CFD Study on MR-deDuster As a Fine Particle Emission Control

Abstract

Reynolds stress model (RSM) is known to demonstrate the closest approximation with experimental data and appraised as the most reliable turbulent model for the flow field in a cyclone simulation against other models. In this study, computational fluid dynamics (CFD) with RSM was used to simulate MR-deDuster, which is a newly developed multi-cyclone for fine particle emission control with axial inlet entry. The flow field characteristics of velocity contour and particle trajectories inside the cyclone were profiled while particle cut-diameter (dpc) of the unit was predicted based on fractional collection efficiency. Throughout the simulation, three different inlet velocities represented by their respective volumetric air flow rates (Q) ranging from 0.27 to 0.35 m3/s were applied and the results were plotted accordingly by referring to the value of these Qs. The velocity flows are almost identical for every unit of the cyclone in MR-deDuster system with the highest velocity of 33 m/s was found at the end of the vortex finder operated at 0.35 m3/s. This is crucial to ensure that each unit is operated in similar mode to get optimum performance of the system. Meanwhile, discrete phase model was used to estimate the particle trajectories in the flow. By releasing a certain number of inert particles at the cyclone’s inlet, fractional collection efficiency plot could be determined using the ratio of tracked particles trapped and released through the outlet. This study indicated that the predicted dpc of MR-deDuster was 1.85 and 2.05 μm for the highest and lowest Q respectively. The former represents the highest inlet velocity which is able to attract more of fine particle deep inside the bottom of the cyclone compared to the latter. Thus, the predicted fractional collection efficiency of 100% was achieved for particle larger and equal to 10 μm in size at the highest volumetric air flow rate or inlet velocity. The importance of inlet velocity in reducing the dpc which simultaneously increase the performance of a cyclone is illustrated in this study as successfully demonstrated by the CFD software.