Computational Fluid Dynamics
Computational Fluid Dynamics (CFD) is the use of numerical methods to simulate fluid flows. The flow problems studied in the Thermofluids Group include both fundamental turbulent flows, and problems that are relevant to industrial applications. When used in conjunction with experimental and analytical tools, CFD can be used to improve the efficiency of aerodynamic components in a cost-effective manner. On a more fundamental level, CFD allows insights into turbulence that are otherwise difficult to obtain experimentally.
The numerical methods used in the Thermofluids Group include Reynolds-Averaged Navier Stokes (RANS), its unsteady variant (uRANS), Large Eddy Simulation (LES), and Direct Numerical Simulation (DNS). Bespoke codes are routinely used for fundamental fluids research, and the commercial CFD package FLUENT is used for more industrially-focused projects.
Current research in the group includes:
- Development of Computational Aeroacoustics methods
- Development of the COSMIC CFD code
- CFD of flow and heat transfer in thermoacoustic devices
- Simulation of subsonic jets and mixing layers using LES and DNS
- Simulation of multiphase flows using LES
The University of Leicester High Performance Computing (HPC) service, named ALICE, provides a valuable tool for CFD research in the Thermofluids Group. This cluster has, for example, allowed Large Eddy Simulation of turbulent flows to be performed to an extremely high level of detail. A sample animation of output from an LES of a plane turbulent mixing layer is shown below.
The plane mixing layer is a fundamental turbulent flow type, and it is a simple way in which to study turbulence. It also has practical applications in diffusion flames, jet flows, aeroacoustics, and many other aerodynamic configurations. A comprehensive understanding of the flow is therefore vital.
Elliptical coherent structures are visible in the side-view visualisation - the lower part of the animation. These structures have also been observed experimentally, but their properties are extremely difficult to measure experimentally owing to their short life spans. Large Eddy Simulation can provide detailed statistical information on the topography of the coherent structures, and hence improve our understanding of the flow. More information on this research can be found here.
