Computational fluid dynamics-based design optimization of water scooping system in amphibious aircraft by a parallel active learning surrogate model
Mengchuang Zhang,Zhizhuo Zhang,Shasha Xia,Xiaochuan Li,Qin Yao,Zhiping Yin
DOI: https://doi.org/10.1016/j.oceaneng.2024.116820
IF: 5
2024-02-02
Ocean Engineering
Abstract:Amphibious aircraft offer advantages in forest firefighting near coastal areas due to the maneuverability, precise delivery, and sufficient water source. However, when aircraft skim over the complex sea surfaces to scoop water, the fluid dynamics become highly unstable, resulting in an imbalance in water distribution on four tanks, which is susceptible to the design of water-bailing bucket and connecting tubes, so-called scooping system. In this study, the fluid-structural properties are investigated through transient one-way CFD analysis. Then the optimization design is conducted by introducing a novel active learning Kriging (ALK) model, which is coupled with parallel computing, novel constrained expected improvement (CEI) learning function and adaptive candidate region (ACR), to conduct the CFD transient analysis with an acceptable error. The results show that the optimal asymmetric design can reduce mass flow rate differences by 70.5 % and maximal von Mises stress by 23.1 %, while the values are 71.8 % and 20.9 % for symmetric design, respectively. Additionally, topological optimization for a simplified tube model reveals that the design of a spiral blade inside the tube can further enhance amphibious aircraft performance. This study offers valuable insights into the engineering of amphibious aircraft for maritime applications, laying the groundwork for future research endeavors.
engineering, civil, ocean, marine,oceanography