SPH方法气-液界面边界条件研究

针对界面附近粒子光滑函数截断和非物理穿透问题,提出一种气-液界面边界条件的处理方法.当界面附近支持域出现不同材料粒子,每步计算可在支持域设置虚粒子,按照密度分配方法给虚粒子物理量赋值,并对界面附近粒子引入气-液两相阻力.采用SPH方法和Level-Set方法,计算运动激波对气-液界面作用问题,两者计算结果一致,初步验证了气-液界面边界条件处理的适用性.用SPH方法分别计算超声速气流中的圆截面液柱绕流和下落问题,界面两侧粒子压力和法向速度连续,给出弓形激波、回流区和下游回流区等定性合理结果.表明本文方法可适度避免界面附近流体粒子光滑截断和粒子非物理穿透现象、界面附近流场数值振荡.     

一种新型光滑粒子动力学固壁边界施加模型

由于Lagrange粒子法的本质, 固壁边界条件的施加一直是光滑粒子动力学方法的难点之一. 本文从固壁边界的物理原理出发, 应用多层虚粒子表征固壁边界, 提出了一种新型固壁边界施加模型. 将虚粒子看作流体的扩展, 计算中虚粒子密度保持不变, 压力、速度等参数通过对流体粒子的插值获得, 虚粒子有条件的参与控制方程的计算, 对流体的密度/压力产生影响, 通过压力梯度隐式地表征壁面与流体之间的作用强度并对流体粒子施加沿壁面法线方向的斥力作用, 防止流体粒子对壁面的穿透. 数值算例测试结果表明, 与现有固壁边界施加方法相比, 本文方法更加符合流体与固壁边界作用的物理原理, 可以简单、有效地施加固壁边界条件, 方便地应用于具有复杂几何边界的问题, 获得稳定的流场形态、规则的粒子秩序及良好的速度、压力等参量的分布.     

液滴在气固交界面变形移动问题的光滑粒子流体动力学模拟

为准确模拟液滴在气固交界面变形移动问题,对基于连续表面张力模型的表面张力光滑粒子流体动力学方法进行了改进.改进方法采用新的边界处理方式和界面法向修正方法,即将固体边界虚粒子色函数值根据液面的位置进行相应设定以保证气-液-固三相交界处流体粒子的界面法向沿接触线法线方向,引入Brackbill提出的壁面附着力边界条件处理方法,对在气-液-固三相交界处的流体粒子及部分固体边界虚粒子的界面法向进行修正,修正前后保持法向模值不变,得到了含壁面附着力边界条件的表面张力算法.模拟了受壁面附着力影响的水槽中液面的变化过程、液滴润湿壁面过程和剪切气流驱动液滴在固体表面变形脱落过程,并与流体体积函数方法进行了对比.结果表明,该方法在处理壁面附着力问题时精度较高,稳定性较好,适合处理工程中液滴在气固交界面变形移动问题.     

耗散粒子动力学处理复杂固体壁面的一种有效方法

耗散粒子动力学(dissipative particle dynamics,DPD)作为一种介观尺度拉格朗日型粒子方法,已经成功地应用于微纳米流动和生化科技的研究中.复杂固体壁面的处理和壁面边界条件的实施一直是DPD方法发展及应用的一个障碍.提出了处理复杂固体壁面的一种新的方法.复杂固体区域通过冻结随机分布并且达到平衡状态的DPD粒子代表;所冻结的DPD粒子位于临近流动区域的一个截距内;在靠近固体壁面的流动区域中设置流动反弹层,当流动DPD粒子进入此流动层后反弹回流动区域.应用这种固体壁面处理方法对简单流动区域的Poiseuille流动和复杂多孔介质内的流动进行了分析.研究表明,这种新的固体壁面处理方法能够有效模拟复杂固体区域,准确实施壁面边界条件.     

