基于ENO格式的三阶修正系数格式

不增加基点,仅摄动二阶ENO格式的系数(简记为MCENO),得到一类求解双曲型守恒律方程的三阶MCENO格式．由MCENO格式的构造过程可以看出,MCENO格式保留了ENO格式的许多性质,例如本质无振荡性、TVB性质等,且能提高一阶精度．进一步,利用MCENO格式模拟二维Rayleigh-Taylor(RT)不稳定性和Lax激波管的数值求解问题．数值结果表明,t=2.0时,MCENO格式的密度曲线处于三阶WENO格式和五阶WENO格式之间,是一个高效高精度格式．值得注意的是,三阶MCENO格式,三阶WENO格式和五阶WENO格式的CPU时间之比为0.62:1:2.19．表明相对于原始ENO格式,MCENO格式在光滑区域有较高精度,能提高格式精度．     

一种新的求解双曲型守恒律的三阶中心差分格式

本文提出了一种求解双曲型守恒律新的三阶中心差分格式，主要是引入了一种推广的三阶重构，并证明了这种重构在网格边界无振荡．所提的格式保持了中心差分格式简单的优点，不需用Riemann解算器，避免了进行特征解耦．数值试验结果表明本文格式是高精度、高分辨率的。     

粘性机制下一类熵相容格式的对比研究

研究了一种人工和物理耗散机制下的离散熵相容格式,探讨数值粘性和物理粘性的大小以及它们所起的作用.所得结论是:在激波捕捉的过程中,粘性系数越大,则无需加入人工粘性项;粘性系数较小时,除了物理粘性项,还需要加入人工粘性项来得到熵相容格式.首先研究了一维粘性Burgers方程离散熵相容格式,再将其推广至Navier-Stokes方程.数值算例采用空间半离散格式,并结合显式三步三阶Runge-Kutta(RK3)方法进行时间推进.这两类方程的数值结果表明,最终选取的熵相容格式能够准确地捕捉到激波.     

高精度隐式参差光滑格式的构造

参照Lax-Wendroff格式的构造方法,就双曲型方程、抛物型方程和双曲-抛物型方程,构造了一种新的IRS(implicit residual smoothing)格式。该IRS格式有二阶或三阶时间精度且大大地拓宽了解的稳定区域和CFL数。这种新的IRS格式有中心加权型和迎风偏向型两种,并用von-Neumann分析方法分析了格式的稳定范围。讨论了在透平机械中广泛应用的Dawes方法的局限性,发现该方法对稳态问题得出的解与时间步长的选取有关,对粘性问题求解时,时间步长受严格限制。最后,结合TVD(total variation diminishing)格式和四阶Runge-Kutta技术,用IRS格式和Dawes方法对二维反射激波场进行了数值模拟,数值结果支持本文的分析结论。     

基于映射函数的三阶WENO改进格式及其应用

低耗散、高分辨率激波捕捉格式对含激波流场的数值模拟具有重要意义.在传统三阶WENO格式（WENO-JS3）和三阶WENO Z格式（WENO-Z3）基础上,基于映射函数,给出WENO-M3、WENO MZ3格式.选用Sod激波管、激波与熵波相互作用、双爆轰波碰撞及双Mach(马赫)反射等经典算例,考察上述格式的计算性能.数值结果表明,WENO-MZ3格式相较其他格式具有耗散低、对流场结构分辨率高的特性.为了进一步扩展WENO-MZ3格式的应用范围,采用该格式数值研究封闭方形舱室内柱形高压、高密度气体爆炸波传播过程,波系演化规律以及壁面典型测点压力载荷.数值计算结果表明WENO-MZ3格式能够较好地模拟包含高压比、高密度比的爆炸波且给出数值耗散较小的壁面压力载荷.     

求解Euler方程的高精度熵相容格式

熵相容格式相比于一般的熵稳定格式进一步控制了激波处的熵增量,一维情况下能有效消除膨胀激波及间断处的伪振荡等现象.对于Euler方程,可以通过对特征变量进行WENO重构以获得高阶熵相容格式的数值粘性项,然后与高阶熵守恒格式结合得到高精度熵相容格式,在WENO重构过程中的权重关于特征变量是非线性的,这导致了大量的向量内积运算.通过用压强和熵代替特征变量来计算权重,可以显著减少重构的计算量,并且数值算例表明这种权重的计算方式能很好地保持数值格式的高阶精度和基本无振荡的效果.     

