Application and analysis of a two-layer rough wall, near wall treatment for bluff body aerodynamics

Owing to various limitations in the simulation of atmospheric flow conditions in the wind tunnels, computational wind engineering is gaining importance. Numerous models have been proposed and validated over the years and RANS-based two-equation turbulence models are widely used by the engineering community. In the early phase of this work, five turbulence models, as implemented in Fluent, were evaluated based on the predicted flow structure and pressure distribution on a cubical model placed in an atmospheric boundary layer (ABL) using the rough-wall log-law and compared to experimental data. This work sought to improve the accuracy between predicted and experimental data by utilizing a two-layer, rough near wall treatment. The realizable k-ε model was used in conjunction with the two-layer, rough wall formulation to predict pressure data, velocity distributions, and various turbulence parameters. The predicted data is compared with available experimental data is shown to improve the predicted pressure distribution.