Shared-Memory Parallelization for Two-Way Coupled Euler-Lagrange Modeling of Bubbly Flows

2014 ◽  
Author(s):  
Jingsen Ma ◽  
Chao-Tsung Hsiao ◽  
Georges L. Chahine
Keyword(s):  
Shared Memory ◽  
Bubble Dynamics ◽  
Bubbly Flows ◽  
Computational Domain ◽  
Time Step ◽  
Two Phase ◽  
Local Memory ◽  
Lagrangian Solution ◽  
Non Local ◽  
The Continuum

Cavitating bubbly flows are encountered in many engineering problems involving propellers, pumps, valves, ultrasonic biomedical applications, … etc. In this contribution an OpenMP parallelized Euler-Lagrange model of two-phase flow problems and cavitation is presented. The two-phase medium is treated as a continuum and solved on an Eulerian grid, while the discrete bubbles are tracked in a Lagrangian fashion with their dynamics computed. The intimate coupling between the two description levels is realized through the local void fraction, which is computed from the instantaneous bubble volumes and locations, and provides the continuum properties. Since, in practice, any such flows will involve large numbers of bubbles, schemes for significant speedup are needed to reduce computation times. We present here a shared-memory parallelization scheme combining domain decomposition for the continuum domain and number decomposition for the bubbles; both selected to realize maximum speed up and good load balance. The Eulerian computational domain is subdivided based on geometry into several subdomains, while for the Lagrangian computations, the bubbles are subdivided based on their indices into several subsets. The number of fluid subdomains and bubble subsets are matched with the number of CPU cores available in a share-memory system. Computation of the continuum solution and the bubble dynamics proceeds sequentially. During each computation time step, all selected OpenMP threads are first used to evolve the fluid solution, with each handling one subdomain. Upon completion, the OpenMP threads selected for the Lagrangian solution are then used to execute the bubble computations. All data exchanges are executed through the shared memory. Extra steps are taken to localize the memory access pattern to minimize non-local data fetch latency, since severe performance penalty may occur on a Non-Uniform Memory Architecture multiprocessing system where thread access to non-local memory is much slower than to local memory. This parallelization scheme is illustrated on a typical non-uniform bubbly flow problem, cloud bubble dynamics near a rigid wall driven by an imposed pressure function.

Shared-Memory Parallelization for Two-Way Coupled Euler–Lagrange Modeling of Cavitating Bubbly Flows

2015 ◽  
Vol 137 (12) ◽  
Author(s):  
Jingsen Ma ◽  
Chao-Tsung Hsiao ◽  
Georges L. Chahine
Keyword(s):  
Shared Memory ◽  
Bubble Dynamics ◽  
Bubbly Flows ◽  
Processing Unit ◽  
Computational Domain ◽  
Time Step ◽  
Two Phase ◽  
Bubble Cloud ◽  
Cloud Dynamics ◽  
The Continuum

Cavitating and bubbly flows are encountered in many engineering problems involving propellers, pumps, valves, ultrasonic biomedical applications, etc. In this contribution, an openmp parallelized Euler–Lagrange model of two-phase flow problems and cavitation is presented. The two-phase medium is treated as a continuum and solved on an Eulerian grid, while the discrete bubbles are tracked in a Lagrangian fashion with their dynamics computed. The intimate coupling between the two description levels is realized through the local void fraction, which is computed from the instantaneous bubble volumes and locations, and provides the continuum properties. Since, in practice, any such flows will involve large numbers of bubbles, schemes for significant speedup are needed to reduce computation times. We present here a shared-memory parallelization scheme combining domain decomposition for the continuum domain and number decomposition for the bubbles; both selected to realize maximum speedup and good load balance. The Eulerian computational domain is subdivided based on geometry into several subdomains, while for the Lagrangian computations, the bubbles are subdivided based on their indices into several subsets. The number of fluid subdomains and bubble subsets matches with the number of central processing unit (CPU) cores available in a shared-memory system. Computation of the continuum solution and the bubble dynamics proceeds sequentially. During each computation time step, all selected openmp threads are first used to evolve the fluid solution, with each handling one subdomain. Upon completion, the openmp threads selected for the Lagrangian solution are then used to execute the bubble computations. All data exchanges are executed through the shared memory. Extra steps are taken to localize the memory access pattern to minimize nonlocal data fetch latency, since severe performance penalty may occur on a nonuniform memory architecture (NUMA) multiprocessing system where thread access to nonlocal memory is much slower than to local memory. This parallelization scheme is illustrated on a typical nonuniform bubbly flow problem, cloud bubble dynamics near a rigid wall driven by an imposed pressure function (Ma et al., 2013, “Euler–Lagrange Simulations of Bubble Cloud Dynamics Near a Wall,” International Mechanical Engineering Congress and Exposition, San Diego, CA, Nov. 15–21, Paper No. IMECE2013-65191 and Ma et al., 2015, “Euler–Lagrange Simulations of Bubble Cloud Dynamics Near a Wall,” ASME J. Fluids Eng., 137(4), p. 041301).

