二维稳定流形的自适应推进算法

稳定和不稳定流形是研究动力系统全局特性的重要工具.一般系统的稳定和不稳定流形的曲率在全局范围内会有明显变化,应根据流形曲率的变化采用不同尺寸的网格单元计算全局流形.然而在现有二维流形算法中,流形网格单元的尺寸在全局范围内是统一的.为持续有效地计算全局稳定流形,提高计算网格对流形曲率变化的适应性.本文在偏微分方程算法的基础上提出一种二维稳定流形的自适应推进算法.该算法的基本思想是根据稳定流形曲率的变化自适应地调整网格单元的尺寸.该算法首先在系统的稳定特征子空间中确定稳定流形的一个初始估计,该初始估计的网格单元尺寸设置为初始大小.然后根据稳定流形网格前沿的曲率特点自适应地产生新的备选网格单元,继而根据相切性条件更新备选点的坐标,并将距离平衡点最近的备选点接受为已知点,最后更新稳定流形网格的前沿并自适应地产生新的备选网格单元,通过这个迭代过程使流形网格自适应地向前推进.本文算法通过引入流形单元尺寸自适应,成功实现了洛伦兹流形和类球面流形的计算,并与偏微分方程算法进行了对比,结果表明自适应推进算法的流形计算单元的尺寸可在全局范围内根据流形曲率自适应地调整.利用自适应推进算法计算二维稳定流形,可实现稳定流形的自适应推进.     

二维自适应网格生成的改进AFT与背景网格法

在基于重生成的自适应有限元网格生成算法研究中,将推进波前法（AFT）与背景网格法结合并提出改进方法,有效地解决了网格生成和单元尺寸计算这两个关键问题。改进的AFT方法,将前沿区分为活跃前沿和非活跃前沿两类,在选取目标前沿时既考虑前沿尺寸又考虑前沿分类。改进的背景网格法,利用结构化栅格对背景网格进行管理,在栅格中直接存放背景网格中的单元,既提高了新单元尺寸的计算速度,又从数值上保证了新生成网格中单元之间尺寸合理过渡。     

数值流形单元法数学网格自适应

基于数值流形方法和有限覆盖技术,将有限元法的后验误差估计理论及h型网格自适应技术推广应用到数值流形单元法中,提出了数值流形单元法的后验误差估计方法和数学网格自适应技术,并编制了相应的程序。数值算例表明,经过网格自适应,可以在粗糙的初始网格基础上得到质量比较理想的网格,计算结果可达到用户要求的精度。     

前沿推进曲面四边形网格生成算法

将一类前沿推进三角形曲面网格生成算法拓展到曲面四边形网格生成。新算法逐个单元推进前沿,避免了铺路法（Paving）逐排推进方式的鲁棒性问题,针对四边形算法在理想点计算、候选点列表构建和新单元生成等环节存在的特殊技术问题给出了系统的解决方案。数值实验表明,本文算法能针对复杂的组合参数曲面自动生成全四边形网格,网格质量优于...     

用逐点插入法生成Delaunay四面体自适应网格

介绍一种基于Delaunay算法的四面体自适应网格的自动划分方法。该方法用单元尺度场控制生成网格的疏密分布,在不满足尺度场要求的单元面形心处插入新节点,同时计算新节点单元尺寸参数,实现三维实体的Delaunay四面体自动划分。此方法具有几个特点:一是表面网格与体内网格同步划分,无需区分两者;二是结点与单元同时生成;三是生成网格自适应性好,疏密分布任意。另外,还介绍了三维网格划分中两个相关算法:一个是约束面恢复算法,该算法基于约束面不允许有单元边与之相交的性质而提出的;另一个是将二维射线法推广至三维空间,判断一个点是否在一多面体内,实现了凹多面体的划分。最后通过算例对单元质量进行了评价。本文所述方法是一种有效的四面体自适应单元生成算法。     

结合自适应网格的基于特征线方程的分离算法

提出一种新的网格自适应方法:在需要加密的网格单元中心加入新结点,并对加密后的相邻三角形网格单元进行公共边变换, 构成新的网格单元. 与传统的在网格边界中点加入新节点的自适应方法相比,新方法可以更加灵活地控制网格密度,加密后的网格继承原先的网格质量不发生畸变,并且算法编程简便,容易实现. 将自适应网格生成方法和基于特征线方程的分离算法相结合,对空腔内不可压缩黏性流动进行了计算. 在特征线方向上进行时间步离散,动量方程求解过程中采用非增量型分离算法. 计算中,把求解变量梯度值作为判定准则,在变化剧烈的区域进行网格局部加密. 计算结果表明该组合算法有很好的计算精度,并有效减少了计算时间和存储量.      

