非线性动力方程精细积分法的自适应步长研究

基于Adams显式和隐式预估公式实现对时间步长的 自适应选择,利用当前时刻v(tk),采用预估公式的两种形式(显式与隐式),对v(tk+1)进行两次预估,利用两公式局部截断误差关系,得出误差估计值ξ(tk+1),并根据其大小 自适应调节时间步长.将该思想应用于预估型(求解过程需要用到预估公式)精细积分算法中,使精细积分算法的时间步长依赖于给定的每步误差限值,提高计算精度,且使算法具有很好的稳定性,对刚度硬化和软化问题均有很好的效果.数值算例验证了本文思想的有效性与适用性.     

黏性不可压缩流体流动前沿的数值模拟

提出了模拟注射成型中黏性、不可压缩流体流动前沿的新方法. 将Hele-Shaw流动应用于非 等温条件下的黏性、不可压缩流体，建立了流动分析模型，用充填因子的输运方程描述流动 前沿. 应用高阶Taylor展开式计算每一时间步长的充填因子，用Galerkin方法导出了计算 充填因子各阶导数的递推公式. 给出了时间增量的选取方法，证明了它的稳定性. 针对Han 设计的试验模具，用相同的材料及工艺条件模拟充填过程，比较了传统方法和该方法的模 拟结果与实验结果的差异. 算例分析表明，该方法可以有效地提高注射成型中流动前沿的 模拟精度和计算效率.     

推进－迭代算法计算效率的研究

本文对求解三维定常超音速动性流场的一次空间推进，在每一个推进站沿伪时间层局部迭代的推进－迭代算法作了进一步的研究．在每一推进站（侧向平面）沿伪时间层局部迭代时，给出了四种不同的隐式迭代方法，即沿侧面两个方向（法向和周向）全用隐式；法向隐式而周向采用Ｇａｕｓｓ－Ｓｉｌｄｌｅ来回扫描迭代；法向隐式而周向显式及以系数矩阵谱半径代替系数矩阵的简化标量隐式算法．用这四种算法模拟了三维球锥黏性绕流，给出了四种不同算法的计算效率和收敛特性比较．     

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. 该方法是一种重要的求解流动问题的有限元方法, 可以使用等阶有限元. 该方法的准隐格式与其他格式相比, 具有时间步长大的特点. 将稳定项中的时间步长与全局时间步长分开, 改进了准隐格式的稳定性. 基于改进的特征分裂有限元方法的准隐格式, 发展了一种层流共轭传热数值模拟的整体耦合方法. 采用这种方法可以将流体计算域和固体计算域作为一个整体划分有限元网格, 并且所有变量都可以采用相同的插值函数, 从而有利于程序的实现. 通过对典型问题的模拟, 验证了这种方法的准确性. 本工作还研究了固体区域时间步长对定常共轭传热问题数值模拟收敛性的影响.      

RTM充模过程数值模拟的隐式有限元算法

建立了基于欧拉方法描述树脂传递模塑(RTM)工艺充模过程的基本数学方程,并采用有限元隐式时间积分方法对基本方程进行了数值求解．编制了基于隐式有限元算法及传统有限元控制体算法的程序,通过具体算例比较了这两种算法的优缺点．与传统的有限元控制体法相比,该文提出的隐式有限元算法能节省计算时间,特别适合于单元、节点数目多的情况．隐式有限元算法是一种纯有限元方法,不需要使用控制体积技术,采用该算法计算出的流动前沿与时间步长无关。     

基于非结构/混合网格模拟黏性流的高阶精度DDG/FV混合方法

间断Galerkin有限元方法 (discontinuous Galerkin method, DGM)因具有计算精度高、模板紧致、易于并行等优点,近年来已成为非结构/混合网格上广泛研究的高阶精度数值方法.但其计算量和内存需求量巨大,特别是对于网格规模达到百万甚至数千万的大型三维实际复杂外形问题,其计算量和存储量对计算资源的消耗是难以承受的.基于"混合重构"的DG/FV格式可以有效降低DGM的计算量和存储量.本文将DDG黏性项离散方法推广应用于DG/FV混合算法,得到新的DDG/FV混合格式,以进一步提高DG/FV混合算法对于黏性流动模拟的计算效率.通过Couette流动、层流平板边界层、定常圆柱绕流,非定常圆柱绕流和NACA0012翼型绕流等二维黏性流算例,优化了DDG通量公式中的参数选择,验证了DDG/FV混合格式对定常和非定常黏性流模拟的精度和计算效率,并与广泛使用的BR2-DG格式的计算结果和效率进行对比研究.一系列数值实验结果表明,本文构造的DDG/FV混合格式在二维非结构/混合网格的Navier-Stokes方程求解中,在达到相同的数值精度阶的前提下,相比BR2-DG格式,对于隐式时间离散的定常问题计算效率提高了2倍以上,对于显式时间离散的非定常问题计算效率提高1.6倍,并且在一些算例中,混合格式具有更优良的计算稳定性.DDG/FV混合格式提升了计算效率和稳定性,具有良好的应用前景.     

