To improve ground-testing tools (e.g., wind-tunnel testing and computational fluid dynamics) that predict the aerodynamic characteristics of real aircraft, it is necessary to explore the differences between data obtained from ground tests and those obtained from flight tests. However, it is difficult to accurately analyze the causes of differences between flight test data and ground test data because there is larger scatter in static aerodynamic data from flight tests compared with those from computational fluid dynamics and wind-tunnel testing. One reason for this is that flight conditions and measurements change according to the flight environment, leading to the data having a multimodal distribution. In this study, monophasic data extracted from flight test data are examined, using a cluster analysis to obtain data with small scatter. Hierarchical cluster analysis was applied to the dataset for several parameters obtained in flight testing in order to consider changes in multiple variables. The results show that proper categorization was possible for cases with large changes during measurement, providing monophasic data with drag coefficient variation of approximately 15–20 drag counts and demonstrating the utility of cluster analysis.
ASJC Scopus subject areas
- Aerospace Engineering