CFD based evaluation of the lock-in phenomenon of a bridge under wind load

The analysis of the dynamic behaviour of long span bridges under wind excitation is usually performed on the basis of experimental tests on physical models in wind tunnels. As an alternative to such procedure, some numerical methodologies have been developed, namely in terms of flutter analysis, though they are still based on some coefficients (flutter derivatives) whose evaluation still involves usually the use of experimental tests.

An attempt to overcome such limitations consists in using different algorithms of Computational Fluid Dynamics (CFD) for the evaluation of force and Scanlan coefficients.

After recent progress in computer technology, the authors could develop and implement a new numerical methodology for the aeroelastic analysis of slender structures. This computational algorithm is a time incremental approach based on two numerical algorithms working together: one of them determines the fluid flow action and the other one evaluates the structural response.

However, most of the studies performed deal with the evaluation of the critical velocity, also known as critical flutter velocity. This procedure can be understood as a verification of the structural safety in terms of ultimate limit state. But, as mentioned by the Eurocodes, it is also needed to verify the serviceability limit states of vibrations caused by wind action, unless the effective span is short enough. In the particular case of very flexible bridges under wind action, this verification can be done in terms of undesirable effects for users (discomfort), comparing the evaluated acceleration (or velocity) peak values of movements with human body acceptance criteria for vibrations. These unknown peak values are achieved by checking the maximum amplitude in the range of synchronized phenomena (lock-in) due to vortex-shedding. In this context, this paper presents the application of the above mentioned computer algorithm to the evaluation of the serviceability conditions of a simply supported bridge with a rectangular cross-section (B/D=6), under wind load considering their fundamental frequency. In particular, it will be evaluated the range of this synchronized phenomena, the peak value of acceleration obtained and: i) the time to start the phenomena; ii) the fully developed time; iii) and the corresponding time step.. Some of the most interesting results associated with the evaluation of the corresponding acceleration peak values of movements are presented, and compared with available human body acceptance criteria for vibrations (comfort evaluation) listed in the bibliography.