Collaborative Research: Flow Instabilities in Gas-solid Flows

固体流在自然界（山体滑坡、雪崩、行星环等）和工业（制药、食品、化工和石油工业）中无处不在。高速气固流在广泛的应用中（循环流化床、气动输送管道、沙流建模、防侵蚀、行星环等），特别是经常产生不稳定性，称为“簇”。，这对系统性能有很大的影响。准确预测集群的发生和演化对相关系统的设计、放大和优化至关重要。结果表明，非弹性粒子间碰撞和气固阻力都能独立地导致不稳定性；然而，在实际系统中，它们总是合作的，它们的相对重要性还没有得到检验。本研究的目的是双重的：(i)阐明高速气固流动中不稳定性的各种来源的相对重要性，以及（ii）批判性地评估新的动力学理论（连续体）模型预测这种不稳定性定量性质的能力。这项研究将从一个简单的、依赖于时间的冷却系统开始，在这个系统中没有动能输入，然后转向更复杂的、实际相关的流化系统，有和没有固体边界，以研究与时间无关的统计数据。将使用四种不同的计算方法。晶格玻尔兹曼模拟解决了粒子周围的详细流动，将用于提供基于第一性原理的数据集，以评估碰撞和气相效应以及星团形成和演化的相对重要性。在连续体水平上，首先采用基于双流体动力学理论的线性稳定性分析来预测稳定性边界；欧拉-拉格朗日方法，其中只有气相被视为连续体，粒子是离散的，欧拉-欧拉模型，其中气相和粒子相都被视为连续体，然后将用于模拟团簇的演化。这些结果将与晶格玻尔兹曼数据进行比较，以对各种连续体模型的预测能力进行关键评估。本研究将生成基于第一性原理的高速气固流团簇形成和演化模拟数据，这些数据将用于建立一个精确的理论，能够在连续体水平上预测团簇不稳定性的发生和演化。气体和固体颗粒的同时流动发生在风暴、滑坡、用于能源生产的反应器和制药工业使用的混合装置中，仅举几例。这些系统中发生的复杂的物理相互作用使得仅从过去的实验中很难预测它们。在这个项目中，没有拟合参数的数学模型将被开发并用于预测这些系统特有的流动现象。这种建模工具的可用性有望在更短的周转时间内以比目前可能的更小的成本改进反应堆的设计。该模型将通过现有的开源代码提供给全球的研究人员，因此预计未来将在许多领域得到应用，包括制药、化学加工工业、能源生产、地质学和天体物理学。

1236157/1236490PI: Hrenya/YinGas-solid flows are ubiquitous in both nature (landslides, avalanches, planetary rings, etc.) and industry (pharmaceuticals, food products, chemical and petroleum industries). High-velocity gas-solids flows that are found in a wide range of applications (circulating fluidized beds, pneumatic transport lines, sand flow modeling, erosion prevention, planetary rings, etc.), in particular, often develop instabilities that are referred to as ?clusters?, which are known to have a large impact on system performance. Accurate prediction of onset and evolution of the clusters is critical to the design, scale-up, and optimization of related systems. It is shown that inelastic inter-particle collisions and gas-solid drag can both independently lead to instabilities; in real systems, however, they always cooperate and their relative importance has not been examined. The objective of this research is twofold: (i) to elucidate the relative importance of the various origins of the instabilities in high-velocity, gas-solid flows, and (ii) to critically assess the ability of a new kinetic-theory (continuum) model to predict the quantitative nature of such instabilities. This research will begin with a simplistic, time-dependent cooling system where kinetic energy input is absent, and then move to more complex and practically relevant fluidization systems with and without solid boundaries to investigate time-independent statistics. Four different computational methods will be used. Lattice Boltzmann simulations that solve the detailed flow around particles will be used to provide first-principle-based data sets needed to assess the relative importance of collisions and gas phase effects as well as cluster formation and evolution. On the continuum level, linear stability analyses based on two-fluid kinetic theories will first be used to predict the stability boundary; Euler-Lagrangian method where only gas phase is treated as a continuum and particles are discrete, and Euler-Euler models where both gas and particle phases are treated as continua, will then be used to simulate the evolution of the clusters. These results will be compared to the lattice Boltzmann data for a critical assessment of the predicative ability of the various continuum models. This research will generate first-principle-based simulation data on cluster formation and evolution for high-velocity gas-solid flows, and these data will be used to establish an accurate theory able to predict both onset and evolution of the clustering instability on the continuum level.The simultaneous flow of gas and solid particles occurs in windstorms, landslides, reactors used for energy production, and mixing units used by the pharmaceutical industry, to mention just a few. The complex physical interactions occurring in these systems make them difficult to predict from past experiments alone. In this project, mathematical models with no fitting parameters will be developed and used to predict flow phenomenon unique to these systems. The availability of such a modeling tool is expected to reduce to improved design of reactors in shorter turn-around times and at smaller costs than is currently possible. The model will be made available to researchers worldwide via an existing open-source code, and thus is expected to find future use in numerous sectors, including pharmaceuticals, chemical process industries, energy production, geology, and astrophysics.