Convection-Diffusion Phenomena and a Navier-Stokes Processor
- 28 Downloads
For heat and mass transport in complex flow situations computational methods are very important. Many technological processes can be simulated by a set of convection-diffusion equations. These equations can numerically be solved using a single algorithm based on the finite domain or control volume method. For turbulent transport a k-ɛ model is often used. This requires that in some cases an experimental validation for a completely new flow situation has to be done.
For two examples the application will be shown. The natural convection in a square cavity both for laminar and turbulent cases will be discussed. For flows and heat transfer in living spaces good predictions including radiative exchange can be given. The second example is the simulation model “Furnace”. The flow, combustion and heat transfer in a glass furnace can be predicted. A full 3-dimensional model has been developed. For fine grids, and for time dependent or 3-D situations the computational effort is large. The elliptic flows and the coupling of a large set of partial differential equations give a slow convergence. CPU time on a main-frame computer may run in many hours. This has led us to the development of a processor to directly solve the convection-diffusion algorithm for the finite control volume method as well as the transport equations. This will be applied in a special purpose dedicated Navier-Stokes computer with the capabilities of a super-computer for this special algorithm. It can be expected that this tool will enhance the application of numerical transport phenomena studies strongly.
Unable to display preview. Download preview PDF.
- Linthorst, S.J.M., Hoogendoorn, C.J.: “Numerical calculations of heat transfer by natural convection in a cubical enclosure”, Proc. 2nd Int. Conf. Num. Methods in Laminar and Turbulent Flow, Venice (1981) pp. 1069–1078.Google Scholar
- Linthorst, S.J.M., Hoogendoorn, C.J.: “Natural convective heat transfer in three dimgjisional inclined small aspect ratio enclosures”, Proc. 8 Int. Heat Tr. Conf., San Francisco (1986) pp. 1501–1505.Google Scholar
- Patel, V.C., Rodi, W., Scheuerer, G.: “Evaluation of turbulence models for near wall and low-Reynolds number flows”, Proc.3rd Symp. Turbulent Shear Flows, Davis, Cal. (1981) pp. 1–11 8.Google Scholar
- Jones, I.P.: “A comparison problem for numerical methods in fluid dynamics, the double glazing problem”, Num. Meth. in Therm. Probl., Pineridge Press, UK (1979) pp. 338–348.Google Scholar
- Euser, H., Hoogendoorn, C.J., Ooyen, H. van: “Airflow in a room as induced by natural convection streams”, Energy Cons, in Heating, Cooling, Ventil. Build., Ed. Hoogendoorn, C.J., Afgan, N.H., Hemisphere Publ., USA, 1 (1978) pp. 259–270.Google Scholar
- Vermogen van radiatoren bij niet genormeerde opstelling. Publ. nr.l Stichting ISSO, Rotterdam, Holland (1980) pp. 29–31.Google Scholar
- Simonis, F., Waal, H. de, Beerkens, R.C.G.: “Influence of furnace design and operation parameters on the residence time distribution of glass tanks, predicted by 3-D computer simulations”, Proc. 14 Int. Conf. on Glass, India (1986).Google Scholar
- Peterson, V.L.: “Impact of computers on aerodynamics research and development”, Proc. I.E.E.E. 72 (1984) no.l pp. 68–79.Google Scholar