History-matching
of core-flood experimental data through numerical
modeling is a powerful tool to get insight into the relevant physical
parameters and mechanisms that control fluid flow in enhanced oil
recovery processes. We conducted a mechanistic numerical simulation
study aiming at modeling previously performed water-alternating-gas
and foam-assisted chemical flooding core-flood experiments. For each
experiment, a one-dimensional model was built. The obtained computed
tomography scan data was used to assign varying porosity, and permeability,
values to each grid block. The main goal of this study was to history-match
measured phase saturation profiles along the core length, pressure
drops, produced phase cuts, and the oil recovery history for each
of the experiments conducted. The results show that, to obtain a good
match for the water-alternating-gas experiment, gas relative permeability
needs to be reduced as a function of injection time due to gas trapping.
The surfactant phase behavior, for the aid of foam-assisted chemical
flooding, was successfully simulated and its robustness was verified
by effectively applying the same phase behavior model to the two different
salinity conditions studied. It resulted in the oil mobilization,
through the injection of a surfactant slug, being properly modeled.
The mechanistic simulation of foam using the steady-state foam model
built in UTCHEM proved inadequate for the mechanistic modeling of
a foam drive in the presence of oil. An alternative heuristic approach
was adopted to overcome this limitation.