Abstract:

In large-eddy simulations (LES) of gas-solid two phase turbulent flows, the absence of subgrid-scale (SGS) turbulence information introduces significant inaccuracies in predicting the motion characteristics of the dispersed phase. To date, all particle subgrid models are based on one-way coupling conditions. With the adaptive mesh algorithm based on wavelet discrete errors, this study addresses this limitation by developing a novel gas-solid turbulence large eddy simulation platform under two-way coupled conditions. The subgrid velocity model seen by particles based on differential filter (DF) was then applied to the simulation of decaying gas-solid homogeneous isotropic turbulence. Comparative analysis of particle-pair statistics in pre-/post-model implementation simulations demonstrates that the proposed approach effectively restores dispersed-phase accuracy while validating its efficacy under high-volume-loading two-way coupling conditions. The inclusion of the DF model achieves enhanced simulation accuracy for the dispersed phase while maintaining relatively low computational costs. It reduces the relative error of the particle pairs radial distribution function with moderate inertia by about 28%, and the relative error of the radial relative velocity by about 13%. This study establishes critical theoretical foundations for advancing engineering applications in industrial particulate systems, particularly enhancing predictive capabilities for fluidized bed dynamics and multiphase reactor optimization.

Key words: two-phase flow, turbulence flow, numerical simulation, particle sub-grid model, two-way coupled

摘要：

两相湍流大涡模拟方法中缺失的亚网格尺度湍流信息会对离散相运动特性的预测造成较大误差。迄今为止，颗粒亚格子模型均是基于单向耦合条件构建的。为了提高气固两相湍流大涡模拟的模拟精度，结合基于小波离散误差的自适应网格算法，搭建了欧拉-拉格朗日框架双向耦合条件下的气固两相湍流模拟平台。在该平台下对衰减均匀各向同性湍流进行模拟，结果表明，基于差分过滤（DF）的颗粒所见流体亚格子速度模型的加入使中小惯性范围内的颗粒对径向分布函数相对误差减小约28%，径向相对速度相对误差减小约13%，能在高载荷下以较低的计算成本补偿离散相模拟精度，为该模型在大型工业反应器及流化床等化工设备实际工程应用的数值模拟中的发展及优化奠定了理论基础。

关键词: 两相流, 湍流, 数值模拟, 颗粒亚格子模型, 双向耦合