The University of Electro-Communications
Muroran Institute of Technology
Yokohama National University
FUJITSU LABORATORIES LTD.
FUJITSU LABORATORIES LTD.
Japan Aerospace Exploration Agency (JAXA)
Japan Aerospace Exploration Agency (JAXA)
Japan Aerospace Exploration Agency (JAXA)
Japan Aerospace Exploration Agency (JAXA)
Japan Aerospace Exploration Agency (JAXA)
49th Fluid Dynamics Conference /the 35th Aerospace Numerical Simulation Symposium (June 28-30, 2017. National Olympics Memorial Youth Center), Shibuya-ku, Tokyo, Japan
抄録（英）
The transonic buffet degrades the aerodynamic performance of the aircraft during cruise. It is a phenomenon that should be avoided absolutely as it may lead to accidents. However, the mechanism of occurrence has yet to be elucidated. To understand this phenomenon, large-scale unsteady data is accumulated using computational fluid dynamics. In contrast, data mining of time series data such as unsteady data is a recent topic in that field. In this study, we have attempted to perform unsteady data mining with capacity exceeding Tera’s order. As a result, the behavior of the physical quantity before the transonic buffet arises was suggested to be different from the data just before the transonic buffet occurs. Moreover, a topological data analysis revealed abnormalities before the temporal origination determined by visualizing flow structure. Based on the results, we visualized the data over time, and found the characteristic change of the viscosity distribution on the wing surface. This should be a clue to elucidate this phenomenon.
内容記述
形態: カラー図版あり
内容記述（英）
Physical characteristics: Original contains color illustrations
A first attempt for detecting transonic buffet signature via unsteady-data mining
AA1730011020.pdf
(11.7MB)
