| Abstract | Mode switching is often observed in the flutter analysis of complex aircraft configurations when using the PK-method implemented in commonly applied commercial software such as NASTRAN. In NASTRAN, the computed eigenroots are sorted in increasing frequencies. Therefore, the appearance of many mode-switching instances is not avoidable in the PK-method flutter results, especially for real-world complex engineering applications. In this study, an extensive in-house sorting capability was created in order to remedy NASTRAN’s lack of a mode-tracking (hence mode sorting) procedure between the airspeed steps. The mode-sorting capabilities were developed based on both the complex eigenvalues and eigenvectors. Other numerical techniques were further developed for the traditional PK-method used in flutter analysis. A hybrid approach was introduced for the initial guess of the reduced frequency, and a deferred correction scheme was employed for the PK iteration process. Also, special care was taken for mode matching when locking eigenroots onto aerodynamics within the PK iterations. These special techniques effectively improved the numerical stability of the eigensolution process and significantly reduced the probability of mode switching. As a result, the eigensolution minimized the severity of the misleading mode switching observed in flutter analyses reducing risks in flight. |
|---|
