Predicting Dynamic Instability Boundaries Using Lattice Filters

This paper discusses the use of lattice filter formulations for computing instability boundaries from subcritical test data. Lattice filters have the inherent properties such that the instability mechanism does not have to be known a priori and the model order can be self adjusting for the data being observed. The filters are numerically efficient and there exist lattice filter forms where if the excitation force on the structure is Gaussian noise over the bandwidth of interest, then the input forces do not need to be measured. Resuits from an application of the lattice filter formulation on a high speed wind tunnel flutter model test show that the projected boundaries compare accurately with actual flutter boundaries encountered during the test.