Direct numerical test of the B-G-K model equation by the DSMC method

In the present paper the rarefied gas flow caused by the sudden change of the wall temperature and the Rayleigh problem are simulated by the DSMC method which has been validated by experiments both in global flow field and velocity distribution function level. The comparison of the simulated results with the accurate numerical solution of the B-G-K model equation shows that near equilibrium the B-G-K equation with corrected collision frequency can give accurate result but as farther away from equilibrium the B-G-K equation is not accurate. This is for the first time that the error caused by the B-G-K model equation has been revealed. The project supported by the National Natural Science Foundation of China (19772059, 19889209)     

二维自适应非结构网格DSMC并行算法研究

研究了二维自适应非结构网格DSMC并行算法实现的过程.首先提出了一类非结构网格自适应策略,有效降低了网格尺度对计算结果的影响,提高了流场的分辨率;然后基于PC-CLUSTER群机并行体系结构与消息传递库MPI并行环境,利用分区并行思想,设计了非结构网格DSMC并行算法,节约了计算时间.利用For-tran90的动态分配内存技术编制了通用计算程序;最后对过渡流域高超声绕流进行了数值模拟,计算结果初步验证了算法的可行性与有效性.     

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。     

Simulations for gas flows in microgeometries using the direct simulation Monte Carlo method

Micro gas flows are often encountered in MEMS devices and classical CFD could not accurately predict the flow and thermal behavior due to the high Knudsen number. Therefore, the gas flow in microgeometries was investigated using the direct simulation Monte Carlo (DSMC) method. New treatments for boundary conditions are verified by simulations of micro-Poiseuille flow, compared with the previous boundary treatments, and slip analytical solutions of the continuum theory. The orifice flow and the corner flow in microchannels are simulated using the modified DSMC codes. The predictions were compared with existing experimental phenomena as well as predictions using continuum theory. The results showed that the channel geometry significantly affects the microgas flow. In the orifice flow, the flow separation occurred at very small Reynolds numbers. In the corner flow, no flow separation occurred even with a high driving pressure. The DSMC results agreed well with existing experimental information.     

三维非结构网格DSMC方法的实现及其应用

研究了三维非结构网格DSMC方法实现的过程。将Bird位置元方案中的子网格思想引入到非结构网格上来,只存储子网格的总体标识号,利用较少的计算网格提高了分子的分辨率与计算精度。提出了将体积元坐标搜索算法与交替数字二叉树搜索算法(ADT)相结合的方法来跟踪模拟分子在网格之间的迁移,使用ADT方法判别分子与物面是否作用,避免了分子表面反射的非确定论判据。利用Fortran 90的动态分配内存技术编制了通用计算程序。最后对高超声速过渡流域航天飞机头部外形绕流进行了数值模拟,数值结果初步验证了算法的可行性。     

Rayleigh-Benard系统不稳定性的DSMC方法研究

本文应用DSMC方法初步研究了稀薄气体状态下Rayleigh-Benard系统的不稳定性特征，着重考察了气体密度梯度对Rayleihg-Benard系统不稳定性特征的影响。研究表明DSMC方法能够成功地模拟Rayleigh-Benard系统内涡的运动，得到了与外力方向相反的密度梯度会增加Rayleigh-Benard系统流动不稳定性的结果。     

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。     

Unified macro-to-microscale method to predict two-phase frictional pressure drops of annular flows

The study considers the prediction of pressure gradients in adiabatic gas–liquid annular two-phase flow in the macro-to-microscale range. Twenty-four empirical correlations have been tested against an experimental data bank drawn together in this study containing 3908 points for eight different gas–liquid combinations and 22 different tube diameters, covering microscale and macroscale channels from 0.517 to 31.7 mm in diameter. The correlations of Lombardi, Friedel and Baroczy-Chisholm were found to be the best existing methods when considering macroscale data only, while the microscale database was best predicted by the correlations of Lombardi, Müller-Steinhagen and Heck and the homogeneous model with the two-phase viscosity defined according to Cicchitti. A new correlating approach based on the vapor core Weber number, capable of providing physical insight into the flow, was proposed and worked better than any of the existing methods for the macroscale database. This new macroscale method was then extended to cover microscale conditions, resulting in one unified method for predicting annular flows from the macroscale to the microscale covering both laminar and turbulent liquid films. The macroscale method optimized for microchannels worked better than any of the other methods considered.     