福建海岸带地质环境背景与海水入侵的形成条件探讨

提出一种新的网格自适应方法：在需要加密的网格单元中心加入新结点,并对加密后的相邻 三角形网格单元进行公共边变换, 构成新的网格单元. 与传统的在网格边界中点加入新节点的自 适应方法相比,新方法可以更加灵活地控制网格密度,加密后的网格继承原先的网格质量不 发生畸变,并且算法编程简便,容易实现. 将自适应网格生成方法和基于特征线方程的分离 算法相结合,对空腔内不可压缩黏性流动进行了计算. 在特征线方向上进行时间步离散,动 量方程求解过程中采用非增量型分离算法. 计算中,把求解变量梯度值作为判定准则,在 变化剧烈的区域进行网格局部加密. 计算结果表明该组合算法有很好的计算精度,并有效减 少了计算时间和存储量.     

黏性边界层网格自动生成

高雷诺数黏性流动在壁面附近存在边界层,在计算模拟中自动生成可靠且有效的计算网格仍然是计算流体力学存在的瓶颈.三棱柱/四面体混合网格技术在一定程度上缓解了这个困难.然而,对于复杂外形的情况,在边界层内自动高效生成高质量的三棱柱单元仍然十分困难.常用的层推进法在凹凸区域及角点处生成的边界层网格单元质量较差,边界层网格最外层尺寸不均匀.针对这些问题,发展了一种黏性边界层网格自动生成方法,通过顶点周围边的二面角识别物面网格特征确定多生长方向,预估并调整生长高度处理相交情况.同时提出一种三维前沿尺寸调节方式,提高了边界层网格单元的正交性,保证了边界层网格与远场网格尺寸的光滑过渡.通过ONERA M6翼型以及带发动机短舱的DLR-F6翼身组合体等外形的网格生成实例及绕流数值模拟,将计算值与标准实验值进行对比,结果表明:该方法能够自动高效地生成满足数值计算需求的混合网格.     

对流占优问题的无网格自适应布点方案

应用常规数值方法求解对流占优的对流扩散方程时会出现非物理的数值伪振荡现象.因此本文提出了一种基于无网格径向点插值法的自适应布点方案,并成功地解决了对流占优时的数值伪振荡问题.在自适应布点的实施过程中,该方案将无网格方法中的背景积分单元作为自适应控制的梯度计算单元,并将该控制单元场函数梯度的大小作为自适应的梯度控制指标,然后给定相应的梯度控制限,通过控制指标和梯度限的比较来指示高梯度区域进行自适应中心加点和梯度计算单元的分解.数值结果表明:这种基于无网格径向点插值法的自适应布点方案不仅能有效地消除对流占优时的数值伪振荡现象,而且它还具有计算精度高、数值稳定性好、算法实施简单、前后处理方便的优点.     

岩土材料应变局部化的有限元分析方法

采用有限单元法分析岩土材料的应变局部化时经常会遇到单元尺寸敏感性问题和网格锁定问题。自适应网格技术能够有效地解决网格锁定问题,但仍然无法克服计算结果对单元尺寸的依赖性,尽管在一维情况下被证明是可行的。复合体理论(均匀化理论)和弱非连续有限元方法可以成功地解决岩土材料的单元尺寸敏感性问题,在一维情况下两类方法实际上是一致的。本文针对岩土材料应变局部化的有限元新技术所存在的若干问题进行了详细的讨论,并给出了有关算例。     

On the Global Geometric Structure of the Dynamics of the Elastic Pendulum

We approach the planar elastic pendulum as a singular perturbation of the pendulum to show that its dynamics are governed by global two-dimensional invariant manifolds of motion. One of the manifolds is nonlinear and carries purely slow periodic oscillations. The other one, on the other hand, is linear and carries purely fast radial oscillations. For sufficiently small coupling between the angular and radial degrees of freedom, both manifolds are global and orbitally stable up to energy levels exceeding that of the unstable equilibrium of the system. For fixed value of coupling, the fast invariant manifold bifurcates transversely to create unstable radial oscillations exhibiting energy transfer. Poincaré sections of iso-energetic manifolds reveal that only motions on and near a separatrix emanating from the unstable region of the fast invariant manifold exhibit energy transfer.     

An adaptive finite element method for forced convection

This paper presents an adaptive finite element method to solve forced convective heat transfer. Solutions are obtained in primitive variables using a high-order finite element approximation on unstructured grids. Two general-purpose error estimators are developed to analyse finite element solutions and to determine the characteristics of an improved mesh which is adaptively regenerated by the advancing front method. The adaptive methodology is validated on a problem with a known analytical solution. The methodology is then applied to heat transfer predictions for two cases of practical interest. Predictions of the Nusselt number compare well with measurements and constitute an improvement over previous results. © 1997 John Wiley & Sons, Ltd.     

Center manifold and stability for skew-product flows

We study the existence and smoothness of global center, center-stable, and center-unstable manifolds for skew-product flows. Smooth invariant foliations to the center stable and center unstable manifolds are also discussed.     

