首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到19条相似文献,搜索用时 296 毫秒
1.
在激波数值计算中,容易出现数值振荡的问题,振荡激烈时会掩盖真实解,为此提出了许多高精度复杂计算格式或采用人工粘性抑制数值振荡.从信号处理的角度,提出双重小波收缩方法,它能自适应提取激波数值振荡解中的真实物理解.先用局部微分求积法求解浅水波方程和理想流体Euler运动方程中的激波问题,发现其数值振荡现象严重,然后采用双重小波收缩方法对其处理,获得了无数值振荡解,它能准确捕捉激波的位置并且保持激波结构.相比于复杂的Riemann(黎曼)求解格式,借助小波收缩方法,可以采用相对简单的计算格式如微分求积法求解激波问题.  相似文献   

2.
翟传磊  勇珩 《应用数学和力学》2015,36(10):1045-1057
在交错网格型Lagrange(拉格朗日)流体力学算法中,通常采用人为粘性捕捉激波,人为粘性的好坏对于计算结果至关重要.研究了一种基于子网格边界处近似Riemann解的新型人为粘性.新人为粘性能够满足动量守恒和熵不等式.利用子网格边界速度差中引入的限制器,新人为粘性能够区分激波和等熵压缩,并能满足球对称问题中的波面不变性.新人为粘性在典型模型数值模拟及惯性约束聚变黑腔整体数值模拟中取得了较好的结果.  相似文献   

3.
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.  相似文献   

4.
恒温平行板间多孔介质通道中的分层耗散流动   总被引:1,自引:1,他引:0  
利用Darcy模型,研究了平行板间充填饱和多孔介质的通道中,在热量入口处传热的粘性耗散效应.讨论了等温边界情况.求得热量入口处局部温度和体积计算平均温度随Nusselt数的分布.给出了独立于Brimkman数的经充分发展的Nusselt数应为6A·D2并观察到,若忽略粘性耗散影响,将导致熟知的内流现象,此时Nusselt数等于4.93.还给出了有限差分数值解.结果表明解析法和数值法的结果吻合很好.  相似文献   

5.
微分求积法(DQM)能以较少的网格点求得微分方程的高精度数值解,但采用单纯的微分求积法求解二维不可压缩Navier_Stokes 方程时,只能对低雷诺数流动获得较好的数值解,当雷诺数较高时会导致数值解不收敛· 为此,提出了一种微分求积法与迎风差分法混合求解二维不可压缩Navier_Stokes 方程的预估_校正数值格式,用伪时间相关算法以较少的网格点获得了较高雷诺数流动的数值解· 作为算例,对1∶1 和1∶2 驱动方腔内的流动进行了计算,得到了较好的数值结果·  相似文献   

6.
豆艳萍 《数学年刊A辑》2004,25(4):531-540
本文考虑一维可压缩Navier-Stokes方程有关初边值问题粘性激波解的渐近稳定性,通过L2-能量估计,证明了在小扰动情况下,粘性激波是稳定的.  相似文献   

7.
本文讨论单个粘性守恒律方程与具有粘性的p方程组的Cauchy问题.根据初始资料的不同情形,其相应的Riemann问题以疏散波,激波或它们的迭加为弱解.本文的目的是指出Cauchy问题的解将分别趋于疏散波,激波或它们的迭加.本文基本方法是能量积分法.文中综述了现有的成果,也提出了一些未解决的问题.  相似文献   

8.
本文考虑一维可压缩Navier-Stokes方程有关初边值问题粘性激波解的渐近稳定性,通过L~2-能量估计,证明了在小扰动情况下,粘性激波是稳定的。  相似文献   

