Quantifying the sensitivity of suspension bridge sectional buffeting to Non-Gaussian wind statistics: A quasi-static approach

Abstract

Traditional Gaussian wind models are widely applied in bridge design; however, they often ignore non-Gaussian statistical features such as skewness and kurtosis found in severe winds. This paper investigates the fundamental sensitivity of the aerodynamic response of a two-dimensional suspension bridge deck section to these higher-order statistics. An optimization procedure is implemented to generate non-Gaussian wind time series, enabling precise control of target skewness and kurtosis while preserving the energy spectrum. By employing a quasi-static model with unity aerodynamic admittance, this study intentionally establishes a conservative upper-bound estimate of the buffeting response to isolate the effects of input probability distribution from aerodynamic damping and spatial correlation factors. The findings indicate that wind non-Gaussianity can amplify the maximum vibration amplitude and theoretically increase the fatigue damage index of the torsional response. These results highlight the potential risks in extreme scenarios and suggest that higher-order statistics should be considered in conjunction with comprehensive aeroelastic models in future design assessments.