Experimental and numerical study of the effect of appurtenances on the wind-induced vibrations of a cable-stayed bridge

This study addresses a significant gap in the comparative assessment of crosswind-induced dynamic responses in both the construction and operation stages of cable-stayed bridges. Although numerical modeling investigation has progressively improved, experimental evaluations of the effects of predominant components (e.g., railing systems, track systems, and ancillary attachments) have not been given considerable attention. In this research, crosswind-induced vibrations in both the construction and completion stages are investigated, and static wind loads, turbulence-induced buffeting, and self-excited forces transmitted to the bridge girder are incorporated through an integrated numerical and experimental approach. Quantitative assessment of the impact of these attachments on the dynamic and aerodynamic behavior of the bridge is essential for the structural safety and stability of cable-stayed bridges. The analysis results suggest that, as the bridge shifts from the construction phase to the completion phase, it becomes increasingly sensitive to wind-induced forces. This is evidenced by an increase in the drag force of 20%, a decrease in the lift force of 64% and a drastic increase in the rolling moment of 52%. Moreover, the bridge's dynamic response to wind increases dramatically with increasing wind speed. More importantly, the susceptibility of the bridge's dynamic responses to wind forces changes markedly before and after the completion stage. During construction, the vertical displacement and acceleration decreased by up to 79% and 40%, respectively, indicating that the vertical stability of the construction phase decreased. Conversely, the completed bridge exhibits lower lateral stability, with the lateral acceleration increasing by up to 69% and the lateral displacement increasing slightly by 23% at most. Comparison of numerical and wind tunnel vertical displacements reveals stage-dependent behavior. During the construction stage, good agreement between the displacements is observed, particularly at higher wind speeds. For the completion stage, agreement is reasonable at low to moderate wind speeds, whereas notable divergence is observed at high wind speeds. Direct wind tunnel measurements for the completed bridge revealed no conventiona flutter or galloping. Soft flutter was observed at an angle of attack of -3°. The critical wind speeds exceed 108 m/s, substantially surpassing the required validation values. These findings show that stage-specific vulnerability assessments that consider component-level details and dynamic force interactions are needed. This study proposes the use of complex technologies such as wind tunnels and stochastic inputs to improve safety and resilience during the entire life cycle of cables-stayed bridges under strong wind conditions.