制造约束下变厚度变截面车身结构的碰撞吸能机理及可靠性优化设计

Lightweight and crashworthiness are the heart for the development of automotive body. Automotive body structures with variable rolled blank and variable cross-sectional shape (VRB-VCS) can improve the lightweight and crashworthiness performances of vehicle. However, literatures on the crashing energy absorption mechanics and reliability-based design optimization of VRB-VCS structures is very limited. This project is performed under axial impact loading and manufacturing constraints (e.g., flexible rolling process, stamping and assembling). Firstly, front longitudinal beam of automotive body is taken as the research object, and a highly accurate parametric model of VRB-VCS structures is established, the accuracy and effectiveness of the model are validated by performing axial impact tests. Secondly, an enhanced optimal Latin hypercube sampling (E-OLHS) algorithm is proposed to meet the highly requirement of manufacturing constraints. Design of experiment is performed by the proposed E-OLHS algorithm to study influences of thickness distribution parameters and cross-sectional shape parameters on crashworthiness of VRB-VCS structures, and the axial deformed mechanics and energy absorption characteristics of VRB-VCS structures are then revealed. Finally, mathematical model of multi-objective reliability-based design optimization (MORBDO) for VRB-VCS structures is established under manufacturing constraints. A new efficient algorithm to solve the model is proposed by integrating multi-modal radial-based importance sampling algorithm and improved hybrid meta-models based pareto frontier pursuing algorithm. To sum up, an efficient MORBDO methodology for VRB-VCS structures is established. The researches of this project can not only enrich the existing theory of crashworthiness design for automotive body under frontal impact loading, but also promote the application of the VRB-VCS structures in engineering.

轻量化和耐撞性是车身开发过程中的关键问题。变厚度变截面车身结构能有效提升汽车的轻量化水平和耐撞性能，但对其碰撞吸能机理及可靠性优化设计方法的研究较少。本课题在轴向碰撞工况和制造约束条件下，以乘用车车身前纵梁为研究对象，建立变厚度变截面车身结构的高精度参数化模型，通过碰撞试验验证其精度和有效性；针对变厚度变截面车身结构的制造约束要求高的特点，提出带约束设计域的改进最优拉丁方抽样算法，研究不同设计参数及其组合对变厚度变截面车身结构耐撞性的影响规律，揭示其结构的轴向变形机理和吸能特性；建立制造约束下变厚度变截面车身结构多目标可靠性优化设计模型，探索多模式径向基重要性抽样算法与基于混合元模型的帕累托前沿搜索算法的高效协同机制，建立面向变厚度变截面车身结构多目标可靠性优化设计的高效求解方法。本课题研究成果对于丰富和完善汽车正面碰撞耐撞性设计方法体系和促进变厚度变截面车身结构的工程应用具有重要意义。

针对制造约束下变厚度变截面（VRB-VCS）车身结构存在的瓶颈问题和技术难点，提出了一种双幂函数型VRB板，揭示了可轧制约束下幂指数参数对双幂函数型VRB板刚度特性的影响规律；构建了制造约束下VRB-VCS车身结构的高精度参数化模型，制备了若干组等厚度、变厚度、变截面、变厚度变截面帽形梁样件，通过落锤冲击试验，验证了数值仿真模型的精度；提出了改进的约束域拉丁方抽样算法（ICD-LHS），解决了VRB-VCS车身结构在复杂约束域内的抽样难题，为开展VRB-VCS车身结构的吸能机理及耐撞性优化设计奠定了基础；利用ICD-LHS算法制定了不同厚度分布参数与不同截面形状参数的实验设计方案，揭示了VRB-VCS车身结构的轴向变形机理与吸能特性；为求解VRB-VCS前纵梁的碰撞安全可靠性优化问题，构建了可制造约束下VRB-VCS前纵梁的正面碰撞局部参数化子模型，提出了兼顾精度和效率的多模式径向基重要抽样方法（MRBIS），并将MRBIS集成于VRB-VCS前纵梁的多目标系统可靠性优化设计，成功解决了制造约束下VRB-VCS前纵梁的多目标系统可靠优化问题。针对VRB结构在弯曲碰撞工况下的优化问题，提出了求解可轧制约束下VRB薄壁结构最优厚度分布的混合元胞自动机算法（eHCA-VRB），开展了100个厚度变量的VRB结构耐撞性优化设计，结果表明：eHCA-VRB算法对于求解可轧制约束下包含大量厚度变量的VRB结构耐撞性优化问题，具有出色的收敛性和优化效率；为求解侧面碰撞工况下车身骨架的厚度优化问题，提出了求解车身厚度优化的子区域混合元胞自动机方法（T-SHCA）。此外，在本课题资助下，提出了钢铝混合单帽型梁的弯曲吸能理论预测模型，揭示了钢铝混合单帽型梁的弯曲吸能机理。本课题的研究成果为VRB-VCS车身结构在汽车上的推广应用，提供了强有力的理论基础及技术指导。