scholarly journals Bubble Dynamics in a Two-Phase Bubbly Mixture

2010 ◽  
Author(s):  
Arvind Jayaprakash ◽  
Sowmitra Singh ◽  
Georges Chahine
Keyword(s):  
Void Fraction ◽  
High Speed ◽  
Bubble Dynamics ◽  
Bubble Radius ◽  
Discharge Voltage ◽  
Large Bubble ◽  
Water Mixture ◽  
Two Phase ◽  
Mixture Density ◽  
Bubbly Medium

The dynamics of a primary relatively large bubble in a water mixture including very fine bubbles is investigated experimentally and the results are provided to several parallel on-going analytical and numerical approaches. The main/primary bubble is produced by an underwater spark discharge from two concentric electrodes placed in the bubbly medium, which is generated using electrolysis. A grid of thin perpendicular wires is used to generate bubble distributions of varying intensities. The size of the main bubble is controlled by the discharge voltage, the capacitors size, and the pressure imposed in the container. The size and concentration of the fine bubbles can be controlled by the electrolysis voltage, the length, diameter, and type of the wires, and also by the pressure imposed in the container. This enables parametric study of the factors controlling the dynamics of the primary bubble and development of relationships between the bubble characteristic quantities such as maximum bubble radius and bubble period and the characteristics of the surrounding two-phase medium: micro bubble sizes and void fraction. The dynamics of the main bubble and the mixture is observed using high speed video photography. The void fraction/density of the bubbly mixture in the fluid domain is measured as a function of time and space using image analysis of the high speed movies. The interaction between the primary bubble and the bubbly medium is analyzed using both field pressure measurements and high-speed videography. Parameters such as the primary bubble energy and the bubble mixture density (void fraction) are varied, and their effects studied. The experimental data is then compared to simple compressible equations employed for spherical bubbles including a modified Gilmore Equation. Suggestions for improvement of the modeling are then presented.

A Developed Numerical Method for Turbulent Unsteady Fluid Flow in Two-Phase Systems with Moving Interface

2017 ◽  
Vol 14 (06) ◽  
pp. 1750063 ◽  
Author(s):  
A. M. Hegab ◽  
S. A. Gutub ◽  
A. Balabel
Keyword(s):  
Numerical Model ◽  
Gas Phase ◽  
Level Set ◽  
The Body ◽  
Phase System ◽  
Computational Domain ◽  
Two Phase ◽  
Moving Interface ◽  
Level Set Function ◽  
Gas System

This paper presents the development of an accurate and robust numerical modeling of instability of an interface separating two-phase system, such as liquid–gas and/or solid–gas systems. The instability of the interface can be refereed to the buoyancy and capillary effects in liquid–gas system. The governing unsteady Navier–Stokes along with the stress balance and kinematic conditions at the interface are solved separately in each fluid using the finite-volume approach for the liquid–gas system and the Hamilton–Jacobi equation for the solid–gas phase. The developed numerical model represents the surface and the body forces as boundary value conditions on the interface. The adapted approaches enable accurate modeling of fluid flows driven by either body or surface forces. The moving interface is tracked and captured using the level set function that initially defined for both fluids in the computational domain. To asses the developed numerical model and its versatility, a selection of different unsteady test cases including oscillation of a capillary wave, sloshing in a rectangular tank, the broken-dam problem involving different density fluids, simulation of air/water flow, and finally the moving interface between the solid and gas phases of solid rocket propellant combustion were examined. The latter case model allowed for the complete coupling between the gas-phase physics, the condensed-phase physics, and the unsteady nonuniform regression of either liquid or the propellant solid surfaces. The propagation of the unsteady nonplanar regression surface is described, using the Essentially-Non-Oscillatory (ENO) scheme with the aid of the level set strategy. The computational results demonstrate a remarkable capability of the developed numerical model to predict the dynamical characteristics of the liquid–gas and solid–gas flows, which is of great importance in many civilian and military industrial and engineering applications.

scholarly journals Simulation of the Effect of Oil Volume Fractions in an Oil-Water Flows Along a Circular Pipe: A Finite Element Approach

2018 ◽  
Vol 10 (5) ◽  
pp. 19
Author(s):  
Ferdusee Akter ◽  
Md. Bhuyan ◽  
Ujjwal Deb
Keyword(s):  
Finite Element ◽  
Volume Fraction ◽  
Fluid Velocity ◽  
Galerkin Approximation ◽  
Computational Domain ◽  
Two Phase Flows ◽  
Two Phase ◽  
Volume Fractions ◽  
Oil In Water ◽  
Element Approach

Two phase flows in pipelines are very common in industries for the oil transportations. The aim of our work is to observe the effect of oil volume fraction in the oil in water two phase flows. The study has been accomplished using a computational model which is based on a Finite Element Method (FEM) named Galerkin approximation. The velocity profiles and volume fractions are performed by numerical simulations and we have considered the COMSOL Multiphysics Software version 4.2a for our simulation. The computational domain is 8m in length and 0.05m in radius. The results show that the velocity of the mixture decreases as the oil volume fraction increases. It should be noted that if we gradually increase the volume fractions of oil, the fluid velocity also changes and the saturated level of the volume fraction is 22.3%.