9.
微通道周期流动电位势及电粘性效应   总被引:1,自引:0,他引:1  
求解了双电层的Poisson-Boltzmann方程和流体运动的Navier-Stokes方程,得到在周期压差作用下,二维微通道的周期流动电位势,流动诱导电场和液体流动速度的解析解.量纲分析表明,流体电粘性力与以下3个参数有关:1) 电粘性数,它表示定常流动时,通道最大电粘性力与压力梯度的比;2) 形状函数,它表示电粘性力在通道横截面的分布形态; 3) 耦合系数,它表示电粘性力的振幅衰减特征和相位差.分析结果表明,微通道周期流动诱导电场、流动速度与频率Reynolds数有关.在频率Reynolds数小于1时,流动诱导电场随频率Reynolds数变化很慢.在频率Reynolds数大于1时,流动诱导电场随频率Reynolds数的增加快速衰减.在通道宽度与双电层厚度比值较小情况下,电粘性效应对周期流动速度和流动诱导电场有重要影响.  相似文献   

10.
从气体动力学基本方程组出发,根据数值计算方法的理论,分析了近年来在跨声速流动计算中广泛使用的人工密度法,指出流函数计算中采用人工密度法,理论上是有问题的,进而提出一种正确的人工粘性表达式。数值计算表明,它扩大了流函数方法可计算的Mach数范围,使激波位置接近于实验结果。  相似文献   

11.
We study the zero-dissipation problem for a one-dimensional model system for the isentropic flow of a compressible viscous gas, the so-called p-system with viscosity. When the solution of the inviscid problem is piecewise smooth and having finitely many noninteracting shocks satisfying the entropy condition, there exists unique solution to the viscous problem which converges to the given inviscid solution away from shock discontinuities at a rate of order ε as the viscosity coefficient ε goes to zero. The proof is given by a matched asymptotic analysis and an elementary energy method. And we do not need the smallness condition on the shock strength.  相似文献   

12.
The plane and axisymmetric problems of super- and hypersonic flow of a homogeneous viscous heat-conducting perfect gas over a blunt body are considered. Generalized viscous shock layer equations that take into account all the second-order effects of boundary-layer theory, i.e., the terms O(Re?1/2), are derived from the Navier–Stokes equations by the asymptotic method, and all the out-of-order third-order terms O(Re?1) and higher-order terms are also retained, except terms with second derivations in the marching coordinate (Re is Reynolds number, determined from the free-stream density and velocity the linear dimension, which is equal to the nose radius of the blunt Body, and the free-stream shear viscosity at the stagnation temperature). Thus, only the presence of terms with second derivatives in the marching coordinate, which specify the elliptical properties of the complete system of Navier–Stokes equations, distinguish it from the generalized viscous shock layer equations, which do not contain these terms. Slip and a temperature jump conditions on a body surface are presented with the same degree of accuracy, and generalized Rankine–Hugoniot conditions on a head shock, which take into account the effects of the viscosity and heat conduction, including their influence on the determination of the pressure, are derived. The incorrect and unfounded approximations used in preceding studies and the efficiency of iterative marching techniques for solving the generalized viscous shock layer equations, as well as the ability of the latter to provide a correct solution for the drag and heat-transfer coefficients in the transitional flow regime if the solution is constructed taking the slip and temperature jump on a surface and on a head shock into account, are noted.  相似文献   

13.
邢秀侠 《数学进展》2005,34(1):43-53
本文首先利用几何奇异摄动方法,证明了粘性系数充分小时一类非凸粘性平衡律方程的粘性冲击波的存在性,推广了原来在非线性项严格凸的条件下得到的结果.进而,利用谱分析和上下解方法,证明了对固定的小的粘性系数此类波是全局渐近指数稳定的,推广了反应扩散方程中经典的全局稳定性结果.  相似文献   

