气固鼓泡床的全尺度直接数值模拟

颗粒全尺度直接数值模拟由于不引入任何相间作用力封闭模型、同时又考虑了颗粒相与流体相之间的四向耦合,因而具有极高的计算精度和准度。本文采用内嵌边界多重直接力算法结合软球碰撞模型对实验室尺度的气固鼓泡流化床装置进行了全尺度直接数值模拟。计算结果真实还原了鼓泡流态化的流动情形,且能有效地捕捉装置内部流场的详细运动情况和涡结构。对流化床内颗粒在传统的离散元模型框架下进行曳力统计发现,平均曳力比全尺度模拟结果小约20%~30%。其具体数值因选取比较的曳力模型和统计网格而异。改进传统曳力模型需考虑颗粒群的非均匀性以及颗粒拟温度。     

中温循环流化床烟气脱硫基础-曳力模型与流动的数值研究

为了准确模拟中温烟气脱硫的循环流化床(CFB)反应器内的脱硫过程,本文首先采用欧拉双流体方法研究了床内的稠密气固两相流动,讨论了表征气固两相相互作用的关键参数-曳力模型,并对适于均匀流动的经典曳力模型进行了颗粒团聚所致的非均匀效应的修正.计算的床内存料量和总压降与实测值之间的误差均小于5%.本研究为中温脱硫过程的优化和降低能耗奠定了流体力学基础.     

过热蒸汽/热载体/生物质三相混合数值模拟

建立了工业尺度数值试验的过热蒸汽/热载体/生物质三相欧拉数理模型。水蒸气设为主连续相,采用k-ε方程描述;热载体和生物质设为两种不同的拟流体相,其黏度和压力采用颗粒动理论方程进行封闭;气固相互作用力采用Symbol-O'Brein曳力模型描述,固固之间作用力采用morsi-alexander模型描述;气固之间传热采用Gunn模型描述。基于该模型的数值试验,实现了喷动流化床混合器的优化设计使热载体与生物质的混合程度达70%以上;并研究了物料性质对生物质终温的影响,生物质终温受生物质粒径的影响最为显著。     

风沙运动问题的SPH-FVM耦合方法数值模拟研究

针对离散颗粒模型和欧拉-欧拉双流体模型在求解气粒两相流动问题中存在的不足,提出了一种新方法——SPH-FVM耦合方法,并应用于风沙运动过程的数值模拟计算.新方法基于拟流体模型,采用光滑粒子流体动力学方法(SPH)对离散颗粒相进行求解,追踪单颗粒运动轨迹,采用有限体积法(FVM)求解连续气体相,捕捉流场特性,两相间通过曳力、压力梯度、体积分数等参量进行耦合,建立了两种不同坐标系下方法间的耦合框架.对SPH粒子所承载的物质属性进行了重定义,改造成了适用于离散颗粒相求解的光滑离散颗粒流体动力学模型(SDPH),阐述了SPH粒子与离散相中颗粒之间的关系,推导得到了拟流体SPH离散方程组.模拟了风沙运动中沙粒跃移过程和自由来流风速作用下沙粒的运动过程以及沙丘在风力作用下缓慢向前蠕移的过程,分析了颗粒的运动轨迹,流场中沙粒水平速度分布规律以及气体速度场在沙粒反作用下的变化情况等,与实验结果相符合,结果表明该方法不仅精度较高,而且计算量较小,适于求解风沙运动问题乃至其他气粒两相流动问题.     

