中国科学院过程工程研究所机构知识库

循环流化床是一种重要的工业装置，其内部的颗粒常常具有较宽的粒径分布，同时气固两相流是一种多尺度复杂流体系统，存在颗粒聚团/气泡等介尺度结构。颗粒粒径分布特性和介尺度结构对气固相间传递特性、对反应效率等工业生产过程中的关键参数具有显著影响。为了更好地探究此类气固两相流系统的流动现象，除了直接通过实验进行研究外，计算流体力学模拟也逐渐成为了一种不可或缺的重要研究手段。本论文的研究就集中在对循环流化床提升管内气固两相流的模拟计算。为了同时考虑介尺度结构和粒径分布的影响，本论文针对提升管反应器内的气固两相流，在能量最小多尺度（EMMS）模型的基础上对双粒度及多粒度颗粒体系进行分析并建立了模型，并将双粒度和多粒度EMMS曳力模型与连续介质模型耦合，对具有宽粒径分布的气固两相流实验进行了大量的三维模拟计算，验证模型的有效性。本论文的具体工作及主要结论如下：第二章首先对含有两种不同颗粒的气固两相流系统开展研究，在前人工作的基础上，优化拓展了双粒度EMMS曳力模型。并与连续介质模型耦合，实现了双粒度颗粒流体体系的三维计算流体力学模拟计算。结果表明因为双粒度EMMS曳力模型同时考虑了不同颗粒物性和介尺度结构的影响，因而相较于目前被广泛使用的均匀曳力模型能更加准确的预测粗、细颗粒的流动状态。在模拟验证过程中还发现不同壁面边界条件对于提升管内的气固两相流有较大的影响。第三章进一步建立了适用于宽粒径分布气固两相流的多粒度EMMS曳力模型。为了在EMMS模型的多尺度框架中描述具有宽粒径分布的颗粒体系，需要首先将连续分布的粒径离散为多个具有不同特征粒径的颗粒相。然后通过对多尺度框架中的密相、稀相和虚拟相间相中的流动及各相之间的相互作用进行描述以构建方程组，最后再利用单位质量颗粒悬浮输送能耗最小这一能量判据对上述方程组进行封闭。通过求解多粒度EMMS曳力模型可以得出不同颗粒所受到的气固相间曳力系数。对多个具有宽粒径分布的实验进行的三维模拟计算结果表明，通过多粒度EMMS曳力模型的修正可以减小被均匀曳力模型高估的气固相间曳力，从而使小颗粒在提升管底部的占比更加合理。此外，在模拟计算中还对比了不同固相应力和固-固相间曳力模型的影响，发现采用不同的固-固相间曳力模型所计算出的颗粒在轴向和径向的分布存在很大的差异，而不同的固相应力模型对计算结果几乎没有影响。为了进一步研究双粒度和宽粒径分布颗粒体系在提升管内的流动情况，第四章对具有上述颗粒体系的停留分布实验进行了模拟计算。在模拟验证过程中对比了双粒度和多粒度EMMS曳力模型与均匀曳力模型计算出的颗粒停留时间分布曲线的差别。发现均匀曳力模型算出的停留时间分布曲线峰值出现的时间明显靠前，而且单峰更高，说明颗粒停留时间短，混合程度较小。而采用双粒度和多粒度EMMS曳力模型计算出的停留时间分布曲线峰值出现较晚，且峰的高度更低，说明颗粒停留时间更长，不同颗粒间的混合程度更大。在对比研究壁面边界条件的影响时发现，当对颗粒采用自由滑移边界条件时，可以计算出颗粒在壁面附近下滑，并形成上稀下浓的浓度分布结构，而且求出颗粒停留时间较长。当壁面边界变粗糙时，可能出现上浓下稀的结构，求出的颗粒停留时间较短。而通过对不同固-固曳力模型计算结果的对比发现，固固相间曳力越小则不同颗粒的流动状态差异越大，反之则流动状态更为接近。最后，第五章总结了本论文的主要工作和创新点，并在现有发现和结论的基础上提出了对下一步工作的展望。;The circulating fluidized bed (CFB) is an important equipment in the chemical industry. The particles within the CFB usually has a wide particle size distribution (PSD). Meanwhile, the gas-solid flow is a multi-scale complex system with meso-scale structures such as clusters and bubbles. The PSD and the meso-scale structures have significant influences on the critical factors such as gas-solid interacting and reaction rates. As an important measure to investigate the gas-solid flow with a wide PSD, CFD (computational fluid dynamic) simulation nowadays has become as indispensable as experiment, and this thesis is focused on the simulation of the bio- and polydisperse gas-solid flows in the riser of circulating fluidized bed. In order to take the PSD and meso-scale structure into consideration, the EMMS (The Energy-Minimization Multi-Scale) drag model for polydisperse gas-solid flow is proposed in this thesis. Then the EMMS drag model is combined with the continuum model to conduct the CFD simulation for the bio- and polydisperse gas-solid flows. A great deal of 3D (3 dimensional) simulations were conduted to verify the effectiveness of the proposed models. The the main points of this thesis are sketched out in the ensuing paragraphs.The EMMS drag model for biodispersed gas-solid flow is constructed in the second chapter. Then the EMMS model is combined with the continuum model to conducted 3D simulations for the bio-dispersed gas-solid flow within the risers of CFB experiments. The results of simulations indicate that a more reasonable gas-solid drag can be calculated by the bio-dispersed EMMS drag model, which will predict the mixing and segregation of the fine and coarse particles better. Besides, the wall boundry condition are investigated at the same time.In the third chapter, the EMMS drag model is extended to fit for the polydisperse gas-solid flow. At first, the continuous particle size is discretized into several characteristic sizes, by which the polydisperse particles are classified into several discrete groups. Then the dense phase, dilute phase and virtual inter-phase of the EMMS drag model are constructed by theses different particle groups. Finally, the minimization of the energy consumed in suspending and transporting per unit mass of particles is used to close the equations of the polydisperse EMMS drag model. Many 3D CFD simulations are conducted by coupling gas-solid drag model and kinetic theories into continuum models. The simulation results reveal that the polydisperse EMMS drag model can modify the overestimated gas-solid drag calculated by classic drag models to predict the segregation between different particles better. In addition, the particulate phase stresses and particle-particle drag calculated by different kinetic theories are investigated, and the results indicate that the particle-particle drag plays an important role in the mixing and segration of different particles. However, the influence of the particulate phase stresses can be ignored.Solids residence time distributions (RTD) of the bio- and polydisperse particle system in CFB risers are simulated in the forth chapter. The RTD curves indicate that the bio- and polydisperse EMMS drag model will lead to the longer particle residence time than the one calculated by the classic drag model. Moreover, the influence of the wall boundry condition and the particle-particle drag models are further investigated by the RTD simulations. The final chapter summarizes the main conclusions and the innovations of the thesis and makes a try to ouline the prospect of the work.