TY - GEN
T1 - Modeling compressible gas flow in anisotropic reservoirs using a nonlinear finite volume method
AU - Zhang, W.
AU - Al Kobaisi, M.
N1 - Publisher Copyright:
Copyright © ECMOR 2020. All rights reserved.
PY - 2020
Y1 - 2020
N2 - A nonlinear two-point flux approximation (NTPFA) finite volume method is applied to the modeling of compressible gas flow in anisotropic reservoirs. Gas compressibility factor and gas density are calculated by the Peng-Robinson equation of state. The governing equations are discretized by NTPFA in space and first-order backward Euler method in time. Newton-Raphson iteration is used as the nonlinear solver during each time step. The NTPFA method employs the harmonic averaging points as auxiliary points during the construction of onesided fluxes. A unique nonlinear flux approximation is obtained by a convex combination of the one-sided fluxes. Since a Newton-Raphson nonlinear solver is used, NTPFA will have a denser discretized coefficient matrix compared to the widely used Two-Point Flux Approximation (TPFA) method on grids that are not K-orthogonal. However, its coefficient matrix is still much sparser than the classical Multi-Point Flux Approximation O (MPFAO) method. Results of numerical examples demonstrate that the pressure profile and gas production rate of NTPFA is in close agreement with that of MPFA-O for most cases while TPFA is inconsistent since the grid is not Korthogonal. The MPFA-O method is well known to suffer from monotonicity issues for highly anisotropic reservoirs and our numerical experiments show that MPFA-O can fail to converge during the Newton-Raphson iterations when the permeability anisotropy is very high while NTPFA still enjoys good performance.
AB - A nonlinear two-point flux approximation (NTPFA) finite volume method is applied to the modeling of compressible gas flow in anisotropic reservoirs. Gas compressibility factor and gas density are calculated by the Peng-Robinson equation of state. The governing equations are discretized by NTPFA in space and first-order backward Euler method in time. Newton-Raphson iteration is used as the nonlinear solver during each time step. The NTPFA method employs the harmonic averaging points as auxiliary points during the construction of onesided fluxes. A unique nonlinear flux approximation is obtained by a convex combination of the one-sided fluxes. Since a Newton-Raphson nonlinear solver is used, NTPFA will have a denser discretized coefficient matrix compared to the widely used Two-Point Flux Approximation (TPFA) method on grids that are not K-orthogonal. However, its coefficient matrix is still much sparser than the classical Multi-Point Flux Approximation O (MPFAO) method. Results of numerical examples demonstrate that the pressure profile and gas production rate of NTPFA is in close agreement with that of MPFA-O for most cases while TPFA is inconsistent since the grid is not Korthogonal. The MPFA-O method is well known to suffer from monotonicity issues for highly anisotropic reservoirs and our numerical experiments show that MPFA-O can fail to converge during the Newton-Raphson iterations when the permeability anisotropy is very high while NTPFA still enjoys good performance.
UR - https://www.scopus.com/pages/publications/85099557256
U2 - 10.3997/2214-4609.202035002
DO - 10.3997/2214-4609.202035002
M3 - Conference contribution
AN - SCOPUS:85099557256
T3 - ECMOR 2020 - 17th European Conference on the Mathematics of Oil Recovery
BT - ECMOR 2020 - 17th European Conference on the Mathematics of Oil Recovery
T2 - 17th European Conference on the Mathematics of Oil Recovery, ECMOR 2020
Y2 - 14 September 2020 through 17 September 2020
ER -