Numerical Study of Wheat Conveying in Separator Cyclone using Computational Fluid Dynamics

Abstract

IntroductionCyclones are widely used to separate solid particles from the fluid phase. Due to the ease of construction, low running costs, and hard-working conditions at high temperatures, people's interest in using cyclones is increasing day by day. Engineers are generally interested in two parameters to perform a complete evaluation of the design and operation of a cyclone. These parameters are the particle collecting efficiency and the pressure drop inside the cyclone. The precise prediction of the pressure drop in cyclone is very important which it is directly related to operating costs.Computational Fluid Dynamics (CFD) is a diversified tool for predicting flow behavior in a wide range of design and operational conditions. Numerical solution of Navier-Stokes equations is the basis of all CFD techniques, which is the result of fast computer upgrades and a better understanding of the numerical resolution of turbulence.Materials and MethodsRegarding preliminary experimental tests and understanding the fluid flow, the flow rate of 0.08 kg s-1 was selected as the flow rate. Six levels of inlet velocities 10, 12, 14, 16, 18, and 20 m s-1 were selected for understanding the effect of inlet velocity on the cyclone performance. The measurements were carried out using a hot-air anemometer (TSI-8484model with a resolution of 0.07 m s-1 and an operating range of 0.125 to 50 m s-1), and a pressure differential meter instrument (CPE310s-KIMO model) with an accuracy of 0.1 Pa.The region is discretized as a finite volume in a set, called the region grid or mesh after discretization. For incompressible fluids, pressure-based and density-based solvers are used, respectively. Regarding the velocity of the material entering the cyclone and low Mach number, a pressure-based solver could be used in this study.The shear stress transport model (SST) is a modified version of the k-ω 2-equation model. This model combines the two turbulence k-ω and k-ε models. The Lagrangian discrete phase model in Ansys Fluent follows to the Euler-Lagrangian model.Defining the best type of boundary condition is important for solving the problem and extracting solving fields. The boundary conditions used in this study include the inlet velocity in the entrance of cyclone and output pressure in both the upper and lower output sections.Results and DiscussionIn the results section, the results are initially validated by experimental results. Then, the parameters relating to separation efficiency and pressure drop are discussed. Finally, the tangential and axial velocities are considered as important parameters in the cyclone performance.One of the important issues in the cyclones is the static pressure because it completely affects the phenomenon of separation in the cyclone. The velocities of 16 m s-1 and 18 m s-1 have a good potential for use as the base velocity of the inlet fluid to the cyclone. The velocity of 20 m s-1 is not suitable for separation due to high-pressure drop related to high static pressure.The separation efficiency in the cyclone was 92 to 99% at all levels, the highest separation efficiency of 99% occurred at the velocity of 16 m s-1 and the lowest separation efficiency of 9% happened at the velocity of 20 m s-1.An increasing trend in axial and radial velocities occurred and the highest tangential velocity occurring in the input section. Considering the working conditions, the inlet velocities of 10 m s-1 to 16 m s-1 are appropriate for the turbulence intensity viewpoint.Conclusions(1): The speeds 16 m s-1 and 18 m s-1 showed a good potential for use as a base velocity of the fluid to the cyclone.(2): The highest separation efficiency for the velocity of 16 m s-1 (99%) and lower isolation efficiency was obtained at velocity of 20 m s-1 (92%).(3): The velocities of 10 m s-1 to 16 m s-1 are suitable input rates from the point of view of turbulence intensity.(4): It is concluded that from the point of view of wear to the velocity of 10 to 16 m s-1, practical use is possible, and the velocity of 18 m s-1 and 20 m s-1 require the reinforcement of the relevant sections.

Keywords

• cyclone
• computational fluid dynamics
• pressure drop
• stairmand