Implementation and performance of the time integration of a 3D numerical transport model

The total solution of a three-dimensional model for computing the transport of salinity, pollutants, suspended material (such as sediment or mud), etc. in shallow seas involves many aspects, each of which has to be treated in an optimal way in order to cope with the tremendous computational task involved. In this paper we focus on one of these aspects, i.e. on the time integration, and discuss two numerical solution methods. The emphasis in this paper is on the performance of the methods when implemented on a vector/parallel, shared memory computer such as a Cray-type machine. The first method is an explicit time integrator and can straightforwardly be vectorized and parallelized. Although a stabilizing technique has been applied to this method, it still suffers from a severe time step restriction. The second method is partly implicit, resulting in much beter stability characteristics; however, as a consequence of the implicitness, it requires in each step the solution of a large number of tridiagonal systems. When implemented in a standard way, the recursive nature would prevent vectorization, resulting in a very long solution time. Following a suggestion of Golub and Van Loan, this part of the algorithm has been tuned for use on the Cray C98/4256. On the basis of a large-scale test problem, performance results will be presented for various implementations.     

An analysis and comparison of the time accuracy of fractional‐step methods for the Navier–Stokes equations on staggered grids

Fractional‐step methods solve the unsteady Navier–Stokes equations in a segregated manner, and can be implemented with only a single solution of the momentum/pressure equations being obtained at each time step, or with the momentum/pressure system being iterated until a convergence criterion is attained.The time accuracy of such methods can be determined by the accuracy of the momentum/pressure coupling, irrespective of the accuracy to which the momentum equations are solved. It is shown that the time accuracy of the basic projection method is first‐order as a result of the momentum/pressure coupling, but that by modifying the coupling directly, or by modifying the intermediate velocity boundary conditions, it is possible to recover second‐order behaviour. It is also shown that pressure correction methods, implemented in non‐iterative or iterative form and without special boundary conditions, are second‐order in time, and that a form of the non‐iterative pressure correction method is the most efficient for the problems considered. Copyright © 2002 John Wiley & Sons, Ltd.     

基于Wilson-θ和Newmark-β法的非线性动力学方程改进算法

Wilson-θ法和Newmark-β法是非线性动力学方程求解的常用方法。它们的一个基本步骤是,将方程改写为增量平衡的形式,在每一个积分步长内用状态参量修正平衡方程的系数矩阵,其本质是在单个步长内对系统的非线性环节进行了线性化处理。本文基于增量思想分别改进了Wilson-θ法和Newmark-β法,根据即时解给出下一步的猜测解,然后对猜测解进行迭代校正,最终得到收敛的近似解。算例表明,改进算法的精度更高,且收敛准则简单。更为重要的是,本文方法无须对非线性项进行线性化处理,因而计算效率更高,适应范围更广。     

Semi-implicit finite difference methods for three-dimensional shallow water flow

A semi-implicit finite difference method for the numerical solution of three-dimensional shallow water flows is presented and discussed. The governing equations are the primitive three-dimensional turbulent mean flow equations where the pressure distribution in the vertical has been assumed to be hydrostatic. In the method of solution a minimal degree of implicitness has been adopted in such a fashion that the resulting algorithm is stable and gives a maximal computational efficiency at a minimal computational cost. At each time step the numerical method requires the solution of one large linear system which can be formally decomposed into a set of small three-diagonal systems coupled with one five-diagonal system. All these linear systems are symmetric and positive definite. Thus the existence and uniquencess of the numerical solution are assured. When only one vertical layer is specified, this method reduces as a special case to a semi-implicit scheme for solving the corresponding two-dimensional shallow water equations. The resulting two- and three-dimensional algorithm has been shown to be fast, accurate and mass-conservative and can also be applied to simulate flooding and drying of tidal mud-flats in conjunction with three-dimensional flows. Furthermore, the resulting algorithm is fully vectorizable for an efficient implementation on modern vector computers.     

