Buffeting response of long span bridges: numerical-experimental validation of fluid-structure interaction models

Estimates of the buffeting response of long-span bridges rely on numerical models that require as input the dynamic finite element model of the bridge, the turbulence characteristics of the incoming wind, and the model for the fluid-structure interaction. Nowadays the first two inputs are reliable and do not require deeper investigations, but fluid-structure interaction models have margins for improvement in their accuracy. While from a numerical and analytical point of view these models have been developed and studied in detail, their experimental validation is still an open issue. Indeed, the numerical-experimental comparison is often done against the response wind-tunnel models, either full-aeroelastic or sectional models, which have some intrinsic limitations: for example, uncertainties related to the small geometrical scale for full-aeroelastic models, or the non-deformability of the deck in sectional models. To overcome some of the current limitations, in this paper a large-scale deformable aeroelastic model is proposed. The experimental results should be used as a benchmark study for the accuracy of numerical models for the buffeting response of long-span bridges.