Session H3: Multiphase Flows V

A Robust Numerical Method for Compressible Dense Granular Flows

Dense granular flows are important for problems such as coal mine explosions or interior ballistics in which flow compressibility and the presence of shocks are important. Numerical solutions of such flows has been plagued with difficulties arising from non-conservative nozzling terms, which are often neglected for numerical convenience. The ``cure" has been to use highly dissipative numerical methods to avoid instability when the non-conservative terms physically must be included. Second-order methods or even refining the grid can reintroduce these numerical instabilities. We developed a robust and high-order numerical method for solving dense granular flows in highly compressible situations that circumvents these problems. The technique has been verified on a number of test problems including advection of a granular material interface, granular shocks, and transmission angles of oblique compaction waves. The method has been demonstrated in challenging situations where a shock impacts a dense layer of dust on very fine meshes approaching the continuum limit of the granular phase.   

Study of Influence of Experimental Technique on Measured Particle Velocity Distributions in Fluidized Bed

Fluid flows that are loaded with high concentration of solid particles are common in oil and chemical processing industries. However, the opaque nature of the flow fields and the complex nature of the flow have hampered the experimental and computational study of these processes. This has led to the development of a number of customized experimental techniques for high concentration particle flows for evaluation and improvement of CFD models. This includes techniques that track few individual particles, measures average particle velocity over a small sample volume and those over a large sample volume. In this work novel high speed PIV (HsPIV), with individual particle tracking, was utilized to measure velocities of individual particles in gas-particle flow fields at the walls circulating and bubbling fluidized bed. The HsPIV measurement technique has the ability to simultaneously recognize and track thousands of individual particles in flows of high particle concentration. To determine the effect of the size of the sample volume on particle velocity measurements, the PDF of Lagrangian particle velocity was compared with the PDF of Eulerian for different domain sizes over a range of flow conditions. The results will show that measured particle velocity distribution can vary from technique to technique and this bias has to be accounted in comparison with CFD simulations.   

A novel finite element framework for numerical simulation of fluidization processes and multiphase granular flow

We present results from a new formulation of a numerical model for direct simulation of bed fluidization and multiphase granular flow. The model is based on a consistent application of continuous-discontinuous mixed control volume finite element methods applied to fully unstructured meshes. The unstructured mesh framework allows for both a mesh adaptive capability, modifying the computational geometry in order to bound the error in the numerical solution while maximizing computational efficiency, and a simple scripting interface embedded in the model which allows fast prototyping of correlation models and parameterizations in intercomparison experiments. The model is applied to standard test problems for fluidized beds.   

Material Point Method and Multi-velocity Formulation for History Dependent Phase Transitions

Phase transition has been used to describe both physical and chemical phenomena ranging from melting of ice, erosion of dirt and sand by storm water, combustion of fuel and pulverization of materials. Many of these processes involve effect of history. Tracing history in cases of extreme material deformation has been a significant issue for many numerical methods, especially in cases of phase transitions. The material point method (MPM) is a numerical method based on the wake solution principle of solving a set of partial differential equations. The starting point in this talk is a set of multi-velocity equations obtained based on statistical description of the materials. The equations and the numerical method allow for the use of realistic material models in cases of extreme deformation. The advantage of this statistical description is its convenience in the consideration of the transition between interacting continua to disperse debris flows. MPM uses Eulerian velocity field to describe the extreme material deformation, while uses Lagrangian material points to track material deformation history. We show comparisons of this multi-velocity formulation with traditional approaches and new capabilities of this formation and the numerical method.   

Computational model and simulations of gas-liquid-solid three-phase interactions

A computational technique to model three-phase (gas-liquid-solid) interactions is proposed in this study. This numerical algorithm couples a connectivity-free front-tracking method that treats gas-liquid multi-fluid interface to the immersed finite element method that treats fully-coupled fluid-solid interactions. The numerical framework is based on a non-boundary-fitted meshing technique where the background grid is fixed where no mesh-updating or re-meshing is required. An indicator function is used to identify the gas from the liquid, and the fluid (gas or liquid) from the solid. Several 2-D and 3-D validation cases are demonstrated to show the accuracy and the robustness of the method.   