A Hybrid Domain Decomposition and Multigrid Method for the Acceleration of Compressible Viscous Flow Calculations on Unstructured Triangular Meshes

This paper is concerned with the formulation and the evaluation of a hybrid solution method that makes use of domain decomposition and multigrid principles for the calculation of two-dimensional compressible viscous flows on unstructured triangular meshes. More precisely, a non-overlapping additive domain decomposition method is used to coordinate concurrent subdomain solutions with a multigrid method. This hybrid method is developed in the context of a flow solver for the Navier-Stokes equations which is based on a combined finite element/finite volume formulation on unstructured triangular meshes. Time integration of the resulting semi-discrete equations is performed using a linearized backward Euler implicit scheme. As a result, each pseudo time step requires the solution of a sparse linear system. In this study, a non-overlapping domain decomposition algorithm is used for advancing the solution at each implicit time step. Algebraically, the Schwarz algorithm is equivalent to a Jacobi iteration on a linear system whose matrix has a block structure. A substructuring technique can be applied to this matrix in order to obtain a fully implicit scheme in terms of interface unknowns. In the present approach, the interface unknowns are numerical fluxes. The interface system is solved by means of a full GMRES method. Here, the local system solves that are induced by matrix-vector products with the interface operator, are performed using a multigrid by volume agglomeration method. The resulting hybrid domain decomposition and multigrid solver is applied to the computation of several steady flows around a geometry of NACA0012 airfoil.     

Lattice-Boltzmann method for yield-stress liquids

In the present study we propose a new version of the lattice-Boltzmann (LB) method for the simulation of flow of yield-stress liquids. Unlike traditional LB methods, collisions are treated implicitly, i.e., the collision term is chosen in such a way that the stress and strain rate tensors satisfy the constitutive equation after the collision. This approach requires the solution of a (one-dimensional) non-linear algebraic equation at each point and at each time step. In the practically important cases of a Bingham liquid this equation can be solved analytically. We calculated the flow of Bingham fluid through a channel and periodic mesh of cylinders.     

刚性源项处理方法在爆震模拟中的应用对比

为解决爆震燃烧模拟中出现的刚性问题,对处理刚性源项问题常见的一步法、逼近法、拟稳态逼近法(α qusai steady state, αQSS)和点隐方法进行对比,从稳定性等方面分析源项处理方法应满足的时间步长要求,并探索各方法之间的联系以及适应化学反应特征变化的能力,进一步通过球头激波诱导燃烧算例比较每种方法的计算效率。理论分析和数值计算表明:一步法在积分刚性源项时,积分步长需小于或等于2倍最短反应特征时间,而逼近法、αQSS法和点隐方法对时间步长取值没有影响;αQSS法可根据化学反应特征的变化自动调整α值和时间步长,适用范围较广,而一步法和逼近法则是αQSS方法的特例。点隐等隐式方法在求解数学意义上的刚性问题时稳定性很好,但计算效率较低。相比而言,αQSS法在计算稳定性和适应化学反应变化方面都具有良好的性能,且针对激波诱导燃烧算例,αQSS法消耗的CPU时间仅为点隐方法的一半,是处理刚性源项较好的选择。     

Modification of an implicit approach based on nonstandard rules for structural dynamics analysis

The present study aims to modify a recently suggested implicit approach consisted of the approximate Euler method and closed-form exponential mapping (herein referred to as the Liu scheme) for the dynamic analysis of structures. Such modification has been developed based upon nonstandard rules. The equation of motion is formulated in the augmented dynamic space to apply the exponential mapping as a group preserving scheme. The formulation of the proposed method involves the hyperbolic sine and cosine functions. The method is therefore prone to divergence due to the behavior of the hyperbolic functions in structures with a high ratio of stiffness to mass. In the present study, to consider the properties of the structural equation into the formulation of the time step size and thereby avoid the divergence, a parameter, known as stability parameter, is thus derived from the exact solution of the equation of motion based on nonstandard rules. Embedding this parameter into the proposed method improves its stability. Afterward, for evaluating the performance of the proposed method, it is applied to several structures with different loading patterns while implemented in programing environment of the Matlab software. The results are compared to those of several commonly used numerical methods in structural applications. It is found that the proposed method has acceptable convergence and accuracy, and low time consumption compared to several commonly used methods. Furthermore, its stability is guaranteed by embedding the stability parameter into the proposed method.     

