Numerical investigations have been carried out for steady flow through an idealised model of an arterial stenosis at physiological Reynolds numbers of500 and 2000 using the large-eddy simulation turbulence methodology and quantitative comparisons have been made with published experimental data. The results indicate that the largeeddy simulation, methodology is suitable for modelling arterial stenosis flows, which reinforces other published work that came to the same conclusion but from a purely qualitative perspective. Moreover, investigations into simpler, classical, problems in the field ofseparated flows, exhibiting similar behaviour to arterial stenosis flows, e.g. confined jet flow through a sudden expansion, have highlighted inaccuracies in an existing publication ofthe Journal ofFluid Mechanics. Our work has revealed that the experimental data, pertaining to the steady confined jet flow through a symmetric sudden expansion, contained in the aforementioned paper, is erroneous and should not be used for comparison with other experimental or numerical results. For over two decades this paper has never been challenged despite the inaccuracies contained therein. Furthermore, higher-order one-sided finite-difference formulae for the approximation ofvelocity gradients and therefore wall shear stresses have been developed for nonuniform grids. It is shown by application to plane-Poiseuille flow, that the 3-point formula yields greater accuracy than the, often used, 2-point formula.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:485945 |
| Date | January 2007 |
| Creators | Onoufriou, Andros |
| Publisher | University of Hertfordshire |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0019 seconds