Abstract
Enhancement of heat transfer to supercritical fluids has drawn the attentions of many researchers within the past few decades. Modeling and predicting heat transfer to turbulent flow of supercritical fluids, however, are very complicated due to severe variations of fluid properties near the critical point. Large discrepancies between available heat transfer data are greatly due to confusion of forced convection and mixed convection data. The data unaffected by buoyancy have been selected cautiously from a large database generated in this study. Such data have been used to develop a 1D numerical model as well as a semi-empirical correlation to predict forced convection heat transfer to turbulent flow of supercritical water. In the numerical model, radial variations of heat flux and shear stress are taken into account. Modifications to turbulent Prandtl number and wall shear stress formulations have been applied to a law of the wall type of model to fit supercritical conditions. The model shows good agreement with experiments. In the experimental part, the extensive database obtained on a full-scale test facility in the present study, plus a new conceptual approach, has been employed together to develop a semi-empirical heat transfer correlation. It accurately predicts the experiments.
