Research On Lightweight Of Body-in-White Based On Multi-objective Optimization

With the development of economy,the number of motor vehicles is also increasing.While being convenient for people,automobile also causes a series of social problems such as environmental pollution and energy crisis.In response to the national strategy of energy conservation and emission reduction,research on lightweight of automobile structure can reduce fuel consumption and achieve the purpose of environmental protection.As one of the main assemblies of a car,the car body has a large space for lightweight optimization.This paper uses computer simulation technology to carry out lightweight design of a certain body in white(BIW)of the electric vehicle,based on finite element theory and multi-objective optimization method.The main research work is as follows:The finite element model has been established in Hypermesh.By analyzing the modal frequency,mode shape and stiffness of body-in-white,the corresponding first-order torsional frequency,first-order bending frequency,torsional stiffness and bending stiffness have been obtained.Comparing with the same model at home and abroad,the low-order modal frequency of the body meets the requirements and will not cause resonance.At the same time,the stiffness also meets the design requirements,and has ability to resist torsion and bending deformation to a certain extent.The concept of relative sensitivity has been introduced and the sensitivity analysis has been carried out on 38 groups of components with large mass on BIW.The relative sensitivity values of first-order torsion,first-order bending,torsional stiffness and bending stiffness to the thickness of parts have been calculated,and the thickness of 21 groups of components whose absolute value of relative sensitivity are smaller has been selected as the optimal design variable.A multi-objective optimization mathematical model has been constructed based on the theory of body lightweight coefficient.Parameterizing the source files of modes and rigidity and importing them into Hyperstudy,The modified extensible lattice sequence method has been applied to sample.Based on the sample points,the least squares regression,moving least squares method,hyperkriging and radial basis function approximation model have been constructed.Comparing with the precisiones of each approximation model,the higher precision moving least square method,hyperkriging and radial basis function approximation models have been kept and obtained six groups of optimization schemes by cross-combination with global response surface method and multi-objective genetic algorithm.The optimal combination of approximate model and optimization algorithm have been obtained through optimization calculation of BIW respectively.The results show that the combination of moving least square method and multi-objective genetic algorithm has the best lightweight effect.The optimized results show that the mass of BIW has been reduced from 129 kg to 124.4kg,and the loss-weight 4.6kg,the lightweight coefficient has been reduced from 3.81 to 3.73.At the same time,the modal frequency and the stiffness performance of the body still meet the design requirements,which proves the feasibility of the optimized scheme and has certain engineering reference value.Figure [44] table [19] reference [83]...