scispace - formally typeset
Search or ask a question
Journal ArticleDOI

A class of finite difference schemes for interface problems with an HOC approach

TL;DR: In this article, the authors proposed a new methodology for numerically solving elliptic and parabolic equations with discontinuous coefficients and singular source terms, obtained by clubbing a recently developed Higher Order Compact (HOC) methodology with special interface treatment for the points just next to the points of discontinuity.
Abstract: In this paper, we propose a new methodology for numerically solving elliptic and parabolic equations with discontinuous coefficients and singular source terms. This new scheme is obtained by clubbing a recently developed Higher Order Compact (HOC) methodology with special interface treatment for the points just next to the points of discontinuity. The overall order of accuracy of the scheme is at least second. We first formulate the scheme for one-dimensional (1D) problems and then extend it directly to two-dimensional (2D) problems in polar coordinates. In the process, we also perform convergence and related analysis for both the cases. Finally, we show a new direction of implementing the methodology to 2D problems in cartesian coordinates. We then conduct numerous numerical studies on a number of problems, both for 1D and 2D cases including the flow past circular cylinder governed by the incompressible Navier-Stokes (N-S) equations. We compare our results with existing numerical and experimental results. In all the cases our formulation is found to produce better results on relatively coarser grids. For the circular cylinder problem, the scheme used is seen to capture all the flow characteristics including the famous von-Karman vortex street. Copyright c © 2010 John Wiley & Sons, Ltd.

Content maybe subject to copyright    Report

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS
Int. J. Numer. Meth. Fluids 2010; 00:137
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/fld
A Class Of Finite Difference Schemes For Interface Prob lems
With An HOC Approach
H. V. R. Mittal
1
, Jiten C. Kalita
2
, Rajendra K. Ray
1
1
School of Basic Sciences, Indian Institute of Technology Mandi, India.
2
Department of Mathematics, Indian Institute of Technology Guwahati, India.
SUMMARY
In this paper, we propose a new methodology for numerically solving elliptic and parabolic equations with
discontinuous coefficients and singular source terms. This new scheme is obtained by clubbing a recently
developed Higher Order Compact (HOC) methodology with special interface treatment for the points just
next to the points of discontinuity. The overall order of accuracy of the scheme is at least second. We first
formulate the scheme for one-dimensional (1D) problems and then extend it directly to two-dimensional
(2D) problems in polar coordinates. In the process, we also perform convergence and related analysis for
both the cases. Finally, we show a new direction of implementing the methodology to 2D problems i n
cartesian coordinates. We then conduct numerous numerical studies on a number of problems, both for 1D
and 2D cases including the flow past circular cylinder governed by the incompressible Navier-Stokes (N-S)
equations. We compare our res ults with existing numerical and experimental results. In all the cases our
formulation is found to produce better res ults on relatively coarser grids. For the circular cylinder problem,
the scheme used is seen to capture all the flow characteristics including t he famous von-K´arm´an vortex
street. Copyright
c
2010 John Wiley & Sons, Ltd.
Received . . .
KEY WORDS: Interface; HOC scheme; Navier-Stokes equation; discontinuous coefficients; non-
uniform grids; von-K´arm´an vortex street.
1. INTRODUCTION
Flow problems involving discon tinuous coefficients and singular source terms still pose a tough
challenge in the field of computational fluid dynam ics (CFD). Such problems g e nerally have
non smooth solutions with jumps across lower dimensional in terfaces. Standard finite difference
approximations fail t o yield correct numerical solutions of such problems because of the fact that
the Taylor’s expansions, which provide the platforms for such approximations, are not valid for non
smooth functions. Of l ate, there has been a spur of i nterest in developing numerical methods for
computing mult iphase flows and multi-physics problems with interfaces. Such problem s arise in
numerous branches of science and engineering such as biochemical processing, free surface flow,
drop deformation, solid mechanics, porous media flow, formation of gas bubbles in liquid, mining
etc. For example, in problems involving dissimilar materials, at the i nterfaces, the m aterial properties
(elastic moduli, permeability, conductivity, etc) are discontinuous.
The last few decades have seen several approaches for numerically solving the interface equat ions
[
18, 19, 20, 31, 39, 40, 45]. Mo st of the earlier numerical works on such problems involved mainl y
the use of immersed boundary (IBM) or immersed interface (IIM) methods on uniform grids and
Correspondence to: School of Basic Sciences, Indian Institute of Technology Mandi, India
Copyright
c
2010 John Wiley & Sons, Ltd.
Prepared using fldauth.cls [Version: 2010/05/13 v2.00]
Citations
More filters
Journal ArticleDOI
TL;DR: In this article, numerical simulations of two-dimensional flow around a rotationally oscillating circular cylinder, placed in a uniform cross flow of a constant property Newtonian fluid, are performed at a fixed Reynolds number of 200.

33 citations

Journal ArticleDOI
TL;DR: In this paper, a heated rotationally oscillating circular cylinder placed in a uniform cross flow of constant properties fluid is investigated and the two-dimensional governing equations of flow motion and energy are solved numerically on non-uniform polar grids using a higher order compact (HOC) formulation.

