The aerodynamic instability of suspension bridges with a rectangular, a trapezoidal and a hexagonal cross-sections respectively were investigated based on the concept of a section model. Measurements of the root-mean-square responses in the vertical and torsional direc-tions of three bridge models at various wind speeds were conducted firstly in a wind tunnel and the results were obtained to confirm those from the parallel numerical simulations. After the validity of the numerical method was verified, both the experimental and numerical results were used to examine the flow effect as well as the aeroelastic behavior of the bridges in de-tail, and further to assess and compare the stability performance of the bridges with the three different shapes. Results show that the predictions of the root-mean-square bridge responses agree well with those from the section model measurements, indicating that the present numerical method can be used to analyze the problem with good accuracy. In addition, based on the numerical re-sults in time series, the variations of net damping ratios and flutter derivatives associated with the three bridge decks are analyzed extensively. Finally, it is found that the rectangular deck leads to the least flutter speed or the worst aerodynamic instability among all the cases. On the other hand, the hexagonal deck, with an increase of about 17 % in terms of the critical flutter speed, possesses the best stability performance.

It is well known that the wake of a bluff-body behaves like a self-excited non-linear fluid oscillator above a critical Reynolds number ≈45, giving rise to periodic vortex formation and shedding at the Strouhal frequency [1]. Beyond the critical point further instabilities lead pro

A semiempirical study is made of the bluff-body problem. Some experiments with interference elements in the wake close behind a cylinder demonstrate the need for considering that region in any complete theory. Dimensional analysis of a simple model of the region leads to a universal Stromal number

for the 2D intermediate and far bluff body wake

Three different PDF algorithms have been applied to investigate a constant-density bluff-body stabilized flow using the same turbulence models and the same boundary conditions. The objectives of this paper are to compare the three algorithms in terms of numerical accuracy and efficiency

Abstract. The understanding and quantitative prediction of velocity and pressure fluctuations in turbulent flows around such bluff bodies have been evolving over the years. The main aim of the present work is to investigate experimentally and numerically the flow field in the wake region

With the rapid increase of bridge span length, bridge structures are becoming more flexible, which requires bluff body aerodynamics studies related to bridge deck flutter instability and vortex induced vibration, as well as stay cable vibration. Aerodynamic stabilization of several suspension

Abstract: Drag reduction of road transport vehicles is a societal need that justifies increased interest in bluff body flows. The suggestion of Hoerner et al that thin boundary layers lower the pressure at the back of bluff bodies has been the inspiration for an experimental campaigne to study

ABSTRACT The effect of several drag reducing devices on the near wake of a generic ground vehicle body was investigated. Drag and base pressure measurements were conducted to identify the effects of the devices on the base drag. A Particle Image Velocimetry (PIV) study was conducted to determine

Two aspects of turbulence modelling were addressed with re-spect to a single cylinder in crossflow. Firstly, the effect of vary-ing the turbulent length scale at the inlet was investigated at a high subcritical Reynolds number of 1.4×105. Variations of up to 14 % were noted in the flow properties such as mean drag and Strouhal number, but significant discrepancy between experi-mental and computational results remained. Secondly, a modifi-cation to the standard k-ω turbulence model was assessed. This time-limit model modified the turbulent viscosity term, inhibit-ing production of turbulent kinetic energy in areas of high strain rates. Tuning of an empirical term was required to match flows at various Reynolds numbers. The time-limit model appeared to offer improvements at some Reynolds numbers compared to the standard k-ω model by reproducing real flow structures such as separation bubbles at a transcritical Reynolds number. Use of the time-limit model is in its infancy, especially in unsteady flows, but appears to have some potential.