George Ilhwan Park

Learn More
Using the recent DNS by Wu and Moin (2010) of a flat-plate boundary layer with a passively heated wall, statistics of the turbulence in transition at Re θ = 500 where spots merge (distributions of the mean velocity, rms velocity and vorticity component fluctuations, kinetic energy production and dissipation rates and enstrophy) have been compared to these(More)
This paper reports on work accomplished subsequent to a pilot study that was presented at the 2001 conference. This current study will eventually involve the training of over 500 novice drivers, and subsequent comparison of real-world accident and violation rates of the simulator trained group with a traditionally trained control group of demographically(More)
Large-Eddy Simulations (LES) of turbulent flows are fundamentally based on resolving eddy sizes larger than a filter width ∆ while modeling the subfilter scales. In practice, ∆ is imposed by the grid resolution and is ideally representative of scales associated with the inertial subrange. Dynamic subgrid-scale (SGS) models for turbulent transport of(More)
In Lagrangian descriptions of turbulent flows laden with heavy particles, the disperse phase is described by the equation of the trajectory dx p,i dt = u p,i (1.1) and the equation of motion 4 3 πρ p a 3 du p,i dt = 6πµa(u i − u p,i), (1.2) where it is assumed that the flow around the particles is dominated by molecular transport. In this formulation, x p,i(More)
Epidemiological studies have demonstrated the exacerbation of respiratory diseases following sandstorm-derived particulate matter (PM) exposure. The presence of anthropogenic and biological agents on the sandstorm PM and the escalation of PM<2.5μm (PM2.5) pollution in China have led to serious concerns regarding the health effects of PM2.5 during Asian(More)
  • By M. MacDonald, D. Chung, +4 authors B. Pierce
  • 2014
A comprehensive direct numerical simulation (DNS) database of turbulent pipe flow was used to investigate the physics of turbulent flow over three-dimensional roughness, with an eye toward developing multiscale wall models for large-eddy simulation (LES). Using cases in which the height of the roughness was varied, we characterize, at various levels of(More)
For the last decade, high-order shock-capturing methods with numerical dissipation controls have been the state-of-the-art numerical approach for direct numerical simulation (DNS) and large eddy simulation (LES) of turbulent flows with shocks. See, for example Yee & Sjögreen (2010, 2007); Sjögreen & Yee (2004); Yee et al. (1999, 2012); Kotov et al. (2013,(More)
  • 1