一种改进的光滑粒子流体动力学前处理方法

为了便于对任意边界形状的计算域快速地布置均匀粒子,提出了一种改进的光滑粒子流体动力学前处理方法.该方法是在2012年Colagrossi等提出的算法基础上进行改进后得到的.Colagrossi等提出的算法能够计算一些简单外形分布比较均匀的粒子.然而当光滑长度与初始粒子间距的比值较大时该方法在计算过程中会出现较强的数值震荡问题,收敛速度慢;而且在计算过程中可能会遭遇流体粒子穿透固体壁面的问题.本文通过引入未知因素修正的平滑粒子动力学模型来提高计算稳定性,并通过对边界附近的流体粒子施加边界力来避免流体粒子穿透固体壁面.算例验证结果表明,利用改进后的光滑粒子流体动力学前处理方法能够快速地对各种边界形状的计算域分布均匀粒子,并且避免了流体粒子穿透固体壁面的问题.     

高精度非定常激波装配法

为正确模拟高超声速绕流中,来流小扰动与弓形激波之间的干扰对流动特征的影响,将弓形激波作为动边界,利用非定常特征关系处理激波处的边界条件.应用五阶精度迎风紧致格式和六阶精度的对称格式与三阶精度的R-K方法相结合,建立高精度非定常激波装配方法.采用该方法数值模拟钝锥高超声速定常流场和二维抛物外形高超声速边界层流动的感受性问题,数值模拟来流小扰动与弓形激波干扰激波后非定常扰动流场,研究扰动波进入边界层产生边界层不稳定波的特征.     

一种基于二维光滑粒子法的流体仿真方法

针对大场景下流体仿真计算复杂度高的问题,本文以浅水方程为基础,提出一种改进的二维光滑粒子方法.该方法中使用光滑粒子法离散二维浅水方程,将水深作为粒子的属性,把计算复杂度降到二维的程度;同时为了提高邻域粒子的搜索效率,提出一种基于动态网格的邻近粒子搜索方法;并使用虚粒子和惩罚力相结合的方法处理边界条件以高效率应对复杂边界;渲染时,首先将粒子映射并插值到规则网格内得到流体表面,避免三维流体表面重构复杂度高的问题,最后利用OpenGL着色语言实现加速渲染,从而达到大场景下流体实时仿真.     

耗散粒子动力学中施加无滑移边界条件的新方法

在耗散粒子动力学(DPD)中施加无滑移边界条件是较困难的,本文提出了一种新的处理方法来实现无滑移边界条件.通过选取合适的壁面与流体粒子之间的保守力系数可以控制近壁面区的粒子数,从而可以使密度波动降低至极小值,实现无滑移边界条件.通过该方法模拟泊肃叶流动求得的流体性质与通过使用Lees-Edwards边界条件模拟剪切流动所求得的流体性质一致,证明了该方法的可靠性.     

Van der Waals状态方程可压缩多流体流动的PPM方法

基于流体体积分数的混合型多流体数值模型，将Piecewise Parabolic Method（PPM）方法应用于可压缩多流体流动的数值模拟，采用双波近似求解多流体van der Waals状态方程的Riemann问题．模拟高密度比且含有激波的可压缩多流体流动，典型的纯界面平移问题模拟结果表明，在接触间断的界面附近，压力和速度没有任何的振荡且界面数值耗散都被控制在2—3个网格之内；一维和二维算例表明，该数值方法可以有效地处理接触间断、激波和多维滑移线等物理问题，并能够比其它多流体数值方法更精细地模拟多流体交界面．     

三角形网格上多介质流动问题界面处理方法

多介质流动问题的求解一般是在结构网格上实现,而三角形网格对于复杂计算区域具有更好的适应性,本文结合rGFM方法,给出三角形网格上多介质流动问题界面处理方法.利用level-set方法跟踪界面,在界面处构造Riemann问题,得到界面处流体准确的流动状态.通过定义界面边界条件,将多介质流动问题转化为单介质流动问题,利用高精度RKDG方法求解.采用多个算例验证该方法的稳健性和有效性,结果表明该方法能准确捕捉界面和激波的位置,保持界面清晰.     

