skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Porous media: fluid distributions and transport with applications to petroleum recovery

Thesis/Dissertation ·
OSTI ID:5371198

Multiphase flows of the sort encountered in petroleum recovery processes are analyzed from microscopic and continuum points of view. On the microscopic level, a statistical, percolation-based, pore-network theory is developed to model the relative permeabilities and capillary pressures that characterize the Darcy transport and capillarity-controlled spatial distributions of two phases, e.g. oil and water, or three phases, e.g. gas, oil and water, in porespace. Inputs to the theory are a disordered bond network or tree to model the chaotic connectivity of porespace. The bonds, which represent lumped pore-segments, are characterized by a single distributed parameter, throat-radius, with which all other pore dimensions are regarded as perfectly correlated. Phases are assigned to the bonds according to capillary pressure levels, wettabilities, and past assignments of fluids to the bonds. These criteria and accessibilities evaluated with percolation theory in turn define pore-filling sequences of occupancy and allowability from which follow predictions of saturations and relative permeabilities. On the continuum level, the statistical theory is combined with an adaptive finite element method to simulate the one-dimensional flow of two and three phases, as in cores taken from reservoir rock. The combined simulation approach draws needed predictions of relative permeabilities and capillary pressures from the statistical theory, so that the need for experimental data is reduced. The adaptive finite element method used to solve the equations that describe the transport of the phases automatically refines subdomains in zones where fluid content changes rapidly. This makes it possible to resolve accurately any fronts that form.

Research Organization:
Minnesota Univ., Minneapolis (USA)
OSTI ID:
5371198
Resource Relation:
Other Information: Thesis (Ph. D.)
Country of Publication:
United States
Language:
English