scholarly journals Two-Phase Dusty Fluid Flow Along a Rotating Axisymmetric Round-Nosed Body

2017 ◽  
Vol 139 (8) ◽  
Author(s):  
Sadia Siddiqa ◽  
Naheed Begum ◽  
M. A. Hossain ◽  
Rama Subba Reddy Gorla
Keyword(s):  
Boundary Layer ◽  
Flow Characteristics ◽  
Leading Edge ◽  
Solid Particles ◽  
Dust Particles ◽  
Computational Domain ◽  
Two Phase ◽  
Carrier Fluid ◽  
Boundary Layer Equations ◽  
Implicit Finite Difference

This article is concerned with the class of solutions of gas boundary layer containing uniform, spherical solid particles over the surface of rotating axisymmetric round-nosed body. By using the method of transformed coordinates, the boundary layer equations for two-phase flow are mapped into a regular and stationary computational domain and then solved numerically by using implicit finite difference method. In this study, a rotating hemisphere is used as a particular example to elucidate the heat transfer mechanism near the surface of round-nosed bodies. We will investigate whether the presence of dust particles in carrier fluid disturbs the flow characteristics associated with rotating hemisphere or not. A comprehensive parametric analysis is presented to show the influence of the particle loading, the buoyancy ratio parameter, and the surface of rotating hemisphere on the numerical findings. In the absence of dust particles, the results are graphically compared with existing data in the open literature, and an excellent agreement has been found. It is noted that the concentration of dust particles’ parameter, Dρ, strongly influences the heat transport rate near the leading edge.

scholarly journals The Two-Phase Hell-Shaw Flow: Construction of an Exact Solution

2013 ◽  
Vol 18 (1) ◽  
pp. 249-257
Author(s):  
K.R. Malaikah
Keyword(s):  
Exact Solution ◽  
Velocity Potential ◽  
Time Dependent ◽  
Computational Domain ◽  
Phase Problem ◽  
Complex Velocity ◽  
Two Phase ◽  
Interface Evolution ◽  
Complex Velocity Potential ◽  
Branch Cuts

We consider a two-phase Hele-Shaw cell whether or not the gap thickness is time-dependent. We construct an exact solution in terms of the Schwarz function of the interface for the two-phase Hele-Shaw flow. The derivation is based upon the single-valued complex velocity potential instead of the multiple-valued complex potential. As a result, the construction is applicable to the case of the time-dependent gap. In addition, there is no need to introduce branch cuts in the computational domain. Furthermore, the interface evolution in a two-phase problem is closely linked to its counterpart in a one-phase problem

scholarly journals Algorithms for Solving Darcian Flow in Structured Porous Media

10.14311/1829 ◽  
2013 ◽  
Vol 53 (4) ◽  
Author(s):  
Michal Kuráž ◽  
Petr Mayer
Keyword(s):  
Proper Time ◽  
Time Integration ◽  
Mass Term ◽  
Step Length ◽  
Richards Equation ◽  
Parabolic Problems ◽  
Computational Domain ◽  
Wetting Front ◽  
Time Step ◽  
The Matrix

This paper presents several algorithms that were implemented in DRUtES [1], a new open source project. DRUtES is a finite element solver for coupled nonlinear parabolic problems, namely the Richards equation with the dual porosity approach (modeling the flow of liquids in a porous medium). Mass balance consistency is crucial in any hydrological balance and contaminant transportation evaluations. An incorrect approximation of the mass term greatly depreciates the results that are obtained. An algorithm for automatic time step selection is presented, as the proper time step length is crucial for achieving accuracy of the Euler time integration method. Various problems arise with poor conditioning of the Richards equation: the computational domain is clustered into subregions separated by a wetting front, and the nonlinear constitutive functions cover a high range of values, while a very simple diagonal preconditioning method greatly improves the matrix properties. The results are presented here, together with an analysis.

A Meshfree-Based Lattice Boltzmann Approach for Simulation of Fluid Flows Within Complex Geometries

2018 ◽  
pp. 188-222 ◽  
Author(s):  
Sonam Tanwar
Keyword(s):  
Lattice Boltzmann Method ◽  
Lattice Boltzmann ◽  
Moving Boundary ◽  
Fluid Flows ◽  
Computational Domain ◽  
Time Step ◽  
Boundary Problems ◽  
Complex Process ◽  
Element Free ◽  
Boltzmann Method

This chapter develops a meshless formulation of lattice Boltzmann method for simulation of fluid flows within complex and irregular geometries. The meshless feature of proposed technique will improve the accuracy of standard lattice Boltzmann method within complicated fluid domains. Discretization of such domains itself may introduce significant numerical errors into the solution. Specifically, in phase transition or moving boundary problems, discretization of the domain is a time-consuming and complex process. In these problems, at each time step, the computational domain may change its shape and need to be re-meshed accordingly for the purpose of accuracy and stability of the solution. The author proposes to combine lattice Boltzmann method with a Galerkin meshfree technique popularly known as element-free Galerkin method in this chapter to remove the difficulties associated with traditional grid-based methods.

Sign in / Sign up

Export Citation Format

Share Document