Embedded large eddy simulation approach for pollutant dispersion around a model building in atmospheric boundary layer

Numerical analysis of pollutant dispersion around elongated buildings: An embedded large eddy simulation approach

“…where v t is the turbulent eddy viscosity, k is the turbulent kinetic energy, and Sc t =0.7 is the turbulent Schmidt number. This value has been chosen in this study based on a series of sensitivity tests confirming previous reports in the literature (Salim et al, 2011;Jadidi et al, 2016a). The eddy viscosity v t is obtained by solving two transport equations for the turbulent kinetic energy k and for the specific dissipation rate ω (rate of dissipation per unit of turbulence kinetic energy, often called turbulent frequency).…”

“…Considering the characteristics of the flow and pollutant dispersion around buildings (highly unsteady wakes, downwash effects, recirculation regions, and corner vortices at the vicinity of the building and a relatively steady flow upstream and far downstream), the embedded LES (ELES) seems to be an attractive choice balancing the computational accuracy and efficiency. It has been used to study complex turbulent flows including those around the Ahmed body (a simplified vehicle geometry) (Mathey et al, 2005), a hump (Davidson and Peng, 2013), and a cubical building (Jadidi et al, 2016a). Jadidi et al (2016a) investigated the performance of ELES in comparison with LES in simulating the pollutant dispersion around a cubical building.…”

“…It has been used to study complex turbulent flows including those around the Ahmed body (a simplified vehicle geometry) (Mathey et al, 2005), a hump (Davidson and Peng, 2013), and a cubical building (Jadidi et al, 2016a). Jadidi et al (2016a) investigated the performance of ELES in comparison with LES in simulating the pollutant dispersion around a cubical building. They concluded that ELES provides comparable results with those of LES with almost 50% lower computational costs (Their LES and ELES simulations took 310 and 158 hours wall-clock time to complete, respectively, using 16 processors).…”

“…Although the RANS simulations were generally able to predict the gross flow and concentration behavior, they fail to correctly describe the anisotropic and unsteady nature of the problem, which is manifested in the significant underestimation of the lateral plume spread (Tominaga and Stathopoulos, 2009). Therefore, scale-resolving methods including large eddy simulation (LES) and hybrid RANS/LES, which directly resolve (to some extent) the more influential, energy carrying eddies and models only the small (subgrid) scales, have been used and promising results were obtained (Tominaga and Stathopoulos, 2010;Gousseau et al, 2011;Jadidi et al, 2016a). Nevertheless, these studies are generally limited to cubical buildings perpendicular to the wind direction (Tominaga and Stathopoulos, 2010;Jadidi et al, 2016a), thus not addressing the complexities arising in the case of an elongated/rotated building.…”

Instantaneous transport of a passive scalar in a turbulent separated flow

“…The total viscosity, , results from the addition of the fluid kinematic viscosity, , and the turbulent viscosity obtained from the sub-grid scale (SGS) model, . The latter is computed using the Smagorinsky [ 38 ] SGS model, where the constant is set constant with a value of 0.1, as it has been proved to provide successful predictions in scalar transport [ 24 , 28 ].…”