耦合格子Boltzmann的聚合物结晶相场方法

在模拟枝晶生长的相场方法上,耦合多松弛格式的格子Boltzmann方法,建立了相场-格子动力学耦合模型,对聚合物结晶过程中出现的枝晶生长形貌进行了数值模拟。耦合模型规避了传统相场法需要高分辨率场精确解决固液界面的难点,其中高精确的解析度会严重影响时间步长的选取。运用LBM方法,提出了一种自适应的时间步长计算方法,允许在改变时间步长的同时不需要相应地改变网格大小,时间步长允许设置得较大而数值模拟结果不会出现发散现象,也避免了高精度差分格式的使用。使用FORTRAN语言对相场-格子动力学耦合模型进行了求解,通过对聚合物等规聚苯乙烯结晶的数值模拟,研究了在不同实验温度下等规聚苯乙烯晶体的数值模拟形貌,并与实验结果进行了比较,验证了模型的有效性。     

A pore-scale model for microfibrous ammonia cracking microreactors via lattice Boltzmann method

Microfibrous microreactors with high reactive surface-to-volume ratio are good choices for ammonia cracking, which is one of the main strategies for CO-free hydrogen production. In the current study, a numerical model based on the lattice Boltzmann method (LBM) is presented to investigate ammonia cracking microreactors with coupled physiochemical thermal processes at the pore scale. Several sets of transport phenomena such as fluid flow, species transport, heat transfer and chemical reaction are taken into account. Moreover, to model the species transport in the ammonia cracking microreactor an active approach is applied for the first time. The model is validated and then employed to simulate the reactive transport in five different microreactors with dissimilar structural parameters. Comparison of the results shows that the fibers orientation is an effective geometric parameter that can greatly influence the hydrogen production efficiency.     

A unified interaction model for multiphase flows with the lattice Boltzmann method

The lattice Boltzmann method (LBM) has been increasingly adopted for modelling multiphase fluid simulations in engineering problems. Although relatively easy to implement, the ubiquitous Shan–Chen pseudopotential model suffers from limitations such as thermodynamic consistency and the formation of spurious currents. In the literature, the Zhang–Chen, Kupershtokh et al., the β-scheme, and the Yang–He alternative models seek to mitigate these effects. Here, through analytical manipulations, we call attention to a unified model from which these multiphase interaction forces can be recovered. Isothermal phase-transition simulations of single-component in stationary and oscillating droplet conditions, as well as spinodal decomposition calculations, validate the model numerically and reinforce that the multiphase forces are essentially equivalent. Parameters are selected based on the vapour densities at low temperatures in the Maxwell coexistence curve, where there is a narrow range of optimal values. We find that expressing the model parameters as functions of the reduced temperature further enhances the thermodynamic consistency without losing stability or increasing spurious velocities.     

基于多GPU并行格子Boltzmann方法的方管湍流模拟

采用CUDA (Compute Unified Device Architecture)和MPI (Message-Passing-Interface)在超级计算机Mole-8.5E上实现了格子Boltzmann方法(LBM)多GPU并行算法,通过三维顶盖驱动方腔流算例验证了多GPU并行LBM算法的准确性和有效性,利用该并行算法分别对雷诺数Reτ为300, 600, 1200下的充分发展的方管湍流进行了大规模模拟。研究发现,当计算网格尺寸小于黏性底层厚度(即Δ+<5)时,在壁面附近的相关传递特性统计误差较小,预测精度满足工程应用范围;Reτ为300, 600时,不同网格尺寸Δ+下的模拟结果表明LBM在方管流中心湍流特性统计具有网格弱相关性,Reτ为600时,与DNS (Direct Numerical Simulation)相比,Δ+=1.667, 3.750, 6.250时平均流向速度的平均误差分别是1.357%, 2.994%和4.766%;Reτ分别为300, 600和1200时,对应网格尺寸Δ+分别为0.833, 1.667和3.333时的方管湍流模拟中,成功捕捉到了二次流特性,预测得到的中心面流向速度、脉动均方根速度等的规律与文献基本吻合,进一步验证了单松弛LBM的可靠性,相关计算结果也为理解高Reτ下的方管湍流特性提供了参考。方管湍流的模拟验证了单松弛LBM多GPU并行算法在超大规模网格计算中的潜力,为进一步实现实际工程流动所需更大规模的数值模拟奠定了基础。     

Gas flow simulations in a structured packing by lattice Boltzmann method

Numerical simulations of gas flow between two sheets of plastic Mellapak^TM 250 Y are performed using Lattice Boltzmann methods in laminar and turbulent regimes. Results are compared with experimental measurements and with known correlations. They are also compared with simulations using a classical CFD code. In all cases, the agreement is very good.     

