Speaker
Description
Water-Alternating-Gas (WAG) injection is a critical technique for Enhanced Oil Recovery (EOR) and Carbon Capture, Utilization, and Storage (CCUS). However, accurate prediction of gas mobility remains a significant challenge due to the complex hysteresis and cycle-dependent nature of the gas relative permeability krg in three-phase flow systems. Conventional empirical models often fail to capture the reduction in gas mobility observed in mixed-wet or oil-wet carbonate reservoirs in the presence of mobile water, leading to significant errors in injectivity and recovery forecasts. This study integrates high-quality steady-state coreflooding data with a novel, physics-based modification of existing hysteresis models to address these limitations.
We conducted a series of steady-state WAG experiments on mixed-wet carbonate samples under reservoir conditions. Unlike unsteady-state methods, the steady-state approach provided discrete, high-resolution krg data points across multiple drainage and imbibition cycles, revealing distinct irreversible hysteresis loops. Experimental results confirmed two primary mechanisms governing gas flow: (1) a non-monotonic trapping behavior that deviates from the classical Land relation, characteristic of non-water-wet systems, and (2) a cycle-dependent reduction in gas mobility driven by the redistribution of fluid phases and pore-throat occupancy.
To model these phenomena, we evaluated several industry-standard models, including Stone, Baker, and Jerauld, but found them insufficient for capturing the observed hysteresis. Consequently, we propose an improved hybrid modeling framework based on the Larsen and Skauge (L&S) model. While the original L&S model introduces a reduction factor to account for hysteresis, it treats this factor as a static constant and relies on Land’s trapping theory, which proved inadequate for our mixed-wet samples.
Our innovation lies in a two-fold modification: First, we replaced the static Land trapping function with a quadratic trapping model (inspired by Spiteri et al.) to accurately match the experimental residual gas saturation (Sgr) endpoints in mixed-wet media. Second, we developed a dynamic mobility reduction function. Instead of a constant exponent, we formulated the L&S reduction factor (α) as a dynamic function of the capillary number (Nc) and cycle number (N). This modification explicitly links the macroscopic decay of krg to the microscopic competition between viscous and capillary forces.
The proposed dynamic model demonstrates a superior match with experimental data compared to the original L&S and WAG-HW models, particularly in predicting the sharp decline in gas injectivity during later WAG cycles. By decoupling the trapping mechanism from mobility reduction, this framework provides a robust tool for reservoir simulators, offering improved accuracy for designing WAG and CO2 storage projects in complex wettability systems.
| Country | United Kingdom |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