14.
We establish existence and stability of multidimensional shock fronts in the vanishing viscosity limit for a general class of conservation laws with “real”, or partially parabolic, viscosity including the Navier-Stokes equations of compressible gas dynamics with standard or van der Waals-type equation of state. More precisely, given a curved Lax shock solution u0 of the corresponding inviscid equations for which (i) each of the associated planar shocks tangent to the shock front possesses a smooth viscous profile and (ii) each of these viscous profiles satisfies a uniform spectral stability condition expressed in terms of an Evans function, we construct nearby smooth viscous shock solutions uε of the viscous equations converging to u0 as viscosity ε→0, and establish for these sharp linearized stability estimates generalizing those of Majda in the inviscid case. Conditions (i)-(ii) hold always for shock waves of sufficiently small amplitude, but in general may fail for large amplitudes.We treat the viscous shock problem considered here as a representative of a larger class of multidimensional boundary problems arising in the study of viscous fluids, characterized by sharp spectral conditions rather than symmetry hypotheses, which can be analyzed by Kreiss-type symmetrizers.Compared to the strictly parabolic (artificial viscosity) case, the main new features of the analysis appear in the high frequency estimates for the linearized problem. In that regime we use frequency-dependent conjugators to decouple parabolic components that are smoothed from hyperbolic components (like density in Navier-Stokes) that are not. The construction of the conjugators and the subsequent estimates depend on a careful spectral analysis of the linearized operator.  相似文献   

15.
Two models based on the hydrostatic primitive equations are proposed. The first model is the primitive equations with partial viscosity only, and is oriented towards large-scale wave structures in the ocean and atmosphere. The second model is the viscous primitive equations with spectral eddy viscosity, and is oriented towards turbulent geophysical flows. For both models, the existence and uniqueness of global strong solutions are established. For the second model, the convergence of the solutions to the solutions of the classical primitive equations as eddy viscosity parameters tend to zero is also established.  相似文献   

16.
The method of combined asymptotic expansions is used to solve the problem of the focusing of a shock wave (in a weakly compressible medium of high viscosity. Asymptotic forms of the solution are constructed in a number of spatial zones. The focusing zone is described by its asymptotic form obtained by combining it with the solution corresponding to viscous geometrical acoustics. The reflection of a shock wave formed as a result of velocity jump near one of the foci of the ellipsoid of revolution is discussed as an example. Analytical relationships descrbing the focusing zone around the second focus are obtained. It is shown that at the focus itself the wave profile has an antisymmetric form, and the compression wave is followed by a rarefaction wave of the same form.  相似文献   

17.
In this paper, we study the inviscid limit problem for the scalar viscous conservation laws on half plane. We prove that if the solution of the corresponding inviscid equation on half plane is piecewise smooth with a single shock satisfying the entropy condition, then there exist solutions to the viscous conservation laws which converge to the inviscid solution away fromthe shock discontinuity and the boundary at a rate of ε^1 as the viscosity ε tends to zero.  相似文献   

18.
In this paper we prove the existence of curved, multidimensional viscous shocks and also justify the small‐viscosity limit. Starting with a curved, multidimensional (inviscid) shock solution to a system of hyperbolic conservation laws, we show that the shock can be obtained as a small‐viscosity limit of solutions to an associated parabolic problem (viscous shocks). The two main hypotheses are a natural Evans function assumption on the viscous profile, together with a restriction on how much the shock can deviate from flatness. The main tools are a conjugation lemma that removes xN/? dependence from the linearization of the parabolic problem about the viscous profile, new degenerate Kreiss‐type symmetrizers used to prove an L2 estimate for the linearized problem, and a finite‐regularity calculus of semiclassical and mixed type (classical‐semiclassical) pseudodifferential operators. © 2003 Wiley Periodicals, Inc.  相似文献   

19.
In this note we analyze a model for a unidirectional unsteady flow of a viscous incompressible fluid with time dependent viscosity. A possible way to take into account such behaviour is to introduce a memory formalism, including thus the time dependent viscosity by using an integro-differential term and therefore generalizing the classical equation of a Newtonian viscous fluid. A possible useful choice, in this framework, is to use a rheology based on stress/strain relation generalized by fractional calculus modelling. This is a model that can be used in applied problems, taking into account a power law time variability of the viscosity coefficient. We find analytic solutions of initial value problems in an unbounded and bounded domain. Furthermore, we discuss the explicit solution in a meaningful particular case.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号