Advanced numerical methods for the simulation of flows in heterogeneous porous media and their application to parallel computing
Flows in highly heterogeneous porous media arise in a variety of processes including enhanced oil recovery, in situ bioremediation of underground contaminants, transport in underground aquifers and transport through biological membranes. The common denominator of these processes is the transport (and possibly reaction) of a multi-component fluid in several phases. A new numerical methodology for the analysis of flows in heterogeneous porous media is presented. Cases of miscible and immiscible displacement are simulated to investigate the influence of the local heterogeneities on the flow paths. This numerical scheme allows for a fine description of the flowing medium and the concentration and saturation distributions thus generated show low numerical dispersion. If the size of the area of interest is a square of a thousand feet per side, geological information on the porous medium can be incorporated to a length scale of about one to two feet. The technique here introduced, Operator Splitting on Multiple Grids, solves the elliptic operators by a higher-order finite-element technique on a coarse grid that proves efficient and accurate in incorporating different scales of heterogeneities. This coarse solution is interpolated to a fine grid by a splines-under-tension technique. The equations for the conservation of species are solved on this fine grid (of approximately half a million cells) by a finite-difference technique yielding numerical dispersions of less than ten feet. Cases presented herein involve a single phase miscible flow, and liquid-phase immiscible displacements. Cases are presented for model distributions of physical properties and for porosity and permeability data taken from a real reservoir. Techniques for the extension of the methods to compressible flow situations and compositional simulations are discussed.
- Research Organization:
- Houston Univ., TX (United States)
- OSTI ID:
- 7271309
- Resource Relation:
- Other Information: Thesis (Ph.D.)
- Country of Publication:
- United States
- Language:
- English
Similar Records
Incorporation of mixed finite element methods in compositional simulation for reduction of numerical dispersion
An ELLAM-MFEM solution technique for compressible fluid flows in porous media with point sources and sinks