In recent years, computational fluid dynamics has emerged as one method that is able to improve our ability to predict subcooled and saturated boiling flows. In the nuclear field, accurate modelling of the critical heat flux (CHF), which is perhaps the main threat to the integrity of reactor fuel rods, has stimulated much interest. In predicting the CHF, one short-term objective is to develop models that can be applied with confidence to both bubbly and boiling flows. In this paper, the accuracy of an Eulerian-Eulerian averaged two-fluid model implemented in the STAR-CCM+ code is evaluated against air-water bubbly and subcooled boiling flow data. The model includes a Reynolds stress turbulence model and a population balance-based approach with newly implemented source terms for bubble break-up and coalescence. The boiling model is based on the heat flux partitioning approach and accommodates the heat flux due to single-phase convection, quenching and evaporation. Despite achieving a satisfactory accuracy for air-water bubbly flows, predictions of subcooled boiling are less accurate in areas such as the average bubble diameter and the velocity profile close to the wall. However, the accuracy obtained for void fraction, turbulence levels and temperature encourages future efforts to improve the model described. Further developments to increase accuracy and general applicability, in particular in relation to the boiling and population balance models, are identified.
展开▼
