Numerical investigation on flow and heat transfer characteristics of various special-shaped narrow channels at high Reynolds number

This paper introduces three innovative special-shaped narrow channels: the transverse sinusoidal wavy channel, the longitudinal sinusoidal wavy channel, and the scale-roughened channel. These channels are designed for the heat transfer enhancement (HTC) of the heat transfer components within nuclear industry under high Reynolds number conditions. Numerical simulations using ANSYS Fluent 2023R2 are conducted to investigate the thermal-hydraulic characteristics of these channels for turbulent water flow, incorporating three-dimensional conjugate heat transfer. The simulations are performed at different mass fluxes (G = 500 kg/m2s and 3000 kg/m2s, corresponding to Reynolds numbers of 22638 and 135827, respectively). The results are compared with and validated against the classic Dittus-Boelter correlation and the conventional friction correlations proposed by Blasius and MacAdams. It has been demonstrated that, compared to the rectangular channel, these three proposed channels can effectively reduce the temperatures of solid regions (external claddings and exothermic cores) only under low mass fluxes. Furthermore, at mass fluxes of 3000 kg/m2s and 500 kg/m2s, the longitudinal sinusoidal wavy channel exhibits average Nusselt numbers significantly higher than those of the rectangular channel, by 24.76 % and 51.24 %, respectively. Although the scale-roughened channel also demonstrates higher average Nusselt numbers (exceeding those of the rectangular channel by 39.90 % and 42.94 % at the same mass fluxes), its Darcy friction factors are significantly greater (4.95 times and 9.6 times greater than those of the rectangular channel). This substantial increase in friction factors significantly diminishes its overall performance. Therefore, the longitudinal sinusoidal wavy channel has been identified as the optimal design due to its ability to enhance comprehensive performance across a wide range of mass fluxes and will be employed in subsequent numerical simulations of multiphase flow.