ABSTRACT
Wind-induced unexpected events for cable-supported bridges are introduced for the purpose of sharing authors’ experiences regarding to data utilization for cause investigation, securing countermeasures, and design feedbacks. The events include two cases of vortex-induced vibration and one case of rollover accident of running vehicle over the bridge deck. The utilized data include the basic acceleration and wind speed data, but wind tunnel tests, data processing techniques, system identification with operational modal analysis (OMA), Measure-Correlate-Predict analysis, and Long Short-term Memory were applied to increase the value of data obtained. The results demonstrate that the damping identification from OMA is an essential issue as a serviceability performance index for flexible long-span bridge. Also, the vulnerability assessment procedure of running vehicle to strong side wind can be utilized for the effective traffic control by utilizing long-term and short-term wind speed data. By these way, examples of increasing the value of data for long-span bridge operation are demonstrated in an aerodynamic point of view.