Numerical simulation of bubble interactions using an adaptive lattice Boltzmann method

Bubble interactions have significant impact on the shape and motion of bubbles, and therefore the dynamics of bubbles in a swarm may be considerably different from that of an isolated bubble. This research presents a numerical study of bubble interactions using a novel lattice Boltzmann method (LBM). By using the adaptive mesh refinement (AMR) and the multiple-relaxation-time (MRT) algorithm, this technique is able to accurately capture the deformation of the interface, and can remain numerically stable for low Morton number and high Reynolds number flows. The numerical approach is briefly illustrated, and validated with the experimental results of the buoyant rise of an isolated bubble in the literature. Then the method is applied to simulate the interaction between multiple bubbles during their buoyant rise. A pair of bubbles with spherical or ellipsoidal shapes is first simulated under different configurations and rise velocities. Both attractive and repulsive interactions are observed in the simulations depending on the relative position and the Reynolds number. When the Reynolds number is sufficiently high, the bubble–wake interaction is found to be the main interaction mechanism, which results in strong attraction between the ellipsoidal bubbles in vertical or oblique arrangement. Simulations for a group of 14 bubbles are also carried out, and effects of the bubble shape and Reynolds number on the spatial distribution of the bubbles are briefly discussed. In general, a good agreement is found between the current simulation and the experimental and numerical results reported in the literature.     

多相/多组分LBM模型及其在微流体领域的应用

多相/多组分流体在化工领域中广泛应用,通常伴随有组分传递、相变、界面的产生与运动、化学反应等复杂的物理、化学过程,建立描述相应复杂体系的理论模型和模拟方法面临着极大的挑战。格子Boltzmann方法（lattice Boltzmann method,LBM）是近年来备受关注的一种介观模拟方法;多相/多组分格子Boltzmann模型具有易于追踪多相动态界面、壁面性质设定简单等优点,从物理模型到计算精度和速度都有很大的优势,尤其适合于模拟微化工系统下的多相多组分流动问题。以格子Boltzmann方法为基础,重点介绍了气液及液液体系多相/多组分LBM模型的研究进展以及其在微流体系统中的应用。     

REV尺度多孔介质格子Boltzmann方法的数学模型及应用的研究进展

综述了多孔介质表征体元尺度(REV)格子Boltzmann模型的研究进展,根据对多孔介质处理方式主要分为部分反弹模型和阻力模型两类,分析归纳了各类模型的优缺点。由于阻力模型中渗流的广义格子Boltzmann方程(GLBE)的作用力是基于GUO等的作用力模型,可以准确得到宏观方程,不存在离散误差,且模型的平衡分布函数和作用力项中都包含反应介质特性的孔隙率,因而应用最为广泛。本文还重点介绍了REV尺度多孔介质LBE模型在流动、传热、传质、化学反应及相变等过程中的具体应用,认为REV尺度多孔介质内的三传一反数学模型中需要加入孔隙尺度因素,在更大工程尺度上应该考虑过程参数的各向异性,展望了REV尺度多孔介质LBE模型的发展和应用前景。     

基于格子Boltzmann方法的单颗粒绕流数值模拟

采用格子Boltzmann方法（LBM）研究了单颗粒绕流流动过程。通过使用LBM中的LBGK（lattice Bhatnagar-Gross-Krook）模型和二阶精度的曲线边界条件处理方法,实现了对单颗粒绕流问题的定常及非定常流动过程中涡结构的模拟。采用动量交换法分别计算了Reynolds数在0.1～200范围内27个不同Reynolds数时的曳力系数,并将计算结果拟合得到基于LBM数值模拟的曳力曲线。计算结果表明,LBM在气固两相流的模拟计算中具有精确、可靠的优点,使用LBM模拟计算曳力曲线的方法经济、易行,并且可以克服由传统实验方法获得曳力曲线的局限性。     

