Large-eddy simulation (LES) is a technique used to simulate turbulent flows. In LES, larger, energy-containing scales of fluid motion are directly solved on the computational grid, while smaller scales are modeled using a subgrid-scale (SGS) model. Separation between larger and smaller scales is achieved using a low-pass filter. In this project, LES will be performed to predict turbulent flow structure inside an internal combustion engine. A widely used, open source code OpenFOAM will be used in this project. Predictions of the code will be compared with available experimental or numerical data.
The project will provide experience in using state-of-the-art computational fluid dynamics (CFD) code that is used by many in industry and research; running CFD simulations in parallel at the Pittsburgh Supercomputing Center (PSC) and elsewhere on the NSF’s XSEDE network; and analyzing turbulent flow data with engineering relevance.