| Flight flutter testing is an essential part of the certification process for all new and significantly modified aircraft. As flutter is a potentially catastrophic phenomenon, it is vital that accurate modal parameter estimates are obtained at each flight condition, in order that the aircraft can be shown to be flutter-free throughout the flight envelope. However, the analysis of flight flutter test data is made difficult by the high levels of noise encountered and the limited time available for flutter clearance. The aim of this work is to develop a number of approaches to improve the safety, efficiency and accuracy of the flight flutter test procedure. In this thesis, the effects of additional wing-tip exciter vanes on the dynamic characteristics of a wing are evaluated. Wind-tunnel tests of a new design of flutter exciter vane have been undertaken and the results are discussed. The application of normal mode theory to aeroelastic systems is also described in order that the u.se of multiple-input force appropriation techniques during flight flutter tests may be considered. The analysis of the response to turbulence excitation is discussed and a comparison is presented, using real and simulated data, of a number of methods which estimate modal parameters purely from the response to an unknown white input. Extensions of the most successful methods are described for cases of non-white excitation and corruption signals. Finally, two techniques are presented which aim to increase the efficiency with which flutter clearance is achieved. The first is an automated approach to the tracking of modal estimates between progressive flight test points; the second is an on-line method for monitoring the overall stability of an aircraft at changing flight conditions. |
