Abstract
A general state-based peridynamics model is developed to simulate transport of fluids in an arbitrary heterogeneous porous medium. The generality encompasses modeling of multiphase, multi-component flow of non-Newtonian and compressible fluids, which is often encountered in but not limited to subsurface reservoirs. Peridynamic model is especially useful for solving non-local problems, such as crack propagation, since it does not assume spatial continuity of field variables. Thus, the formulation presented here, combined with peridynamics-based damage model, can be used to simulate hydraulic fracturing with complex fluids. To demonstrate its capability to simulate multi-phase flow in porous media, the derived model is verified against the analytical Buckley-Leverett solution for immiscible Newtonian two-phase flow. Further, the non-Newtonian two-phase fluid flow in porous media is verified by simulating the polymer flood process involving immiscible displacement of a Newtonian fluid by a non-Newtonian fluid against a generalized solution obtained by Wu et al. [22]. The non-local solutions are shown to be consistent with the corresponding local solutions in limiting cases. Moreover, mass conservation of all the phases is satisfied, irrespective of discretization and extent of non-locality.
| Original language | English |
|---|---|
| Article number | 109075 |
| Journal | Journal of Computational Physics |
| Volume | 402 |
| DOIs | |
| State | Published - Feb 1 2020 |
Funding
This work is supported by DOE Grant No. DE-FOA-0000724 and by the member companies participating in the Joint Industry Program on Hydraulic Fracturing and Sand Control at the University of Texas at Austin. P. Seleson would like to acknowledge support provided by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U. S. Department of Energy. S. Agrawal would like to thank the anonymous reviewers for their useful and encouraging feedback. This work is supported by DOE Grant No. DE-FOA-0000724 and by the member companies participating in the Joint Industry Program on Hydraulic Fracturing and Sand Control at the University of Texas at Austin. P. Seleson would like to acknowledge support provided by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U. S. Department of Energy. S. Agrawal would like to thank the anonymous reviewers for their useful and encouraging feedback.
Keywords
- Fracture modeling
- Heterogeneity
- Multiphase flow
- Non-local model
- Peridynamic theory
- Transport in porous media
