Aeroelastic flutter is a destructive instability phenomenon for which dedicated flight test campaigns are considered compulsory according to airworthiness regulations. In these tests, the dynamic characteristics of the aeroelastic system are interpreted from measurements of the aircraft’s structural responses to external excitations. Over the years, several flutter identification and prediction techniques have been suggested to increase the efficiency of flutter flight tests and to enable better prediction of the flutter boundary. While most of the methods rely on external mechanical excitation accessories, the Autoregressive Moving-Average (ARMA) flutter prediction method attempts to identify the aeroelastic system based on the aircraft structural response to random air turbulence excitation.
This study investigates the application aspects and the methodology of flutter prediction via ARMA system identification in a dedicated wind-tunnel experiment. For that purpose, an elastic wing was designed and manufactured using rapid prototyping. The wing was tested at the Technion’s subsonic wind-tunnel at subcritical points and all the way to flutter.
The current study presents the flutter prediction methodology and process as conducted numerically and in two wind tunnel experiments. The structural responses used for ARMA system identification are acceleration and strains (measured by strain gauges) in the first wind-tunnel test, and strains measured via optical fibers in the second test. The study presents comparison of the flutter prediction capability when based on discrete, spars measurement (accelerometers, strain gauges) versus that from more dense measurement (optical fibers). The study focuses on the practical aspects of the application of the method in wind-tunnel testing, which is also of great importance for flight testing.
