Aerodynamic stability optimisation of twin-box bridge girders

Suspension bridges are preferred for long spans due to their capability of achieving some of the longest free spans in bridge engineering. While single streamlined box girders are commonly utilised in these structures, the increasing trend towards longer spans demands enhanced consideration of the aerodynamic stability of the bridge girder. Due to the bridge structure flexibility, suspension bridges may become aerodynamically unstable and enter into flutter at high winds if not designed properly.

To maintain aerodynamic stability in such cases, the adoption of a twinbox girder configuration has emerged as a viable solution. The twinbox deck has demonstrated superior structural and aerodynamic performance, compared to a monobox girder of equivalent width of the carriageways. The design of the 1915 Çanakkale Bridge has further illuminated that the aerodynamic stability of a twinbox deck can exceed theoretical predictions, if the girder assumes a ”noseup” twist under high wind conditions. A phenomenon not previously recognised within the bridge engineering community. A desirable ”noseup” twist can be achieved through adjustments to the girders crosssection geometry or by the addition of wind screens or appendages, ensuring a significant aerodynamic moment coefficient. Proper exploitation of the ”noseup” effect in twinbox bridge girders is crucial for design of future long span twinbox suspension bridges, and is thus the focus of this Ph.D. project.

The objectives of this Ph.D. project are to explore the implications of the ”noseup” effect in twinbox bridge girder designs, enhancing the understanding of aerodynamic stability mechanisms. Through comprehensive research, including wind tunnel testing and computational fluid dynamics simulations, this thesis clarifies some of the critical phenomena contributing to improved aerodynamic stability. Key findings establishes that the critical wind speed for onset of flutter increases with the degree of noseup rotation of the twinbox girder. The presence of a positive moment coefficient at zero angle of attack, relates to reduced suction on the bottom surface of the upwind girder, ensuring that the elastically supported deck will always meet the mean wind at ever increasing angles of attack. Additionally, the inner gantry rails markedly affect of the aerodynamic stability of the twinbox girder by promoting the noseup rotation, which enhances flutter stability and VortexInduced Vibrations (VIV). The precise positioning of the inner gantry rails have a pronounce effect on the VIV response.

In conclusion, the research underlines the importance of geometric details in twinbox girder designs and presents important insights for future developments in longspan suspension bridge engineering, for enhancing the aerodynamic stability.