基于高阶空间精度格式和嵌套网格对旋翼尾迹捕捉的改进

发展了一种基于高阶迎风格式和嵌套网格捕捉直升机悬停旋翼涡尾迹的方法．无粘通量采用Roe Reimann求解器,使用改进的5阶加权基本无振荡(WENO)格式对交界面左右状态进行高阶插值,并与MUSCL插值进行比较．为便于捕捉尾迹和实施周期性边界条件,计算采用结构嵌套网格,其中高质量的旋翼网格完全嵌套于背景网格中．当解达到近似收敛后在桨尖涡分布区域对背景网格进行加密,如此经过3次得到优化的背景网格．考虑到WENO格式插值的特点,提出了搜索3层洞边界和人工外边界的方法以便插值的直接进行．用该方法对一跨音速和一亚音速悬停旋翼粘性流场进行了数值计算．数值结果表明:所发展方法对涡尾迹具有很高的捕捉能力;与MUSCL格式相比,WENO格式具有较低的数值耗散．     

求解浅水波方程的熵相容格式

提出了一种求解浅水波方程组的熵相容格式．在熵稳定通量中添加特征速度差分绝对值的项来抵消解在跨过激波时所产生的熵增,从而实现熵相容．新的数值差分格式具有形式简单、计算效率高、无需添加任何的人工数值粘性的特点．数值算例充分说明了其显著的优点．利用新格式成功地模拟了不同类型溃坝问题的激波、稀疏波传播及溃坝两侧旋涡的形成,是求解浅水波方程组较为理想的方法.     

可压缩平面流中尖角处不定常涡产生的初始阶段的数值模拟

本文采用有限体积法结合高分辨率的ＴＶＤ（总变差不增）格式，对平面激波遇矩形、三角形障碍物问题进行数值模拟，采用Ｓｃｈｗａｒｚ变换生成计算网格，可使网格线在尖角处较密集且网格设置符合流场物理特性。初始阶段的流场复杂且粘性影响的尺度很小，模型中可以忽略流体的粘性。计算结果描述了平面无粘可压流中，由于弓形激波后熵与温度的不均匀性引起的尖角下游不定常集中涡的产生过程。数值计算与应我们要求所做激波管实验的光测结果相符合。     

高浓度悬浮固粒对射流界面的粘性稳定性影响

该文首次利用双流体模型和扰动速度势理论,推得含高浓度悬浮固粒的射流界面粘性稳定性方程和对应的固气扰动速度比值方程．通过数值计算,得到了不同雷诺数及固粒属性的射流界面粘性稳定性曲线和对应的固气扰动速度比值曲线．在分析和比较所得的粘性稳定性曲线的基础上,得到了流场雷诺数及固粒特性对射流界面粘性稳定性影响的结论．同时,通过分析所得的固气扰动速度比值曲线,得到了流场雷诺数及固粒等效斯托克斯数对固粒跟随气流的扰动 性能的影响的结论．这些结论是首次在计入气流的粘性的条件下得到的,不同于文献［８］和文献［１０］相关的囿于无粘情形的研究,对于两相射流发展的认识和工程实际中实施对两相射流场的人工控制有重要意义．     

The Runge‐Kutta DG finite element method and the KFVS scheme for compressible flow simulations

This article presents a new type of second‐order scheme for solving the system of Euler equations, which combines the Runge‐Kutta discontinuous Galerkin (DG) finite element method and the kinetic flux vector splitting (KFVS) scheme. We first discretize the Euler equations in space with the DG method and then the resulting system from the method‐of‐lines approach will be discretized using a Runge‐Kutta method. Finally, a second‐order KFVS method is used to construct the numerical flux. The proposed scheme preserves the main advantages of the DG finite element method including its flexibility in handling irregular solution domains and in parallelization. The efficiency and effectiveness of the proposed method are illustrated by several numerical examples in one‐ and two‐dimensional spaces. © 2006 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 2006     

An investigation of chaos in reaction-diffusion equations

The purpose of this article is to investigate graphically and numerically the topic of chaos in reaction-diffusion equations. This article is based on the article by Mitchell and Bruch [1]. One- and two-dimensional forms of the reaction-diffusion equation are discretized using the explicit Euler finite difference scheme. Plots are presented to show the effect of bifurcation parameters on the difference equations. Varying these parameters produce single point, periodic, chaotic, intermittent, and divergent solutions.     

ZERO DISSIPATION LIMIT OF THE COMPRESSIBLE HEAT-CONDUCTING NAVIER-STOKES EQUATIONS IN HE PRESENCE OF THE SHOCK

The zero dissipation limit of the compressible heat-conducting Navier-Stokes equations in the presence of the shock is investigated. It is shown that when the heat ε→ 0 (see (1.3)), if the solution of the corresponding Euler equations is piecewise smooth with shock wave satisfying the Lax entropy condition, then there exists a smooth solution to the Navier-Stokes equations, which converges to the piecewise smooth shock solution of the Euler equations away from the shock discontinuity at a rate of ε. The proof is given by a combination of the energy estimates and the matched asymptotic analysis introduced in [3].     

Asymptotic Expansions and Numerical Methods for Compressible and Low Mach Number Fluid Flow

The results of a formal asymptotic low Mach number analysis [5, 6] of the Euler equations of gas dynamics are used to extend the validity of a numerical method from the simulation of compressible inviscid flow fields to the low Mach number regime. Although, different strategies are applicable [7, 8, 5, 9] in this context we focus our view to a preconditioning technique recently proposed by Guillard and Viozat [16]. We present a finite volume approximation of the governing equations using a Lax‐Friedrichs scheme whereby a preconditioning of the incorporated numerical dissipation is employed. A discrete asymptotic analysis proves the validity of the scheme in the low Mach number regime.     