Implicit time accurate simulation of unsteady flow

Implicit time integration was studied in the context of unsteady shock‐boundary layer interaction flow. With an explicit second‐order Runge–Kutta scheme, a reference solution to compare with the implicit second‐order Crank–Nicolson scheme was determined. The time step in the explicit scheme is restricted by both temporal accuracy as well as stability requirements, whereas in the A‐stable implicit scheme, the time step has to obey temporal resolution requirements and numerical convergence conditions. The non‐linear discrete equations for each time step are solved iteratively by adding a pseudo‐time derivative. The quasi‐Newton approach is adopted and the linear systems that arise are approximately solved with a symmetric block Gauss–Seidel solver. As a guiding principle for properly setting numerical time integration parameters that yield an efficient time accurate capturing of the solution, the global error caused by the temporal integration is compared with the error resulting from the spatial discretization. Focus is on the sensitivity of properties of the solution in relation to the time step. Numerical simulations show that the time step needed for acceptable accuracy can be considerably larger than the explicit stability time step; typical ratios range from 20 to 80. At large time steps, convergence problems that are closely related to a highly complex structure of the basins of attraction of the iterative method may occur. Copyright © 2001 John Wiley & Sons, Ltd.     

A high accurate time-domain-advance integration method for transient elastodynamic problems

All step-by-step integration methods available at present for structural dynamic analysis use the displacement, velocity, and acceleration vectors computed at a previous interval for evaluating those at an advanced time step. Hence, an accumulated error will be definitely introduced after such integration. This paper presents a novel time-domain-advance integration method for transient elastodynamic problems in which the exact initial conditions are strictly satisfied for the solutions for each time step. In this way, the accumulated error can be eliminated and the approximate solutions will converge to the exact ones uniformly on the whole time domain. Therefore, the new method is more accurate. When applying to a structural dynamic problem, the present mehtod does not have to use the initial acceleration as is required by most other algorithms and the corresponding computation can be avoided. The present method is simple in representation, easy to be programmed, and especially suitable for accurate analyses of long-time problems. The comparison of numerical results with exact ones shows that the present method is much more accurate than some most widely used algorithms.     

基于特征分裂有限元准隐格式的共轭传热整体耦合数值模拟方法

共轭传热现象在科学和工程领域中大量存在.随着计算能力的发展,对共轭传热现象进行准确有效的数值模拟,成为科学研究和工程设计上的重要挑战.共轭传热数值模拟的方法可以分为两大类:分区耦合和整体耦合.本文采用有限元法对共轭传热问题进行整体耦合模拟.固体传热求解采用标准的伽辽金有限元方法.流动求解采用基于特征分裂的有限元方法.该...     

Segregated finite element algorithms for the numerical solution of large-scale incompressible flow problems

This paper presents results of an ongoing research program directed towards developing fast and efficient finite element solution algorithms for the simulation of large-scale flow problems. Two main steps were taken towards achieving this goal. The first step was to employ segregated solution schemes as opposed to the fully coupled solution approach traditionally used in many finite element solution algorithms. The second step was to replace the direct Gaussian elimination linear equation solvers used in the first step with iterative solvers of the conjugate gradient and conjugate residual type. The three segregated solution algorithms developed in step one are first presented and their integrity and relative performance demonstrated by way of a few examples. Next, the four types of iterative solvers (i.e. two options for solving the symmetric pressure type equations and two options for solving the non-symmetric advection–diffusion type equations resulting from the segregated algorithms) together with the two preconditioning strategies employed in our study are presented. Finally, using examples of practical relevance the paper documents the large gains which result in computational efficiency, over fully coupled solution algorithms, as each of the above two main steps are introduced. It is shown that these gains become increasingly more dramatic as the complexity and size of the problem is increased.     