球形胶囊内约束熔化过程的LBM模拟

为了准确预测球形胶囊内部材料的相变性能,为后续相变微胶囊浆体的多尺度研究提供微观相变信息,采用格子Boltzmann方法（LBM）,引入浸入式移动边界处理方案,借鉴糊状区和热焓理论,构建了适于相变模拟的数值模型,模拟了球形胶囊内部固液相变过程,讨论了不同粒径尺度下熔化机制的区别。结果表明,利用LBM方法得到的预测结果与可视化实验数据吻合较好,清晰地呈现出球体上部温度热分层和下部液相强对流共存的特性。随着粒径尺度的不断减小,胶囊内部对流作用逐渐减弱,甚至当粒径小于3 mm时,其内部对流作用可忽略。     

Modeling of ionic diffusivity in non-saturated cement-based materials using lattice Boltzmann method

The ionic diffusivity in non-saturated cement-based materials is quantitatively studied using lattice Boltzmann method. The Shan–Chen multi-phase lattice Boltzmann model modified by incorporating the Carnahan–Starling equation of state into the model is used to simulate the equilibrium distribution of water and gas phases with density ratio as high as 775 in the three-dimensional microstructure of cement paste at various degrees of water saturation. The lattice Boltzmann diffusion model is subsequently applied to simulate the diffusion process of the ionic species through the partially saturated cement paste. Before the application of the modified Shan–Chen model, the benchmark tests including bubble test and contact angle test are carried out to validate it. It is shown that the ionic diffusivity is greatly influenced by the degree of water saturation. The simulated relative ionic diffusivity as a function degree of water saturation is in good agreement with the experimental data obtained from literature. In addition, the effect of w/c ratio on the ionic diffusivity through cement paste under non-saturated conditions is investigated. It indicates that there is a significant difference between the relative diffusivity for cement paste with w/c ratios of 0.4 and 0.5 over the whole range of water saturation degree. However, the difference of relative diffusivity for cement paste with w/c ratios of 0.5 and 0.6 is not obvious.     

基于格子Boltzmann方法模拟纳米流体强化传质过程

使用Boltzmann方法对纳米流体中的传质过程进行了模拟,给出了修改后的Boltzmann方程和计算纳米流体中扩散系数的方程,通过两种方法,基于有限体积颗粒的LBE方法和基于点源颗粒的LBE方法进行了模拟,并与宣益民的实验进行了对比。最后计算了CO2在纳米流体中的扩散系数,计算结果表明,纳米流体因为其纳米颗粒的微扰动对传质有着很大的强化效果,为CO2吸收提供了一种新思路,并且证明了纳米流体的强化传质主要是靠对流传质。     

并列和错列纤维过滤器稳态除尘过程的 格子Boltzmann模拟

大部分纤维捕集效率和压降的理论模型认为纤维性能仅取决于来流速度、颗粒粒径、纤维体积分数、过滤层厚度、纤维直径等因素。实际上,布袋除尘器的性能还与纤维配置方式直接相关。利用LB(lattice Boltzmann)两相流模型对多层纤维捕集颗粒物过程进行了数值模拟,研究了不同纤维配置方式下系统压降与捕集效率的变化。结果表明,错列纤维的性能参数优于并列纤维;纤维排列间距增大,压降增幅大于捕集效率,导致性能参数下降。通过比较不同位置纤维的捕集能力发现,在布朗扩散和拦截捕集机制主导下,前方纤维捕集能力略强于后方纤维;而在惯性碰撞捕集机制主导时,对捕集贡献最大的主要是前两排纤维,后方纤维对捕集效率的贡献非常小,可以忽略。这些研究结果可以对布袋除尘器的多层纤维配置方式的优化提供理论依据和工程建议。     

功能性多孔介质中气泡输运调控的格子Boltzmann分析

为提高毛细蒸发海水淡化技术中的蒸发效率,多孔介质层需要维持一定的毛细压力,同时还要确保气泡能够快速通过。基于此背景,本文建立了多孔介质参数化模型,探究了气泡穿越多孔介质间隙过程的运动特征,研究在保持一定孔隙毛细压力的同时,通过调控孔道尺寸及排布从而使气泡能够更快速地通过多孔介质层。基于格子Boltzmann伪势模型分析了多孔介质孔隙率、壁面润湿特性、孔道排布及气泡水平方向初速度等对气泡形貌、上升速度、与壁面平均接触面积及孔隙毛细力的影响,获得了多孔介质的孔隙率设计范围,骨架润湿特性调控以及孔道排布方式的选择依据。同时还获得了在实际蒸发过程中,可以使气泡存在一定的水平方向初速度,从而能够更快地脱离多孔介质的策略。     

