Expanding a Flutter Envelope Using Data From Accelerating Flight: Application to the F-16 Fighter Aircraft

Due to the destructive nature of flutter, flutter testing is a mandatory requirement for certification of both civilian and military aircraft. However, along with the complexity of newer aircraft, the time and cost associated with flutter testing has increased dramatically. Considering that many of the test techniques and analysis methods used to perform flutter testing date back to the 1950s and 1960's it may be time to take a fresh look at how flutter testing can best be accomplished. This thesis revisits flutter testing techniques and proposes an alternative to traditional flutter testing. The alternative uses flight test data from an aircraft that is performing an acceleration to clear the flutter envelope of the aircraft. Four academic issues arise from this new test approach. 1) Are frequencies and damping affected by the acceleration of the aircraft? 2) Can parameter identification algorithms extract frequency and damping values from the time varying data? 3) Can the vibration response at airspeeds (or Mach numbers) beyond which the aircraft has accelerated be anticipated? 4) What formal criteria can be used to determine when the aircraft needs to end the acceleration and terminate the test point? The academic contribution of this thesis is to address these issues. It is shown that although the frequencies and damping values do change the change is so small that it is irrelevant. It is also shown that by taking small windows of data, within which the change in parameters is small, it is possible to accurately identify parameters from the time varying data. Finally it is shown that at least in principal parameters can be predicted using data from sub-critical airspeeds, and that testing can be discontinued before an unstable flight condition is reached.