方腔内Cu/Al2O3水混合纳米流体自然对流的格子Boltzmann模拟

纳米流体作为一种较高的导热介质,广泛应用于各个传热领域.鉴于纳米颗粒导热系数和成本之间的矛盾,本文提出了一种混合纳米流体.为了研究混合纳米流体颗粒间相互作用机理和自然对流换热特性,在考虑颗粒间相互作用力的基础上,利用多尺度技术推导了纳米流体流场和温度场的格子Boltzmann方程,通过耦合流动和温度场的演化方程,建立了Cu/Al2O3水混合纳米流体的格子Boltzmann模型,研究了混合纳米流体颗粒间的相互作用机理和纳米颗粒在腔体内的分布.发现在颗粒间相互作用力中,布朗力远远大于其他作用力,温差驱动力和布朗力对纳米颗粒的分布影响最大.分析了纳米颗粒组分、瑞利数对自然对流换热的影响,对比了混合纳米流体(Cu/Al2O3-水)与单一金属颗粒纳米流体(Al2O3-水)的自然对流换热特性,发现混合纳米流体具有更强的换热特性.     

可压缩均匀湍流中颗粒对湍流调制的数值研究

本文采用欧拉-拉格朗日点源方法对颗粒调制可压缩均匀各向同性衰减湍流进行了直接数值模拟,颗粒和流场之间通过双向耦合相互作用,初始湍流马赫数达到1.2,研究了不同密度(St_0)颗粒对湍流调制的影响,轻颗粒(St_0=0.1)略微增强流体的湍动能和旋转运动;临界颗粒(St_0=1.0)也增强流体的旋转运动;而重颗粒(St_0=5.0)则会削弱流体的湍动能和旋转运动。对于流场的可压缩性,颗粒由于其惯性都会削弱可压缩性,并且颗粒越重,削弱越强。     

方腔内Cu/Al2O3水混合纳米流体自然对流的格子Boltzmann模拟

纳米流体作为一种较高的导热介质, 广泛应用于各个传热领域. 鉴于纳米颗粒导热系数和成本之间的矛盾, 本文提出了一种混合纳米流体. 为了研究混合纳米流体颗粒间相互作用机理和自然对流换热特性, 在考虑颗粒间相互作用力的基础上, 利用多尺度技术推导了纳米流体流场和温度场的格子Boltzmann方程, 通过耦合流动和温度场的演化方程, 建立了Cu/Al₂O₃水混合纳米流体的格子Boltzmann模型, 研究了混合纳米流体颗粒间的相互作用机理和纳米颗粒在腔体内的分布. 发现在颗粒间相互作用力中, 布朗力远远大于其他作用力, 温差驱动力和布朗力对纳米颗粒的分布影响最大. 分析了纳米颗粒组分、瑞利数对自然对流换热的影响, 对比了混合纳米流体(Cu/Al₂O₃-水)与单一金属颗粒纳米流体(Al₂O₃-水)的自然对流换热特性, 发现混合纳米流体具有更强的换热特性.     

细颗粒间相互作用力的研究

对微米、近亚微米尺度的细颗粒间相互作用过程的显微观察发现,细颗粒间具有"吸引-旋绕-排斥"的相互作用行为.受力分析表明,包含范德华力、静电库仑力和电像力的传统颗粒间作用力模型不能解释这种相互作用行为.根据细颗粒的荷电特性,提出细颗粒间还具有诱导偶极子作用力.通过引入偶极子作用力改进了细颗粒间作用力模型,利用新的模型对细颗粒间相互作用进行了模拟,得到了和实验相同的相互作用行为,并且对影响细颗粒相互作用的参数进行了分析.提出投入大颗粒和增加外静电场等都是促进颗粒凝并的有效措施.     

二维颗粒运动以及传热的数值模拟

本文提出了直接热源算法(Direct heat source)使得颗粒壁面温度满足Dirichlet边界条件,同时采用Wang等提出的内嵌边界方法(Immersed boundary method)和多重直接力算法(Multi-direct forcing)全尺度模拟二维颗粒的运动.本文计算了运动的冷颗粒在不同Grashof数时的运动状态.本文的计算结果与其他学者的计算结果符合得很好,从而验证了本文采用的方法的精确性与有效性.     

