Surrogate-based aerodynamic shape optimization of a morphing wing considering a wide Mach-number range

Aerospace Science and Technology(2022)

引用 11|浏览15
暂无评分
摘要
Aerodynamic shape design considering both subsonic and hypersonic performance is a challenge in the field of aerospace. Many studies have shown that morphing wing is a promising technology to address this challenge. In this article, a new morphing mechanism which changes both the plane shape (such as span and sweep angle) and the profile (such as chord length and relative thickness) of the wing is proposed and a multi-objective optimization design for this new morphing mechanism is conducted based on surrogate-based optimization (SBO) algorithm. In the optimization design, three different configurations of the morphing wing which are the 36-, 45- and 70-degree sweep angle state, respectively and four different flow conditions which are subsonic flow (0.25 Ma), transonic flow (0.85 Ma), supersonic flow (1.2 Ma) and hypersonic flow (6 Ma) are considered. 36-degree sweep angle state is used for subsonic flow, 45-degree sweep angle state is used for transonic flow and supersonic flow and 70-degree sweep angle state is used for hypersonic flow. Subsonic lift and transonic, supersonic, hypersonic lift-to-drag ratios are objectives of this optimization design which has up to 64 design variables. To reduce computational cost of high-fidelity CFD (computational fluid dynamics) simulations and find the global optimum, Kriging surrogate model combined with a parallel infill-sampling method and a multi-round strategy are employed in this study. At last, the optimized morphing wing is evaluated and compared with the baseline and a fixed wing which is a moderate sweep wing. Results show that the optimized morphing wing features a very good aerodynamic performance improvement over the flow regimes from subsonic to hypersonic flow conditions.
更多
查看译文
关键词
Morphing wing,Deformation structure,Aerodynamic optimization,Kriging surrogate model
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要