Analysis on the formation and growth of condensing aerosol particles in a turbulent mixing layer

A simulation of the formation and growth of dibutyl phthalate (DBP) particles in a three-dimensional turbulent mixing layer is performed to investigate the effects of turbulence on the aerosol evolution. A fast, hot stream with DBP vapor is mixed with a slow, cold stream achieving supersaturation by turbulent mixing. The aerosol dynamics are solved with the quadrature method of moments, and the moments are transported via a Lagrangian particles scheme. The results show that aerosol particles are formed in the cold stream, while they grow rapidly in the hot stream. The differential diffusion of temperature/vapor concentration and aerosol particles is investigated through conditional statistics in the mixture fraction space. Aerosol particles formed in the cold stream tend to drift towards the hot stream and grow substantially there.   

Time resolved measurements of rigid fiber dispersion in near homogeneous isotropic turbulence

Time resolved, planar particle image velocimetry (PIV, 3kHz) and two-orthogonal view, digital holographic cinematography (2kHz) was used to measure 3D fiber trajectories/orientation dynamics in near homogeneous isotropic air turbulence (HIT) with dilute suspended fibers. The PIV covered a field of view of 6x12 mm$^{\mathrm{2}}$ and the holography a volume of interest of 17$^{\mathrm{3}}$ mm$^{\mathrm{3}}$, positioned at the center of the chamber. HIT (Re$_{\lambda }=$144) was generated in the center of a 40$^{\mathrm{3}}$ cm$^{\mathrm{3}}$ cube by eight woofers mounted on each of its corners. Three different nylon fibers having a length of 0.5 mm and diameter of 10, 14 and 19$\mu $m were released from the top of the chamber. Fibers had Stokes numbers of order one and are expected to accumulate in regions of low vorticity and settle along a path of local minimal drag. Fiber 3D trajectories/orientations have been obtained from the holography measurements and orientational/translational dispersion coefficients will be presented. In addition the flow field in the vicinity of tracked fibers has been resolved by the PIV, and results on fluid and fiber accelerations and position correlation with in-plane strain rate and out-of-plane vorticity will be presented.   

Fine Structure in Energy Dissipation at the onset of turbulence under oscillatory Flow

Fine structure formation studies have been an active area of reserach in recent past and is very much associated in turbulence phenomena. The large scale structures are of ow width and contain most of the energy which dominates transport of mass, momentum and heat. The small scales include dissipative range responsible for most of the energy dissipation and inertial range. Since small scales contain most of interesting formation of structures in terms of physics and simulation that is the most obvious reason we are interested in fine structure formation of the small scale turbulence structure. The analytical side of studies focus mostly on the singularities of Navier Stokes equation have natural connection to far away dissipation range. Small scale turbulence is a fertile ground for studying on vortex breakdown and reconnection that carries many interesting physics. Research on small scale turbulence have remarkable influence heat transfer and particle/chemical phenomena. In this work, we intend to focus on numerical investigation of dynamics of vortex structure of small scale at the onset of turbulence at various Reynold number.   

Development of iterative algorithms of increased convergence and accuracy for multiphase flow simulation

Newton's method is commonly used in reservoir simulation problems. However, it doesn't have the property of globally convergence (Younis et al., 2008). Most of the convergence problems are generally related to change of the nonlinear equation being solved during iteration process (Maksimov et al., 2010): 1) change of phase flow direction; 2) change of the well working target (fixed rate - limit bottom hole pressure - shut); 3) appearance and disappearance of a phase; 4) appearance of movable phase; 5) others, connected with problem formulation improvement. Note that the form of the approximating equations is unknown in advance and is determined by the solution itself. We have considered approaches of improving convergence of Newton's method for reservoir flow problems in general problem formulation, characterized by dependence of the equation approximation form on the solution itself and thus changing of the approximating equation during the iteration process. The approach is based on the universal principle of decreasing of each residual component in all cells. For determination of chopping level, points of approximation form change are employed. As an addition to basic approach, controlled violation of approximation rules was considered, not affecting material balance of system.