TY - JOUR

T1 - An Experimenting Field Approach for the Numerical Solution of Multiphase Flow in Porous Media

AU - Salama, Amgad

AU - Sun, Shuyu

AU - Bao, Kai

N1 - KAUST Repository Item: Exported on 2020-10-01

PY - 2015/7/14

Y1 - 2015/7/14

N2 - In this work, we apply the experimenting pressure field technique to the problem of the flow of two or more immiscible phases in porous media. In this technique, a set of predefined pressure fields are introduced to the governing partial differential equations. This implies that the velocity vector field and the divergence at each cell of the solution mesh can be determined. However, since none of these fields is the true pressure field entailed by the boundary conditions and/or the source terms, the divergence at each cell will not be the correct one. Rather the residue which is the difference between the true divergence and the calculated one is obtained. These fields are designed such that these residuals are used to construct the matrix of coefficients of the pressure equation and the right-hand side. The experimenting pressure fields are generated in the solver routine and are fed to the different routines, which may be called physics routines, which return to the solver the elements of the matrix of coefficients. Therefore, this methodology separates the solver routines from the physics routines and therefore results in simpler, easy to construct, maintain, and update algorithms.

AB - In this work, we apply the experimenting pressure field technique to the problem of the flow of two or more immiscible phases in porous media. In this technique, a set of predefined pressure fields are introduced to the governing partial differential equations. This implies that the velocity vector field and the divergence at each cell of the solution mesh can be determined. However, since none of these fields is the true pressure field entailed by the boundary conditions and/or the source terms, the divergence at each cell will not be the correct one. Rather the residue which is the difference between the true divergence and the calculated one is obtained. These fields are designed such that these residuals are used to construct the matrix of coefficients of the pressure equation and the right-hand side. The experimenting pressure fields are generated in the solver routine and are fed to the different routines, which may be called physics routines, which return to the solver the elements of the matrix of coefficients. Therefore, this methodology separates the solver routines from the physics routines and therefore results in simpler, easy to construct, maintain, and update algorithms.

UR - http://hdl.handle.net/10754/579849

UR - http://doi.wiley.com/10.1111/gwat.12353

UR - http://www.scopus.com/inward/record.url?scp=84961207948&partnerID=8YFLogxK

U2 - 10.1111/gwat.12353

DO - 10.1111/gwat.12353

M3 - Article

C2 - 26171913

SN - 0017-467X

VL - 54

SP - 262

EP - 273

JO - Groundwater

JF - Groundwater

IS - 2

ER -