Analytical solution to a mode of mixed elastohydrodynamic lubrication with mixed contact regimes: Part II—Considering the contact adhering layer effect in the inlet zone

The present paper presents the researches succeeding the first part of the paper [Y.B. Zhang, Analytical Solution to A Mode of Mixed elastohydrodynamic lubrication with Mixed Contact Regimes: Part I—Without Consideration of Contact Adhering Layer in the Inlet Zone. Journal of Molecular Liquids, 2006, Vol.117, (10.1016/j.molliq.2006.04.006)], which analyzed one mode of mixed elastohydrodynamic lubrication with mixed contact regimes for the relatively heavy load and low rolling speed which make the conventional hydrodynamic lubrication occur in the inlet zone while make the physical adsorbed layer boundary lubrication occur in the Hertzian zone, based on the Newtonian fluid model. The present paper presents analysis to other two modes of mixed EHL with mixed contact regimes for relatively heavy loads, low rolling speeds and Newtonian fluids, where the conventional hydrodynamic lubrication, physical adsorbed layer boundary lubrication and oxidized chemical layer boundary lubrication can simultaneously occur in the inlet zone while the oxidized chemical layer boundary lubrication or the fresh metal-oxidized chemical boundary layer dry contact occur in the Hertzian zone, considering the contact adhering layer effect in the inlet zone. The present analysis is also extended to the first mode of mixed EHL with mixed contact regimes as analyzed in Part I [Y.B. Zhang, Analytical Solution to A Mode of Mixed elastohydrodynamic lubrication with Mixed Contact Regimes: Part I—Without Consideration of Contact Adhering Layer in the Inlet Zone. Journal of Molecular Liquids, 2006, Vol.117, (10.1016/j.molliq.2006.04.006)] when the contact adhering layer effect in the inlet zone is considered. Results of contact pressures, film thicknesses, load partitions in the contact and characteristic rolling speeds for approaching to zero averaged hydrodynamic film thickness in the Hertzian zone are obtained from this analysis respectively as functions of the contact adhering layer thickness in the inlet zone. The results show that the contact adhering layer effect in the inlet zone in the present EHL is reduced with the increase of load; At large loads, this effect may be negligible; At small loads, it may be very significant. The results also show that at low rolling speeds, when the contact adhering layer effect in the inlet zone is considered, the load-carrying capacity of the present EHL contact is increased especially for small loads. This means that at low rolling speeds the contact adhering layer effect in the inlet zone may reduce the elastohydrodynamic lubrication deviation from classical EHL theory predictions especially for small loads.     

Direct numerical simulation of transitional flow in a finite length curved pipe

Transition to turbulent flow in a curved pipe has been well studied through experiments and numerical simulations. Numerical simulations often use a helical pipe with an infinite length such that the inlet and outlet boundary conditions can be modelled as periodic which greatly reduces computational time. In this study, we examined a finite length curved pipe with Poiseuille flow imposed at the inlet and a stress-free boundary condition at the outlet. Direct numerical simulation of the Navier-Stokes equations for rigid walls and a Newtonian fluid was performed using nek5000. Straight extensions were added to the inlet and outlet such to diminish the impact of boundary conditions on the flow field in the region with curvature. The examined model has a pipe radius of curvature that is three times the pipe radius. The model has ～355 million nodes and required an order of magnitude greater computational time when compared with an infinite length curved pipe. Results show that the critical Reynolds number, the lowest value with instabilities present in the flow, is much greater than that of a straight pipe and occurs near Re=5000–5200. This is larger than the critical Reynolds number typically reported for an infinite length curved pipe (Re=4200–4300).     

Numerical investigation of supersonic flow breakdown at the inlet duct throttling

The work presents the results of investigating the process of supersonic flow deceleration in a duct of the two-dimensional inlet throttled by variation of the outlet cross-sectional area. An inlet with three external compression shock waves designed for the freestream Mach number M_d = 7 was considered as an example for the investigation. A one-dimensional analysis of the conditions for realization of the supersonic flow deceleration regimes in the inlet duct with two throats — in the inlet entrance and at the inlet duct outlet, has been carried out. The parametric numerical computations of two-dimensional inviscid or turbulent flows in the inlet were performed with the use of the Euler and Navier—Stokes codes of the program package FLUENT. The critical conditions for the nonuniform flow in the outlet throat bringing to choking the inlet duct were determined.     

