The coupling of bem and fem for the two-dimensional viscous flow problem

In this paper we propose a numerical scheme for treating the problem of sJow viscous flow past an obstacle in the plane. This scheme is a combination of boundary element and finite element methods. By introducing an auxiliary boundary curve, we divide the region under consideration into two subregions, an inner and an outer region. In the inner region, we employ a finite element method (FEM) for solving a system of simplified field equations with proper natural boundary conditions. In the outer region, the solution is expressed in the form of a simple-layer potential with density function satisfying a system of modified integral equations of the first kind. The latter are solved by a boundary element method (BEM). Both solutions are matched on the common auxiliary boundary curve. Error estimates in suitable function spaces are derived in terms of the mesh widths as well as the small parameters, the Reynolds numbers     

High Hartmann number flow through a rectangular duct with unequally and finitely conducting walls

Nonsymmetric Hartmann flow through a rectangular duct is investigated for thin duct walls with, generally, unequal but finite conductivities. A high Hartmann number is adopted. Consistent with known phenomena, both Hartmann layers transverse to the applied magnetic field are assumed to be separated from the two side boundary layers by four corner regions plus four inner corner regions. The method of singular perturbations and matched asymptotic expansions is applied to the coupled system. The equations governing the core and Hartmann layers are first partially resolved for leading terms. This is then followed by tackling equations governing one side layer and two adjacent corner regions. The latters' incorporation secures, for the former, only those boundary conditions that are compatible along the transverse walls. Both corner regions are denied access to non-required boundary conditions along the neighbouring side wall by the adjoining inner corner regions. However, the latters' boundary value problems need not be tackled for the acquirement of only dominant terms beyond all four inner corner regions. The complementary side layer and associated corners are accounted for by a non-symmetric reflection principle. Results reveal that a difference between conductivities in the transverse walls together with at least one finitely conducting side wall impart to disturbances within the core and Hartmann layers (i) a nontrivial dependence on the transverse coordinate relative to the magnetic field and flow in addition to the (usual) dependence on the field aligned coordinate, (ii) a dependence on side wall parameters in addition to the dependence on transverse wall parameters. Applications to related situations are considered. These include the case for a perfectly conducting lower wall, a finitely conducting upper wall, and equally and finitely conducting side walls.     

Stokes flow in closed,rectangular domains

Three practically relevant, Stokes flows in closed, rectangular cavities are considered. The first involves a solid-walled cavity where flow is driven by the motion of either one or both of its horizontal bounding walls; the other two have an upper free surface and are driven either by the motion of vertical side walls or by a horizontally-moving lower wall. Each problem is formulated as a biharmonic boundary value problem (bvp) for the streamfunction. The relative merits of two different coefficient determination methods for the corresponding analytical solutions are assessed and, in addition, each solution is compared with its numerical counterpart obtained using a finite element formulation of the governing equations. It is shown that, provided the number N of terms in each solution is sufficiently large, they are in extremely good agreement and, similarly, they compare well with work from other published theoretical and experimental studies. Streamlines are presented, over a wide range of operating parameters, for the geometries containing an upper free surface. For the flow generated by two moving vertical side walls two flow transformation mechanisms are identified. For cavities with small and decreasing width to depth (aspect) ratios, there is a sequence of critical aspect ratios at which flow bifurcations arise with a centre becoming a saddle point and vice versa, whereas for large aspect ratios increasing the ratio further leads to eddy growth from the lower wall, resulting in a regular sequence of separatrices along the cavity width. In the case of flow generated by a horizontally-moving lower wall the streamlines are simpler and exhibit the regular array of separatrices reported previously for flow in a solid-walled cavity with a single moving wall.     

Simultaneously developing flow and heat transfer of non-Newtonian fluids in equilateral triangular duct

A numerical investigation based on the Galerkin finite element method was carried out to solve the full three-dimensional governing equations for simultaneously developing steady laminar flow and heat transfer to a purely viscous non-Newtonian fluid described by a power law model flowing in equilateral triangular ducts. Two commonly used thermal boundary conditions, constant wall temperature (T boundary condition) and constant wall heat flux both axially and peripherally (H2 boundary condition) were examined. It is shown that the Nusselt number distribution along the walls is affected appreciably by the variation of the power law index. Results are presented and discussed for a wide range of power law indices and Prandtl numbers for T and H2 boundary conditions.     