Minimal geodesics on groups of volume‐preserving maps and generalized solutions of the Euler equations

The three‐dimensional motion of an incompressible inviscid fluid is classically described by the Euler equations but can also be seen, following Arnold [1], as a geodesic on a group of volume‐preserving maps. Local existence and uniqueness of minimal geodesics have been established by Ebin and Marsden [16]. In the large, for a large class of data, the existence of minimal geodesics may fail, as shown by Shnirelman [26]. For such data, we show that the limits of approximate solutions are solutions of a suitable extension of the Euler equations or, equivalently, are sharp measure‐valued solutions to the Euler equations in the sense of DiPerna and Majda [14]. © 1999 John Wiley & Sons, Inc.     

High order upwind scheme for modelling turbulent shallow water flow in hydraulic structures

An unstructured finite volume model for quasi-2D free surface flow with wet-dry fronts and turbulence modelling is presented. The convective flux is discretised with either a an hybrid second-order/first-order scheme, or a fully second order scheme, both of them upwind Godunov's schemes based on Roe's average. The hybrid scheme uses a second order discretisation for the two unit discharge components, whilst keeping a first order discretisation for the water depth [2]. In such a way the numerical diffusion is much reduced, without a significant reduction on the numerical stability of the scheme, obtaining in such a way accurate and stable results. It is important to keep the numerical diffusion to a minimum level without loss of numerical stability, specially when modelling turbulent flows, because the numerical diffusion may interfere with the real turbulent diffusion. In order to avoid spurious oscillations of the free surface when the bathymetry is irregular, an upwind discretisation of the bed slope source term [4] with second order corrections is used [2]. In this way a fully second order scheme which gives an exact balance between convective flux and bed slope in the hydrostatic case is obtained. The k – ε equations are solved with either an hybrid or a second order scheme. In all the numerical simulations the importance of using a second order upwind spatial discretisation has been checked [1]. A first order scheme may give rather good predictions for the water depth, but it introduces too much numerical diffusion and therefore, it excessively smooths the velocity profiles. This is specially important when comparing different turbulence models, since the numerical diffusion introduced by a first order upwind scheme may be of the same order of magnitude as the turbulent diffusion. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

A Well-Balanced HLL Scheme for Hyperbolic Conservation Systems With Source Terms北大核心CSCD

针对含源项的双曲守恒方程给出了一种新的有限体积格式.经典的有限体积格式不能正确地模拟对流通量项和外力之间的平衡所产生的动力学问题.为解决这个问题,仿照经典的HLL近似Riemann求解器设计思路设计了含源项的近似Riemann求解器.针对含重力源项的一维流体Euler方程和理想磁流体方程,通过对通量计算格式的修正得到了保平衡HLL格式（WB-HLL）,并给出了保平衡的证明.针对一维Euler方程和理想磁流体给出了两个算例,比较了传统HLL格式和提出的WB-HLL格式的计算精度.计算结果表明,WB-HLL格式精度更高,收敛更快.     

Hydrodynamic limits of the nonlinear Klein–Gordon equation

We perform the mathematical derivation of the compressible and incompressible Euler equations from the modulated nonlinear Klein–Gordon equation. Before the formation of singularities in the limit system, the nonrelativistic-semiclassical limit is shown to be the compressible Euler equations. If we further rescale the time variable, then in the semiclassical limit (the light speed kept fixed), the incompressible Euler equations are recovered. The proof involves the modulated energy introduced by Brenier (2000) [1].     

两类变时间步长的非线性Galerkin算法的稳定性

1．引言近年来,随着计算机的飞速发展,人们越来越关心非线性发展方程解的渐进行为．为了较精确地描述解在时间t→∞时的渐进行为,人们发展了一类惯性算法,即非线性Galerkin算法．该算法是将来解空间分解为低维部分和高维部分,相应的方程可以分别投影到它们上面,它的解也相应地分解为两部分,大涡分量和小涡分量;然后核算法给出大涡分量和小涡分量之间依赖关系的一种近似,以便容易求出相应的近似解．许多研究表明,非线性Galerkin算法比通常的Galerkin算法节省可观的计算量．当数值求解微分方程时,计算机只能对已知数据进行有限位…     

Quasi-acoustic scheme for the Euler equations of gas dynamics

We suggest an original scheme and an algorithm for the numerical solution of the Euler equations of gas dynamics. The construction of the scheme is based on the mass, momentum, and energy conservation laws. The flux computation is carried out by summation of elementary fluxes formed by small-amplitude running waves that satisfy the linearized equations of gas dynamics. The scheme contains no artificial regularizers, has second-order accuracy on smooth solutions, and is quasimonotone in a neighborhood of the discontinuities. Examples of one- and two-dimensional computations are given.