Flutter prediction based on dynamic eigen decomposition of coupled CFD-CSD flight simulation
During the design and analysis of aircraft flutter boundaries are typically computed using the p-k iterations, k iterations, p iterations, or eigenvalue analysis, that require computationally expensive and numerically sensitive procedures especially for complex configurations. In this work, the Dynamic Eigen Decomposition (DED) and a frequency domain stability theorem developed previously will be applied to a coupled CFD-CSD simulation to predict the aircraft flutter. It is known that the dynamic eigenmodes (DE) of the aeroelastic system are an intrinsic property of the system independent of dynamic pressure and flutter mode is identical to one of the DEs. Thus, the aeroelastic instability can be predicted by simply extrapolating the corresponding dynamic eigenvalues obtained at low, subcritical dynamic pressures. We will extend the theory to show that this is also true in the case of two parameters where air speed as well as air density vary. The proposed scheme is applied to computationally simulated flight data of F-16 wing model with a bending and a torsion mode shape subjected to a vertical gust in a transonic flow. It is shown that the new approach produces very accurate flutter predictions with the gust response. It is also shown that the single parameter variation of dynamic pressure based on a mean, constant speed of sound can yield sufficiently accurate flutter results.