气泡群上升过程相互作用的格子 Boltzmann三维数值模拟

应用格子Boltzmann自由能模型,在三维空间里模拟了大密度比气泡群在静止的黏性不可压缩流体中上升过程以及它们之间的相互作用。为了避免气液密度比过大造成数值不稳定问题,采用八点差分和十八点差分格式分别求解一阶1586Φ和二阶1586²Φ。模拟结果表明,气泡的初始大小和初始位置影响上升气泡周围的流场以及形状变化。当直径相同的气泡群上升时,位置靠下的气泡会受到位置靠上气泡尾迹的影响,并有很明显的形状变化。气泡之间的影响程度取决于两气泡之间的距离及相对位置,并且随着距离增加而逐渐减小。然而直径不同的气泡群上升时,不管初始位置如何,大气泡总会对小气泡造成强烈的影响。     

Pore-scale study of reactive transfer process involving coke deposition via lattice Boltzmann method

Coke deposition during catalytic industrial processes can lead to the narrowing and blockage of pore structures inside particles so as to influence the catalyst performance. In this work, a direct pore-scale simulation is carried out by the lattice Boltzmann method to investigate the change of pore structures caused by coke formation. Coke deposition processes in the single pore and complex porous medium are evaluated. The interaction of pore structure and mass transfer under the condition of coke deposition is analyzed. The results reveal that the pore narrowing and blockage can hinder the viscous flow and Knudsen diffusion, resulting in the change of mass transfer mechanism, which is sensitive to gas rarefaction effect and coke deposition rate as well as surface roughness. For the complex porous medium, the uniformity of pore size distribution becomes weak under the gas rarefaction effect and coke deposition.     

颗粒团绕流曳力系数的LBM计算

采用格子Boltzmann方法(Lattice Boltzmann method, LBM) 中的LBGK(Lattice Bhatnagar- Gross-Krook)模型和二阶精度的曲线边界条件处理方法对二维颗粒团绕流现象进行了数值模拟,并同时使用动量交换法计算了两种颗粒团构型中不同颗粒的曳力系数。结果表明：颗粒团曳力系数与颗粒聚团的构型有着密切联系,颗粒聚团的形成将导致颗粒团曳力系数大幅度减小。除颗粒团构型因素外,颗粒间距和流动Reynolds数也是导致颗粒团曳力系数发生改变的主要因素。当颗粒聚团存在时,颗粒团中不同颗粒的受力有较大差异,若忽略颗粒聚团效应,则颗粒团曳力系数的计算必然将产生偏差。     

Organic crystallization processes

Crystallization has been the most important separation and purification process in the pharmaceutical industry throughout its history. Many parallels exist in the fine chemicals industry also. Over the past several decades the study of crystallization operations has taken on increasing levels of importance because of several factors that require effective control of crystallization processes. These levels of control require better understanding of the complex interactions of nucleation and growth as well as the operating characteristics of crystallization equipment including the critical issue of scale-up.In the pharmaceutical industry, the issue of better control, desirable in and of itself, is reinforced by the need to satisfy both company internal and governmental regulatory authorities on the consistency of chemical and physical properties of active pharmaceutical ingredients (APIs). Control of crystallization operations and choice of equipment both for pilot plant and manufacturing are thereby critical. The objective of this paper is to summarize the critical issues that must be addressed in order to achieve this level of control.     

Thermal behaviour of multi-walled carbon nanotube/polymer composites using lattice Boltzmann method

A three-dimensional computational model was developed to study the thermal behaviour of multi-walled carbon nanotube (MWNT)-polymer composites. A lattice Boltzmann algorithm has been introduced to tackle the conjugate heat transfer among different phases. The model was validated by comparing with the theoretical solutions and simulation results. The effects of the orientation, volume fraction and diameter of MWNTs on the effective thermal conductivity (λ_e) of composites were quantified. It is found that the effective thermal conductivity of the composites weakens with the variation of MWNTs orientation angle, and the decrease extent deduces gradually. Moreover, with the same orientation angle, the λ_e of the composites increases with the volume fraction of MWNTs. It also indicates that MWNTs with a small diameter at a constant volume fraction can promote the thermal conductivity of the composites more efficiently than that with a larger diameter.