概率功能度量求解的共轭梯度步长调节法

功能度量法(PMA)由于其稳定高效的特点,适用于概率结构优化设计中概率约束的评定。PMA中改进均值法常用于求解概率功能度量,针对其求解高度非线性功能函数时出现周期振荡和混沌等不收敛现象,提出了一种新的共轭梯度步长调节法(CGS)。该方法基于RMIL共轭搜索方向和自适应步长调节策略提出,新的共轭搜索方向在保证收敛性的前提下加速了迭代进程,而自适应步长调节策略无需了解功能函数凹凸性及非线性程度等先验信息,无需确定步长的合适取值。通过限定步长准则自动选取初始步长,并随迭代过程不断调节,直至最终收敛。多个算例表明,与其他求解方法相比,本文的共轭梯度步长调节法更加高效且稳健。     

The convergent condition and united formula of step reduction method

The step reduction method was first suggested by Prof. Yeh Kai-yuan^[1]. This method has more advantages than other numerical methods. By this method, the analytic expression of solution can he obtained for solving nonuniform elastic mechanics. At the same time, its calculating time is very short and convergent speed very fast. In this paper. the convergent condition and united formula of step reduction method are given by mathematical method. It is proved that the solution of displacement and stress resultants obtained by this method can converge to exact solution uniformly. when the convergent condition is satisfied. By united formula, the analytic solution can be expressed as matrix form, and therefore the former complicated expression can be avoided. Two numerical examples are given at the end of this paper which indicate that by the theory in this paper, a right model can be obtained for step reduction method.Project Supported by Science and Technic Fund of the National Education Committee.The author would like to thank Prof. Yeh Kai-yuan for his directing.     

A matrix‐free preconditioned Newton/GMRES method for unsteady Navier–Stokes solutions

The unsteady compressible Reynolds‐averaged Navier–Stokes equations are discretized using the Osher approximate Riemann solver with fully implicit time stepping. The resulting non‐linear system at each time step is solved iteratively using a Newton/GMRES method. In the solution process, the Jacobian matrix–vector products are replaced by directional derivatives so that the evaluation and storage of the Jacobian matrix is removed from the procedure. An effective matrix‐free preconditioner is proposed to fully avoid matrix storage. Convergence rates, computational costs and computer memory requirements of the present method are compared with those of a matrix Newton/GMRES method, a four stage Runge–Kutta explicit method, and an approximate factorization sub‐iteration method. Effects of convergence tolerances for the GMRES linear solver on the convergence and the efficiency of the Newton iteration for the non‐linear system at each time step are analysed for both matrix‐free and matrix methods. Differences in the performance of the matrix‐free method for laminar and turbulent flows are highlighted and analysed. Unsteady turbulent Navier–Stokes solutions of pitching and combined translation–pitching aerofoil oscillations are presented for unsteady shock‐induced separation problems associated with the rotor blade flows of forward flying helicopters. Copyright © 2000 John Wiley & Sons, Ltd.     

Coupling the finite element method and molecular dynamics in the framework of the heterogeneous multiscale method for quasi-static isothermal problems

Multiscale models are designed to handle problems with different length scales and time scales in a suitable and efficient manner. Such problems include inelastic deformation or failure of materials. In particular, hierarchical multiscale methods are computationally powerful as no direct coupling between the scales is given. This paper proposes a hierarchical two-scale setting appropriate for isothermal quasi-static problems: a macroscale treated by continuum mechanics and the finite element method and a microscale modelled by a canonical ensemble of statistical mechanics solved with molecular dynamics. This model will be implemented into the framework of the heterogeneous multiscale method. The focus is laid on an efficient coupling of the macro- and micro-solvers. An iterative solution algorithm presents the macroscopic solver, which invokes for each iteration an atomistic computation. As the microscopic computation is considered to be very time consuming, two optimisation strategies are proposed. Firstly, the macroscopic solver is chosen to reduce the number of required iterations to a minimum. Secondly, the number of time steps used for the time average on the microscale will be increased with each iteration. As a result, the molecular dynamics cell will be allowed to reach its state of thermodynamic equilibrium only in the last macroscopic iteration step. In the preceding iteration steps, the molecular dynamics cell will reach a state close to equilibrium by using considerably fewer microscopic time steps. This adapted number of microsteps will result in an accelerated algorithm (aFE-MD-HMM) obtaining the same accuracy of results at significantly reduced computational cost. Numerical examples demonstrate the performance of the proposed scheme.     

