Abstract
Gas-particle flows in circulating fluidized beds (CFB) with wide particle size distributions were simulated using the Eulerian-Eulerian approach to analyze the effects of the particle phase division and the applicability of the particle-particle drag model. The results indicate that the simulation is not accurate by just using a single average particle diameter when the particle size distribution includes a critical particle diameter. A binary particle phase division criterion was then developed to establish two particle phases representing two types of particles with different flow patterns. Coupling the Eulerian-Eulerian approach with the new criterion enabled accurate predictions of the pressures, particle volume fractions, and particle mass circulation rates that were in agreement with experimental data. The influences of different particle-particle drag models were also investigated to show that the simulation using the Syamlal model was not accurate due to the overestimated particle-particle drag, while the results without particle-particle drag and with the Manger model were similar and much more accurate. Moreover, the flow mechanism for the non-uniformity of particle circulation rates in the parallel circulating loops of the CFB boiler was revealed. This study improves the Eulerian-Eulerian simulations of fluidization with wide particle size distributions and further deepens the understanding of flow characteristics in CFB.
| Original language | English |
|---|---|
| Pages (from-to) | 357-372 |
| Number of pages | 16 |
| Journal | International Journal of Chemical Reactor Engineering |
| Volume | 20 |
| Issue number | 3 |
| DOIs | |
| State | Published - 1 Mar 2022 |
| Externally published | Yes |
Keywords
- Eulerian-Eulerian approach
- Parallel circulating loops
- Particle phase division criterion
- Particle-particle drag
- Wide particle size distributions
Fingerprint
Dive into the research topics of 'Numerical study on key issues in the Eulerian-Eulerian simulation of fluidization with wide particle size distributions'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver