Abstract We present a computational cell-based dynamic cluster structure-dependent (DCSD) drag model which consists of eight balance equations and one extreme value equation for energy dissipation rate with the bivariate extreme value (BEV) method. The criterion of cluster existence is proposed to identify the heterogeneous flow and homogenous flow of gas-particles suspension. The occurrence of clusters is intermittent, and the mean intermittency factor of clusters is from 0.233 to 0.266. The life of the clusters is predicted. The occurrence of heterogeneous structures contribute to the hydrodynamic-dominant mechanism (HD), collision-dominant mechanism (CD) and compromise in competition of collision-hydrodynamic dominant mechanism (CHD). The CHD for particle clustering is a central mechanism, and CD is next to HD. The clusters diameter, volume fraction and velocity are compared with calculations using empirical correlations. The calculated large diameter clusters using empirical correlations are characteristic of flow of collective clusters which are identified using velocity correlation coefficient.

中文翻译：

---

在气体颗粒立管中使用动态集群结构相关阻力模型进行集群间歇模拟

摘要 我们提出了一种基于计算单元的动态簇结构相关（DCSD）阻力模型，该模型由八个平衡方程和一个能量耗散率极值方程组成，采用双变量极值（BEV）方法。提出了簇存在判据来识别气粒悬浮液的非均相流和均相流。簇的出现是间歇性的，簇的平均间歇因子在0.233到0.266之间。预测簇的寿命。异质结构的出现有助于流体动力主导机制（HD）、碰撞主导机制（CD）和碰撞-流体动力主导机制（CHD）竞争的妥协。粒子聚类的 CHD 是一个中心机制，CD 紧随 HD 之后。簇直径，体积分数和速度与使用经验相关性的计算进行比较。使用经验相关性计算的大直径簇是使用速度相关系数识别的集体簇流的特征。