scholarly journals A VARIABLE-DENSITY PROJECTION METHOD FOR INTERFACIAL FLOWS

2003 ◽  
Vol 44 (6) ◽  
pp. 553-574 ◽  
Author(s):  
Ming-Jiu Ni ◽  
Mohamed Abdou ◽  
Satoru Komori
Author(s):  
F. N. Fard ◽  
B. Lessani

A time-accurate numerical algorithm is proposed for low Mach number variable density flows in curvilinear coordinate systems. In order to increase the stability of the method, a predictor-corrector time integration scheme, coupled with the projection method, is employed. The projection method results in a constant-coefficient Poisson equation for the pressure in both the predictor and corrector steps. The continuity equation is fully satisfied at each step. To prevent the pressure odd-even decoupling typically encountered in collocated grids, a flux interpolation technique is developed. The spatial discretization method offers computational simplicity and straightforward extension to 3D curvilinear coordinate systems, which are essential in the simulation of turbulent flows in complex geometries. The accuracy and stability of the algorithm are tested with a series of numerical experiments, and the results are validated against the available data in the literature.


1998 ◽  
Vol 142 (1) ◽  
pp. 1-46 ◽  
Author(s):  
Ann S. Almgren ◽  
John B. Bell ◽  
Phillip Colella ◽  
Louis H. Howell ◽  
Michael L. Welcome

2018 ◽  
Vol 56 (4) ◽  
pp. 2461-2483
Author(s):  
Rénald Chalayer ◽  
Laurent Chupin ◽  
Thierry Dubois

1992 ◽  
Vol 101 (2) ◽  
pp. 334-348 ◽  
Author(s):  
John B Bell ◽  
Daniel L Marcus

Author(s):  
Pablo Gómez ◽  
Claudio Zanzi ◽  
Julián Palacios ◽  
Julio Hernández ◽  
Joaquín López

A graded-adaptive grid projection method to solve the Navier-Stokes equations for incompressible interfacial flows characterized by large density ratios is presented. The numerical model is similar to the one we proposed in [7] and extended to 3D problems in [8]. The free surface is described using a level set method. A Godunov-type method and a Crank-Nicholson temporal discretization scheme are used to solve the advection equation of the level set function and to update of the momentum equation. The reinitialization procedure of the distance function is based on solving a hyperbolic equation to steady state using third-order Runge-Kutta and fifth-order WENO schemes. The conservation equations are discretized on a rectangular adaptive grid with an octree data structure and the pressure stored at the grid cell vertices. In order to avoid spurious pressure oscillations, the velocity components are stored at the cell edges. This new storage scheme combines the advantages of vertex-based schemes, in which the nodes where the pressure is stored are aligned, and cell center-based schemes, which avoid pressure-velocity coupling problems. The numerical model incorporates a continuous surface tension model based on the balanced-force algorithm proposed in [6]. A special treatment of T-nodes (nodes located at vertices, edges or faces of cells belonging to two different refinement levels) is proposed that considerably improves the efficiency of the method. Several tests in two and three dimensions have been carried out to assess the accuracy and efficiency of the proposed method. In this work we present some numerical results for a 3D kinematic test, which are compared with those obtained by other authors. We also present results for the impact of a drop of water onto a liquid surface, which are compared with experimental visualization results.


1997 ◽  
Vol 130 (2) ◽  
pp. 269-282 ◽  
Author(s):  
Elbridge Gerry Puckett ◽  
Ann S. Almgren ◽  
John B. Bell ◽  
Daniel L. Marcus ◽  
William J. Rider

Sign in / Sign up

Export Citation Format

Share Document