EMU Vehicle Body Structure Optimization Design Based On Improved Particle Swarm Optimization

As a new type of transportation tools,High-speed EMU palys an important role in China's railway passenger transport,which lightweight optimization design of the vehicle body has always been an impotant part in the vehicle design.Based on the analysis of the performance of the CRH380B EMU middle vehicle body,this paper put forward the lightweight optimization model which take the underframe as the main research object.An improved simplified particle swarm optimization algorithm for solving constrained problems is proposed and compiled corresponding procedures based on the study of PSO algorithm and related improvement methods.An lightweight optimization design of vehicle body has been put forward which use the improved PSO algorithm combined with MATLAB and ISIGHT.The specific contents of the paper are as follows:(1)In order to improve the optimization efficiency of PSO,a population information fusion strategy is proposed based on the simplified PSO.Only use the individual optimal position and population optimal location to update the particle position to make the algorithm description more simple.Constructing a random term rely on population information to increase population diversity.Design adaptive contronk function equation to balance the global detection and local development.The numerical experiment results show that the improved slgorithm has obvious advantages compared with the original algorithm in solving constrained optimization problems interms of searching ability and convergence speed.(2)The vehicle body geometrical model has been simplified based on the analysisi of CRH380B EMU structure.The shell element is used to describe the force characteristics of vehicle body.And a sophisticated vehicle body finite element model has been established.According to the European standard EN12663:2000,calculate 16 cases of the vehicle body finite element model.complete the modal caculation and check the static strength,stiffness and modal of the vehicle body.All results show that the vehicle body structure meets the standard requirements.(3)The thickness of the underframe side-beam and cover plate is the design variable.The parameters corresponding to restrain the static strength,stifness and modal are extracted as the output response use the ISIGHT.Using the optimal Latin hypercube experiment design,the corresponding sanple points between the output response and the design variables are obtained.The approximate model of the vehicle body structure is established by using the second order polynomial response surface model.The parameter sensitivity and the approximate model error are analyzed.The multiple quadratic regression equation of the output response is obtained according to the coefficient table.(4)Based on the approximate model,the vehicle body is seen as the objective function to lightweight the vehicle body with the constraint of the stiffness,strenght and first order vertical resonace frequency.Respectively using the PSO in ISIGHT and improved algorithm to solve the optimzation problem.Compared with the two optimizaiton results and verified,result shows that the two algorithms can achieve the purpose of lightweigh the vehicle body under the premise of ensuring the static strength,stifness and modal,but the improved algorithm has advantages in terms of searching ability and convergence speed.And an optimizaiton scheme has been proposed according to the improved algorithm optimization result.