Back to Search Start Over

Accurate multiscale finite element method for numerical simulation of two-phase flow in fractured media using discrete-fracture model.

Authors :
Zhang, Na
Yao, Jun
Huang, Zhaoqin
Wang, Yueying
Source :
Journal of Computational Physics. Jun2013, Vol. 242, p420-438. 19p.
Publication Year :
2013

Abstract

Abstract: Numerical simulation in naturally fractured media is challenging because of the coexistence of porous media and fractures on multiple scales that need to be coupled. We present a new approach to reservoir simulation that gives accurate resolution of both large-scale and fine-scale flow patterns. Multiscale methods are suitable for this type of modeling, because it enables capturing the large scale behavior of the solution without solving all the small features. Dual-porosity models in view of their strength and simplicity can be mainly used for sugar-cube representation of fractured media. In such a representation, the transfer function between the fracture and the matrix block can be readily calculated for water-wet media. For a mixed-wet system, the evaluation of the transfer function becomes complicated due to the effect of gravity. In this work, we use a multiscale finite element method (MsFEM) for two-phase flow in fractured media using the discrete-fracture model. By combining MsFEM with the discrete-fracture model, we aim towards a numerical scheme that facilitates fractured reservoir simulation without upscaling. MsFEM uses a standard Darcy model to approximate the pressure and saturation on a coarse grid, whereas fine scale effects are captured through basis functions constructed by solving local flow problems using the discrete-fracture model. The accuracy and the robustness of MsFEM are shown through several examples. In the first example, we consider several small fractures in a matrix and then compare the results solved by the finite element method. Then, we use the MsFEM in more complex models. The results indicate that the MsFEM is a promising path toward direct simulation of highly resolution geomodels. [Copyright &y& Elsevier]

Details

Language :
English
ISSN :
00219991
Volume :
242
Database :
Academic Search Index
Journal :
Journal of Computational Physics
Publication Type :
Academic Journal
Accession number :
89032981
Full Text :
https://doi.org/10.1016/j.jcp.2012.12.006