Study on extraction parameters of flutter derivatives for the development of a time-domain formulation of self-excited forces

Abstract eng:
This paper addresses the evaluation of the dependence of bridge-deck flutter derivatives to the extraction parameters used in wind tunnel tests. The studied parameters are the model scale, model velocity and amplitude of motion. The measurements of flutter derivatives for the Great Belt Bridge were done through forced-vibration tests. This research shows that the flutter derivatives measured for different model scales have similar trends. Some derivatives show a dependence with respect to the amplitude. When studied in their dimensional form, the flutter derivatives H *5 , A*5 , H *2 and A*2 exhibit nonlinearities for the velocity. Therefore, the study of a new nondimensionalization for the flutter derivatives based on the velocity would be interesting for the development of a new time-domain formulation of self-excited forces. INTRODUCTION Slender bridges, such as suspension and cable-stayed bridges, are flexible structures that are effective to resist earthquakes, but their flexibility makes them vulnerable to wind actions, such as the flutter instability. Flutter is an aeroelastic instability caused by an interaction between wind and bridge-deck motions. This instability can cause major damages and even a total collapse. The assessment of flutter is done by the analysis of self-excited forces, i.e., forces resulting from deck motions. Scanlan's formulation of self-excited forces is widely used for this purpose [1]. This force model is based on the flutter derivatives, which are experimental coefficients obtained from wind tunnel tests. From the fact that the flutter derivatives are functions of the reduced frequency, the use of Scanlan’s formulation is limited to linear analyses. However, slender bridges are nonlinear structures because of the geometrical nonlinearities of the cables. Material nonlinearities of the bridge structure can also arise from the large displacements of the deck associated with wind loads. Additionally, the self-excited forces show aerodynamic nonlinearities. Some researchers demonstrate the dependence of flutter derivatives with respect to the amplitude of motion of the bridge deck [2-4]. The flutter derivatives also exhibit a nonlinear behavior for the frequency of oscillation [4]. Including these nonlinearities in flutter analysis would lead to more realistic flutter predictions for cable-supported bridges. To do so, the development of a new time-domain model of self-excited forces will be pertinent. Prior to the development of a new time-domain approach, it is relevant to have a better understanding of the dependence of flutter derivatives with respect to the extraction parameters.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket