Navier-Stokes方程二阶速度滑移边界条件的检验

对微尺度气体流动,Navier-Stokes方程和一阶速度滑移边界条件的结果与实验数据相比,在滑移区相互符合,在过渡领域则显著偏离.为改善Navier-Stokes方程在过渡领域的表现,有些研究者尝试引入二阶速度滑移边界条件,如Cercignani模型,Deissler模型和Beskok-Karniadakis模型.以微槽道气体流动为例,将Navier-Stokes方程在不同的二阶速度滑移模型下的结果与动理论的直接模拟Monte Carlo(DSMC)方法和信息保存(IP)方法以及实验数据进行比较.在所考察的3种具有代表性的二阶速度滑移模型中,Cercignani模型表现最好,其所给出的质量流率在Knudsen数为0.4时仍与DSMC和IP结果相符;然而,细致比较表明,Cercignani模型给出的物面滑移速度及其附近的速度分布在滑流区和过渡领域的分界处(Kn=0.1)已明显偏离DSMC和IP的结果.     

微尺度气体流动

了解微尺度气体流动特点是微机电系统设计和优化的基础.有关的研究可以上溯到20世纪初Knudsen的平面槽道流动质量流量的测量和Millikan的小球阻力系数的测量,实验结果揭示了稀薄气体效应即尺度效应对气体运动的重要影响.由于流动特征长度很小,微尺度气流经常处于滑流区甚至过渡领域,流动的相似参数为Knudsen数和Mach数.因此可以考虑利用相似准则,通过增大几何尺寸、减小压力的途径,解决微机电系统实验观测遇到的困难.为解决直接模拟Monte Carlo方法分析微机电系统中低速稀薄气流遇到的统计涨落困难,我们提出了信息保存法(IP),该方法能够有效克服统计散布,并已成功用于多种微尺度气流.      

气体运动论数值算法在微槽道流中的应用研究

简要介绍基于Boltzmann模型方程的气体运动论数值算法基本思想及其对二维微槽道流动问题数值计算的推广，并阐述适用于微尺度流动问题的气体运动论边界条件数值处理方法。通过对压力驱动的二维微槽道流动问题进行数值模拟，将不同Knudsen数下的微槽道流计算结果分别与有关DSMC模拟值和经滑移流理论修正的N—S方程解进行比较分析，表明基于Boltzmann模型方程的气体运动论数值算法对微槽道气体流动问题具有很好的模拟能力。     

低速二维流动的粒子仿真研究

本文运用信息保存法对低速二维的流动现象进行模拟，考察了低速条件下的有限平板绕流以及微槽道气体流动问题。研究表明：在对低速流动的模拟过程中，运用IP法在能够获得较好的结果的同时，具有比DSMC方法更高的计算效率。     

基于Boltzmann模型方程的气体运动论统一算法研究

模型方程出发,研究确立含流态控制参数可描述不同流域气体流动特征的气体分子速度分布函数方程; 研究发展气体运动论离散速度坐标法, 借助非定常时间分裂数值计算方法和NND差分格式, 结合DSMC方法关于分子运动与碰撞去耦技术, 发展直接求解速度分布函数的气体运动论耦合迭代数值格式; 研制可用于物理空间各点宏观流动取矩的离散速度数值积分方法, 由此提出一套能有效模拟稀薄流到连续流不同流域气体流动问题统一算法. 通过对不同Knudsen数下一维激波内流动、二维圆柱、三维球体绕流数值计算表明, 计算结果与有关实验数据及其它途径研究结果(如DSMC模拟值、N-S数值解)吻合较好, 证实气体运动论统一算法求解各流域气体流动问题的可行性. 尝试将统一算法进行HPF并行化程序设计, 基于对球体绕流及类``神舟'返回舱外形绕流问题进行HPF初步并行试算, 显示出统一算法具有很好的并行可扩展性, 可望建立起新型的能有效模拟各流域飞行器绕流HPF并行算法研究方向. 通过将气体运动论统一算法推广应用于微槽道流动计算研究, 已初步发展起可靠模拟二维短微槽道流动数值算法; 通过对Couette流、Poiseuille流、压力驱动的二维短槽道流数值模拟, 证实该算法对微槽道气体流动问题具有较强的模拟能力, 可望发展起基于Boltzmann模型方程能可靠模拟MEMS微流动问题气体运动论数值计算方法研究途径.      

MEMS稀薄气体内部流动模拟中的信息保存法

首先综述了处理低速稀薄气体流动的一些方法: 线化Boltzmann方程方法、Lattice Boltzmann方法(LBM)、加滑移边界的Navier-Stokes方程、以及DSMC方法, 并讨论它们在模拟MEMS中过渡领域低速流动特别是内部流动所遇到的困难, 其中表明了LBM现有方案不适合模拟过渡领域中的MEMS流动问题. 信息保存(IP)法通过保存一个模拟分子所代表的大量分子的平均信息,克服了流速低使得信息噪声比小而引起统计模拟的困难. 本文给出了方法的一些理论证实. MEMS中内部流动的特点, 即流速低和大的长宽比的特点, 引起椭圆性问题, 即出入口边界条件相互影响需要协调的问题. 通过对(长约几千微米的)微槽道流动应用IP方法的算例,演示了采用守恒形式的质量守恒方程和超松弛法可成功地解决这一问题. 借助同样的方法,用IP方法求解了真实长度(1\,000\,$\mu$m)硬盘驱动器读写头在过渡领域的薄膜支撑问题, 压力分布与具有严格气体动理论基础的概括化Reynolds方程完全相符, 而在此之前, DSMC方法只对短的读写头(5\,$\mu$m)与Reynolds方程做了校验. 作者建议将原来用于求解读写头润滑问题的Reynolds方程退化来求解过渡领域中的微槽道流动问题, 从而提供了一个有严格气体动理论品性的检验方法来验证求解MEMS内部流动的各种方法.      

开式空腔流动的蒙特卡罗直接模拟

蒙特卡罗直接模拟(Direct Simulation Monte-Carlo)方法是一种基于分子动力论的随机性数值模拟方法，它在各类稀薄气体流动模拟中得到了广泛应用。本文首先讨论了DSMC方法的基本原理，引用文献结果论述了DSMC方法与Boltzmann方程的内在一致性，采用DSMC方法从分子运动论层次对过渡区开式空腔流动进行精细模拟，再现空腔内旋涡的形成及发展过程，了解复杂流场的流场特性。模拟和分析了空腔的形状(展弦比)、壁温以及Knudsen数(稀薄性)等因素对空腔内流动和旋涡结构的影响。研究表明，空腔的展弦比、壁面温度以及流动Knudsen数等因素对空腔内旋涡的大小、形状、位置、个数都有极大的影响，应用DSMC方法可以真实再现稀薄条件下的开式空腔流动。     

气体化学反应流动的DSMC／EPSM混合算法研究

发展了平衡粒子模拟方法(EPSM)，建立了与高温气体化学反应动力学理论相匹配的：EPSM耦合模型，并通过混合参数进行流区的自动识别，将：EPSM方法与蒙特卡罗直接模拟方法(OSMC)结合，构造了可模拟化学反应流动的DSMC／EPSM混合算法。应用该算法对汲及化学反应的二维高超音速竖板绕流流场进行模拟，将结果与DSMC方法的结果进行比较，验证了新算法对求解化学反应流动的可行性。将混合算法的计算效率与DSMC方法的计算效率进行比较，发现混合算法能够大大提高计算效率。     

微尺度泊肃叶流的高阶连续模型分析

分别从分子运动论及连续流理论出发,对体积力驱动的微尺度平面泊肃叶(Poiseuille)流的横向分布特征进行了分析. 分子水平模拟采用直接模拟蒙特卡罗(direct simulation Monte Carlo, DSMC)方法;连续流理论则主要考察了伯内特(Burnett)及超伯内特(Super-Burnett)等高阶连续模型,在平行流假设下,获得一组高阶非线性常微分方程,补充完整的边界条件,并应用龙格-库塔(Runge-Kutta)方法求解. 结果表明,即使对于过渡领域流动,高阶连续模型可以给出与DSMC 结果完全相符的压力分布,而速度分布当努森(Knudsen)数约为0.2时即在壁面开始出现偏差;对于温度的横向分布,伯内特模型回复到纳维-斯托克斯(Navier-Stokes)水平,不能得到与DSMC一致的双峰结构,而超伯内特模型在滑移流动领域与DSMC定性相符,在过渡领域却仅能正确预测主流区温度分布,壁面附近差异明显;横向热流与纳维-斯托克斯模型预测接近,但机理上存在本质区别. 本文结果提示选用连续模型时,不仅要根据流动参数来判断,还可以根据所关注的物理量来进行调整,适度扩大连续模型的适用范围. 但即使采用高阶本构关系,连续模型仍然不能完全描述壁面附近区域的非平衡效应(如努森层效应),这是试图扩大连续模型适用范围时必然会遇到的困难.     

微喷管流动粒子仿真与Navier-Stokes数值模拟比较研究

运用DSMC（Direct Simulation Monte—Carlo）方法从分子运动论层次对大膨胀比、喉部转角为尖角的微喷管流动现象进行模拟,考察来流总压对喷管性能的影响,并与Navier—Stokes方程运算结果、实验结果进行比较。研究表明：在模拟微型喷管的流动现象时,DSMC方法比N—S方程更加适用。     

EFFECTS OF LID-DRIVEN CAVITY SHAPE ON THE FLOW ESTABLISHMENT PHASE

Experiments are carried out to study the flow establishment phase inside closed cavities submitted to the impulsive translation from rest, of one of their walls at a Reynolds number of 1000. Three standard industrially machined or molded cylindrical cavity shapes are studied and are compared with respect to the efficiency of mixing process: square, rectangular and semi-circular of length-to-width ratio of 2:1. The flow structures in the mid-cross-section are analysed by means of fine topological and kinematic visualization series using two complementary techniques: continuous dye filament and discrete solid tracers both coupled with a laser sheet illumination. Particular attention is given to vorticity propagation and primary/secondary eddy formations. Although a roughly similar vortex generation is observed in all examined cavities, important differences appear with time. The semi-circular cavity flow results in a much more homogeneous and uniform recirculation with no secondary flow recirculation zone. On the contrary, the square and rectangular cavity flows develop a better flow mass dispersion and, respectively, one and two secondary eddies. At the final time of observation (t^*=12), both semi-circular and rectangular cavity flows seem to reach their steady state whereas the square one continues to evolve. Comparisons with 2-D computational results of other authors illustrate the three-dimensional flow aspect present in experiments.     

Mass flow rate prediction of pressure–temperature-driven gas flows through micro/nanoscale channels

In this paper, we study mass flow rate of rarefied gas flow through micro/nanoscale channels under simultaneous thermal and pressure gradients using the direct simulation Monte Carlo (DSMC) method. We first compare our DSMC solutions for mass flow rate of pure temperature-driven flow with those of Boltzmann-Krook-Walender equation and Bhatnagar-Gross-Krook solutions. Then, we focus on pressure–temperature-driven flows. The effects of different parameters such as flow rarefaction, channel pressure ratio, wall temperature gradient and flow bulk temperature on the thermal mass flow rate of the pressure–temperature-driven flow are examined. Based on our analysis, we propose a correlated relation that expresses normalized mass flow rate increment due to thermal creep as a function of flow rarefaction, normalized wall temperature gradient and pressure ratio over a wide range of Knudsen number. We examine our predictive relation by simulation of pressure-driven flows under uniform wall heat flux (UWH) boundary condition. Walls under UWH condition have non-uniform temperature distribution, that is, thermal creep effects exist. Our investigation shows that developed analytical relation could predict mass flow rate of rarefied pressure-driven gas flows under UWH condition at early transition regime, that is, up to Knudsen numbers of 0.5.     

An explicit transient algorithm for predicting incompressible viscous flows in Arbitrary Geometry

A numerical method for predicting viscous flows in complex geometries has been presented. Integral mass and momentum conservation equations are deploved and these are discretized into algebraic form through numerical quadrature. The physical domain is divided into a number of non-orthogonal control volumes which are isoparametrically mapped on to standard rectangular cells. Numerical integration for unsteady mementum equations is performed over such non-orthogonal cells. The explicitly advanced velocity components obtained from unsteady momentum equations may not necessarily satisfy the mass conservation condition in each cell. Compliance of the mass conservation equation and the consequent evolution of correct pressure distribution are accomplished through an iterative correction of pressure and velocity till divergence-free condition is obtained in each cell. The algorithm is applied on a few test problems, namely, lid-driven square and oblique cavities, developing flow in a rectangular channel and flow over square and circular cylinders placed in rectangular channels. The results exhibit good accuracy and justify the applicability of the algorithm. This Explicit Transient Algorithm for Flows in Arbitrary Geometry is given a generic name EXTRAFLAG.     

On drag, Strouhal number and vortex-street structure

A phenomenological model for the vortex-shedding process behind bluff cylindrical bodies is proposed. Relationships between Strouhal frequency St, drag coefficient c_D, Reynolds number Re and geometric wake parameters are obtained from mass conservation, momentum conservation in the transverse direction and energy considerations. For the first time, Roshko's (Technical Report TN 3169, NACA, US Government Printing Office, Washington DC, 1954) experimental discovery of vortex-street similarity behind different cylinders is analytically derived. In addition, the empirically obtained Strouhal-frequency laws of Roshko (Technical Report TN1191, NACA, US Government Printing Office, Washington DC, 1954) and Fey (Phys. fluids A 10 (1998) 1547) are also reproduced. Measurements of St and c_D including their Re dependency for flows around cylinders with circular, square, triangular, semi-circular and other cross sections agree favorably with the proposed model.     

DSMC／EPSM混合算法研究

将EPSM算法与DSMC方法结合，构造了可模拟含近连续流区及过渡流区的DSMC/EPSM混合算法。运用混合算法模拟了马赫数等于5时超音速竖板绕流及马赫数等于4时超音速平板绕流，并将结果与DSMC算法的结果进行比较，证明了DSMC／EPSM混合算法的有效性，同时将EPSM算法与DSMC算法的效率进行了比较。     

A novel efficient hybrid DSMC–dynamic collision limiter algorithm for multiscale transitional flows

A new 2D parallel multispecies polyatomic particle–based hybrid flow solver is developed by coupling the Direct Simulation Monte Carlo (DSMC) method with a novel Dynamic Collision Limiter (DCL) approach to solve multiscale transitional flows. The hybrid DSMC‐DCL solver can solve nonequilibrium multiscale flows with length scales ranging from continuum to rarefied. The DCL method, developed in this work, dynamically assigns different number of collisions in cells, which is based on the local value of K‐S parameter such that the number of collisions per time step is limited in near‐equilibrium flow regions. Present hybrid solver uses the Kolmogorov‐Smirnov statistical test as the continuum breakdown parameter, based on which, the solution domain is decomposed into near‐equilibrium and nonequilibrium flow regions. Direct Simulation Monte Carlo is used where nonequilibrium flow regions are encountered, while the DCL method is used where flow regions are found to be in near‐equilibrium state. In this work, we have studied hypersonic flow of nitrogen over a blunt body with an aerospike and supersonic flow of argon through a micronozzle. The results obtained by the hybrid DSMC‐DCL solver are compared and shown to agree well with the experimental data and with those obtained from DSMC, with significant savings in the computational cost.     

Lattice Boltzmann simulations of viscoplastic fluid flows through complex flow channels

We present the results of lattice Boltzmann (LB) simulations for the planar-flow of viscoplastic fluids through complex flow channels. In this study, the Bingham and Casson model fluids are covered as viscoplastic fluid. The Papanastasiou (modified Bingham) model and the modified Casson model are employed in our LB simulations. The Bingham number is an essential physical parameter when considering viscoplastic fluid flows and the modified Bingham number is proposed for modified viscoplastic models. When the value of the modified Bingham number agrees with that of the “normal” Bingham number, viscoplastic fluid flows formulated by modified viscoplastic models strictly reproduce the flow behavior of the ideal viscoplastic fluids. LB simulations are extensively performed for viscoplastic fluid flows through complex flow channels with rectangular and circular obstacles. It is shown that the LB method (LBM) allows us to successfully compute the flow behavior of viscoplastic fluids in various complicated-flow channels with rectangular and circular obstacles. For even low Re and high Bn numbers corresponding to plastic-property dominant condition, it is clearly manifested that the viscosity for both the viscoplastic fluids is largely decreased around solid obstacles. Also, it is shown that the viscosity profile is quite different between both the viscoplastic fluids due to the inherent nature of the models. The viscosity of the Bingham fluid sharply drops down close to the plastic viscosity, whereas the viscosity of the Casson fluid does not rapidly fall. From this study, it is demonstrated that the LBM can be also an effective methodology for computing viscoplastic fluid flows through complex channels including circular obstacles.     

Study of heat transfer of a helium–xenon mixture in heated channels with different cross-sectional shapes

This paper describes the results of an experimental study of heat transfer in the case of the flow of a helium–xenon mixture with a Prandtl number approximately equal to 0.23 and the flows of pure helium and air in heated tubes of circular or triangular cross sections with a constant density of the heat flow. The region of thermal stability is studied. The law of heat transfer on the stabilized region is compared with known relationships. The approach that helps obtaining an expression for the calculation of heat transfer in heat transfer devices with circular and triangular cross sections, which operate in a mixture heating mode on the initial region, is developed.