Coarse-grain DEM-CFD modelling of dense particle flow in Gas–Solid cyclone

Numerical Investigation of Cyclone Separators: Physical Mechanisms and Theoretical Algorithms

“…However, the DEM model fully describes the particle interactions during contacts such as particle rotation, rolling, and sliding. It faces high computational requirements cost, especially with high particle numbers, which is considered a critical problem [216]. Therefore, the existing DEM is difficult to apply in large-scale cyclonic systems where over a billion particles are typically used.…”

“…Mirzaei et al [1] grain parcel is defined by [233,234]. With this scaling up, the time step used for the simulation also increases, leading to much faster computations and guaranteeing numerical accuracy by resolving inter-particle collisions [165,216,235]. The coarse-graining concept is solved at grain (parcel) scale with a Lagrangian method, which has a similar governing equations algorithm to the pure DEM of translational and rotational motions [230,236,237].…”

“…However, it should be noted that the one-way coupling approach may not be suitable for highly concentrated particleladen flows, where the particles can significantly affect the fluid flow. In such cases, two-way or four-way coupling methods may be necessary to model the gas-particle flow accurately [166,216].…”

Hydrodynamic Full-Loop Simulation of an Industrial Fluid Catalytic Cracking Fluidized Bed System with Mechanical Valves

“…As the computational ability and multiphase flow theory have developed by leaps and bounds, computational fluid dynamics (CFD) has become a powerful tool for revealing complex hydrodynamics in multiphase flows. − Among many simulation approaches, at present, only two methods are feasible for simulations of large-scale/industrial-scale gas–solid fluidized beds, i.e., the Eulerian–Eulerian method (also namely, two-fluid model, TFM) and coarse-grained Eurerian–Lagrangian method like the multiphase-particle-in-cell (MP-PIC), coarse-grained discrete particle model (DPM), etc. In recent years, the coarse-grained method has developed rapidly and gained more ground in large-scale simulations; − however, it still shows inadequacy in predicting complex industrial reactors like those involving a huge number of particles as well as equipped with internal baffles and perforated distributors since the usage of a high coarse-graining ratio in treating millions of particles requires additional modeling in parcel-to-parcel collisions and parcel–gas interaction and meanwhile greatly increases the collision probability within orifices of distributors. The contradictory requirements, respectively, from the computational burden and the complex geometry restrict the wide application of coarse-grained methods nowadays. ,− By comparison, the TFM simulation with the help of its moderate computational expense and improved accuracy by mesoscale modeling is still the most popular approach in industrial applications, especially in predicting complex, large-scale ones, such as the full-loop CFB system. − …”

Cfd Simulation and Experimental Validation of Multiphase Flow in Industrial Cyclone Preheaters

“…However, conventional four-way coupling models namely the Discrete Element Method (DEM) model and the Eulerian-Eulerian model might not be suitable options. DEM is too expensive to be implemented in industrial-scale simulations (Although coarse-graining strategies are available to reduce computational expenses of DEM [30][31][32], its current application in cyclones is still limited to lab-scale cases [22][23][24][25][26][27]), and the Eulerian-Eulerian model cannot easily handle the polydispersity of particles which is quite crucial in the performance of cyclones. On the other hand, a more novel hybrid multiphase model, the Dense Discrete Phase Model, that has been successfully validated for the simulation of hydrodynamics of multiphase flow in a highly loaded large-scale cyclone in our previous work [33,34] is a promising and feasible model.…”