Lightweight Design Of Commercial Vehicles Based On Combination Approximation Model

In order to meet the needs of enterprises,this paper aims to seek a lightweight optimization design method of commercial vehicle cab,which can improve optimization efficiency and enhance the precision of the surrogates and short the design period.Aiming at the cab of a H series commercial vehicle in the enterprise,the lightweight research is carried out by combining implicit parametric modeling,static and dynamic performance analysis,sensitivity analysis,experimental design,surrogates and intelligent optimization algorithm.(1)An implicit parametric model for commercial vehicle cab body-in-white was constructed and the precision of the model was confirmed.Firstly,the implicit parametric modeling of the cab was carried out by using SFE-Concept software,which was converted into a finite element model,and the basic static and dynamic performance analysis was carried out.Secondly the commercial vehicle cab was analyzed by using the physical test platform,and the same boundary conditions as the simulation were applied.Finally,the simulation values and experimental values were compared to verify the precision of the model.(2)The multidisciplinary integrated optimization design of commercial vehicle cab body-in-white was carried out.Firstly,in the cause of improving the efficiency of the later optimization,cross-sectional variables with a high impact on the cab structure were screened by cross-sectional sensitivity analysis,moreover,the relative sensitivity analysis was used to further screen the cross-sectional variables.furthermore,combined with the actual needs of the enterprise,the key thickness variables were also considered,and the key optimization variables were determined.Secondly,the Optimal Latin Hypercube method was used for obtaining the sample point data.Finally,the test function was used to verify the precision of common approximate models,and a research framework based on support vector machine approximate model was constructed.(3)The parameters with high dependence and difficulty in finding optimal solution of support vector machine were improved.Firstly,because of the speedy convergence,the particle swarm optimization algorithm was used to optimize the parameters.Secondly,the optimal solution found by the particle swarm optimization algorithm was likely to be a local optimal solution,therefore,the mutation factor was introduced to increase the probability of the global optimal solution.After that,the precision of the approximate model before and after the introduction of the mutation factor was compared by using the obtained sample point data,and the results indicated that the particle swarm-support vector machine approximation model technique with mutation factors was more advantageous.Finally,in cause of obtaining a more accurate approximation model,the ensemble of surrogates based on improved support vector machine was proposed,and it was convenient for lightweight optimization of commercial vehicle cab.(4)The lightweight design of commercial vehicle cab using the ensemble of surrogates was carried out.In order to improve the accuracy of the single approximation model,especially to solve the low accuracy in the first-order bending modal frequency of the research team,a combined approximation model of response surface-improved support vector machine was constructed,and the weight coefficients of quality,basic static and dynamic performance were solved respectively,it was verified that the precision of the ensemble of surrogates met the requriements,and the deterministic coefficient of the first-order bending modal frequency was higher than 0.95.Finally,the second-generation genetic algorithm was applied to optimize the lightweight design of the commercial vehicle cab,and the results indicated that the mass of the cab was reduced by 6.76%,the bending stiffness was increased by 7.07%,the torsional stiffness was increased by 4.44%,the first-order torsional modal frequency was increased by 6.14% and the first-order bending modal frequency was increased by 6.62%.It had achieved good lightweight effect and had certain research value in practical engineering application.