Improved ERA Based Identification of Flutter Derivatives of a Thin Plate

Abstract

In this paper, improved Eigensystem Realization Algorithm (ERA/DC—Eigensystem Realization Algorithm with Data Correlation) from literature has been applied for the identification of flutter derivatives of a thin plate section under free decay (transient) responses, to overcome the problem of poor noise handling capability of original Eigen Realization Algorithm (ERA). A thin plate section has been considered in the present study as theoretical solutions of flutter derivatives are available for it from Theodorsen’s functions and also most of the bridge deck cross-sections have large side ratio. The 2-DOF (heave and pitch) aeroelastic responses of thin plate section have been numerically simulated using Newmark-beta time integration method for various reduced wind speeds (U/fB; U—oncoming wind speed, f—frequency of oscillation, B—width of the plate), using flutter derivatives obtained from Theodorsen’s theoretical solution. Then, ERA and ERA/DC, have been applied towards identification of frequencies and damping ratios for the chosen range of reduced wind speeds and for 5, 10 and 20 % noise levels. Further, flutter derivatives have also been identified and compared with theoretical values to prove the robustness of the improved Eigen Realization Algorithm (ERA/DC) in identification of flutter derivatives and in handling noise.