Abstract Summary
The role of atmospheric turbulence in the onset mechanism of bridge aeroelastic instabilities has not been fully understood yet, and it often leads to controversial results. Indeed, concerning torsional and coupled flutter, experimental and numerical results have shown that atmospheric turbulence may have either a stabilising or a destabilising influence. In particular, the parametric excitation induced by the variation of the self-excited forces due to the angle of attack associated with large-scale atmospheric turbulence was found to anticipate the coupled flutter onset of about 20% in the case of the Hardanger Bridge, in Norway (Barni et al., 2022a). In that work, the so-called 2D RFA model was employed to predict the parametric effect of large-scale turbulence. Moreover, after simplifying the full bridge structure to a three-degree-of-freedom 2D model, a first attempt to formally study the stability of the bridge under a periodic parametric excitation according to the Floquet theory was reported in Barni et al. (2022b). Along this research line, the present work numerically investigates the stochastic stability of both a 2D and a full bridge model of the Hardanger Bridge exposed to a realistic turbulent wind field. Even if the governing equations of the system result in a linear time-variant state-space model, a Monte Carlo approach is followed due to the lack of validity of the Markovian process assumption, which does not allow the definition of a diffusion problem. Self-excited forces are again modelled according to the 2D RFA model, and the sample time histories of the slowly-varying angle of attack are derived from the realisations of the turbulent random wind field. The flutter stability of the system is statistically assessed by looking at its free-vibration behaviour given some initial conditions and a large number of realisations of the time histories of the angle of attack. Although the Monte Carlo approach does not foster any physical interpretations of the stability of the system, the numerical simulations highlight the sensitivity of the flutter stability to the standard deviation of the slowly-varying angle of attack, which is directly linked to turbulence intensity, band-superposition cut-off frequencies and integral length scale. For the considered case study, the paper emphasises a significant destabilising effect of the turbulence intensity, as well as the strongly stochastic nature of the Hopf bifurcation. References: Barni, N., Øiseth, O. A., Mannini, C. (2022a). Buffeting response of a suspension bridge based on the 2D rational function approximation model for self-excited forces. Engineering Structures, 261, 114267. Barni, N., Gioffrè, M., Mannini, C. (2022b). Bridge flutter stability in turbulent flow. Proceedings of the XVII Conference of the Italian Association for Wind Engineering, IN-VENTO 2022 (under review)
