Study On Optimization Design For Automobile Safety Based On Variable Complexity Approximate Model

With the continuous updating of computer software and the development of the computer simulation theory,the accuracy of the finite element model of the body is increasingly high,the problems of sampling time and hardware upgrades are especially prominent.Nowadays there are two methods for optimization design of the body structure.One method is the directly sampling of the vehicle.It has higher accuracy and reliable response but longer time consuming.Another is the qualitative optimization of the part by the simplified model method,and the optimization result is substituted into the vehicle model to be verified.It is short time consuming and targeted,but the quality of the results depends for the most part on the experience of researchers because of the imperfection theory.Aiming at these problems,a new method used the optimization design of the body combines the advantages of both methods by the variable complexity optimization.the main contents are as following:(1)We compare the fitting accuracy of three approximate models for different optimization problems by numerical examples and engineering examples.The results show the fitting accuracy of the RBF response surface model is higher than others in case of linear,weakly nonlinear and strongly nonlinear.And the robustness strongly of that is more suitable for the engineering examples.(2)Based on the application conditions of the analysis and optimization of the vehicle frontal impact by the simplified model method,we study the effects of three factors in the deformation of the front stringer and the passenger compartment on a collision course.The results show the change of material and cross-section has greater impact on the deformation mode of the front stringer,while the change of thickness is more effective for the B column peak acceleration and lightweight design during the collision.Ultimately we choose the change of material to improve the front stringer under the principle of cost control.(3)Aiming at the optimization design of the vehicle safety,we proposed a new method that is variable complexity model of RBF metamodeling technique.a compensation response surface model was created by generating optimal Latin square design through the small sample data spanning the simplify model and high accuracy model.And then a new RBF response surface model was established on the basis of new test data which was received through the simplify model and compensation response surface model.The method is applied to vehicle 100% frontal impact multi-objective optimization design for the lightest quality and minimum body B column peak acceleration through selecting 7 parts which influence greater than the others of the front structure.The results show that the method can ensure the modelprecision,rapid convergence to the optimal solution.The results of this study show that,besides the performance of the body structure can be improved,the method by using variable complexity model of RBF metamodeling technique can reduce the sampling time and improve computational efficiency,and it has a strong engineering practicality.This method also provides a reference method for body lightweight design.