Liang Cheng's dissertation
by
Liang Cheng, Ph.D.
University of Texas at Austin, 2002
Supervisors: William R. Rossen
Foam has application in gas diversion in improved oil recovery, acid diversion in well
stimulation, and environmental remediations. Successful foam application in these processes
requires predictive foam modeling. In this study, a foam simulator based on the "fixed-Pc*"
model is implemented into a chemical-flood reservoir simulator, UTCHEM, and successfully
validated against an analytical method. This foam simulator can describe foam flow in both
low- and high-quality regimes. Two numerical artifacts resulting from numerical dispersion
of the surfactant front are examined; both can be controlled by using a higher-order
numerical scheme and grid refinement. A study of the effect of surfactant on foam strength
and propagation reinforces Zeilinger’s (1996) finding that, in the absence of dispersion,
the foam front propagates as an indifferent wave; independent of how foam strength depends
on surfactant concentration.
Two-dimensional simulations of foam flow in adjacent layers show that capillary-crossflow
does weaken foam in the high-permeability layer, but it does not fatally hurt sweep
efficiency between adjacent layers. Viscous cross-flow gives an even frontal advance of
foam in all cases we examined.
A gas-trapping model for foam is implemented into UTCHEM. The model fits steady-state
foam mobility and the non-Newtonian rheology of liquid injected after foam. When applied
to model dynamic coreflood data, the simulator qualitatively fits the transition period
between foam and liquid injection in laboratory corefloods.
Back to theses index
