The transport of nanoparticles in subsurface with fractured, anisotropic porous media: Numerical simulations and parallelization

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS

卷:350

Article

Chen, Meng-Huo¹  Salama, Amgad²  Sun, Shuyu³ 

[1] Univ Leeds, Sch Comp, EC Stoner Bldg, Leeds LS2 9JT, W Yorkshire, England

[2] Univ Regina, Fac Engn, Regina, SK, Canada

[3] King Abdullah Univ Sci & Technol, Phys Sci & Engn Div, Thuwal 23955, Saudi Arabia

关键词: Multipoint flux approximation;    Nanoparticle transport;    Algebraic multigrid;    Parallel computation;   

DOI  :  10.1016/j.cam.2018.09.019

来源: Elsevier

【 摘 要 】

The flow of fluids through fractured porous media has been an important topic in the research of subsurface flow. The several orders of magnitude in size between the fractures and the rock matrix causes difficulties for simulating such flow scenario. The fluid velocities in fractures are also several orders of magnitude higher than that in the rock matrix due to high permeability and porosity. If there exists pollutant such as nanoparticles in the fluids, the pollutant may be transported rapidly and the rock matrix's properties near the fractures are hence changed. In this research, we simulate the transport phenomena of nanoparticles in the fluid flow through fractured porous media. The permeability fields which contain different anisotropy angles are considered in the simulation. Fractures are represented explicitly by volumetric grid cells and the numerical algorithm is parallelized in order to reduce the simulation time. We investigate the effect of the appearance of fractures and rotated anisotropy on the transport of nanoparticles, particles deposition, entrapment and detachment. The results show that flow directions are affected by the direction of anisotropy and the transport of nanoparticles in the fractures is significantly faster than that in rock matrix due to high fluid velocities. The direction of anisotropy distorted the pressure field and changed the fluid flow directions, which determined the time needed for the pollutant front to reach the fractures. The parallel efficiency of the overall algorithm is also discussed and the experimental results show that it is deeply affected by the performance of the multigrid solver. (C) 2018 Elsevier B.V. All rights reserved.

【 授权许可】

Free