Influence of variable heat flux on natural convection along a corrugated wall in porous media

In this work the coupled non-linear partial differential equations, governing the free convection from a wavy vertical wall under a power law heat flux condition, are solved numerically. For both Darcy and Forchheimer extended non-Darcy models, a wavy to flat surface transformation is applied and the governing equations are reduced to boundary layer equations. A finite difference scheme based on the Keller Box approach has been used in conjunction with a block tri-diagonal solver for obtaining the solution. Detailed simulations are carried out to investigate the effect of varying parameters such as power law heat flux exponent m, wavelength–amplitude ratio a and the transformed Grashof number Gr′. Both surface undulations and inertial forces increase the temperature of the vertical surface while increasing m reduces it. The wavy pattern observed in surface temperature plots, become more prominent with increasing m or a but reduces as Gr′ increases.     

A mapping finite difference model for infinite elastic media

A simple mapping finite difference model is presented for the solution of boundary-value problems in the theory of time-harmonic elastic vibrations. The finite problem domain is condensed by mapping into a smaller finite domain using a suitable coordinate transformation. The field equations and the boundary conditions are also appropriately transformed. The radiation condition at infinity is satisfied through a change of the dependent variable. Finite difference forms of the transformed equations are then solved in the mapped domain, subject to the transformed boundary conditions.     

基于二重网格的定常Navier-Stokes方程的局部和并行有限元算法

对二维定常的不可压缩的Navier-Stokes方程的局部和并行算法进行了研究．给出的算法是多重网格和区域分解相结合的算法,它是基于两个有限元空间:粗网格上的函数空间和子区域的细网格上的函数空间．局部算法是在粗网格上求一个非线性问题,然后在细网格上求一个线性问题,并舍掉内部边界附近的误差相对较大的解．最后,基于局部算法,通过有重叠的区域分解而构造了并行算法,并且做了算法的误差分析,得到了比标准有限元方法更好的误差估计,也对算法做了数值试验,数值结果通过比较验证了本算法的高效性和合理性．     

ON THE COUPLING OF BOUNDARY INTEGRAL AND FINITE ELEMENT METHODS FOR SIGNORINI PROBLEMS

1.IntroductionPartialdifferentialequationssubjecttounilateralboundaryconditionsareusuallycalledSignoriniproblemsintheliterature.TheseproblemshavebeenstudiedbymanyauthodssincetheappearenceofthehistoricalpaperbyA.Signoriniin1933[25].Signoriniproblemsaroseinmanyareasofapplicationse.g.,theelasticitywithunilateralconditions[lo],thefluidmechnicsproblemsinmediawithsemipermeableboundaries[8,12],theelectropaintprocess[1]etc.Fortheexistence,uniquenessandregularityresultsforSignorinitypeproblemswerefer…     

A full multigrid method for semilinear elliptic equation

A full multigrid finite element method is proposed for semilinear elliptic equations. The main idea is to transform the solution of the semilinear problem into a series of solutions of the corresponding linear boundary value problems on the sequence of finite element spaces and semilinear problems on a very low dimensional space. The linearized boundary value problems are solved by some multigrid iterations. Besides the multigrid iteration, all other efficient numerical methods can also serve as the linear solver for solving boundary value problems. The optimality of the computational work is also proved. Compared with the existing multigrid methods which need the bounded second order derivatives of the nonlinear term, the proposed method only needs the Lipschitz continuation in some sense of the nonlinear term.     

