MODELING OF THE FLUID FLOW AND HEAT TRANSFER IN A PEBBLE BED MODULAR REACTOR CORE WITH A COMPUTATIONAL FLUID DYNAMICS CODE

摘要

The Pebble Bed Modular Reactor (PBMR), a promising Generation IV nuclear reactor design, raises many novel technological issues for which new experience and techniques must be developed. This brief study explores a few of these issues, utilizes a computational fluid dynamics code to model some simple phenomena, and points out deficiencies in current knowledge that should be addressed by future research and experimentation. A highly simplified representation of the PBMR core is analyzed with FLUENT, a commercial computational fluid dynamics code. The applied models examine laminar and turbulent flow in the vicinity of a single spherical fuel pebble near the center of the core, accounting for the effects of the immediately adjacent fuel pebbles. Several important fluid flow and heat transfer parameters are examined, including heat transfer coefficient, Nusselt number, and pressure drop, as well as the temperature, pressure, and velocity profiles near the fuel pebble. The results of these "unit cell" calculations are also compared to empirical correlations available in the literature. As FLUENT is especially sensitive to geometry during the generation of a computational mesh, the sensitivity of code results to pebble spacing is also examined. The results of this study show that while a PBMR presents a novel and complex geometry, a code such as FLUENT is suitable for calculation of both local and global flow characteristics, and can be a valuable tool for the thermal-hydraulic study of this new reactor design. FLUENT results for pressure drop deviate from the Darcy correlation by several orders of magnitude in all cases. When determining the heat transfer coefficient, FLUENT is again much lower than Robinson's correlation. Results for Nusselt number show better agreement, with FLUENT predicting results that are 10 or 20 times as large as those from the Robinson and Lancashire correlations. These differences may arise because the empirical correlations concern mainly integral parameters, while the FLUENT model focuses on local flow behaviors. Local phenomena are significant in the case of local heat transfer characteristics, fine temperature distribution calculations to identify hot spots, and fission product transport phenomena. All of these are important to a safety analysis of the PBMR reactor during normal operation, as well as during transient circumstances, and should be the focus of future research efforts.