Zero Viscosity Limit for Analytic Solutions of the Navier-Stokes Equation on a Half-Space.¶ II. Construction of the Navier-Stokes Solution

This is the second of two papers on the zero-viscosity limit for the incompressible Navier-Stokes equations in a half-space in either 2D or 3D. Under the assumption of analytic initial data, we construct solutions of Navier-Stokes for a short time which is independent of the viscosity. The Navier-Stokes solution is constructed through a composite asymptotic expansion involving the solutions of the Euler and Prandtl equations, which were constructed in the first paper, plus an error term. This shows that the Navier-Stokes solution goes to an Euler solution outside a boundary layer and to a solution of the Prandtl equations within the boundary layer. The error term is written as a sum of first order Euler and Prandtl corrections plus a further error term. The equation for the error term is weakly nonlinear; its linear part is the time dependent Stokes equation. This error equation is solved by inversion of the Stokes equation, through expressing the solution as a regular (Euler-like) part plus a boundary layer (Prandtl-like) part. The main technical tool in this analysis is the Abstract Cauchy-Kowalewski Theorem. Received: 5 September 1996 / Accepted: 14 July 1997     

基于格子Boltzmann方法的超声速非平衡流模拟

基于D1Q4可压缩格子Boltzmann模型,按照流通矢量分裂方法的思路,采用坐标旋转技术构造求解三维带化学反应Navier-Stokes方程对流通量求解器.结合有限体积法求解三维化学非平衡流Navier-Stokes方程,采用时间算子分裂算法解决化学反应刚性问题,数值模拟超声速化学非平衡流的三个经典算例.数值结果表明:在高马赫数下,采用D1Q4可压缩格子Boltzmann模型构造的三维对流通量求解器数值模拟中没有出现非物理解,同时在超声速化学非平衡流场中正确分辨激波、燃烧波等物理现象,精度和分辨率均较高,验证了本文构造的三维对流通量求解器的可靠性,拓宽了D1Q4可压缩格子Boltzmann模型的应用范围,为计算超声速化学非平衡流提供一种新方法.     

Kinetics of the Raman instability in laser plasma

The phase-space evolution in a non-relativistic and homogeneous laser plasma in the presence of the stimulated Raman scattering is studied. Transform method is used for a solution of the set of partial differential equations which consists of the Vlasov equation and of the full set of Maxwell equations in a 1D model. Numerical instability of the Fourier-Hermite representation is described and discussed. To overcome numerical instabilities during the simulation, a simplified Fokker-Planck collision term is employed. In the collisionless case the solution is pushed to the practicable limit and the initial phase of particle trapping and acceleration in the potential wells of the electrostatic wave accompanying the Raman backscattered wave was recorded. Also the growth of the electrostatic partner of the Raman forward scattered wave was observed.     

Partial regularity of a generalized solution to the Navier-Stokes equations in exterior domain

In this note, we prove two regularity theorems for solutions to the Navier-Stokes equations of an I.B.V.P. in exterior domains. Namely, we prove that the setS of the singular points of a solution, if not empty, has at most 1-Hausdorff measureH ¹(S)=0. Moreover, the setS is enclosed in a sphere of rayR for anyt>0. These results are obtained as corollaries to the partial regularity results furnished in [2].     

Blowup Criterion for Viscous Baratropic Flows with Vacuum States

We prove that the maximum norm of the deformation tensor of velocity gradients controls the possible breakdown of smooth(strong) solutions for the 3-dimensional (3D) barotropic compressible Navier-Stokes equations. More precisely, if a solution of the 3D barotropic compressible Navier-Stokes equations is initially regular and loses its regularity at some later time, then the loss of regularity implies the growth without bound of the deformation tensor as the critical time approaches. Our result is the same as Ponce’s criterion for 3-dimensional incompressible Euler equations (Ponce in Commun Math Phys 98:349–353, 1985). In addition, initial vacuum states are allowed in our cases.     

Modeling of heat explosion with convection

The work is devoted to numerical simulations of the interaction of heat explosion with natural convection. The model consists of the heat equation with a nonlinear source term describing heat production due to an exothermic chemical reaction coupled with the Navier-Stokes equations under the Boussinesq approximation. We show how complex regimes appear through successive bifurcations leading from a stable stationary temperature distribution without convection to a stationary symmetric convective solution, stationary asymmetric convection, periodic in time oscillations, and finally aperiodic oscillations. A simplified model problem is suggested. It describes the main features of solutions of the complete problem.     

Local Collocation Approach for Solving Turbulent Combined Forced and Natural Convection Problems

An application of the meshless Local Radial Basis Function Collocation Method (LRBFCM) [22, 30–33] in solution of incompressible turbulent combined forced and natural convection is for the first time explored in the present paper. The turbulent flow equations are described by the low-Re number $k − ε$ model with Launder and Sharma [23] and Abe et al. [1] closure coefficients. The involved temperature, velocity, pressure, turbulent kinetic energy and dissipation fields are represented on overlapping 5-noded sub-domains through the collocation by using multiquadrics Radial Basis Functions (RBF). The involved first and second order partial derivatives of the fields are calculated from the respective derivatives of the RBF's. The involved equations are solved through the explicit time stepping. The pressure-velocity coupling is based on Chorin's fractional step method [11]. The adaptive upwinding technique, proposed by Lin and Atluri [27], is used because of the convection dominated situation. The solution procedure is represented for a 2D upward channel flow with differentially heated walls. The results have been assessed by achieving a reasonable agreement with the direct numerical simulation of Kasagi and Nishimura [20] for Reynolds number 4494, based on the channel width, and Grashof number 9.6 × 10⁵. The advantages of the represented mesh-free approach are its simplicity, accuracy, similar coding in 2D and 3D, and straightforward applicability in non-uniform node arrangements.     

Global regularity of the 4D restricted Euler equations

We are concerned with the critical threshold phenomena in the restricted Euler (RE) equations. Using the spectral and trace dynamics we identify the critical thresholds for the 3D and 4D restricted Euler equations. It is well known that the 3D RE solutions blow up. Projected on the 3-sphere, the set of initial eigenvalues which give rise to bounded stable solutions is reduced to a single point, which confirms that the 3D RE blowup is generic. In contrast, we identify a surprisingly rich set of the initial spectrum on the 4-sphere which yields global smooth solutions; thus, 4D regularity is generic.     

Discrete data assimilation in the Lorenz and 2D Navier-Stokes equations

Consider a continuous dynamical system for which partial information about its current state is observed at a sequence of discrete times. Discrete data assimilation inserts these observational measurements of the reference dynamical system into an approximate solution by means of an impulsive forcing. In this way the approximating solution is coupled to the reference solution at a discrete sequence of points in time. This paper studies discrete data assimilation for the Lorenz equations and the incompressible two-dimensional Navier-Stokes equations. In both cases we obtain bounds on the time interval h between subsequent observations which guarantee the convergence of the approximating solution obtained by discrete data assimilation to the reference solution.     

On the sharp front-type solution of the Nagumo equation with nonlinear diffusion and convection

This paper is concerned with the Nagumo equation with nonlinear degenerate diffusion and convection which arises in several problems of population dynamics, chemical reactions and others. A sharp front-type solution with a minimum speed to this model equation is analysed using different methods. One of the methods is to solve the travelling wave equations and compute an exact solution which describes the sharp travelling wavefront. The second method is to solve numerically an initial-moving boundary-value problem for the partial differential equation and obtain an approximation for this sharp front-type solution.     

A sharp lower bound for the Hausdorff dimension of the global attractors of the 2D Navier-Stokes equations

For a special class of the Navier-Stokes equations on the two-dimensional torus, we give a lower bound in the formG ^2/3 (whereG is the Grashof number) for the Hausdorff dimension of its global attractor which is optimal up to a logarithmic term.Dedicated to the memory of Professor Nicholas D. Kazarinoff     

超声速后台阶流动/射流相互作用的数值模拟

采用高精度格式求解二维Navier-Stokes方程研究超声速射流与同向超声速后台阶流动相互作用的流场基本结构及规律,分别应用5阶WENO格式、6阶中心差分格式离散对流项和黏性项,时间推进采用3阶Runge-Kutta格式,并应用消息传递接口（message passing interface,MPI）非阻塞式通信实现并行化.分别研究了超声速后台阶流动、超声速射流的基本结构特征,以此讨论和分析超声速后台阶流动/射流相互作用的特征,以及不同来流条件对波系结构、涡结构、剪切层、膨胀扇等的影响,尤其是来流剪切层和射流剪切层的相互作用,形成复杂的波系结构及相互干扰的流动现象.      

On Depletion of the Vortex-Stretching Term in the 3D Navier-Stokes Equations

Certain new cancellation properties in the vortex-stretching term are detected leading to new geometric criteria for preventing finite-time blow-up in the 3D Navier-Stokes equations.     

