Application of Coupled Solver Technology to CFD Modelling of Multiphase Flows with CFX

Abstract

Reactor beds have been investigated using simplified methods, the reason being restrictions in resolving the reactor bed geometry. Such methods require empirical data which essentially restrict the generality and reliability of the predictions.It is proposed to resolve the reactor bed geometry using a digitalisation method, importing the 3-D geometric information into the CFD code, generating a computational grid automatically and conducting detailed analysis of the fluid flow regime, heat transfer and chemical reactions tacking place.Preliminary simulations of a reactor bed ± void fraction 72 % ± predict an overall heat transfer coefficient within 15 % of the experimental value.Multiphase flows are involved in almost all of the processes in chemical engineering; for example: gas bubbles in liquid columns, solid particles in mixing vessels, liquid droplets in spray dryers. Multiphase flow problems are inherently very complex, many of the physical processes (heat, mass and momentum) are linked and coupled between the phases. Despite the difficulties, many successful CFD calculations have been made and reported in the literature. In most of the current CFD algorithm for multiphase flows a segregated solution method is used. The momentum, continuity and other equations are solved in separated steps. The whole calculation sequent must be repeated or iterated many times (typically thousands of iterations) for the solution to address all the coupling terms properly and converge to the final solution. For single-phase flows convergence speed can be improved by using what is known as a coupled solver technology. Within this technique the governing transport equations are solved all at a time. This strategy has been implemented in CFX-5 to model and simulate two-phase flows. This paper describes the coupled solver technology implemented in the CFX-5 software and examines the advances of this technology in terms of CFD performance by way of engineering examples. Validation tests modeling bubbly flow in an airlift reactor demonstrate good convergence characteristics and linear scalability of the solver with respect to both problem size and parallel computation. Numerical results of the simulation are in very good agreement with a previously published study.Kerosene pool fires with pool diameters 8 m d 25 m have been modeled using the CFX-4.3 code by AEATechnology. The employed software includes a low Reynolds number k-e turbulence model and, as far as the combustion process is concerned, a version of the eddy break up model. Calculations of transient flow velocities and temperature fields were made and axisymmetry was assumed. The results show a periodically rise of vortices from the flame base Figure.Digitalised segment of catalytic reactor bed.638 C o m p u t a t i o n a l E n g i n e e r i n g Chemie Ingenieur Technik (73) 6 I 2001

OpenFOAM Computational Fluid Dynamic Simulations of Two-Phase Flow and Mass Transfer in an Advanced-Flow Reactor

“…12,13 Capturing detachment, deformation, breakup, and coalescence of bubbles and drops becomes a challenging task. An additional challenge is to capture accurately capillary effects, which are important on small scales.…”

“…Efforts over the years have been directed toward the simulation of multiphase flow with the aim of capturing the dynamic events that are observed experimentally. , Capturing detachment, deformation, breakup, and coalescence of bubbles and drops becomes a challenging task. An additional challenge is to capture accurately capillary effects, which are important on small scales.…”

“…Among the continuum methods used to model two-phase flow, there are two main approaches: − the Lagrangian and the Eulerian. The use of a moving (unstructured) mesh is included within the first approach with a Lagrangian movement of the interface.…”

Buoyancy driven flow of light gases in atmosphere compared to that of hot gases

“…The accuracy of both methods was chosen to be of second order to reduce the effect of numerical diffusion. The system of algebraic equations was then solved to find the value of the variables using the coupled scheme (Burns et al (2001)) that solves the equations in a single system. The convergence criterion was set at 10 À5 and the number of iterations per time step was set at 500.…”

Fully-coupled pressure-based two-fluid solver for the solution of turbulent fluid-particle systems

“…The approach shows dramatic improvements in convergence on both structured and unstructured grids in comparison to the SIMPLE algorithm and shows superior performance on denser meshes. This framework [16,26,36,37] has since been extended to multiphase applications -namely the two-fluid model [3]. In this approach the entire system is coupled i.e.…”