Content area
Abstract
The existence of nonlinearities in aircraft causes a phenomenon known as Limit Cycle Oscillation (LCO) to occur instead of the classical flutter phenomenon which is observed if the aircraft is linear. Flutter leads to a structural failure, while LCO can lead to fatigue problems. In the context of flight flutter testing it is very difficult to distinguish between the two phenomena, as both are first characterised by bounded oscillations. Due to safety reasons one cannot risk flying the aircraft at higher airspeeds to check whether the bounded oscillations will give rise to flutter or LCO. The current flight flutter testing procedure is based on verifying that these bounded oscillations will not occur inside the designed flight envelope. The methodology consists of continuously predicting the conditions at which the bounded oscillations would occur until the whole flight envelope is cleared. This thesis evaluates the performance of standard linear methods of data analysis to perform flight flutter tests on nonlinear aircraft and proposes a new method of analysis to give more accurate results. The possible approaches for the implementation of the proposed method are evaluated and compared. A simulated idealised aeroelastic system containing cubic stiffness nonlinearity is used to evaluate the performance of the existing and proposed methods. Methods for detection and quantification of nonlinearities are considered as tools to substantiate the use of linear or nonlinear methods for data analysis. Some of the methods evaluated on simulated systems are applied to flight flutter test data from a suspected nonlinear aircraft and the existence of nonlinearities is confirmed at some portions of the flight envelope. A novel flutter exciter is designed, manufactured and tested with the aim of providing random excitation in flight in order to improve the quality of the measured signals and allow the application of different identification methods.