A semi-implicit finite volume scheme for a simplified hydrostatic model for fluid-structure interaction

Simulating fluid-structure interaction problems usually requires a considerable computational effort. In this article, a novel semi-implicit finite volume scheme is developed for the coupled solution of free surface shallow water flow and the movement of one or more floating rigid structures. The model is well-suited for geophysical flows, as it is based on the hydrostatic pressure assumption and the shallow water equations. The coupling is achieved via a nonlinear volume function in the mass conservation equation that depends on the coordinates of the floating structures. Furthermore, the nonlinear volume function allows for the simultaneous existence of wet, dry and pressurized cells in the computational domain. The resulting mildly nonlinear pressure system is solved using a nested Newton method. The accuracy of the volume computation is improved by using a subgrid, and time accuracy is increased via the application of the theta method. Additionally, mass is always conserved to machine precision. At each time step, the volume function is updated in each cell according to the position of the floating objects, whose dynamics is computed by solving a set of ordinary differential equations for their six degrees of freedom. The simulated moving objects may for example represent ships, and the forces considered here are simply gravity and the hydrostatic pressure on the hull. For a set of test cases, the model has been applied and compared with available exact solutions to verify the correctness and accuracy of the proposed algorithm. The model is able to treat fluid-structure interaction in the context of hydrostatic geophysical free surface flows in an efficient and flexible way, and the employed nested Newton method rapidly converges to a solution. The proposed algorithm may be useful for hydraulic engineering, such as for the simulation of ships moving in inland waterways and coastal regions.     

基于预估-校正的Generalized-α法及其在非线性结构动力学中的应用

直接积分法是求解动力学方程的一种有效方法。应用一种预估-校正的Generalized-α法对结构大变形动力学问题进行分析求解,并与Newmark法和Bathe法进行对比研究。首先预估当前计算步的解,然后以预估值作为起始值进行非线性迭代计算,并对解不断校正,直到满足收敛条件,进入下一时间步的计算。在保证Generalized-α法性能的基础上,简化了非线性迭代公式,便于编程实现。通过壳和实体的大变形动力学算例,证明了本文方法具有较高的稳定性和精度。     

Size reduction methods for the implicit time-dependent simulation of micro–macro viscoelastic flow problems

Traditionally, micro–macro simulations have been performed using simple explicit time-marching algorithms, which lack the desirable stability of implicit methods. In this study, a fully implicit time integration scheme is presented and implemented for the first time for the solution of time-dependent complex flows using the Brownian Configuration Fields approach. Special techniques need to be applied to deal with the very large size of the resulting linear systems. A novel size-reduction scheme is used, allowing an independent treatment for each molecular field and suited to parallel hardware architecture. To illustrate the method, a selected number of applications using linear spring chains are presented and the results are compared with their corresponding closed form constitutive equation. The excellent agreement between the results demonstrates the feasibility of the proposed approach.     

一种结构动力时程分析的积分求微方法

传统采用微分求积(differential quadrature,DQ)法求解动力问题时都是以位移响应作为基本未知量,而将速度响应和加速度响应表示为位移响应的加权和的形式.如此做法需要处理线性方程组或者矩阵方程(Sylvester方程)才能求得动力响应,导出的算法一般为有条件稳定算法.本文利用动力响应的Duhamel积分解,逆用DQ原理,提出了一种计算卷积的高精度显式算法.该算法可以逐时段地求解出动力时程响应,当各时段内DQ节点分布完全一致时,仅须进行一次Vandermonde矩阵求逆计算即可应用于各个时段,一次性获得时段内多个时刻的位移响应值,因而具有计算效率高的优点.通过分析动力方程积分格式,证明本文动力算法传递矩阵的谱半径恒等于1,因而该算法具有无条件稳定特性,且计算过程中不会产生数值耗散. 本文算法的数值精度取决于分析时段内布置的DQ节点数量$N$,具有$N-1$阶代数精度.实际操作时可以取10个甚至更多的DQ节点数,从而获得比较高的数值精度.