car body as the research object,simulation of the car body in various working conditions,calculate the stiffness,structural strength and mode of the car body,analyze the structural characteristics of the car body,and compared with the car body test,verify the accuracy of the simulation calculation.Through BP neural network and response surface method,combined with non-dominated multi-objective genetic algorithm NSGA-II,the vehicle structure optimization is carried out.With the thickness of the vehicle section as the design variable and the stress in dangerous working conditions as the constraint,the data is processed in batch through the program,and a relatively ideal optimization model is obtained without considering the weight.The static stress is constrained.To achieve the purpose of reducing the body mass and improving the body mode.The main content of this paper is as follows:Firstly,the finite element model of subway TC car body was established in HyperMesh,and the corresponding load and constraint conditions were applied to the car body according to the relevant provisions of standard BS EN12663-1:2010.According to the standard GB/T7928,the maximum vertical displacement of the middle lower flange of the side beam of the car body under the vertical overloading condition is 7.03mm,which is less than 1/1000 of the vehicle distance and meets the requirements of the vertical stiffness of the car body.According to TB/T3550.1-2019,the torsional rigidity of the vehicle body was evaluated.The calculated equivalent torsional rigidity value was 1.33 ×10~9N·m~2/rad,larger than the standard value of 5.5 × 10~8N·m~2/rad,meeting the requirements of torsional rigidity of the vehicle body.The finite element calculation of 8 working conditions of the car body shows that the safety factor of the static strength of the car body is greater than 1,which basically meets the safety requirements of the static strength.The simulation results of static stiffness and static strength were compared with the experimental results to verify the accuracy of the finite element model.According to the finite element calculation,the first order vertical bending vibration frequency of the aluminum body is 18.70Hz,and the first order vertical bending vibration frequency of the vehicle body in the condition of reconditioning is 6.63Hz.Then,the sensitivity analysis was used to select the parts that have great influence on the performance of the car body structure.The thickness of these parts was selected as the input,and the mass of the steel structure of the car body and the first-order vertical bending mode frequency under the condition of reconditioning were taken as the output.BP neural network was established respectively,and the parameters of the neural network were adjusted to continuously improve the accuracy of the neural network.Three first-order polynomial response functions were established with the thickness of the selected component as the input and the three dangerous Von.Mises stress values under two conditions as the output.Finally,the non-dominated multi-objective genetic algorithm NSGA-Ⅱ was selected for multi-objective optimization of the car body structure.With the selected component thickness as the design variable,the three stresses under two working conditions did not exceed the allowable stress of the material as the constraint,the lightweight of the car body and the improvement of the first-order vertical bending mode frequency of the reconditioning were realized.The optimization results show that the mass of the subway body decreases from 6.1t to 5.94t,and the first-order flexor mode increases from 6.63Hz to 7.28Hz.This paper can provide some reference for relevant designers to carry out the multiobjective optimization of the subway car body structure performance and lay a foundation for further research on the vehicle optimization problem.