Numerical Study of Liquid Metal Turbulent Heat Transfer in Cross-Flow Tube Banks

被引:1
作者
Tassone, Alessandro [1 ]
Meeusen, Jasper [2 ]
Serafini, Andrea [1 ]
Caruso, Gianfranco [1 ]
机构
[1] Sapienza Univ Rome, Dept Astronaut Elect & Energy Engn DIAEE, Nucl Engn Res Grp, Corso Vittorio Emanuele 2,244, I-00186 Rome, Italy
[2] Katholieke Univ Leuven, Dept Mech Engn TME, Celestijnenlaan 300A postbus 2421, B-3001 Heverlee, Belgium
关键词
heat transfer; turbulent Prandtl number; liquid metal; cross-flow; tube bank; BENCHMARK SIMULATION; PRANDTL NUMBER; SUPPORT CFD; BUNDLE; CHALLENGES; MODEL;
D O I
10.3390/en16010387
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Heavy liquid metals (HLM) are attractive coolants for nuclear fission and fusion applications due to their excellent thermal properties. In these reactors, a high coolant flow rate must be processed in compact heat exchangers, and as such, it may be convenient to have the HLM flowing on the shell side of a helical coil steam generator. Technical knowledge about HLM turbulent heat transfer in cross-flow tube bundles is rather limited, and this paper aims to investigate the suitability of Reynolds Average Navier-Stokes (RANS) models for the simulation of this problem. Staggered and in-line finite tube bundles are considered for compact (a = 1.25), medium (a = 1.45), and wide (a = 1.65) pitch ratios. The lead bismuth eutectic alloy with Pr = 2.21 x 10(-2 )is considered as the working fluid. A 2D computational domain is used relying on the k - omega Shear Stress Transport (SST) for the turbulent momentum flux and the Pr-t concept for the turbulent heat flux prediction. The effect of uniform and spatially varying Pr-t assumptions has been investigated. For the in-line bundle, unsteady k - omega SST/Pr-t = 0.85 has been found to significantly underpredict the integral heat transfer with regard to theory, featuring a good to acceptable agreement for wide bundles and Pe >= 1150. For the staggered tube bank, steady k - omega SST and a spatially varying Pr-t has been the best modeling strategy featuring a good to excellent agreement for medium and wide bundles. A poor agreement for compact bundles has been observed for all the models considered.
引用
收藏
页数:19
相关论文
共 33 条