气固广义流态化包括并流上行、并流下行和逆流下行等不同的操作模式，各操作模式下均存在颗粒轴径向非均匀分布。对气固并流上行流动，随气体速度增大，反应器中会出现从鼓泡、湍动、快速流态化到稀相输送的流域转变，快速流态化向稀相输送的转变伴随结构的突变，出现噎塞（Choking）现象。对气固逆流下行流动，当气体速度或者固体循环量过大时，向下运动的颗粒会被阻止甚至从反应器顶部溢出，出现液泛（Flooding）现象。上述动力学特征随操作条件的变化，能以流态化相图系统反映，这对反应器的设计和操作控制具有重要的理论和实际意义。但目前已发表的相图主要针对并流上行各流域且基于实验数据或经验关联，并且不能完全反映噎塞和液泛等重要特征。气固流态化是典型的非线性非平衡系统，具有以颗粒团聚物和气泡为典型形态的动态介尺度结构，对此提出的能量最小多尺度（Energy-Minimization Multi-Scale，EMMS）模型已成功描述了并流上行的流域转变和噎塞现象，并拓展应用于鼓泡流态化和下行床系统。本文基于EMMS逆流下行床模型预测液泛现象，并通过EMMS稳态模型与CFD-DEM模拟结果的相互验证，论证了其合理性，最后集成上述研究得到基于EMMS模型的广义流态化相图。论文的主要内容如下：论文首先将EMMS模型拓展应用于逆流下行床轴向动力学参数和液泛气速的预测，明确了操作条件、颗粒物性、边壁摩擦力和入口空隙率对轴向动力学分布的影响。逆流下行床中完全发展段空隙率随表观气速的变化规律为：随表观气速增大，空隙率先缓慢降低、再剧烈降低、最后微弱下降直至模型无解，总体表现为S形分布。根据该分布规律，提出了液泛气速的预测方法。为了验证上述预测，论文分别采用CFD-DEM模拟与稳态EMMS模型分析了气固广义流态化中的流域转变，两者的预测结果定性一致。对逆流下行流动的CFD-DEM模拟检验了EMMS逆流下行床模型预测的轴向动力学参数分布规律和液泛判定方法，据此定性总结分析了液泛现象的物理含义。同时，对于并流上行噎塞状态饱和夹带量的预测，原始EMMS模型与CFD-DEM模拟的计算结果相近，验证了EMMS模型的正确性。而CFD-DEM模拟精细地展示了噎塞时床层顶部稀相区与底部密相区共存，空隙率轴向S形分布，径向中间稀、边壁浓的环核结构，并与大量实验结果相符。最后，结合原始EMMS模型和各拓展模型，得到了基于EMMS模型的广义流态化相图，以反映各操作模式下床层空隙率、噎塞和液泛气速随操作条件的变化。但基于EMMS模型的相图针对特定颗粒物性计算，实际系统中颗粒物性差异大，为此对相图上的典型特征变量，采用量纲分析将其表达为物性和操作条件组成的无量纲量的函数，并拟合得到它们之间的关联式，从而拓展了相图的应用范围。本论文研究有助于拓展EMMS理论在气固系统稳态模拟中的应用，对气固系统在不同模式下的操作和控制具有重要的理论和实际意义。;Gas-solid fluidized beds operate in different ways such as concurrent upward, concurrent downward and counter-current downward modes, which can be called gas-solid generalized fluidization. No matter in which mode, both axial and radial heterogeneities are always displayed. In gas-solid concurrent upward flow, there are typical regime transitions such as bubbling, turbulent, fast fluidization and pneumatic transport with the increase of gas velocity. The transition from fast fluidization to pneumatic transport is accompanied by a sudden change in bed structure, which is called choking. In gas-solid counter-current downward flow, the downward movement of particles can be prevented and even blown out from the reactor top if the gas velocity or solids flux is high enough, which is called flooding. The above-mentioned hydrodynamic features can be systematically described by using the so-called fluidization phase diagrams, which is of great theoretical and practical significance for the design and operation of fluidized reactors. However, current phase diagrams are generally drawn according to either experimental data or empirical correlations and mainly focused on the regime transitions in concurrent upward flow, failing to reflect the features of choking and flooding completely.Gas-solid fluidization as a kind of nonlinear and non-equilibrium system exhibits typical dynamical meso-scale structures of particle clusters or gas bubbles. The Energy-Minimization Multi-Scale (EMMS) model has successed in predicting the choking as well as other regime transitions in concurrent upward flow, and has been extended to bubbling fluidization and downer systems. In this thesis, the EMMS-based counter-current model was used to predict the flooding phenomenon，which was verified further by using a computational fluid dynamics method coupled with adiscrete element method (CFD-DEM). Finally, the phase diagrams of generalized fluidization for Geldart A and B particles were redrawn according to the calculation results of the EMMS model and its extensions.The main contents of this thesis are listed as follows:The EMMS model was first extended to predict the axial hydrodynamic parameters and the flooding phenomenon and clarify the effects of operating conditions, particle properties, wall-friction and entrance voidage in counter-current downward flow. It was found that with the increase of gas velocity the voidage (efull) in the fully developed region of counter-current downward flow decreases slightly at first, but then decreases significantly, and remains nearly invariable finally until the model cannot be numerically solved, showing an S-shaped profile. Recognizing this, a reasonable method was proposed to predict the flooding phenomenon.In order to validate the foregoing prediction, the regime transitions in generalized fluidization were analyzed by using the CFD-DEM simulation and the steady-state EMMS model. The two models predicted consistent results qualitatively. For counter-current downward flow, the CFD-DEM simulation validates the prediction of the EMMS-based model in the axial heterogeneity and the flooding feature. Based on the both predictions, the physical meaning of flooding was explored qualitatively. Meanwhile, both methods give good predictions for saturation carrying capacity at the choking state in concurrent upward flow. Moreover, the CFD-DEM simulation provides the details of the coexistence of bottom dense region and top dilute in the axial direction and the core-annulus structure in the radial direction, which are inaccordance with well-accepted experimental data.Finally, combining the calculation results of the original EMMS model and its extensions, the EMMS-based phase diagram of generalized fluidization was drawn to describe the variations of voidage, flooding and choking gas velocities with operating conditions under the various operation modes. However, the EMMS-based phase diagram was calculated for specific material properties, and did not capture the influence of variable material properties in actual systems. Therefore, some typical hydrodynamic parameters in the phase diagram were correlated to both operating conditions and material properties in the form of dimensionless groups, so as to broaden the application range of the EMMS-based phase diagram of generalized fluidization.This research is helpful to extending the application of the EMMS model in the steady-state simulation of gas-solid flow, and of great theoretical and practical significance for the operation and control of gas-solid fluidized systems under different operation modes.