Local Equilibrium Mechanistic Simulation of CO2-Foam Flooding

Abstract

Mechanistic modeling of the non-Newtonian CO2-foam flow in porous media is a challenging task that is computationally expensive due to abrupt gas mobility changes. The objective of this paper is to present a local equilibrium (LE) CO2-foam mechanistic model, which could alleviate some of the computational cost, and its implementation in the Matlab Reservoir Simulation Tool (MRST). Interweaving the LE-foam model into MRST enables users quick prototyping and testing of new ideas and/or mechanistic expressions. We use MRST, the open source tool available from SINTEF, to implement our LE-foam model. The model utilizes MRST automatic differentiation capability to compute the fluxes as well as the saturations of the aqueous and the gaseous phases at each Newton iteration. These computed variables and fluxes are then fed into the LE-foam model that estimates the bubble density (number of bubbles per unit volume of gas) in each grid block. Finally, the estimated bubble density at each grid block is used to readjust the gaseous phase mobility until convergence is achieved. Unlike the full-physics model, the LE-foam model does not add a population balance equation for the flowing bubbles. The developed LE-foam model, therefore, does not add much computational cost to solving a black oil system of equations as it uses the information from each Newton iteration to adjust the gas mobility. Our model is able to match experimental transient foam flooding results from the literature. The chosen flowing foam fraction (Xf) formula dictates to a large extent the behavior of the solution. An appropriate formula for Xf needs to be chosen such that our simulations are more predictive. The work described in this paper could help in prototyping various ideas about generation and coalescence of bubbles as well as any other correlations used in any population balance model. The chosen model can then be used to predict foam flow and estimate economic value of any foam pilot project.