基于EMMS的简化双流体模型及其在气固流动模拟中的应用

发布时间：2018-01-01 10:35

本文关键词：基于EMMS的简化双流体模型及其在气固流动模拟中的应用　出处：《中国科学院大学(中国科学院过程工程研究所)》2017年博士论文　论文类型：学位论文

【摘要】：气固两相流系统中普遍存在的颗粒聚团和气泡等介尺度结构对流动有着重要的影响。研究这类结构的数值模拟方法主要分为三类,分别是直接数值模拟(Direct Numerical Simulation,DNS)、离散颗粒模拟(Discrete Particle Model,DPM)和双流体模型(Two-Fluid Model,TFM)。对于工业装置的模拟,DNS和DPM需要的计算量太大,因此主要的模拟方法是TFM。但双流体模型的模拟精度依赖于子模型的准确度,比如衡量固相颗粒之间相互作用的固相应力模型以及气固相间曳力模型等。当前广泛采用的双流体模型通过基于颗粒动理论的(Kinetic Theory of Granular Flow,KTGF)的固相应力模型来考虑固相颗粒之间的作用,该方法需要额外的计算量来迭代求解颗粒温度的输运方程,同时也容易带来计算稳定性的问题。对于相间曳力的计算,目前研究者的一个共识是需要考虑介尺度结构。此外,许多研究发现,对于气固系统的模拟,相间曳力模型比固相应力模型更重要。因此,如果将考虑了介尺度结构的EMMS(Energy-Minimization Multi-Scale)曳力与简化的固相应力模型结合起来,将有望得到计算速度快同时精度也较高的简化双流体模型。如果进一步采用稳定性更好的数值离散方法,将能进一步提高模型的稳定性。在这一思路下,本论文第二章提出了一种简化双流体模型(Simplified Two-Fluid Model,STFM),将固相应力模型作了简化,同时耦合了 EMMS曳力。并对该模型中的两相动量方程进行了恒等变形,将两相的体积分率与相速度进行了分离,以避免固相动量守恒方程中的奇点问题。论文第三章首先介绍了 OpenFOAM(Open source Field Operation And Manipulation)平台下有限体积方法的基础,然后详细介绍了简化双流体模型在OpenFOAM下的具体实现,并重点阐述了求解固相连续性方程的MULES(Multi-dimensional Universal Limiterwith Explicit Solution)算法的原理。第四章用三个不同复杂度的算例对简化双流体模型进行了验证,总体上讲,简化双流体模型耦合EMMS曳力的模拟精度较高,与完整双流体模型(Full Two-Fluid Model,FTFM)耦合EMMS曳力相当,但是STFM的计算速度提高到FTFM的二倍以上。然而,对于复杂的工业装置,可能同时存在着多种流域。不同流域可能呈现出截然不同的流动特点,很难有一套模型能适用于所有流域。为此,本论文第五章提出了一种多区域简化双流体模型,这种模型能将工业装置划分成不同的区域并根据不同区域的流动特点选择不同的子模型。对于双流体框架下的模型,主要的两种子模型是摩擦应力模型和曳力模型。本论文首先考察了不同摩擦应力模型在颗粒堆积和密相鼓泡床模拟中的影响,然后考察了不同的曳力模型在不同流域情况下的表现。在此基础上,本论文开发了能针对不同流域选择不同摩擦应力模型和曳力模型的多区域简化双流体模型。此外,为了提高数值稳定性,本论文实现了一种延迟修正的离散格式,既能保持一阶迎风格式的稳定性,又能达到二阶离散精度。本章最后用此多区域简化双流体模型对一个实验室尺度的全回路循环流化床进行了模拟,并与之前的单区域简化双流体模型模拟的结果进行了比较,对比结果显示多区域简化双流体模型显著提高了模拟的精度。论文第六章总结了所取得的成果,并对简化双流体模型的应用前景以及未来的研究方向提出了展望。
[Abstract]:Common gas-solid two-phase flow system in particle agglomeration and bubble meso structure has an important effect on the flow. The numerical simulation method of this kind of structure is mainly divided into three categories, namely the direct numerical simulation (Direct Numerical Simulation, DNS), discrete particle simulation (Discrete Particle Model, DPM) and the two fluid model (Two-Fluid Model TFM). For the simulation of industrial equipment, computation of DNS and DPM need to be too large, so the simulation method is mainly TFM. but the simulation accuracy of the two fluid model depends on the accuracy of the model, such as the ratio between solid measure solid particles interaction model of stress and gas solid drag force model. Two fluid model widely used based on the granular dynamic theory (Kinetic Theory of Granular Flow, KTGF) of the solid stress model to consider the interaction between solid particles, the method requires Additional computation to iterative particle temperature transport equation, at the same time, it is easy to bring the calculation stability. For the calculation of the drag force, currently a consensus on the need to consider the mesoscale structure. In addition, many studies have found that the gas-solid system simulation, the drag force model is more important than solid the corresponding force model. Therefore, if considering the meso scale structure of EMMS (Energy-Minimization Multi-Scale) solid drag and the simplified stress model combined, will be expected to simplify the calculation speed of the two fluid model has higher precision and faster. If we use more stable numerical discretization method, will further improve the stability of the model. In this way, the second chapter puts forward a simplified two fluid model (Simplified Two-Fluid Model, STFM), the solid stress model was simplified, and at the same time Coupling the EMMS drag force. And the phase momentum equations in the model were identical deformation, the two phase volume fraction and phase velocity were separated to avoid the problem of singularity of conservation of momentum equation of solid phase. The third chapter introduces OpenFOAM (Open source Field Operation And Manipulation) platform based finite volume method then, details the realization of the simplified two fluid model in OpenFOAM, and focuses on solving the continuity equation of solid phase MULES (Multi-dimensional Universal Limiterwith Explicit Solution) algorithm principle. The fourth chapter uses three examples with different complexity of the simplified two fluid model was verified, generally speaking, simulation high precision of simplified two fluid model coupling EMMS drag, and the complete two fluid model (Full Two-Fluid Model, FTFM) EMMS phase coupling drag when, but STFM The calculation speed is increased to two times than FTFM. However, for complex industrial equipment, may also exist in a variety of different river basin. Flow may exhibit different characteristics, it is difficult to have a set of models can be applied to all river basin. Therefore, the fifth chapter puts forward a simplified two fluid model area. This model can be divided into different areas of industrial equipment and according to the flow characteristics of different regions of different sub models. The two fluid under the framework of the model, two seed model mainly is friction stress model and drag model. This thesis investigates different friction stress model in granular and dense drum effect bubble bed simulation, and then examines the performance of different drag force models in different watershed conditions. On this basis, this thesis is developed according to different choice of different basin friction stress model And multi regional drag model simplified two fluid model. In addition, in order to improve the numerical stability, this paper implements a discrete scheme of delay correction, which can maintain the stability of the first order upwind scheme, and can achieve the accuracy of two order discrete. At the end of this chapter, using a simplified two fluid model of the multi area on a laboratory scale the whole loop circulating fluidized bed was simulated, and compared with the previous single area simplified simulation of two fluid model results, the comparison results show that the multi regional simplified two fluid model significantly improves the accuracy of the simulation. The sixth chapter summarizes the achievements, and application prospect of the simplified two fluid model and the future the research direction is put forward.

【学位授予单位】：中国科学院大学(中国科学院过程工程研究所)
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TQ021.1;TQ018

【参考文献】