En
期刊文章
文章精选
当期目录
过刊浏览
高被引
高下载
特邀文章
专刊合集
待刊文章
Volume 117
您当前的位置：首页 > 期刊文章 > 过刊浏览 > Volumes 108-119 (2025) > Volume 117
Adaptive Gaussian mapping for unresolved CFD-DEM coupling
Najmeh Negahdari a, Ramin Khodabandehlou b, Hamid Reza Norouzi a *
a Center of Engineering and Multiscale Modelling of Fluid Flow (CEMF), Department of Chemical Engineering, Amirkabir University of Technology, Tehran, Iran
b Multiphase Systems Research Lab, School of Chemical Engineering, University of Tehran, Tehran, Iran
10.1016/j.partic.2026.06.035
Volume 117, October 2026, Pages 60-75
Received 30 January 2026, Revised 19 June 2026, Accepted 26 June 2026, Available online 18 July 2026, Version of Record 27 July 2026.
E-mail: h.norouzi@aut.ac.ir
Highlights

• Adaptive Gaussian mapping method is proposed for unresolved CFD-DEM.

• It handles varying cell-to-particle size ratio down to one.

• It maintains mapping accuracy for hexahedral and polyhedral meshes with low non-orthogonality.


Abstract

The computational fluid dynamics–discrete element method (CFD-DEM) is widely used for simulating particle–fluid systems. Coupling the fluid and particle phases requires porosity to be calculated, as well as the interphase momentum exchange terms, through mapping the particles into CFD cells. However, in practice, no single mapping approach is suitable for the broad range of CFD cell sizes to particle diameters, particularly when the ratio is close to unity. To address this, a new adaptive Gaussian mapping method has been developed, in which the kernel parameters are automatically adjusted according to the local CFD cell size and particle diameter. The validity of the proposed approach was evaluated under various packing conditions and mesh structures. For orthogonal hexahedral meshes, the mapping error remained below 2% for medium packing density and approximately 2–4% for dense packing (with a cell-to-particle size ratio between 1 and 3). For polyhedral meshes with low non-orthogonality, the error was below 1%. However, highly non-orthogonal meshes exhibited errors of up to 14%, suggesting reduced robustness in these conditions. The method was further validated in coarse-grained fluidized bed simulations, with the ratio of the cell size to the particle diameter varying from 5 to 1. Compared with diffusion-based simulations, the adaptive Gaussian method preserves the hydrodynamic behavior of the bed when coarse-graining is increased. This was achieved by controlling the smoothing locally through adaptive kernel-width adjustment, which avoids excessive smoothing. The results demonstrate that the proposed approach can effectively predict the hydrodynamic behavior of fluidized beds, particularly in coarse-grained CFD-DEM simulations.

Graphical abstract
Keywords
Fluidization; CFD-DEM; Coarse-graining; Porosity; Gaussian distribution
文章导读