一种分子运动和碰撞双重网格DSMC方法研究

基于非结构网格与直角网格相结合的策略,提出一种分子运动和碰撞双重网格DSMC实现方法。通过分子的位置坐标信息,实现运动网格与碰撞网格信息交互,从而结合非结构网格的贴体性和直角网格易于控制子网格数量的优点,同时提出并实现了基于MCS自适应的动态子网格技术以有效限制分子平均碰撞距离,提高了DSMC方法的通用性和计算精度。通过对超声速圆柱绕流和扩张管道的数值模拟,验证该方法的有效性和高精度性。数值结果表明,本文提出的基于分子运动和碰撞双重网格的DSMC方法提高了流场的分辨率,且通过基于MCS自适应的动态子网格技术,有效地降低了流场的平均碰撞距离,提高了碰撞质量和计算精度。     

DSMC方法与稀薄气流计算的发展

直接模拟ＭｏｎｔｅＣａｒｌｏ（ＤＳＭＣ）方法是依赖物理的概率模拟方法．在求解过渡领域流动的众多解析、数值和模拟方法中,目前只有ＤＳＭＣ方法是可以模拟三维复杂真实气体流动的方法．从宏观参量到细观速度分布函数的水平上,该方法均能得到实验的支持．本文综述了ＤＳＭＣ方法的最近发展,包括我们在检验模型、处理内能松弛模拟和化学反应模拟以及求解三维绕流的通用方法等方面的工作。      

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.     

A parallel cell‐based DSMC method on unstructured adaptive meshes

A parallel DSMC method based on a cell‐based data structure is developed for the efficient simulation of rarefied gas flows on PC‐clusters. Parallel computation is made by decomposing the computational domain into several subdomains. Dynamic load balancing between processors is achieved based on the number of simulation particles and the number of cells allocated in each subdomain. Adjustment of cell size is also made through mesh adaptation for the improvement of solution accuracy and the efficient usage of meshes. Applications were made for a two‐dimensional supersonic leading‐edge flow, the axi‐symmetric Rothe's nozzle, and the open hollow cylinder flare flow for validation. It was found that the present method is an efficient tool for the simulation of rarefied gas flows on PC‐based parallel machines. Copyright © 2004 John Wiley & Sons, Ltd.     

A hybrid DSMC/Navier–Stokes frame to solve mixed rarefied/nonrarefied hypersonic flows over nano‐plate and micro‐cylinder

We extend a hybrid DSMC/Navier–Stokes (NS) approach to unify the DSMC and the NS simulators in one framework capable of solving the mixed non‐equilibrium and near‐equilibrium flow regions efficiently. Furthermore, we use a one‐way state‐based coupling (Dirichlet–Dirichlet boundary‐condition coupling) to transfer the required information from the continuum region to the rarefied one. The current hybrid DSMC–NS frame is applied to the hypersonic flows over nanoflat plate and microcylinder cases. The achieved solutions are compared with the pure DSMC and NS solutions. The results show that the current hybrid approach predicts the surface heat transfer rate and shear stress magnitudes very accurately. Some important conclusions can be drawn from this study. For example, although the shock wave region would be a non‐equilibrium region, it is not necessary to use a pure DSMC simulator to solve it entirely. This is important when the researchers wish to predict the surface properties such as velocity slip, temperature jump, wall heat flux rate, and friction drag magnitudes accurately. Our investigation showed that our hybrid solution time would be at least 40% (for the flat plate) and 35% (for the cylinder) of the time that must be spent by a pure DSMC solver to attain the same accuracy.Copyright © 2013 John Wiley & Sons, Ltd.     

DSMC approach to nonstationary Mach reflection of strong incoming shock waves using a smoothing technique

The direct simulation Monte Carlo (DSMC) method is applied to simulation of nonstationary Mach reflection of strong shock waves. Normally the DSMC method is very time consuming in solving unsteady flow field problems especially for high Mach numbers, because of the necessity of iterative calculations to eliminate the inherent statistical fluctuation caused by a finite sample size. A central weighted smoothing technique is introduced to process the DSMC results, so that the iteration time can be significantly reduced. In spite of some relaxations of the shock wave structure, the smoothing technique is verified to be useful to estima te the flow fields qualitatively and even quantitatively by using a relatively small sample size. The comparison between the present approach and the kineticmodel approach (Xu et al. 1991a, 1991b) on the application to unsteady rarefied flow fields was also carried out.This article was processed using Springer-Verlag T_EX Shock Waves macro package 1.0 and the AMS fonts, developed by the American Mathematical Society.     

Local pressure preconditioning method for steady incompressible flows

The convergence and accuracy characteristics of the preconditioned incompressible Euler and Navier–Stokes equations are studied. An object-oriented C++ numerical code has been developed for solving the inviscid and viscous, steady, incompressible flows problems. The code is based on the cell-centred finite volume method. In this scheme, two-dimensional incompressible Euler and Navier–Stokes equations are modified by a robust artificial compressibility (AC) and a local preconditioning matrix of pressure-sensor type. The preconditioned equations are solved with the Jameson's numerical approach, i.e. artificial dissipation and artificial viscosity terms under the form of a fourth- and second-order derivative, respectively. An explicit four-stage Runge–Kutta integration algorithm is applied to obtain the steady-state condition. The computed results include the steady-state solution of flow past the NACA-hydrofoils and a circular cylinder in free stream, for which the numerical results are compared with numerical works of other researchers. Good agreement is observed. The effects of AC parameter, artificial viscosity and dissipation factor, and local preconditioning coefficient on convergence rate and solution accuracy are tested by computing flow over the NACA0012 hydrofoil. In addition, some important design criteria of a preconditioner, such as stiffness reduction, hyperbolicity, symmetrisability, accuracy preservation for M → 0, and M-property have been examined analytically.     

