Stabilisation of AMG solvers for saddle‐point stokes problems

When a block factorisation is used to precondition the saddle‐point equations of the discrete Stokes problem, the stability that this gives for the relaxation of residual errors may not be conserved in the coarse‐grid approximations (CGA) of algebraic multi‐grid (AMG) solvers. If the same first‐order interpolation is used in the inter‐grid transfer operators for the scalar and the vector fields, the conditioning degrades with each coarsening step until eventually a critical coarsening is reached beyond which residual errors are no longer damped and will become divergent with any further coarsening. It is shown that by introducing the same block pre‐conditioner as an integral part of the coarsening algorithm, stable smoothing can be maintained at all levels of the CGA. The pre‐conditioning need only be applied at preselected grid levels, one immediately before the critical threshold and others beyond that level if required. Excessive complexity in the CGA is thereby avoided. The method is purely algebraic and may be used for both classical AMG solvers and for smoothed‐aggregation AMG solvers. It should be applicable to other coupled vector and scalar fields in science and engineering that involve second‐order (block‐diagonal) and first‐order (block‐off‐diagonal) discrete difference operators. Copyright © 2015 John Wiley & Sons, Ltd.     

Development of a parallel computational fluid dynamics algorithm on a hypercube computer

One of the main factors limiting the widespread use of computational fluid dynamics codes for engineering design is their very large requirements both in terms of computer memory and CPU time. Distributed memory parallel computers offer both the potential for a dramatic improvement in cost/performance over conventional supercomputers and the scalability to large numbers of processors that is required if performance beyond that of current supercomputers is to be achieved. As part of an evaluation to explore the potential of such machines for computational fluid mechanics applications, a concurrent algorithm for the solution of the Navier-Stokes equations has been developed and demonstrated on a hypercube parallel computer. The algorithm is based on a domain decomposition of a well-established serial pressure correction algorithm. The algorithm is demonstrated on both a 32-node scalar and eight-node vector Intel iPSC/2 for complicated two-dimensional laminar and turbulent flow problems with different grid sizes and numbers of processors. Speed-ups relative to a single processor of 12.9 with 16 processors and 20.2 with 32 processors are achieved on a scalar iPSC/2, demonstrating the parallel efficiency of the algorithm. Measured performance on a 32-node scalar iPSC/2 exceeds one-sixth that of a Cray X-MP running the original serial algorithm. The performance of the algorithm on an eight-node vector iPSC/2 exceeds that of the larger scalar hypercube and is about one-fifth that of the Cray X-MP. With cost/performance more than 10 times better than the Cray, these results dramatically show the cost effectiveness of vector hypercubes for this class of fluid mechanics algorithm.     

Algebraic multigrid and incompressible fluid flow

This paper is concerned with the development of algebraic multigrid (AMG) solution methods for the coupled vector–scalar fields of incompressible fluid flow. It addresses in particular the problems of unstable smoothing and of maintaining good vector–scalar coupling in the AMG coarse‐grid approximations. Two different approaches have been adopted. The first is a direct approach based on a second‐order discrete‐difference formulation in primitive variables. Here smoothing is stabilized using a minimum residual control harness and velocity–pressure coupling is maintained by employing a special interpolation during the construction of the inter‐grid transfer operators. The second is an indirect approach that avoids the coupling problem altogether by using a fourth‐order discrete‐difference formulation in a single scalar‐field variable, primitive variables being recovered in post‐processing steps. In both approaches the discrete‐difference equations are for the steady‐state limit (infinite time step) with a fully implicit treatment of advection based on central differencing using uniform and non‐uniform unstructured meshes. They are solved by Picard iteration, the AMG solvers being used repeatedly for each linear approximation. Both classical AMG (C‐AMG) and smoothed‐aggregation AMG (SA‐AMG) are used. In the direct approach, the SA‐AMG solver (with inter‐grid transfer operators based on mixed‐order interpolation) provides an almost mesh‐independent convergence. In the indirect approach for uniform meshes, the C‐AMG solver (based on a Jacobi‐relaxed interpolation) provides solutions with an optimum scaling of the convergence rates. For non‐uniform meshes this convergence becomes mesh dependent but the overall solution cost increases relatively slowly with increasing mesh bandwidth. Copyright © 2006 John Wiley & Sons, Ltd.     