On the Global Regularity of a Helical-Decimated Version of the 3D Navier-Stokes Equations

We study the global regularity, for all time and all initial data in H ^1/2, of a recently introduced decimated version of the incompressible 3D Navier-Stokes (dNS) equations. The model is based on a projection of the dynamical evolution of Navier-Stokes (NS) equations into the subspace where helicity (the L ²-scalar product of velocity and vorticity) is sign-definite. The presence of a second (beside energy) sign-definite inviscid conserved quadratic quantity, which is equivalent to the H ^1/2-Sobolev norm, allows us to demonstrate global existence and uniqueness, of space-periodic solutions, together with continuity with respect to the initial conditions, for this decimated 3D model. This is achieved thanks to the establishment of two new estimates, for this 3D model, which show that the H ^1/2 and the time average of the square of the H ^3/2 norms of the velocity field remain finite. Such two additional bounds are known, in the spirit of the work of H. Fujita and T. Kato (Arch. Ration. Mech. Anal. 16:269–315, 1964; Rend. Semin. Mat. Univ. Padova 32:243–260, 1962), to be sufficient for showing well-posedness for the 3D NS equations. Furthermore, they are directly linked to the helicity evolution for the dNS model, and therefore with a clear physical meaning and consequences.     

The concept of minimal attractor and maximal attractors of partial differential equations of the Kuramoto-Sivashinsky type

The first part of the paper is devoted to definition of the smallest set that may be regarded as the attractor of the dissipative system. Simple properties of this set are discussed. In the second part some geometrical facts concerning attractors of the partial differential equations are revealed. They include well-known results for the two-dimensional Navier-Stokes equation and new ones for the Kuramoto-Sivashinsky and similar equations.     

三维湍流Rayleigh-Bénard热对流的高效并行直接求解方法

高Ra数Rayleigh-Bénard热对流的湍流特性研究是当前国际上的一个热门研究课题, DNS模拟计算是研究该课题的重要手段之一. 当计算规模增大而网格数巨大时计算工作难以实现, 高Ra湍流热对流的数值模拟研究面临重大挑战. 本文创建了大规模高效并行计算的三维湍流热对流直接求解方法. 采用FFT变换解耦压力泊松方程, 将其变换成沿z方向上的块三对角方程组, 并利用块三对角方程的MPI与OpenMP联立的大规模高效并行近似解求解方案, 创建了可以高效并行计算的热对流直接求解方法. 通过对该方法并行效率的验证计算, 证明新的直接求解并行计算方法具有很好的并行效率和计算时效. 三维窄方腔热对流的计算结果表明, 本文方法计算的三维热对流特性是合理的. 本文创建的可大规模高效并行计算的三维湍流热对流直接求解方法, 也很可能是关于计算流体力学不可压NS方程大规模高效并行计算在特殊情况中计算技术上的一个突破.     

Lower Bounds for an Integral Involving Fractional Laplacians and the Generalized Navier-Stokes Equations in Besov Spaces

When estimating solutions of dissipative partial differential equations in L^p-related spaces, we often need lower bounds for an integral involving the dissipative term. If the dissipative term is given by the usual Laplacian −Δ, lower bounds can be derived through integration by parts and embedding inequalities. However, when the Laplacian is replaced by the fractional Laplacian (−Δ)^α, the approach of integration by parts no longer applies. In this paper, we obtain lower bounds for the integral involving (−Δ)^α by combining pointwise inequalities for (−Δ)^α with Bernstein's inequalities for fractional derivatives. As an application of these lower bounds, we establish the existence and uniqueness of solutions to the generalized Navier-Stokes equations in Besov spaces. The generalized Navier-Stokes equations are the equations resulting from replacing −Δ in the Navier-Stokes equations by (−Δ)^α.     

Accurate numerical resolution of transients in initial-boundary value problems for the heat equation

If the initial and boundary data for a PDE do not obey an infinite set of compatibility conditions, singularities will arise in the solution at the corners of the initial time–space domain. For dissipative equations, such as the 1-D heat equation or 1-D convection–diffusion equations, the impacts of these singularities are short lived. However, they can cause a very severe loss of numerical accuracy if we are interested in transient solutions. The phenomenon has been described earlier from a theoretical standpoint. Here, we illustrate it graphically and present a simple remedy which, with only little extra cost and effort, restores full numerical accuracy.     

Local coupling of shell models leads to anomalous scaling

We consider cascade models of turbulence which are obtained by restricting the Navier-Stokes equation to local interactions. By combining the results of the method of extended self-similarity and a novel subgrid model, we investigate the inertial range fluctuations of the cascade. Significant corrections to the classical scaling exponents are found. The dynamics of our local Navier-Stokes models is described accurately by a simple set of Langevin equations proposed earlier as a model of turbulence [20]. This allows for a prediction of the intermittency exponents without adjustable parameters. Excellent agreement with numerical simulations is found.