钻探浆液渗透过程中混合非解析CFD-DEM耦合框架开发与验证

Development and validation of a hybrid unresolved CFD-DEM coupling framework for drilling slurry penetration

摘要:

常规非解析CFD?DEM方法受限于网格尺寸与颗粒直径限制，难以应用于大尺度宽粒径岩土介质的高效模拟。为此，本文开展了混合非解析CFD?DEM耦合方法二次开发。该方法以流体网格尺寸与颗粒直径之比作为自动判别准则，对于比值＞3的细颗粒以采样点法进行体积分数映射和颗粒?流体动量交换；而对于比值≤3的粗颗粒，则引入高斯核函数进行体积分数平滑映射，并对颗粒诱导的流体动量源项进行扩展分配。进一步地，针对颗粒靠近壁面时核函数拓展域被流体边界截断导致的映射误差，采用基于网格体积加权的归一化修正，以保证边界附近映射满足守恒要求。基于所开发程序，设计了体积分数守恒验证、单颗粒沉降速度验证和球形颗粒柱Ergun压降验证，以此评价混合非解析CFD?DEM耦合框架在宽粒径模型应用中的可靠性。结果表明，该方法可兼顾宽粒径颗粒体系模拟中的计算效率与耦合稳定性，为土石混合体注浆渗透、细颗粒迁移及颗粒柱渗流等问题提供可行的数值模拟方法。

Abstract:

Conventional unresolved CFD-DEM methods are limited by the constraint between grid size and particle diameter， making it difficult to efficiently simulate large-scale geomaterials with a wide particle size distribution. To address this issue， this paper presents the secondary development of a mixed unresolved CFD-DEM coupling framework. In this method， the ratio of fluid grid size to particle diameter serves as an automatic criterion. For fine particles with a ratio greater than 3， a sampling-point method is employed for volume fraction mapping and particle-fluid momentum exchange； whereas for coarse particles with a ratio less than 3， a Gaussian kernel function is introduced for smoothed volume fraction mapping， and the particle-induced fluid momentum source term is distributed over an extended region. Furthermore， to correct the mapping error caused by truncation of the kernel function''s extended domain at fluid boundaries when particles approach walls， a normalization correction based on grid-volume weighting is adopted to ensure conservation near boundaries. Based on the developed program， three verification tests， namely， volume fraction conservation， single-particle settling velocity， and Ergun pressure drop in a spherical particle column are designed to evaluate the reliability of the mixed unresolved CFD-DEM coupling framework in applications involving wide particle size distributions. The results demonstrate that the proposed method can balance computational efficiency and coupling stability in simulations of polydisperse particle systems， providing a feasible numerical approach for problems such as grouting permeation in soil-rock mixtures， fine particle migration， and seepage through particle columns.