A vector form of Compensation Theorem and its application to boundary-value problems

Summary The paper presents the vector form of the Compensation Theorem and illustrates its application to the formulation of a class of boundary value problems. The above theorem originally stated for electrical networks has previously been extended to electromagnetic problems by Monteath, but only in a scalar form. The subject of discussion of this paper is the generalization of his approach to the vector case.     

风场模拟中AR模型的若干问题

自回归(AR)模型具有计算量小,模拟速度快等优良特性,在风场模拟中得到了广泛应用.本文对AR模型进行了系统的研究,将脉动风场模拟中广泛应用的AR模型归为两大类,对模型中的参数从理论上进行了合理的解释.对两种模型模拟脉动风场时涉及到的Wiener-KJaintchine公武的交换形式,通过分析对其进行了修正,指出算法上可以采用FFT技术来计算互相关矩阵的元素以提高计算效率.提出了AR模型编程中偶然发现的自回归顺序问题,算例表明两种不同方法的风速时程样本及其无偏自相关估计和自功率谱估计均有较大的影响,希望能引起更多同行对该问题的注意.尽管标量过程AR模型简单且易于掌握,但不能考虑时滞问题.相比之下,理论分析和数值实验都证明,向量过程的AR模型在精度总体要高于标量过程的AR模型,但其运算时间也相应增多.     

Performance of algebraic multi‐grid solvers based on unsmoothed and smoothed aggregation schemes

A comparison is made of the performance of two algebraic multi‐grid (AMG0 and AMG1) solvers for the solution of discrete, coupled, elliptic field problems. In AMG0, the basis functions for each coarse grid/level approximation (CGA) are obtained directly by unsmoothed aggregation, an appropriate scaling being applied to each CGA to improve consistency. In AMG1 they are assembled using a smoothed aggregation with a constrained energy optimization method providing the smoothing. Although more costly, smoothed basis functions provide a better (more consistent) CGA. Thus, AMG1 might be viewed as a benchmark for the assessment of the simpler AMG0. Selected test problems for D'Arcy flow in pipe networks, Fick diffusion, plane strain elasticity and Navier–Stokes flow (in a Stokes approximation) are used in making the comparison. They are discretized on the basis of both structured and unstructured finite element meshes. The range of discrete equation sets covers both symmetric positive definite systems and systems that may be non‐symmetric and/or indefinite. Both global and local mesh refinements to at least one order of resolving power are examined. Some of these include anisotropic refinements involving elements of large aspect ratio; in some hydrodynamics cases, the anisotropy is extreme, with aspect ratios exceeding two orders. As expected, AMG1 delivers typical multi‐grid convergence rates, which for all practical purposes are independent of mesh bandwidth. AMG0 rates are slower. They may also be more discernibly mesh‐dependent. However, for the range of mesh bandwidths examined, the overall cost effectiveness of the two solvers is remarkably similar when a full convergence to machine accuracy is demanded. Thus, the shorter solution times for AMG1 do not necessarily compensate for the extra time required for its costly grid generation. This depends on the severity of the problem and the demanded level of convergence. For problems requiring few iterations, where grid generation costs represent a significant penalty, AMG0 has the advantage. For problems requiring a large investment in iterations, AMG1 has the edge. However, for the toughest problems addressed (vector and coupled vector–scalar fields discretized exclusively using finite elements of extreme aspect ratio) AMG1 is more robust: AMG0 has failed on some of these tests. However, but for this deficiency AMG0 would be the preferred linear approximation solver for Navier–Stokes solution algorithms in view of its much lower grid generation costs. Copyright © 2001 John Wiley & Sons, Ltd.     