静电与流体作用下微米颗粒团迁移机理研究

由于带电微米颗粒之间存在静电、流体相互作用,颗粒团的迁移行为与单颗粒截然不同。本文以静电除尘器(ESP)中带电颗粒团迁移沉积为背景,首先探讨流体有限惯性对单颗粒扰流流场的影响,在此基础上利用Oseen动力学计算颗粒团迁移过程中的颗粒团形状演化机理。计算发现,在颗粒间流体相互作用下,颗粒团由初始的球形经历扁化、拖尾、破碎等过程,最后形成子颗粒团。随着颗粒荷电量增加,颗粒团迁移由原本的破碎模式转变为较为稳定的各向同性膨胀模式。迁移模式转变能够用荷电参数κ_q定量描述。最后,我们发现颗粒团破碎形式与颗粒初始分布不均匀性有关,并且利用Lyapunov指数定量分析了颗粒团形状不稳定性。     

基于拉氏概率密度的两相湍流二阶矩模型

用颗粒运动的拉氏分析和PDF方法,改进了颗粒相的二阶矩模型。由拉氏两相运动的随机微分方程出发,采用随机过程分析和信号分析法得到湍流两相流动的PDF输运方程,双流体模型方程和两相脉动速度相关的基本模式的封闭式,和用其它方法导出的方程与封闭式的结果一致,对封闭式作了重要的改进,在分析颗粒轨道上的流体湍流作用时间时,全面地引入拉氏分析的轨道穿越效应、惯性效应、连续效应和湍流的各向异性。     

Aggregation of particles settling in shear-thinning fluids

We have experimentally studied the coaxial settling of three identical non-Brownian spheres in a shear-thinning fluid at small Reynolds numbers. While settling, the particles create corridors of reduced viscosity in their wake and, if they are initially close enough to one another, they can form stable clusters. By analogy with previous results obtained on two-particle interaction in the first part of this work, we show that the particle velocities can be satisfactorily described using a first-order expression and assuming that the reduced viscosity remains constant. We report systematic experiments performed at different initial separation distances between particles and the use of our simple model allows the prediction of the settling behaviour and in particular the conditions for clusters formation. We thus show that particle aggregation can occur even for large initial distances between particles and within times that are small compared to the time scales in Newtonian fluids.     

Transient settling of a dilute spherical suspension droplet at low Reynolds number

The transient settling in a viscous incompressible fluid of a spherical dilute cloud of particles starting from rest under the influence of a small constant applied force is studied in a continuum model on the basis of the linearized Navier-Stokes equations. Explicit expressions are derived for the motion of the cloud and for the flow velocity and pressure of the fluid. Equations of transient Stokesian dynamics are formulated that allow numerical study of the motion of a dilute cloud of particles of arbitrary initial configuration.     

Aggregation of particles settling in shear-thinning fluids

We have experimentally studied the coaxial settling of three identical non-Brownian spheres in a shear-thinning fluid at small Reynolds numbers. While settling, the particles create corridors of reduced viscosity in their wake and, if they are initially close enough to one another, they can form stable clusters. By analogy with previous results obtained on two-particle interaction in the first part of this work, we show that the particle velocities can be satisfactorily described using a first-order expression and assuming that the reduced viscosity remains constant. We report systematic experiments performed at different initial separation distances between particles and the use of our simple model allows the prediction of the settling behaviour and in particular the conditions for clusters formation. We thus show that particle aggregation can occur even for large initial distances between particles and within times that are small compared to the time scales in Newtonian fluids. Received 10 July 2002 RID="a" ID="a"e-mail: talini@fast.u-psud.fr     