25 citations

Journal ArticleDOI
TL;DR: In this article, a mesh-free, radial basis function-generated finite difference (RBF-FD)-based method for designing stencil weights that can be applied directly to data that crosses an interface is presented.
Abstract: When thermal diffusivity does not vary smoothly within a computational domain, standard numerical methods for solving heat equilibrium problems often converge to an inaccurate solution. In the present paper, we discuss a mesh-free, radial basis function-generated finite difference (RBF-FD)-based method for designing stencil weights that can be applied directly to data that crosses an interface. The approach produces a very accurate solution when thermal diffusivity varies smoothly on either side of an interface. It continues to produce high-quality results when a region between two interfaces is much smaller that the distance between adjacent discrete data nodes in the domain (as becomes the case for thin, nearly insulating layers). We give several test cases that demonstrate the method solving heat equilibrium problems to 4th-order accuracy in the presence of smoothly-curved interfaces.

25 citations

Journal ArticleDOI
TL;DR: The efficiency of the scheme is demonstrated not only by its ability to capture smaller scales up to post quaternary level of vortices for the flow in lid-driven cavity, but also its robustness in handling non-rectangular boundaries of immersed bodies.

11 citations

References
More filters
Journal ArticleDOI
TL;DR: In this article, the authors extended previous work on the solution of the Navier-Stokes equations in the presence of moving immersed boundaries which interact with the fluid and introduced an improved numerical representation of the δ-function.

2,968 citations


"A class of finite difference scheme..." refers background in this paper

  • ...70s [8], originally with the purpose of modeling blood flow in arteries where singular forces were smeared out by a discrete delta function....

    [...]

Journal ArticleDOI
TL;DR: In this paper, the authors present a visualisation de l'ecoulement for tourbillon and dynamique des: fluides, aubes, cylindre, instabilite.
Abstract: Keywords: visualisation de l'ecoulement ; tourbillon ; dynamique des : fluides ; aubes ; cylindre ; instabilite ; ecoulement : secondaire Note: moult photos Reference Record created on 2005-11-18, modified on 2016-08-08

1,654 citations

Journal ArticleDOI
TL;DR: In this paper, the authors developed finite difference methods for elliptic equations of the form \[ abla \cdot (\beta (x)) + \kappa (x)u(x) = f(x)) in a region in one or two dimensions.
Abstract: The authors develop finite difference methods for elliptic equations of the form \[ abla \cdot (\beta (x) abla u(x)) + \kappa (x)u(x) = f(x)\] in a region $\Omega $ in one or two space dimension...

1,442 citations


"A class of finite difference scheme..." refers background or methods or result in this paper

  • ...Leveque and Li’s IIM [1] considered the interface that was typically not aligned to the grid but rather cut between grid points....

    [...]

  • ...Most of the earlier numerical works on such problems involved mainly the use of immersed boundary method or immersed interface method (IIM) on uniform grids, and their global order of accuracy was at most second....

    [...]

  • ...Recently, Chen et al. [5] applied explicit jump immersed interface method (EJIIM) to the incompressible N-S equations, when singular forces and discontinuous viscosity might exist along some interface in the solution domain....

    [...]

  • ...This table also confirms that the overall orders of accuracy of our scheme is greater than 2 and is better than those obtained by Leveque and Li’s original IIM [1], Wiegmann’s EJIIM scheme [7], and decomposed IIM of Berthelsen [20]....

    [...]

  • ...To validate our 2D formulation in Cartesian coordinates, we present two examples that were also used by Leveque and Li [1] to validate their second-order IIM method....

    [...]

Journal ArticleDOI
TL;DR: In this paper, a new immersed-boundary method for simulating flows over or inside complex geometries is developed by introducing a mass source/sink as well as a momentum forcing.

1,090 citations


"A class of finite difference scheme..." refers background in this paper

  • ...[18], Russell and Wang [19], Berthelsen [20], Gilmanov and Sotiropoulos [21], Taira and colonius [22], Ge and Sotiropoulos [23] , Borazjani et al....

    [...]

Journal ArticleDOI
TL;DR: In this article, the authors describe measurements of the drag on circular cylinders, made by observing the bending of quartz fibres, in a stream with the Reynolds number range 0·5-100.
Abstract: Part I describes measurements of the drag on circular cylinders, made by observing the bending of quartz fibres, in a stream with the Reynolds number range 0·5-100. Comparisons are made with other experimental values (which cover only the upper part of this range) and with the various theoretical calculations.Part II advances experimental evidence for there being a transition in the mode of the vortex street in the wake of a cylinder at a Reynolds number around 90. Investigations of the nature of this transition and the differences between the flows on either side of it are described. The interpretation that the change is between a vortex street originating in the wake and one originating in the immediate vicinity of the cylinder is suggested.

1,075 citations


"A class of finite difference scheme..." refers result in this paper

  • ...For Re D 20, our drag coefficient is about 9:8% lower than the experimental result of Tritton [34], whereas for Re D 40, it is about 5:1% higher than the experimental results [34]....

    [...]

  • ...Tritton DJ. Experiments on the flow past a circular cylinder at low Reynolds numbers....

    [...]