N-S方程基于投影法的特征线算子分裂有限元求解

提出了一种求解非定常不可压缩纳维-斯托克斯方程（N-S方程）的新型有限元法:基于投影法的特征线算子分裂有限元法.在每一个时间层上将N-S方程分裂成扩散项、对流项、压力修正项.对流项采用多步显式格式,且在每一个对流子时间步内采用更加精确的显式特征线-伽辽金法进行时间离散,空间离散采用标准伽辽金法.应用此算法对平面泊肃叶流、方腔流和圆柱绕流进行数值模拟,所得结果与基准解符合良好.尤其对于Re=10000的方腔流,给出了方腔中分离涡发展和运动的计算结果,并发现在该雷诺数下存在周期解,表明该算法能较好地模拟流体流动中的小尺度物理量以及流场中分离涡的运动.      

基于非结构/混合网格模拟黏性流的高阶精度DDG/FV混合方法

间断Galerkin有限元方法(discontinuous Galerkin method, DGM) 因具有计算精度高、模板紧致、易于并行等优点, 近年来已成为非结构/混合网格上广泛研究的高阶精度数值方法. 但其计算量和内存需求量巨大, 特别是对于网格规模达到百万甚至数千万的大型三维实际复杂外形问题, 其计算量和存储量对计算资源的消耗是难以承受的. 基于“混合重构”的DG/FV 格式可以有效降低DGM 的计算量和存储量. 本文将DDG 黏性项离散方法推广应用于DG/FV 混合算法, 得到新的DDG/FV混合格式, 以进一步提高DG/FV混合算法对于黏性流动模拟的计算效率. 通过Couette流动、层流平板边界层、定常圆柱绕流, 非定常圆柱绕流和NACA0012 翼型绕流等二维黏性流算例, 优化了DDG 通量公式中的参数选择, 验证了DDG/FV 混合格式对定常和非定常黏性流模拟的精度和计算效率, 并与广泛使用的BR2-DG 格式的计算结果和效率进行对比研究. 一系列数值实验结果表明, 本文构造的DDG/FV混合格式在二维非结构/混合网格的Navier-Stokes 方程求解中, 在达到相同的数值精度阶的前提下, 相比BR2-DG格式, 对于隐式时间离散的定常问题计算效率提高了2 倍以上, 对于显式时间离散的非定常问题计算效率提高1.6 倍, 并且在一些算例中, 混合格式具有更优良的计算稳定性. DDG/FV 混合格式提升了计算效率和稳定性, 具有良好的应用前景.      

????????????????????????????

合隐式和显式时间积分技术,对结构非线性动力反应分析提出一种并行混合时间积分算 法. 该算法采用区域分解技术. 将并发性引入到算法中,即利用显式时间积分技术进行界面 节点积分而利用隐式算法求解局部子区域. 为实现并行混合时间积分算法,设计了灵活的并 行数据信息流. 编写了该算法的程序,在工作站机群实现了数值算例,验证了算法的精度和 性能. 计算结果表明该算法具有良好的并行性能,优于隐式算法.     

结构非线性动力分析混合时间积分并行算法

综合隐式和显式时间积分技术,对结构非线性动力反应分析提出一种并行混合时间积分算法.该算法采用区域分解技术.将并发性引入到算法中,即利用显式时间积分技术进行界面节点积分而利用隐式算法求解局部子区域.为实现并行混合时间积分算法,设计了灵活的并行数据信息流.编写了该算法的程序,在工作站机群实现了数值算例,验证了算法的精度和性能.计算结果表明该算法具有良好的并行性能,优于隐式算法.     