无缓冲气COIL扩压器流场数值模拟

 以无缓冲气化学氧碘激光器(COIL)实验器件的数据计算得到的混合喷管出口参数平均值作为光腔入口条件，对6种不同构型的扩压器从光腔入口至扩压器出口的流场进行了数值模拟，得出了各流场参数分布；对不同构型扩压器的流场特点、总压恢复性能进行了分析；研究了扩压器出口背压对流场参数的影响。结果表明：对于主流无缓冲气的COIL，等截面扩压器具有较好的压力恢复性能；增大扩压器出口背压可以使扩压器的压力恢复性能提高，然而，较高的背压使激波串向光腔方向移动，从而使光腔流场受到干扰，影响光腔的光束质量。     

Investigation of annular supersonic inlets with isentropic compression

A technique for designing the supersonic annular inlets with isentropic deceleration surfaces is considered. The contour of an isentropic supersonic nozzle constructed by the method of characteristics for an inviscid gas flow with given uniform parameters at the inlet and at the outlet is used as the basic configuration of the inlet. The reversed flow of a viscous gas is computed with the aid of numerical techniques in the contour under consideration and the real operational characteristics of the obtained inlet of a fixed geometry are determined in the range of the conditions of its application. In the process of computations, the minimum cross-sectional sizes are selected, which ensure the inlet start without a detached bow shock at the entrance.     

氩等离子体射流在空气环境中冲击平板层流流动与传热的数值模拟

本文采用组合扩散系数方法处理不同气体组分之间的扩散，对氩等离子体的流射入空气环境并撞击平板时的层流流动和传热进行了数值模拟．这种新的处理混合气体中质量扩散的方法有助于更准确地描述等离子体条件下的组分扩散与能量输运。文中给出了射流中速度、温度及氩质量分数的分布情况，以及基板处热流密度分布的若干典型的数值模拟结果．     

激波与层流/湍流边界层相互作用实验研究

在超声速风洞中,分别对层流和湍流来流条件下的边界层和斜激波(激波强度足以引起流动分离)相互干扰进行了实验研究,利用纳米粒子示踪平面激光散射(NPLS)技术获得了两种条件下流场的精细结构图像;利用粒子图像测速(PIV)技术获得了两种条件下流场的速度场和涡量场;综合运用NPLS结果和PIV结果对比分析了两种流动的瞬时流动结构和时间相关性,实验结果表明:层流边界层内的分离区呈现出狭长的条状,而湍流边界层内分离区呈现出较规则的椭圆;在入射激波上游距入射点较远的位置,层流边界层外围拟序结构会诱导出一系列压缩波系,进而汇聚成空间位置不稳定的诱导激波,而湍流边界层则是在入射激波上游较近的地方直接形成较强且稳定的诱导激波;在入射激波下游,层流边界层内的膨胀区域较小且急促,膨胀后产生的再附激波很弱,而湍流边界层内的膨胀区域较大,膨胀后产生的激波较强。     

Characteristics of a magnetogasdynamic diffuser under different modes of electric current switching

This paper elaborates upon a previous investigation into the influence of external electric and magnetic fields on a flow through a supersonic diffuser. The aim of the present study is to correlate a change in the configuration of a shock wave emerging near the diffuser inlet at magnetohydrodynamic interaction with the amount of force and energy actions and with total pressure losses. For this purpose, the main parameters of the shock wave structure and the total pressure are measured at the diffuser outlet when the flow is subjected to magnetic and electric fields of various strengths at different routes of current passage. In the experiments, a shock tube with a supersonic nozzle is employed. The shock tube forms a flow behind the shock wave reflecting from the end of the tube, which terminates in the nozzle. The diffuser is located directly downstream of the nozzle. The investigation is carried out in xenon. The flow is subjected to external fields at the inlet of the diffuser. The shock wave structure is visualized by frame sweeping of Schlieren patterns of the flow. The total pressure is measured with a piezoelectric transducer located at the end of the channel. The results obtained make it possible to optimize the action on the flow in terms of power consumption and total pressure losses for a given design of the diffuser.     

嵌套网格技术中的Collar网格和虚拟网格方法

针对嵌套网格技术中的结合部问题,提出了Collar网格和虚拟网格方法.用双曲型微分方程生成的处于结合部的有两个边界分别处于不同固定曲面上的Collar网格,在保证计算网格的生成方便快捷而且网格质量高的前提下,解决了为物体结合部的内外边界点提供有效插值单元的问题.虚拟网格为紧贴物面的面网格,它的作用是将物面边界条件传递给其它网格的边界面,而其本身不作流场计算.计算实践表明,将Collar网格和虚拟网格结合起来应用在嵌套网格技术中能保证几何外形不发生变化,有效地处理各种复杂组合体外形的结合部问题.