Three-dimensional mesh morphing iterative methodology for flow scouring around bridge piers implemented in a commercial CFD code.

摘要

Flow scour is the engineering term used to describe the erosion of a sediment bed due to fluid flow. Local scour occurs around objects placed in the path of flow, such as bridge piers. Severe damage or even failure of structures may occur if the amount of scour is too great. Due to the complexity of the fluid/structure interactions and cost of experiments, computation fluid dynamics (CFD) software is often used to predict the shape and depth of a scour hole. This study extends the previous 3-D iterative methodology, with several improvements to the scouring process, implemented in commercial CFD software called STAR-CCM+ to predict the scour hole formation around circular bridge piers. These improvements are inclusion of a variable critical shear stress (VCSS), scouring normal to the sediment bed and a sand slide model. Reynolds averaged Navier-Stokes (RANS) equations and a realizable k-epsilon 2-layer turbulence model are used to resolve the flow field. The methodology uses a single-phase implicit unsteady approach to obtain sediment bed shear stress values. Two moving boundary relations are employed, with one based upon an empirical correlation for critical shear stress and the other utilizing the slope of the sediment bed, to iteratively deform the sediment bed. This is accomplished by a user-defined routine to move the sediment bed at each time step and a mesh morphing procedure to stretch the existing mesh to maintain cell quality throughout the flow domain. The computational methodology uses Python and Java macro scripts and was implemented using Bash script in a LINIX environment. Simulation results have been validated by comparison to experimental data found in literature. It was found that simulations overpredict the maximum scour depth by up to 25%. Substantial improvements to prediction of the scour hole shape are also seen in the simulations.