Research On Uncertainty Optimization Method And Its Application In Design Of Automotive Body Structure Of Tailor Rolled Blank

With the rapid increasing of automobile production and sales,environmental pollution,energy crisis and traffic safety have become important issues facing the development of automobile industry,and these issues are closely related to automobile lightweight.Tailor rolled blank(TRB)technology can flexibly adjust and optimize the thickness distribution according to the loading force or performance requirements of parts,and truly realize the “tailor made” of thickness distribution,thus providing the best solution for lightweight design of automobile.However,at present the research on the design and manufacturing of TRB structure is still at the starting stage,and the related theoretical model and optimization design method have not been established yet.It means that there is still a long way to go to guide the design and large-scale application of TRB structure in automobile industry.Therefore,it is of great theoretical significance and practical application value to carry out systematic research on the design method for vehicle security structure of tailor rolled blank.In order to guide the light-weight and crashworthiness design of TRB structure more effectively and efficiently,it is essential to understand deeply the mechanical characteristics of the new continuous variable thickness structure.In the thesis,the systematic research on energy absorption mechanism,high efficiency finite element(FE)modeling technology,multi-case reliability design optimization method and the reliability optimization design method coupled process-performance algorithm of TRB structure is performed.Finally,a novel efficient multi-objective discrete robust optimization algorithm is developed for typical discrete optimization problems in the process of vehicle body structure design.The main contents are as follows:(1)Research on the energy absorption mechanisms of TRB structure.In this thesis,the quasi-static/dynamic axial impact characteristics and quasi-static bending characteristics of TRB structure are primarily explored by using experimental method.Then based on the heterogeneity of thickness and material properties of TRB structure,the true stress-strain curves of TRB sheet with different rolling quantities are obtained by uniaxial repeated tensile test,and the effective stress strain field of TRB sheet was constructed.Afterwards,an efficient FE modeling method,which considers the difference of thickness and material properties of TRB structure simultaneously,is proposed and verified by the previous experimental data.Finally,the influence of thickness distribution and location of transition zone on the light-weight and crashworthiness of TRB structure under quasi-static/dynamic axial crushing and quasi-static bending conditions are studied systematically.(2)Research on the multi-case reliability-based design method of TRB structure under typical vehicle collision conditions.To improve the efficiency of light-weight and crashworthiness design of TRB structure and reduce the risk of design failure,the thesis organically combines the non-dominated sorting genetic algorithm(NSGA-II),Monte Carlo Simulation(MCS)and radial basis function(RBF)networks to develop a systematic multi-objective and multi-case reliability-based design optimization(MOMCRBDO)method.Then,the proposed method is applied to the design of TRB front longitudinal beam under typical collision conditions such as frontal impact,offset impact and side impact.The optimization results show that the deterministic optimization design can improve the crashworthiness of TRB structure,but the reliability of the obtained optimal solution is only about 50%.In contrast,although the use of MOMCRBDO results in some sacrifice in the crashworthiness of TRB structure,this method not only improves the reliability of Pareto solutions,but also enhances the robustness of optimal solutions under multiple loading cases.(3)Research on the reliability-based optimization design method coupled process-performance for TRB structure.To realize the efficient transfer of stamping process parameters and stamping effect into the crashworthiness model,the thesis developed a FE-based sequential coupleing method,and the accuracy of the proposed method is verified by comparing the thickness and residual stress/strain distribution before and after mapping.Afterwards,the method is applied to the crashworthiness analysis of front longitudinal beam.The numerical results show that the stamping effect,such as residual strain and thickness reduction,has great influence on the deformation mode and crashworthiness of the front longitudinal beam.Then based on FE sequential coupleing method,the thesis combines the multi-objective particle swarm optimization(MOPSO),Monte carlo simulation with descriptive sampling(MCSDS)and surrogate model technique to develop a general multi-objective reliability-based design optimization method coupled process-performance algorithm(MORBDOCP).Finally,the proposed method is applied to the optimization of the TRB vehicle structures.The results show that the proposed algorithm results in a loss of crashworthiness,but the method can accurately consider the volatility of the design variables and obtain the optimal solution that fully satisfies the process reliability constraints.(4)Research on the multi-objective discrete robust optimization design algorithm for vehicle body structure design.To efficiently solve multi-objective discrete robust optimization problems in real-life engineering problems,the thesis proposed a novel efficient multi-objective discrete robust optimization(MODRO)algorithm based on multi-criteria decision making technology and continuous Taguchi method.In the present MODRO procedure,grey relational analysis(GRA),coupled with principal component analysis(PCA),was used as a multicriteria decision making model for converting multiple conflicting objectives into one unified cost function.The optimization process was iterated using the successive Taguchi approach to avoid the limitation that conventional Taguchi method fails to deal with a large number of design variables and design levels.The proposed method was first verified by a mathematical benchmark example and a ten-bar truss design problem;and then it was applied to a more sophisticated design case of full scale vehicle structure for crashworthiness criteria.The results showed that the algorithm is able to ac hieve an optimal design in a fairly efficient manner attributable to its integration with the multicriteria decision making model.Note that the optimal design can be directly used in practical applications without further design selection.Due to its independence on metamodeling techniques,the proposed algorithm could be fairly promising for engineering design problems with high dimensions.