A simple stabilized finite element method for solving two phase compressible-incompressible interface flows

被引:17
作者
Billaud, Marie [1 ]
Gallice, Gerard [2 ]
Nkonga, Boniface [3 ,4 ]
机构
[1] UMR CELIA, F-33405 Talence, France
[2] CEA, CESTA, F-33114 Le Barp, France
[3] Univ Nice Sophia Antipolis, Lab JA Dieudonne, F-06108 Nice, France
[4] PUMAS, INRIA Sophia Antipolis, F-06902 Sophia Antipolis, France
关键词
Two phase flows; Interface; Unified Navier-Stokes equations; Primitive unknowns; Stabilized finite element method; Level set method; NAVIER-STOKES EQUATIONS; LEAST-SQUARES DISCRETIZATIONS; COMPUTATIONAL FLUID-DYNAMICS; IMPROVED UNIFIED SOLVER; MULTIPHASE FLOWS; EULER EQUATIONS; FREE SURFACES; FORMULATION; CONVECTION; OPERATORS;
D O I
10.1016/j.cma.2010.10.017
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
When computing interface flows between compressible (gas) and incompressible (liquid) fluids, one faces at least to the following difficulties: (1) transition from a gas density linked to the local temperature and pressure by an equation of state to a liquid density mainly constant in space, (2) proper approximation of the divergence constraint in incompressible regions and (3) wave transmission at the interface. The aim of the present paper is to design a global (i.e. the same for each phase) numerical method to address easily this coupling. To this end, the same set of primitive unknowns and equations is used everywhere in the flow, but with a dynamic parameterization that changes from compressible to incompressible regions. On one hand, the compressible Navier-Stokes equations are considered under weakly compressibility assumption so that a non-conservative formulation can be used. On the other hand, the incompressible non-isothermal model is retained. In addition, the level set transport equation is used to capture the interface position needed to identify the local characteristics of the fluid and to recover the adequate local modelling. For space approximation of Navier-Stokes equations, a Galerkin least-squares finite element method is used. Two essential elements for defining this numerical scheme are the stabilization and the computation of element integral of the approximated weak form. Since very different concerns motivate the need for stabilization in compressible and incompressible flows, the first difficulty is to design a stabilization operator suitable for both types of flows especially in mixed elements. In addition, some integral of discontinuous functions must be correctly computed to ensure interfacial wave transmission. To overcome these two difficulties, specific averages are computed especially near the interface. Finally, the level set transport equation is computed by a quadrature free Discontinuous Galerkin method. Numerical strategies are performed and validated for 1D and 2D applications. (C) 2010 Elsevier B.V. All rights reserved.
引用
收藏
页码:1272 / 1290
页数:19
相关论文
共 42 条
[1]   Low mach number limit of the full Navier-Stokes equations [J].
Alazard, T .
ARCHIVE FOR RATIONAL MECHANICS AND ANALYSIS, 2006, 180 (01) :1-73
[2]   A 2-PHASE MIXTURE THEORY FOR THE DEFLAGRATION-TO-DETONATION TRANSITION (DDT) IN REACTIVE ANTIGRANULOCYTES-MATERIALS [J].
BAER, MR ;
NUNZIATO, JW .
INTERNATIONAL JOURNAL OF MULTIPHASE FLOW, 1986, 12 (06) :861-889
[3]   BUBBLES IN VISCOUS-LIQUIDS - SHAPES, WAKES AND VELOCITIES [J].
BHAGA, D ;
WEBER, ME .
JOURNAL OF FLUID MECHANICS, 1981, 105 (APR) :61-85
[4]  
BILLAUD M, 2009, THESIS U BORDEAUX, P1
[5]   STREAMLINE UPWIND PETROV-GALERKIN FORMULATIONS FOR CONVECTION DOMINATED FLOWS WITH PARTICULAR EMPHASIS ON THE INCOMPRESSIBLE NAVIER-STOKES EQUATIONS [J].
BROOKS, AN ;
HUGHES, TJR .
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 1982, 32 (1-3) :199-259
[6]   A numerical method for two-phase flow consisting of separate compressible and incompressible regions [J].
Caiden, R ;
Fedkiw, RP ;
Anderson, C .
JOURNAL OF COMPUTATIONAL PHYSICS, 2001, 166 (01) :1-27
[7]   Runge-Kutta discontinuous Galerkin methods for convection-dominated problems [J].
Cockburn, Bernardo ;
Shu, Chi-Wang .
Journal of Scientific Computing, 2001, 16 (03) :173-261
[8]   Mass preserving finite element implementations of the level set method [J].
Di Pietro, Daniele A. ;
Lo Forte, Stefania ;
Parolini, Nicola .
APPLIED NUMERICAL MATHEMATICS, 2006, 56 (09) :1179-1195
[9]   A higher-order generalized ghost fluid method for the poor for the three-dimensional two-phase flow computation of underwater implosions [J].
Farhat, Charbel ;
Rallu, Arthur ;
Shankaran, Sriram .
JOURNAL OF COMPUTATIONAL PHYSICS, 2008, 227 (16) :7674-7700
[10]  
Fedkiw RP, 1999, J COMPUT PHYS, V152, P457, DOI 10.1006/jcph.1999.6136