Tohoku University Graduate School
Tohoku University Department of Aeronautics and Space Engineering, Faculty of Engineering
Tohoku University Department of Aeronautics and Space Engineering, Faculty of Engineering
出版者
航空宇宙技術研究所
出版者(英)
National Aerospace Laboratory (NAL)
雑誌名
航空宇宙技術研究所特別資料
雑誌名(英)
Special Publication of National Aerospace Laboratory
巻
34
ページ
113 - 118
発行年
1997-01
会議概要(会議名, 開催地, 会期, 主催者等)
航空宇宙技術研究所 6-7 JUN. 1996 東京 日本
会議概要(会議名, 開催地, 会期, 主催者等)(英)
National Aerospace Laboratory 6-7 JUN. 1996 Tokyo Japan
In the field of aerospace engineering, the aircraft design process can be divided into three stages: conceptual, preliminary and detail designs. The conceptual design is the first stage to decide a direction of design from mission requirements. In this process, the initial sizing of the aircraft is estimated from existing database and empirical methods. Since there are various demands for aerodynamics, structures, fuel, payload, and so on, a multiobjective optimization is necessary for the design. Therefore, in this paper, an optimization method is investigated to design wing planforms for this stage. In this work, three objectives are considered for wing planform optimization: aerodynamic performance, wing weight, and fuel tank volume. Aerodynamic performance is evaluated by CFD (Computational Fluid Dynamics) analysis. Wing weight and fuel tank volume are estimated by empirical formulae. To simplify the problem, only the leading-edge sweep angle and two chord lengths at the midspan and wingtip are used as design variables. The leading edge of the wing is kept straight. Thus, a tapered wing with a trailing-edge kink will be designed. As an optimizer, Genetic Algorithm (GA) is employed here. GA simulates evolution by selection. Because GA can find a more global optimum, it is expected to perform better than the conventional gradient-based methods. Major drawback of GA is that it requires a large number of function evaluations. Thus, FLO-22 code based on the full potential equation is used because of its efficiency instead of Euler or Navier-Stokes solver. Planform optimization problems are considered for typical subsonic aircraft. The design results show the feasibility of the present approach for the initial sizing with optimization.