Dynamic Reliability of the Buffeting Responses of Suspension Bridges Considering Non-Gaussian Factors

Abstract

Based on the accurate analysis of the time history of bridge buffeting responses and considering the nonlinearity factors of the structure and wind load, the non-Gaussian response of a structure is derived. The Hermite moment model is introduced to transform the unknown distribution responses into standard Gaussian random processes based on their first four moments, and the dynamic reliability of the structure under strong wind conditions is then numerically analyzed in combination with the Poisson process method. A long-span suspension bridge at the East Sea of China is analyzed and the buffeting dynamic reliability of the stiffening girder is discussed in detail. The results indicate that (1) the middle support section of the stiffening girder is the critical section under fluctuating wind conditions and the dynamic reliability decreases as the wind speed increases, (2) nonlinear effects in the structure are apparent under strong wind conditions, and (3) the common assumption of Gaussian process will produce inaccurate results and unsafe bridge design.