Technical note: A fourth‐order finite difference scheme for a system of a 2D nonlinear elliptic partial differential equations

In this article we present a fourth‐order finite difference scheme, for a system of two‐dimensional, second‐order, nonlinear elliptic partial differential equations with mixed spatial derivative terms, using 13‐point stencils with a uniform mesh size h on a square region R subject to Dirichlet boundary conditions. The scheme of order h⁴ is derived using the local solution of the system on a single stencil. The resulting system of algebraic equations can be solved by iterative methods. The difference scheme can be easily modified to obtain formulae for grid points near the boundary. Computational results are given to demonstrate the performance of the scheme on some problems including Navier‐Stokes equations. © 2001 John Wiley & Sons, Inc. Numer Methods Partial Differential Eq 17: 43–53, 2001     

ADFE METHOD WITH HIGH ACCURACY FOR NONLINEAR PARABOLIC INTEGRO-DIFFERENTIAL SYSTEM WITH NONLINEAR BOUNDARY CONDITIONS

Alternating direction finite element (ADFE) scheme for d-dimensional nonlin-ear system of parabolic integro-differential equations is studied. By using a local approxi-mation based on patches of finite elements to treat the capacity term qi(u), decomposition of the coefficient matrix is realized; by using alternating direction, the multi-dimensional problem is reduced to a family of single space variable problems, calculation work is sim-plified; by using finite element method, high accuracy for space variant is kept; by using inductive hypothesis reasoning, the difficulty coming from the nonlinearity of the coeffi-cients and boundary conditions is treated; by introducing Ritz-Volterra projection, the difficulty coming from the memory term is solved. Finally, by using various techniques for priori estimate for differential equations, the unique resolvability and convergence proper-ties for both FE and ADFE schemes are rigorously demonstrated, and optimal H1 and L2 norm space estimates and O((△t)2) estim     

Some solvability problems for the Boltzmann equation in the framework of the Shakhov model

We consider the nonlinear Boltzmann equation in the framework of the Shakhov model for the classical problem of gas flow in a plane layer. The problem reduces to a system of nonlinear integral equations. The nonlinearity of the studied system can be partially simplified by passing to a new argument depending on the solution of the problem itself. We prove the existence theorem for a unique solution of the linear system and the existence theorem for a positive solution of the nonlinear Urysohn equation. We determine the temperature jumps on the lower and upper walls in the linear and nonlinear cases, and it turns out that the difference between them is rather small.     

The coupling of boundary integral and finite element methods for nonmonotone nonlinear problems ∗

In this paper, we apply the coupling of the boundary integral and finite element methods to study the weak solvability of certain nonmonotone nonlinear exterior boundary value problems. In order to convert the original exterior problem into an equivalent nonlocal boundary value problem on a finite region, we employ two different approaches based on the use of one and two integral equations on the coupling boundary. Existence of a solution for the associated weak formulation, and convergence properties of the corresponding Galerkin approximations are deduced from fundamental results in nonlinear functional analysis. Indeed, the main arguments of our proofs are based on a surjectivity theorem for mappings of type (S) and on the Fredholm alternative for nonlinear A-proper mappings.     

A fast spectral/difference method without pole conditions for Poisson-type equations in cylindrical and spherical geometries

A simple and efficient FFT-based fast direct solver for Poisson-typeequations on 3D cylindrical and spherical geometries is presented.The solver relies on the truncated Fourier series expansion,where the differential equations of Fourier coefficients aresolved using second-order finite difference discretizationswithout pole conditions. Three different boundary conditions(Dirichlet, Neumann and Robin conditions) can be handled withoutsubstantial differences.     

复合材料加筋薄壁圆锥壳体有限变形的混合型理论

本文利用变分原理和平均筋条刚度法,建立了在任意载荷作用下纵向和环向密加筋复合材料圆锥壳体有限变形的Donnell型理论.考虑了面板最一般的弯曲拉伸耦合关系和加筋筋条的偏心效应的影响.导出了平衡条件、边界条件和变形协调方程.给出了以应力函数和挠度函数表示的耦合形式的非性性变系数偏微分方程组.对于一些特殊情况,给出了相应的简化方程.     

An asymptotic analysis of the steady process of saturation of a laminated porous material

The propagation of a steady saturation front in a double-layer porous material, situated between impenetrable walls, is investigated. The closed system of equations and boundary conditions are written on the assumption that the displacing and displaced phases, the viscosities of which differ considerably, are connected, and that the capillary pressure on the interface is constant. The features of the behaviour of the interface in the neighbourhood of the boundary of the layers are investigated in the case when the layer thicknesses differ considerably. When the permeabilities of the layers differ considerably, asymptotic expressions are obtained for the pressure and shape of the interface, and a comparison is made with the results of a numerical solution of the complete problem and with the known asymptotic relations obtained when using a simplified boundary condition at the interface.     

