基于概念设计的车身装配结构拓扑优化理论和方法的研究

In the body concept design phase,in order to realize the optimal design for body and optimal design for body manufacture, the technique is required that can combine the factors of body structure performance, design for assembly (DFA) and design for manufacturability(DFM) to determine the assembly structure and connect arrangement for body and get reasonable layout for gobal structure of the body finally. It is the very important for body of the detailed design and manufacturing.So, this project is to do research on on vehicle assembly topological structure optimization design. Firstly, the assembly structure of body in white (BIW) including beams, panels and connections will be defined as a topology design domain . Then it will be converted to a relation connection diagram that the beams will be expressed as vertices, and connectors will be expressed as the edges. Secondly, with defining body stiffness, the assembly costs and manufacturability as optimization design objectives and defining the beams and connected graphs in relation diagram as design variables, the topological optimization model can be established for white body assembly structure arrangement. At last, the modified fast and elitist multiobjective genetic algorithm(NSGA-II) will be design to solve the optimization model based on the semantic crossover figure decomposition algorithm that can realize the division of parent and child node of the reproductive and mutations.

为了实现车身结构的最优化设计，需要在概念设计阶段结合车身结构性能、车身的可装配性和可制造性等因素，确定车身装配拓扑结构和连接布置，以最终获得合理的车身总体结构布置方案。本项目拟以车身装配拓扑结构的优化设计为研究目标，首先将白车身装配结构定义为由梁、板和连接部件组成的拓扑设计域，并表示为以梁和板为顶点、以连接关系为边的装配联接图；再以车身刚度、装配成本和可制造性的相关指标为优化设计目标，以联接图中板、梁和连接为设计变量，建立车身装配结构布置设计的拓扑优化模型；最后采用修正的非支配分类遗传优化算法NSGA-II进行多目标优化求解，基于图分解语义的交叉算法实现父结点的分裂和子节点的繁殖和突变，最终将得到最优的Pareto解集，实现白车身装配结构的最优设计。申请者还将开发实现车身装配结构设计软件系统。本项目最终将实现对车身合理的装配结构布置，为车身正向设计技术提供有力的支持。

车身装配结构的设计是车身结构设计中的核心技术，需要综合车身结构性能、车身的可装配性和可制造性等因素，来确定车身装配拓扑结构和连接布置，最终获得合理的车身总体结构布置方案，这对于实现车身结构轻量化有重要意义。本项目在国家自然基金资助下，以概念设计阶段车身装配拓扑结构的优化设计为研究目标，首先将白车身装配结构定义为由梁、板和连接部件组成的拓扑设计域，并表示为以梁和板为顶点、以连接关系为边的装配联接图；再以车身刚度、装配成本和可制造性的相关指标为优化设计目标，以联接图中板、梁和连接为设计变量，建立车身装配结构布置设计的拓扑优化模型；最后采用修正的非支配分类遗传优化算法NSGA-II 进行多目标优化求解，基于图分解语义的交叉算法实现父结点的分裂和子节点的繁殖和突变，最终得到最优的Pareto 解集，实现了白车身装配结构的最优设计。本项目的一个重要成果是结合车身结构概念模型提出了改进图分解方法，有效的解决了装配结构离散优化不收敛的问题，同时课题组还将该方法推广到平台模块车身装配结构的定义。相关成果发表在Structural and Multidisciplinary Optimization, International Journal of Vehicle Design等多个重要学术期刊，并申请了相关技术专利。课题组还自主的开发实现了车身装配结构设计软件系统，申请获批了软件著作权，并在企业技术部门进行了试用，效果良好。本项目最终实现了对车身合理的装配结构布置，为可车身正向设计技术提供有力的支持。