An arbitrary Lagrangian–Eulerian method based on the adaptive Riemann solvers for general equations of state

Approximate or exact Riemann solvers play a key role in Godunov‐type methods. In this paper, three approximate Riemann solvers, the MFCAV, DKWZ and weak wave approximation method schemes, are investigated through numerical experiments, and their numerical features, such as the resolution for shock and contact waves, are analyzed and compared. Based on the analysis, two new adaptive Riemann solvers for general equations of state are proposed, which can resolve both shock and contact waves well. As a result, an ALE method based on the adaptive Riemann solvers is formulated. A number of numerical experiments show good performance of the adaptive solvers in resolving both shock waves and contact discontinuities. Copyright © 2008 John Wiley & Sons, Ltd.     

Three-dimensional eddy current calculation in multiply connected regions by using the magnetic vector potential

There are two methods of using the magnetic vector potential for three-dimensional eddy current calculations. The first method uses the continuous magnetic vector potential which accompanies a scalar potential and has the advantage that no cutting is necessary for the multiply connected region problem. The second method uses the discontinuous magnetic vector potential which accompanies no scalar potential and has the disadvantage that cutting is necessary for the multiply connected region problem. In this paper a formulation using the continuous magnetic vector potential and accompanying scalar potential is given, together with computed results for three three-dimensional multiply connected eddy current problems.     

Influence of molecular diffusion on alignment of vector fields: Eulerian analysis

The effect of diffusive processes on the structure of passive vector and scalar gradient fields is investigated by analyzing the corresponding terms in the orientation and norm equations. Numerical simulation is used to solve the transport equations for both vectors in a two-dimensional, parameterized model flow. The study highlights the role of molecular diffusion in the vector orientation process and shows its subsequent action on the geometric features of vector fields.     

Symmetric formulation of tangential stiffnesses for non-associated viscoplasticity with an implicit time integration scheme

A numerical scheme is presented which enables the use of symmetric equationsolvers in tangential stiffness programs for non-associated viscoplastic materials.     

Symmetric formulation of tangential stiffnesses for non-associated plasticity

A numerical scheme is presented which makes it possible to use the symmetric equation solvers in tangential stiffness programs for non-associated materials.     

Exact fully 3D Navier–Stokes solutions for benchmarking

Unsteady analytical solutions to the incompressible Navier–Stokes equations are presented. They are fully three-dimensional vector solutions involving all three Cartesian velocity components, each of which depends non-trivially on all three co-ordinate directions. Although unlikely to be physically realized, they are well suited for benchmarking, testing and validation of three-dimensional incompressible Navier–Stokes solvers. The use of such a solution for benchmarking purposes is described.     

A novel non-primitive Boundary Integral Equation Method for three-dimensional and axisymmetric Stokes flows

A new Boundary Integral Equation (BIE) formulation for Stokes flow is presented for three-dimensional and axisymmetrical problems using non-primitive variables, assuming velocity field is prescribed on the boundary. The formulation involves the vector potential, instead of the classical stream function, and all three components of the vorticity are implied. Furthermore, following the Helmholtz decomposition, a scalar potential is added to represent the solenoidal velocity field. Firstly, the BIEs for three-dimensional flows are formulated for the vector potential and the vorticity by employing the fundamental solutions in free space of vector Laplace and biharmonic equations. The equations for axisymmetric flows are then derived from the three-dimensional formulation in a second step. The outcome is a domain integral free BIE formulation for both three-dimensional and axisymmetric Stokes flows with prescribed velocity boundary condition. Numerical results are included to validate and show the efficiency of the proposed axisymmetric formulation.     