Numerical methods for coupled systems of quasilinear elliptic equations with nonlinear boundary conditions

This paper is concerned with numerical solutions of a coupled system of arbitrary number of quasilinear elliptic equations under combined Dirichlet and nonlinear boundary conditions. A finite difference system for a transformed system of the quasilinear equations is formulated, and three monotone iterative schemes for the computation of numerical solutions are given using the method of upper and lower solutions. It is shown that each of the three monotone iterations converges to a minimal solution or a maximal solution depending on whether the initial iteration is a lower solution or an upper solution. A comparison result among the three iterative schemes is given. Also shown is the convergence of the minimal and maximal discrete solutions to the corresponding minimal and maximal solutions of the continuous system as the mesh size tends to zero. These results are applied to a heat transfer problem with temperature dependent thermal conductivity and a Lotka-Volterra cooperation system with degenerate diffusion. This degenerate property leads to some interesting distinct property of the system when compared with the non-degenerate semilinear systems. Numerical results are given to the above problems, and in each problem an explicit continuous solution is constructed and is used to compare with the computed solution     

Semi‐analytical solution of MHD flow through boundary integrals on the pipe wall

A mathematical model is given for the magnetohydrodynamic (MHD) pipe flow as an inner Dirichlet problem in a 2D circular cross section of the pipe, coupled with an outer Dirichlet or Neumann magnetic problem. Inner Dirichlet problem is given as the coupled convection‐diffusion equations for the velocity and the induced current of the fluid coupling also to the outer problem, which is defined with the Laplace equation for the induced magnetic field of the exterior region with either Dirichlet or Neumann boundary condition. Unique solution of inner Dirichlet problem is obtained theoretically reducing it into two boundary integral equations defined on the boundary by using the corresponding fundamental solutions. Exterior solution is also given theoretically on the pipe wall with Poisson integral, and it is unique with Dirichlet boundary condition but exists with an additive constant obtained through coupled boundary and solvability conditions in Neumann wall condition. The collocation method is used to discretize these boundary integrals on the pipe wall. Thus, the proposed procedure is an improved theoretical analysis for combining the solution methods for the interior and exterior regions, which are consolidated numerically showing the flow behavior. The solution is simulated for several values of problem parameters, and the well‐known MHD characteristics are observed inside the pipe for increasing values of Hartmann number maintaining the continuity of induced currents on the pipe wall.     

Positive solutions and dynamics of a finite difference reaction–diffusion system

We investigate the dynamics and methods of computation for some nonlinear finite difference systems that are the discretized equations of a time-dependent and a steady-state reaction–diffusion problem. The formulation of the discrete equations for the time-dependent problem is based on the implicit method for parabolic equations, and the computational algorithm is based on the method of monotone iterations using upper and lower solutions as the initial iterations. The monotone iterative method yields improved upper and lower bounds of the solution in each iteration, and the sequence of iterations converges monotonically to a solution for both the time-dependent and the steady-state problems. An important consequence of this method is that it leads to a bifurcation point that determines the dynamic behavior of the time-dependent problem in relation to the corresponding steady-state problem. This bifurcation point also determines whether the steady-state problem has one or two non-negative solutions, and is explicitly given in terms of the physical parameters of the system and the type of boundary conditions. Numerical results are presented for both the time-dependent and the steady-state problems under various boundary conditions, including a test problem with known analytical solution. These numerical results exhibit the predicted dynamic behavior of the time-dependent solution given by the theoretical analysis. Also discussed are the numerical stability of the computational algorithm and the convergence of the finite difference solution to the corresponding continuous solution of the reaction–diffusion problem. © 1993 John Wiley & Sons, Inc.     

FINITE ELEMENT NONLINEAR GALERKIN COUPLING METHOD FOR THE EXTERIOR STEADY NAVIER-STOKES PROBLEM

1.IntroductionNonlinearGalerkinmethodsaremultilevelschemesforthedissipativeevolutionpartialdifferentialequations.Theycorrespondtothesplittingsoftheunknownu:u=y z)wherethecomponentsareofdifferentorderofmagnitudewithrespecttoaparameterrelatedtothespati...