Due to the limitations of the bridge deck modeling and wind tunnel installations, the parametric uncertainties of aerostatic displacement-dependent wind loads have become an urgent problem for reliability analysis in bridge engineering. Computational fluid dynamic (CFD) simulations were conducted to investigate the distribution model of aerostatic coefficients. The Monte Carlo simulation (MCS) was used to generate random samples of stream-lined box cross-sections. Among these samples, wind attack angles of −8°–8° were considered, and the sample size of each angle group was set as 30. Then, the statistical properties for the variation and distribution of aerostatic coefficients were discussed. In addition, sub-samples with sizes of 5–20 were randomly sampled from the initial sample group to evaluate the influence of the sample scale. The coefficient of variation (COV) of aerostatic coefficients with respect to geometric uncertainties increased up to 0.217, which occurred at −7° wind attack angle. Lognormal distribution models obtained by hypothesis tests maintained a low deviation for estimating the distribution of aerostatic coefficients. The sample size had a significant effect on quantifying the error between empirical and theoretical distribution models. Simulation method proposed in this paper provides a systematic approach for probabilistic problems involving fluid-structure interaction and parametric uncertainties.
Probabilistic Analysis on Aerostatic Displacement-dependent Wind Loads on a Stream-lined Box Girder
KSCE J Civ Eng
KSCE Journal of Civil Engineering ; 27 , 1 ; 299-312
2023-01-01
14 pages
Article (Journal)
Electronic Resource
English
Transportation Research Record | 2010
|Stream-lined electric welded diesel tug "Davie"
Engineering Index Backfile | 1936