Performance of finite volume solutions to the shallow water equations with shock‐capturing schemes

Numerical methods have become well established as tools for solving problems in hydraulic engineering. In recent years the finite volume method (FVM) with shock capturing capabilities has come to the fore because of its suitability for modelling a variety of types of flow; subcritical and supercritical; steady and unsteady; continuous and discontinuous and its ability to handle complex topography easily. This paper is an assessment and comparison of the performance of finite volume solutions to the shallow water equations with the Riemann solvers; the Osher, HLL, HLLC, flux difference splitting (Roe) and flux vector splitting. In this paper implementation of the FVM including the Riemann solvers, slope limiters and methods used for achieving second order accuracy are described explicitly step by step. The performance of the numerical methods has been investigated by applying them to a number of examples from the literature, providing both comparison of the schemes with each other and with published results. The assessment of each method is based on five criteria; ease of implementation, accuracy, applicability, numerical stability and simulation time. Finally, results, discussion, conclusions and recommendations for further work are presented. Copyright © 2002 John Wiley & Sons, Ltd.     

Integrals of motion for the classical two-body problem with drag

Integrals of motion for the two-body problem with drag are obtained by operating on the second-order vector differential equation describing the motion. The force field consists of an inverse-square gravitational attraction and a drag force proportional to the velocity vector and inversely proportional to the square of the distance to the attracting center. The developed integrals are the analogs of the Keplerian scalar energy, the vector angular momentum, and the Laplace vector.     

Molecular transfer in turbulent flows

For modeling the molecular transfer of a passive scalar in a known turbulent field, the equations for the average scalar value and the correlation function for the scalar field are written in a form which makes it possible to examine the effect of molecular transfer on turbulent transfer and scalar dissipation. For the closure of the equation for the correlation function, the Prandtl hypothesis is used. The statistical reliability of this closure is demonstrated. The system proposed makes it possible to predict the dynamics of a decaying uniform scalar field and to explain why the effect of the real value of the molecular-transfer coefficient on the decaying scalar field is weak. Specific features of the transport process in a plane layer with prescribed scalar values on the layer boundaries are considered.     

DNS and Modelling of Passive Scalar Transport in Turbulent Channel Flow with a Focus on Scalar Dissipation Rate Modelling

A direct numerical simulation of turbulent channel flow with an imposed mean scalar gradient is analyzed with a focus on passive scalar flux modelling and in particular the treatment of the passive scalar dissipation equation. The Prandtl number is 0.71 and the Reynolds number based on the wall friction velocity and the channel half width is 265. Budgets are presented for the passive scalar variance and its dissipation rate, as well as for the individual scalar flux components. These form a basis for a discussion of modelling issues related to explicit algebraic scalar flux modelling. This revised version was published online in July 2006 with corrections to the Cover Date.     

Wideband fast multipole boundary element method: Application to acoustic scattering from aerodynamic bodies

A wideband adaptive multi‐level fast multipole method (MLFMM) is used to accelerate the matrix–vector products arising from a boundary element method (BEM) formulation which solves the Burton–Miller boundary integral equation (BIE). The wideband MLFMM presented here applies a plane wave expansion formulation with fast interpolation and filtering for calculations in the high‐frequency regime and a partial wave expansion formulation with rotation‐coaxial translation in the low‐frequency regime. The iterative solvers GMRES, Bi‐CGSTAB and CGS are tested and compared and a block diagonal preconditioner is used to improve the condition number of the BEM matrices and to accelerate the convergence of the iterative solvers. Details on the implementation of the formulations are described, including the treatment of singular integrals. Results for acoustic scattering from a wing plus engine nacelle configuration for a prescribed source in a subsonic uniform flow are presented for a broad range of frequencies in order to assess the implemented capability. Copyright © 2010 John Wiley & Sons, Ltd.     

The Geometry of Cauchy's Fluxes

A formulation of the Cauchy theory for balance laws of scalar valued quantities is considered from a general geometric point of view. It is assumed only that the ambient space is an orientable m-dimensional manifold. The analog of the usual flux vector field is an (m−1)-differential form. Both the Cauchy theorem and differential version of the balance laws are formulated in this context. Accepted December 9, 1999?Published online July 12, 2000