共 71 条
The role of clusters on heat transfer in sedimenting gas-solid flows
被引:27
作者:
Guo, Lei
[1
]
Capecelatro, Jesse
[1
]
机构:
[1] Univ Michigan, Dept Mech Engn, Ann Arbor, MI 48105 USA
关键词:
Heat transfer;
Particle;
Cluster;
Convection;
Multiphase;
DIRECT NUMERICAL-SIMULATION;
FLUIDIZED-BED;
PREFERENTIAL CONCENTRATION;
PARTICULATE FLOWS;
MASS-TRANSFER;
PARTICLE;
MODELS;
FLUX;
WALL;
COMBUSTION;
D O I:
10.1016/j.ijheatmasstransfer.2018.12.065
中图分类号:
O414.1 [热力学];
学科分类号:
摘要:
In this study, Eulerian-Lagrangian simulations of unbounded sedimenting gas-solid flows are conducted to quantify the effect of two-phase flow interactions on thermal transport. Cold particles are initially randomly distributed in a hot gas and settle under gravity for a wide range of volume fractions and Prandtl numbers. Two-way coupling spontaneously generates clusters of particles with a characteristic length scale much larger than the particle diameter, where fluctuations in particle concentration generate and sustain the carrier-phase turbulence. Simulations reveal that the gas phase is cooled relatively fast in the vicinity of clusters while hot spots persist in regions void of particles. This non-homogeneity was found to delay the overall heat transfer by as much as a factor of four compared to systems with a homogeneous particle distribution. To identify the mechanisms responsible for these variations, the spatially-averaged temperature equation is derived for statistically homogeneous (zero-dimensional) time-dependent particle-laden flows. It is found that the local fluid volume fraction-temperature covariance, referred to here as the fluid temperature seen by the particles, is the key term that accounts for the effect of clusters on heat transfer. Next, a framework for studying thermally fully-developed flows is presented, and statistics at steady state are reported. Consistent with previous work, local variations in the interphase heat transfer coefficient are found to vary by several orders of magnitude. Finally, a statistical modeling approach based on the concept of presumed-shape probability distribution functions is proposed. (C) 2018 Elsevier Ltd. All rights reserved.
引用
收藏
页码:1217 / 1230
页数:14
相关论文