基于序列近似优化方法的加筋圆锥壳结构优化设计

Structure Optimization Design of Stiffened Conical Shells Based on Sequential Approximate Optimization Method

为提高运载火箭圆锥壳舱段承载效率,开展轴压载荷下加筋圆锥壳结构轻量化设计。首先,建立加筋圆锥壳结构有限元模型,采用非线性显式动力学算法开展分析;然后,对比分析加筋圆锥壳结构参数灵敏度指标及中间框径向和垂直安装方式对结构极限承载性能的影响;最后,基于径向基函数代理模型和Kriging代理模型序列近似优化方法开展加筋圆锥壳结构优化设计。结果表明:腹板高度、桁条数量、翼缘厚度及耦合效应是加筋圆锥壳极限承载能力的关键影响因素,而中间框径向和垂直安装方式对结构极限承载性能的影响不显著;采用序列近似优化方法经迭代优化后,相对初始设计减重9.40%,验证了方法的有效性。

To improve the load-carrying capacity of the conical shell structure of launch vehicles,the optimization design of stiffened conical shell structure under axial compression was carried out.First,the finite element model of stiffened conical shell structure was established and analyzed by nonlinear explicit dynamic method.Then,the influence of structural parameter sensitivity index and radial and vertical mounting mode of middle frame on ultimate bearing capacity of stiffened conical shell was analyzed.Finally,the structural optimization of stiffened conical shells was carried out based on radial basis function surrogate model and Kriging surrogate model sequence approximate optimization method respectively.The results showed that the structural parameters and coupling effect of web height,truss number and flange thickness were the key factors influencing the ultimate bearing capacity of the stiffened conical shell,while the radial and vertical installation of the middle frame had no significant influence on the ultimate bearing capacity of the structure.Besides,compared with the initial design,the weight was reduced by 9.40%after optimization using the sequential approximate optimization method,which verified the effectiveness of the method.