Accurate closed-form eigensolutions of three-dimensional panel flutter with arbitrary homogeneous boundary conditions

Abstract

Highly accurate closed-form eigensolutions for flutter of three-dimensional (3D) panel with arbitrary combinations of simply supported (S), glide (G), clamped (C) and free (F) boundary conditions (BCs), such as cantilever panels, are achieved according to the linear thin plate theory and the first-order piston theory as well as the complex modal analysis, and all solutions are in a simple and explicit form. The iterative Separation-of-Variable (iSOV) method proposed by the present authors is employed to obtain the highly accurate eigensolutions. The flutter mechanism is studied with the benefit of eigenvalue properties from mathematical senses. The effects of boundary conditions, chord-thickness ratios, aerodynamic damping, aspect ratios and in-plane loads on flutter properties are examined. The results are compared with those of Kantorovich method and Galerkin method, and also coincide well with analytical solutions in literature, verifying the accuracy of the present closed-form results. It is revealed that, (A) the flutter characteristics are dominated by the cross section properties of panels in the direction of stream flow; (B) two types of flutter, called coupled-mode flutter and zero-frequency flutter which includes zero-frequency single-mode flutter and buckling, are observed; (C) boundary conditions and in-plane loads can affect both flutter boundary and flutter type; (D) the flutter behavior of 3D panel is similar to that of the two-dimensional (2D) panel if the aspect ratio is up to a certain value; (E) four to six modes should be used in the Galerkin method for accurate eigensolutions, and the results converge to that of Kantorovich method which uses the same mode functions in the direction perpendicular to the stream flow. The present analysis method can be used as a reference for other stability issues characterized by complex eigenvalues, and the highly closed-form solutions are useful in parameter designs and can also be taken as benchmarks for the validation of numerical methods.