Simulation of Snow Redistribution on the Surface of a Long-Span Box Girder Bridge Based on CFD-DEM Coupling

Abstract

Bridge structures are typically elevated above the ground, with lower temperatures on the bridge deck. Accumulated snow particles can drift under the influence of strong winds, posing a substantial threat to traffic safety. The study of wind-induced snow hazards on bridge structures is significant for ensuring the safe operation of transportation in high-latitude and cold regions. Due to the complex mechanisms of snow motion, many unresolved issues remain. This study focuses on a large-span highway box girder bridge and investigates wind-induced snow redistribution on the girder surface using a three-dimensional CFD approach. The nondimensional wind-induced redistribution coefficients of the snow particles were obtained. In addition, a detailed analysis of the mechanisms behind the wind-induced snow redistribution was conducted from a flow field perspective. The results indicate that auxiliary components of the bridge model, such as railings and sidewalk pavement layers, directly influence the redistribution of particles on their surfaces. Furthermore, scaled model tests were conducted in a wind tunnel to validate the accuracy of the numerical simulations. Polyethylene particles were used to simulate snow particles. Moreover, to reveal the mechanism of how wind attack angles affect, the redistribution of particles and flow field information were analyzed under different wind attack angles (0° and ± 3°). The results demonstrate that different wind attack angles have a significant impact, particularly on the windward side of the bridge. The erosion extent of snow particles under the negative wind attack angle is higher than that under the 0° attack angle, and the erosion extents under both the negative and 0° attack angles are higher than that under the positive attack angle. The average and maximum differences in the nondimensional distribution coefficients of particles among the three conditions reach 34.5% and 77.8%, respectively. These findings not only provide data support for practical engineering applications but also offer methods and insights for further research on wind–snow interactions on bridge structures.