A computational particle fluid-dynamics model coupled with an energy-minimization multi-scale (EMMS) drag model was applied to investigate the influence of particle-size distribution on the hydrodynamics of a three-dimensional full-loop circulating fluidized bed. Different particle systems, including one monodisperse and two polydisperse cases, were investigated. The numerical model was validated by comparing its results with the experimental axial voidage distribution and solid mass flux. The EMMS drag model had a high accuracy in the computational particle fluid-dynamics simulation of the three-dimensional full-loop circulating fluidized bed. The total number of parcels in the system (Np) influenced the axial voidage distribution in the riser, especially at the lower part of the riser. Additional numerical simulation results showed that axial segregation by size was predicted in the two polydisperse cases and the segregation size increased with an increase in the number of size classes. The axial voidage distribution at the lower portion of the riser was significantly influenced by particle-size distribution. However, radial segregation could only be correctly predicted in the upper region of the riser in the polydisperse case of three solid species.

计算粒子流体动力学模型与能量最小化多尺度（EMMS）阻力模型相结合，用于研究粒度分布对三维全回路循环流化床流体力学的影响。研究了不同的粒子系统，包括一种单分散和两种多分散的情况。通过将其结果与实验轴向空隙率分布和固体质量通量进行比较，对数值模型进行了验证。EMMS阻力模型在三维全回路循环流化床的计算颗粒流体动力学模拟中具有很高的精度。系统中的包裹总数（Np）影响了立管中的轴向空隙分布，特别是在立管的下部。附加的数值模拟结果表明，在两种多分散的情况下，按尺寸预测了轴向偏析，并且偏析尺寸随尺寸类别数量的增加而增加。立管下部的轴向空隙分布受到粒径分布的显着影响。但是，在三种固体物质的多分散情况下，只能在立管的上部区域正确预测径向偏析。