In order to analyze complex turbulent shear flows with heat transfer, accurate predictions of mean flows, temperature and turbulent quantities are of monumental importance. A method of computational simulation has been widely used as a means to predict turbulent flows with heat transfer in engineering and environmental geophysics. In the simulation, though a turbulence model is required to solve the equations, it is well-known that the two-equation turbulence model is notably useful for properly predicting turbulent flows of technological interest. Likewise, the prediction of turbulent heat transfer is also required for analyzing related complex thermal fields. Therefore, a two-equation heat-transfer model must be developed in order to obtain adequate predictions for complex turbulent heat-transfer phenomena. On the other hand, since a Low-Reynolds-Number (LRN) turbulent heat-transfer model should properly reflect the wall effect, the various thermal conditions at the wall and the Prandtl number effect, the wall-reflection function, which includes the distance from the wall, is generally used. Such a wall-reflection function adequately reflects the wall effect, but calculations of the distance from the wall inevitably increase computational load. This study seeks to construct a two-equation heat-transfer model which is applicable to predictions of complex turbulent heat-transfer phenomena, in which the wall-reflection function does not explicitly include the distance from the wall. The proposed new model predicts adequately complex turbulent heat-transfer phenomena including the Prandtl number effects obtained by experiment and direct numerical simulation.
展开▼