Influence of wall motion on particle sedimentation using hybrid LB-IBM scheme

We integrate the lattice Boltzmann method (LBM) and immersed boundary method (IBM) to capture the coupling between a rigid boundary surface and the hydrodynamic response of an enclosed particle laden fluid. We focus on a rigid box filled with a Newtonian fluid where the drag force based on the slip velocity at the wall and settling particles induces the interaction. We impose an external harmonic oscillation on the system boundary and found interesting results in the sedimentation behavior. Our results reveal that the sedimentation and particle locations are sensitive to the boundary walls oscillation amplitude and the subsequent changes on the enclosed flow field. Two different particle distribution analyses were performed and showed the presence of an agglomerate structure of particles. Despite the increase in the amplitude of wall motion, the turbulence level of the flow field and distribution of particles are found to be less in quantity compared to the stationary walls. The integrated LBM-IBM methodology promised the prospect of an efficient and accurate dynamic coupling between a non-compliant bounding surface and flow field in a wide-range of systems. Understanding the dynamics of the fluid-filled box can be particularly important in a simulation of particle deposition within biological systems and other engineering applications.     

A non-equilibrium molecular dynamics approach to fluid transfer through microporous membranes

The colour field non-equilibrium molecular dynamics method is applied to model fluid transfer between two fluid phases separated by microporous membranes (i.e. pore width ≤ 2 nm). The model is simple. All particles show short range spherical interactions, but reproduce the main features of the experimental phenomenology of molecular sieve membranes. An external force is applied to the fluid particles and, due to the presence of the membrane, a stationary flux across the system results. A range of mass transfer coefficients was investigated and the fluxes were found to be linear with force intensities. The density was found to have a small effect on molecular mobility, while temperature had a more significant effect, showing the features of an activated process. In addition, the changes in these fluxes with the membrane size, and the cross interaction parameters between the fluid and the membrane particles were examined.     

两个非磁性颗粒在磁流体中的沉降现象研究

在磁场作用下,在磁流体里添加非磁性颗粒(non-magnetic particles,NPs),可以使得NPs形成不同的结构,操控NPs的运动从而影响磁流体的特性,这种应用逐渐受到了研究者的关注.为了更好地操控磁流体里NPs的运动,本文采用一种多物理模型研究在外加磁场作用下,磁流体中两个NPs沉降的运动过程.其中,用格子玻尔兹曼方法模拟磁流体的运动,外加磁场对磁流体的影响用一种自修正方法求解泊松方程,这个自修正方法可以使欧姆定律满足守恒定律.NPs之间的偶极干扰力采用偶极力模型,同时采用一种相对过渡平滑的共轭边界条件处理NPs与磁流体交界面的流固干扰以避免磁场密度过渡的突变.本文主要探究两个NPs在磁流体中的沉降,揭示磁场作用下NPs的相互干扰原理;同时,对控制NPs运动时的参数进行调节,得到NPs不同的运动轨迹,达到操控颗粒运动的目的.本研究可对NPs在磁流体中的应用提供定量的分析结果,对NPs在工业上的应用提供有力的理论支撑.     

Hydrodynamic screening and axisymmetric turbulence in the transient sedimentation of a dilute suspension at small Reynolds number

A theory of settling of a dilute suspension of identical spherical particles in a viscous incompressible fluid is developed on the basis of the equations of transient Stokesian dynamics. The equations describe hydrodynamic interactions between particles moving under the influence of a constant force, starting to act at a particular instant of time. For a dilute suspension, a monopole approximation can be used. It is argued that the growth of velocity fluctuations is bounded by a combination of two effects, destructive interference of the flow patterns of individual particles, and a rearrangement of particle positions leading to a time-dependent microstructure of the suspension. After a long time, the microstructure tends to a steady state. The corresponding structure factor is described phenomenologically. The corresponding pair correlation function and the velocity correlation functions describing axisymmetric turbulence on the length scale of the mean distance between particles are evaluated.     

Two-Fluid Anisotropic Cosmological Model with Variable G and Λ

Two-fluid anisotropic Bianchi type-III cosmological model is investigated with variable gravitational constant G and cosmological constant Λ in the framework of Einstein’s general relativity. In the two-fluid model, one fluid represents the matter content of the universe and another fluid is chosen to model the cosmic microwave background radiation. The dynamics of the anisotropic universe with variable G and Λ are discussed. We also discussed in detail the behavior of associated fluid parameters and kinematical parameters.