Thanos Papanicolaou
IIHR - Hydroscience & Engineering, The University of Iowa

Home Research Team Vision Research Press Releases Publications

 

Application Note:  3D Hydrodynamic Sediment Transport
   
Three-dimensional models: In many hydraulic engineering applications one has to resource to 3-D models when 2-D models are not suitable to describe certain hydrodynamic/sediment transport processes.  Flows in the vicinity of piers and near hydraulic structures are examples where 3-D flow structures are ubiquitous and 2-D models in this case do not adequately represent the physics.  With the latest developments in computing technology such as computational speed, parallel computing and data storage classification, 3-D hydrodynamic/sediment transport models have become much more attractive to use comparatively to ten years ago.  The majority of the 3-D models solve the continuity and the Navier-Stokes equations along with the sediment mass balance equation via the methods of finite-difference, finite-element or finite-volume. The Reynolds Average Navier-Stokes (RANS) approach has been employed to solve the governing equations.
   
Flow heterogeneity over 3D cluster microform: A study was conducted to examine the flow around a self-occurring cluster bed form and the use of general computation fluid dynamics methods for hydraulic and geophysical flow applications. This is accomplished through a comprehensive experimental/ numerical investigation. In the laboratory, cluster bed forms are first formed from movable sediment, and laser Doppler velocimeter measurements of two-dimensional fluid velocity are then taken around a formed cluster.  A three-dimensional Reynolds averaged Navier-Stokes simulation of the physical cluster and flow conditions is then conducted using CFD package FLUENT, near-wall, and shear stress transport (SST) turbulence modeling with the inclusion of hydraulic roughness.  SST near-wall modeling is advantageous compared to the more widely used wall functions approach for flows with significant roughness and flow separation because the model equations can be integrated down to the wall.  Therefore, SST near-wall modeling makes no a priori assumption that the law of the wall is valid throughout the wall region of the flow.  Additionally, it has the ability to intrinsically handle boundary roughness through the boundary condition for turbulent specific dissipation at the wall, allowing for wall functions to be bypassed in accounting for roughness effects.  The study shows that in the wall region surrounding the cluster, flow is 3D and quite complex, with different scales of embedded flow structures dominating the cluster wake and leading to flow heterogeneities in pressure and bed-shear stress.  Results also indicate that near-wall modeling with SST compared favorably with the experimental flow data without tuning of model constants.