任意平面区域的变尺寸有限元网格划分

利用任意平面域边界表述中的拓年要素定义网格的控制信息,通过引入节点间距函数来控制区域内网格尺寸变化,并将结合前沿生成法和Ｄｅｌａｕｎａｙ三角化方法的优点,优先处理前沿的最长边,尽可能在局部生成边长逐渐减小的Ｄｅｌａｕｎａｙ三角形,最终实验区域内网格的疏密过渡。     

Approximate inertial manifolds for reaction-diffusion equations in high space dimension

The concept of approximate inertial manifolds was introduced by Foiaset al. (1987) in the case of the two-dimensional Navier-Stokes equations. These manifolds are finite dimensional smooth manifolds such that the orbits enter a very thin neighborhood of the manifold after a transient time; this concept replaces the one of inertial manifold when either an inertial manifold does not exist or its existence is not known. Our aim in this paper is to prove that approximate inertial manifolds exist for reaction-diffusion equations in high space dimension by opposition with exact inertial manifolds whose existence has only been proved in low dimension and for which nonexistence results have been obtained in space dimensionn=4.     

Automatic unstructured surface mesh generation for complex configurations

In this paper a new object‐oriented (OO) approach is presented for automatic parallel advancing front based surface mesh generation and adaptive remeshing for complex configurations. Based on the ST++‐system the advantages of the OO design and implementation compared to the traditional structural approach are described. Algorithmic enhancements to the advancing front method are explained, enabling a robust NURBS based triangulation process directly on B‐rep CAD data. The message passing (MPI) parallelization strategy together with the achievable performance improvements are demonstrated. With the outlined parallel geometry analysis/rasterization a powerful method is described to derive automatically a well suited mesh size specification without any user‐interaction from scratch. The application of this method to a complex ‘real world’ example finishes this paper. Copyright © 2004 John Wiley & Sons, Ltd.     

Direct Numerical Simulations of Premixed Turbulent Flames with Reduced Chemistry: Validation and Flamelet Analysis

Direct numerical simulation is a very powerful tool to evaluate the validity of new models and theories for turbulent combustion. In this paper, direct numerical simulations of spherically expanding premixed turbulent flames in the thin reaction zone regime and in the broken reaction zone regime are performed. The flamelet-generated manifold method is used in order to deal with detailed reaction kinetics. The computational results are analyzed by using an extended flame stretch theory. It is investigated whether this theory is able to describe the influence of flame stretch and curvature on the local burning velocity of the flame. It is found that if the full profiles of flame stretch and curvature through the flame front are included in the theory, the local mass burning rate is well predicted. The influence of several approximations, which are used in other existing theories, is studied. When flame stretch is assumed constant through the flame front or when curvature of the flame front is neglected, the theory fails to predict the local mass burning rate. The influence of using a reduced chemistry model is investigated by comparing flamelet simulations with reduced and detailed chemistry.     

Pullback Attracting Inertial Manifolds for Nonautonomous Dynamical Systems

In this paper we present an abstract approach to inertial manifolds for nonautonomous dynamical systems. Our result on the existence of inertial manifolds requires only two geometrical assumptions, called cone invariance and squeezing property, and two additional technical assumptions, called boundedness and coercivity property. Moreover we give conditions which ensure that the global pullback attractor is contained in the inertial manifold. In the second part of the paper we consider special nonautonomous dynamical systems, namely processes (or two-parameter semi-flows). As a first application of our abstract approach and for reason of comparison with known results we verify the assumptions for semilinear nonautonomous evolution equations whose linear part satisfies an exponential dichotomy condition and whose nonlinear part is globally bounded and globally Lipschitz.     

加筋路径驱动的板壳自适应等几何屈曲分析

加筋薄壁结构常被用于航空航天结构的轻量化设计.随着结构尺寸和几何特征的增加,需要更加精细的网格来满足分析精度的要求.传统的等几何方法采用NURBS张量积形式的拓扑结构,使得在分析过程中难以实现局部细化,而全局细化则会增加不必要的自由度.为了提升加筋板壳结构的数值分析精度和效率,提出一种基于RPHT (rational polynomial splines over hierarchical T-meshes)样条的加筋板壳自适应等几何屈曲分析方法.样条网格可以沿着加筋路径进行自适应的局部细化,有效提升低自由度下加筋板壳结构等几何屈曲分析的精度.首先,蒙皮和筋条分别采用RPHT样条曲面和NURBS样条曲线进行建模,几何建模与数值仿真采用统一的几何语言,实现建模与分析的一体化.其次,采用几何投影算法和样条插值算法实现筋条与蒙皮之间的高效高精度强耦合,并建立基于加筋路径驱动自适应网格细化方法.最后,曲线加筋板和网格加筋壳两个算例验证本方法的高效性和鲁棒性,通过与基于NURBS的等几何分析进行对比,本方法能够明显降低分析模型的总自由度.