A bubble structure-based drag model developed for the monodispersed system has been extended to simulate bi-dispersed bubbling fluidized beds. Both dense phase and dilute phase are considered to be comprised of two types of particles with different sizes. The derivation of the structure-based drag coefficient for individual particle based on the force equilibrium principle is proved to be independent of the volume fractions (ε_si) of particle i. The multi-fluid model in the commercial software Ansys Fluent is employed to evaluate the polydispersed drag model for the segregation of binary gas-particle flows in a bubbling fluidized bed. It is shown that the simulation results predicted by the new structure-based drag model are in reasonable agreement with experimental data with a 6.34% root mean square error (RMSE). The new structure-based model can capture the particle distribution at the top region of the fluidized bed well. The bubble behaviour can also be captured by the new model well.

中文翻译：

---

使用从单分散模型扩展的基于多分散结构的阻力模型模拟鼓泡流化床的二元颗粒分离

为单分散系统开发的基于气泡​​结构的阻力模型已扩展到模拟双分散鼓泡流化床。密相和稀相都被认为是由两种不同尺寸的颗粒组成。证明了基于力平衡原理推导单个粒子的基于结构的阻力系数与粒子i的体积分数 ( ε _si )无关。商业软件 Ansys Fluent 中的多流体模型用于评估多分散阻力模型，用于在鼓泡流化床中分离二元气体-颗粒流。结果表明，新的基于结构的阻力模型预测的模拟结果与实验数据具有合理的一致性，均方根误差（RMSE）为 6.34%。新的基于结构的模型可以捕获流化床井顶部区域的颗粒分布。新模型也可以很好地捕获气泡行为。