An iterative stabilized fractional step algorithm for numerical solution of incompressible N–S equations

Stabilized fractional step algorithm has been widely employed for numerical solution of incompressible Navier–Stokes equations. However, smaller time step sizes are required to use for existing explicit and semi‐implicit versions of the algorithm due to their fully or partially explicit nature particularly for highly viscous flow problems. The purpose of this paper is to present two modified versions of the fractional step algorithm using characteristic based split and Taylor–Galerkin like based split. The proposed modified versions of the algorithm are based on introducing an iterative procedure into the algorithm and allow much larger time step sizes than those required to the preceding ones. A numerical study of stability at acceptable convergence rate and accuracy as well as capability in circumventing the restriction imposed by the LBB condition for the proposed iterative versions of the algorithm is carried out with the plane Poisseuille flow problem under different Reynolds numbers ranging from low to high viscosities. Numerical experiments in the plane Poisseuille flow and the lid‐driven cavity flow problems demonstrate the improved performance of the proposed versions of the algorithm, which are further applied to numerical simulation of the polymer injection moulding process. Copyright © 2005 John Wiley & Sons, Ltd.     

不可压缩N-S方程分步算法稳定性与精度的数值研究

众所周知,LBB条件排除了在不可压缩流动N-S方程空间离散中采用速度u和压力p同阶线性插值的简单单元。基于压力泊松(Poisson)方程的分步算法曾被认为可以绕开LBB条件限制,然而近年来研究表明,并非各种类型的分步算法都能有效地避开LBB条件。本文针对不同雷诺数下的平面Poiseuille流动问题模拟,分析对比了当采用不同类型的u-p单元空间插值时增量与非增量迭代分步算法的稳定性与精度,为合理选择分步算法和u-p插值类型提供了依据和参考。     

饱和土动力学有限元分析的改进稳定分步算法

基于Blot理论，控制饱和变形多孔介质中固相位移u和孔隙压力pw演变的场方程的空间半离散化导致u-pw型混合有限元方程。在固体颗粒和孔隙液体不可压缩以及零渗透性情况下，基本未知量u，pw的近似插值函数必须满足Babuska—Brezzi条件或者与之等价的Zienkiewicz和Taylor分片试验。采用相同低阶u-pw插值的有限元(如线性三角形单元和双线性四边形单元)不能满足B—B条件。分步算法作为一种稳定技术的引入可以绕开B—B条件，但现有分步算法在瞬态问题中仍存在虚假数值振荡和不稳定现象。本文在现有分步算法的基础上引入迭代过程，有效地缓解和克服了数值振荡现象，使低阶u-pw单元得以正常应用。应用双线性四节点u-pw单元的数值结果表明了所提出的包含迭代过程的改进分步算法的有效性。     

Steady‐state solution of incompressible Navier–Stokes equations using discrete least‐squares meshless method

A fractional step method for the solution of the steady state incompressible Navier–Stokes equations is proposed in this paper in conjunction with a meshless method, named discrete least‐squares meshless (DLSM). The proposed fractional step method is a first‐order accurate scheme, named semi‐incremental fractional step method, which is a general form of the previous first‐order fractional step methods, i.e. non‐incremental and incremental schemes. One of the most important advantages of the proposed scheme is its capability to use large time step sizes for the solution of incompressible Navier–Stokes equations. DLSM method uses moving least‐squares shape functions for function approximation and discrete least‐squares technique for discretization of the governing differential equations and their boundary conditions. As there is no need for a background mesh, the DLSM method can be called a truly meshless method and enjoys symmetric and positive‐definite properties. Several numerical examples are used to demonstrate the ability and the efficiency of the proposed scheme and the discrete least‐squares meshless method. The results are shown to compare favorably with those of the previously published works. Copyright © 2010 John Wiley & Sons, Ltd.     

A comparison of time integration methods in an unsteady low‐Reynolds‐number flow

This paper describes three different time integration methods for unsteady incompressible Navier–Stokes equations. Explicit Euler and fractional‐step Adams–Bashford methods are compared with an implicit three‐level method based on a steady‐state SIMPLE method. The implicit solver employs a dual time stepping and an iteration within the time step. The spatial discretization is based on a co‐located finite‐volume technique. The influence of the convergence limits and the time‐step size on the accuracy of the predictions are studied. The efficiency of the different solvers is compared in a vortex‐shedding flow over a cylinder in the Reynolds number range of 100–1600. A high‐Reynolds‐number flow over a biconvex airfoil profile is also computed. The computations are performed in two dimensions. At the low‐Reynolds‐number range the explicit methods appear to be faster by a factor from 5 to 10. In the high‐Reynolds‐number case, the explicit Adams–Bashford method and the implicit method appear to be approximately equally fast while yielding similar results. Copyright © 2002 John Wiley & Sons, Ltd.     

