CAREER: Petascale DNS of Evaporating Droplet-Laden Homogeneous Turbulence
University Of Washington, Seattle WA
Investigators
Abstract
The combustion processes in electric power plants, jet engines, gasoline and diesel powered vehicles are the primary sources of the carbon dioxide emissions that grew by about 80% between 1970 and 2004. In the next decades, anthropogenic warming, due mostly to these CO2 emissions, could lead to impacts that are abrupt and irreversible, including increased water stress for hundreds of millions of people, ocean acidification, ecosystems change, and flooding (IPCC, 2007 & 2009). One method to help stabilize, and hopefully reverse, anthropogenic CO2 emissions is to reduce fossil fuel consumption by improving combustion efficiency. To do so, we must better understand the complex physical processes involved in spray combustion of liquid fuels. The vaporization rate of fuel droplets is recognized as a key mechanism in fuel-droplet combustion and most spray combustion devices operate in the turbulent regime. The effects of droplet vaporization on the dynamics of turbulence are therefore important, and the underlying physical mechanisms are largely unknown. The main scientific objective of this research is to explain the physical mechanisms occurring in evaporating droplet-laden turbulence. The research is focused on explaining the effects of droplet/turbulence and droplet/droplet/turbulence interactions. The study is conducted by developing a petascale DNS code to simulate evaporating droplets in homogeneous turbulence. The novelty of the computational methodology stands in the ability to capture the process of heat, momentum and mass transfer of the liquid droplets with the surrounding fluid, while fully resolving the turbulent flow. The research is transformational to the science by performing unprecedented fully-resolved DNS of homogeneous turbulence (with an inertial-range of turbulence scales) laden with millions of vaporizing droplets using the NSF petascale supercomputer, Blue Waters (NCSA).
View original record on NSF Award Search →