A Helmholtz pressure equation method for the calculation of unsteady incompressibl eviscous flows

A time-implicit numerical method for solving unsteady incompressible viscous flow problems is introduced. The method is based on introducing intermediate compressibility into a projection scheme to obtain a Helmholtz equation for a pressure-type variable. The intermediate compressibility increases the diagonal dominance of the discretized pressure equation so that the Helmholtz pressure equation is relatively easy to solve numerically. The Helmholtz pressure equation provides an iterative method for satisfying the continuity equation for time-implicit Navier–Stokes algorithms. An iterative scheme is used to simultaneously satisfy, within a given tolerance, the velocity divergence-free condition and momentum equations at each time step. Collocated primitive variables on a non-staggered finite difference mesh are used. The method is applied to an unsteady Taylor problem and unsteady laminar flow past a circular cylinder.     

An experimental investigation of roll waves in high pressure two-phase inclined pipe flows

This paper examines the effects of small upward inclinations on the formation of roll waves and the properties of fully developed roll waves at high pressure conditions. A total of 984 experiments were conducted at six positive pipe inclinations θ = 0.00°, 0.10°, 0.25°, 1.00°, 2.50° and 5.00° using a 25 m long 10 cm i.d. pipe. Sulfur hexafluoride (SF₆) was used at 8 bara giving a gas density of 50 kg/m³. Two independent mechanisms for the formation of roll waves were identified; (1) interaction between 2D shallow water waves and (2) a visible long wavelength instability of the stratified layer. Viscous long wavelength linear stability analysis predicted the critical liquid flow rate and liquid height for the initiation of roll waves when roll waves were formed due to the second mechanism. A simple equation from shallow water wave theory agreed with measurements for critical liquid flow rate when roll waves were formed due to the first mechanism. Shallow water wave speed agreed with critical wave speeds at transition and nonlinear wave speeds for fully developed roll waves in certain cases. The increase in interfacial friction due to the presence of large waves was compared with models from the literature.     

A random simulation of droplet distribution in nozzle and plume flows

A simple entrainment model is used to estimate droplet streamlines, velocity and mass flux in rocket exhaust plumes. Since droplet mass flux constitutes only about 1% of the exhaust mass flux, the effect of droplet entrainment on the gas flow is neglected. The novelty of the present model is in obtaining the droplet distribution within the nozzle by assuming a small radial random velocity component for droplets at the throat. Gas flow in the nozzle is approximated as isentropic plus a correction for the boundary layer. The computed distribution of droplet mass flux is found to be in good agreement with experimental data. Received 15 January 1996 / Accepted 11 September 1996     

Extension of the local domain-free discretization method to large eddy simulation of turbulent flows

In this work, an immersed boundary method, called the local domain-free discretization (DFD) method, is extended to large eddy simulation (LES) of turbulent flows. The discrete form of partial differential equations at an interior node may involve some nodes outside the solution domain. The flow variables at these exterior dependent nodes are evaluated via linear extrapolation along the direction normal to the wall. To alleviate the requirement of mesh resolution in the near-wall region, a wall model based on the turbulence boundary layer equations is introduced. The wall shear stress yielded by the wall model and the no-penetration condition are enforced at the immersed boundary to evaluate the velocity components at an exterior dependent node. For turbulence closure, a dynamic subgrid scale (SGS) model is adopted and the Lagrangian averaging procedure is used to compute the model coefficient. The SGS eddy viscosity at an exterior dependent node is set to be equal to that at the outer layer. To maintain the mass conservation near the immersed boundary, a mass source/sink term is added into the continuity equation. Numerical experiments on relatively coarse meshes with stationary or moving solid boundaries have been conducted to verify the ability of the present LES-DFD method. The predicted results agree well with the published experimental or numerical data.     

高速流场气动光学RANS/DSMC混合数值模拟算法研究

提出了适用于高速流场气动光学数值模拟的RANS/DSMC混合算法.通过RANS对全局时均流场进行数值模拟,再对其中局部流场的脉动量采用DSMC进行数值模拟,以Maxwellian速度分布实现RANS宏观量信息向DSMC微观量信息的传递.采用超声速环境下尖劈模型对该混合算法进行校验,对比试验结果论证了算法的正确性.在计算...