The characteristic‐based split (CBS) scheme for viscoelastic flow past a circular cylinder

A fully explicit, characteristic‐based split (CBS) method for viscoelastic flow past a circular cylinder, placed in a rectangular channel, is presented. The pressure equation in its explicit form is employed via an artificial compressibility parameter. The constitutive equations used here are based on the Oldroyd‐B model. No loss of convergence to steady state was observed in any of the results presented in this paper. Comparison of the present results with other available numerical data shows that the CBS algorithm is in excellent agreement with them at lower Deborah numbers. However, at higher Deborah numbers, the present results differ from other numerical solutions. This is due to the fact that the positive definitiveness of the conformation matrix is lost between a Deborah number of 0.6 and 0.7. However, the positive definitiveness is retained when an artificial diffusion is added to the discrete constitutive equations at higher Deborah numbers. It appears that the fractional solution stages used in the CBS scheme and the higher‐order time step‐based convection stabilization clearly reduce the instability at higher Deborah numbers. The Deborah number limit reached in the present work is three without artificial dissipation and two with artificial dissipation. Copyright © 2007 John Wiley & Sons, Ltd.     

Improved CVP scheme for laminar incompressible flows

An improved scheme of the continuity vorticity pressure (CVP) variational equations method is presented. The changes from the original version of the CVP method concern the velocity and the pressure correction equations that are used in the solution procedure and the topology of the grid where the method is applied. The improved CVP scheme is faster, simpler and more stable than the original version of the method. The efficiency and the accuracy of the new scheme are tested and validated through comparison of predictions and of computational time, with numerical results obtained with the SIMPLE method. Moreover, we present extensive comparisons of the results of the improved CVP scheme with numerical and experimental data from various researchers that show excellent agreement for a wide range of benchmark 2D and 3D laminar internal flow problems such as flow over a backward facing step, flow in square, circular and elliptical curved ducts and pulsating flow. Copyright © 2010 John Wiley & Sons, Ltd.     

不可压缩N-S方程的稳定分步算法及耦合离散

在有限增量微积分(finite increment calculus, FIC)的理论框架下，通过引入一个附加变量，发展了压力稳定型分步算法，有效改善了经典 分步算法的压力稳定性，同时还避免了标准FIC方法中存在的空间高阶导数的计算. 为保证 数值方法同时具有较快的计算速度和较好的健壮性，发展了有限元与无网格的耦合空间离散 方法. 该方案可在网格发生扭曲的区域采用无网格法空间离散以保证求解的精度和稳定性， 而在网格质量较好的区域以及本质边界上保留使用有限元法空间离散以提高计算效率和便于 施加本质边界条件. 方腔流考题的数值模拟结果突出地显示了所发展的压力稳定型分步算 法比经典分步算法具有更好的压力稳定性，能够有效消除速度-压力插值空间违反LBB条件而 导致的压力场的虚假数值振荡. 平面Poisseuille流动和一个典型型腔充填过程的数值模拟 结果, 表明了发展的耦合离散方案相对于单一的有限元法和单一的无网格法在综合考虑计 算效率和算法健壮性方面的突出优点.     

黏塑性本构计算的稳定性分析

提出本构方程计算方法的稳定性问题,针对黏塑性本构计算的显式精确算法的稳定性进行分析,发现该算法并非无条件稳定,使用小扰动方法给出了其计算稳定的必要条件,稳定性条件对数值计算中的时间步长提出限制要求。通过有限元算例验证了分析的正确性,计算结果也表明理论推导得到的稳定性公式能够准确预测满足计算稳定性条件要求的最大时间步长与各参数之间关系。     

Efficient implicit algorithm for the equations of 2-D viscous compressible flow: application to shock-boundary layer interaction

A fully implicit algorithm has been developed to time integrate the equations of 2-D compressible viscous flow. The algorithm was constructed so as to optimize computational efficiency. The time-consuming block matrix inversions usually associated with implicit algorithms have been reduced to the trivial non-iterative inversion of four sets of scalar bidiagonal matrices. Thus, the algorithm requires virtually no more computer storage than an explicit algorithm. The efficient structure of the implicit algorithm is reflected in comparative timings which slow that it requires only a factor of two more computer time per point per time step than a typical explicit algorithm. Therefore, the algorithm allows more economical solution of given flows than existing explicit methods and also allows more difficult problems to be attempted using available computer resources. Application of the algorithm to the problem of shock-boundary layer interaction produces results consistent with both experimental measurements and other calculations.