Design datamining is demonstrated through multipoint aerodynamic design of supersonic wings. Tradeoff information for the design is gathered by Multiobjective Evolutionary Algorithms (MOEAs) and visualized by Self- Organizing Maps (SOMs). Four design objectives are considered: Aerodynamic drags are minimized at both supersonic and transonic cruise conditions under lift constraints. Bending and pitching moments are also minimized for structure and stability considerations. MOEAs are first performed by using 72 design variables . SOM is then applied to map the resulting Pareto solutions obtained in the four dimensional objective function space to two dimensions. This reveals global tradeoffs between four design objectives. Furthermore, the relations between design variables are mapped onto another SOM. The resulting SOMs are confirmed to perform aerodynamic design datamining properly.
CITATION STYLE
Obayashi, S. (2002). Pareto Solutions of Multipoint Design of Supersonic Wings Using Evolutionary Algorithms. In Adaptive Computing in Design and Manufacture V (pp. 3–15). Springer London. https://doi.org/10.1007/978-0-85729-345-9_1
Mendeley helps you to discover research relevant for your work.