Keywords: fluid-structure interaction, bridge aeroelasticity, flutter, wind tunnel model.

As computational fluid dynamics (CFD) software has been under great development in the past ten years and it seems possible to use them for civil engineering applications. Different kinds of problems were considered in our study. The main goal is to determine the forces on an arbitrary body due to airflow around it. These forces can be quasi-steady or unsteady. Generally structural deformations do not have significant role on the flow field around the structure. In these cases flow forces can be evaluated on fixed bodies. However, in case of flexible buildings, deformations can modify forces induced by the wind, which is a fluid-structure interaction problem. These issues can be handled by the FLUENT and the ANSYS commercial software.

In civil engineering practice the use of codes and standards has a great importance. When we face the problem of determining wind loading on buildings we can use the Eurocode for instance. It has proposals for a wide range of structures but it has also limitations. Thus, CFD codes were involved in our studies in order to analyze complex structures that are not included in the standards.

First of all, in order to validate the CFD software, a basic study was performed; a 2D circle cross section was calculated with uniform airflow around it. The main question was the alternating force due to the vortex shedding. We analyzed flow around the circle and rectangle. Constructing a computational grid is one of the most critical issues in CFD simulations. The grid must be dense enough at the wall so as to be able to capture the high gradients and coarse enough to have an acceptable computational time. We applied the realizable k-epsilon turbulence model in an URANS (unsteady Reynolds averaged Navier Stokes) simulation. We compared the results with that from the literature and the Eurocode. The peak values are the force coefficients and the Strouhal-number can also be obtained.

Secondly oscillating objects were calculated with special focus on the flutter instability. As a fully aeroelastic wind tunnel model was created, it was possible to study its aerodynamic performance. The flutter derivatives for the cross section of the bridge were determined. Alternatively a flat plate with the same width was also investigated. The flutter derivatives of that were evaluated and compared with analytical solutions. Good agreement was found. Special effects of mesh and turbulence models were also studied. With the flutter derivatives the critical wind speed for the wind tunnel model was determined using an updated classical method; by applying the dynamic parameters of the bridge model calculated with the ANSYS software, an eigenvalue problem developed for two degrees of freedom system was extended to three-dimensional systems. The calculated and measured critical wind speeds were in agreement.

In the near future the three-dimensional fluid-structure interaction simulation for bridge flutter assessment will be developed.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £125 +P&P)