This paper presents a numerical strategy for reliability-based design optimisation of an aircraft wing structure using a surrogate-assisted approach. The design problem is set to minimise aircraft wing mass subject to structural and aeroelastic constraints, while design variables are structural dimensions. The problem has uncertainties in the material properties. The Kriging model is used for estimating the values of design functions. Two strategies of sampling technique are used, i.e., optimum Latin hypercube sampling (OLHS) with and without infill sampling. Uncertainty quantification is achieved by means of optimum normal distribution Latin hypercube sampling. The original design problem is converted to be a multiobjective optimisation problem. Optimum results show that OLHS with infill sampling gives a more accurate surrogate model; however, OLHS without infill sampling results in the better design solutions based on actual function evaluations.

中文翻译：

---

具有静态和动态气动弹性约束的机翼代理辅助可靠性优化

本文提出了一种使用代理辅助方法对飞机机翼结构进行基于可靠性的设计优化的数值策略。设计问题旨在将受结构和气动弹性约束的飞机机翼质量最小化，而设计变量是结构尺寸。该问题在材料特性方面具有不确定性。克里金模型用于估计设计函数的值。使用了两种采样技术策略，即具有和不具有填充采样的最佳拉丁超立方体采样（OLHS）。不确定性量化是通过最佳正态分布拉丁超立方体采样实现的。原设计问题转化为多目标优化问题。优化结果表明，带填充采样的 OLHS 提供了更准确的代理模型；然而，