TL;DR
This paper introduces a new model for calculating oil-water relative permeability in low-permeability reservoirs by incorporating a "quasi-threshold pressure gradient." This model offers a more accurate representation of non-Darcy flow behavior, leading to improved predictions of water flooding efficiency and enhanced oil recovery compared to traditional methods.
What Problem Does This Paper Address?
Low-permeability reservoirs are vital for global energy supply, but their efficient development is a significant challenge. These reservoirs are characterized by complex fluid flow behaviors, including strong non-Darcy flow and a pronounced threshold pressure gradient (the minimum pressure needed to initiate flow). Traditional methods for characterizing oil-water flow, like the Johnson-Bossler-Naumann (JBN) method, often struggle to accurately capture these non-linear effects, leading to suboptimal water flooding strategies and lower-than-expected oil recovery. This paper aims to bridge this gap by developing a more robust model that accounts for the dynamic nature of the threshold pressure gradient in these challenging reservoirs.
Methods
The researchers employed a multi-pronged approach:
- Experimental Investigation: They conducted unsteady-state water flooding experiments on native core samples from a low-permeability reservoir. These experiments, performed at a constant temperature of 50°C, involved injecting water into cores with permeabilities ranging from 2.99 to 34.40 mD. Key parameters like cumulative oil and water production, and pressure drop were meticulously recorded.
- Threshold Pressure Gradient Characterization: A significant contribution is the development of a mathematical model to characterize the "quasi-threshold pressure gradient." This model correlates the injected water cut with normalized pressure gradient, taking into account corrected permeability and water saturation.
- Novel Relative Permeability Model: Building upon the threshold pressure gradient model and non-Darcy flow theory, they derived a new computational model for calculating relative permeability during unsteady-state water flooding. This model dynamically integrates the threshold pressure gradient.
- Model Validation: The proposed model was rigorously validated against the conventional JBN method using the experimental data.
Datasets
The study utilized native core samples collected from a specific low-permeability reservoir block. Four representative core samples with varying permeabilities (2.99 to 34.40 mD) were used for the experiments. The data generated from these core floodings served as the basis for developing and validating the new model. No new benchmark datasets were introduced; rather, existing experimental data was used to compare against established methods.
Key Insights and Ideas
- Dynamic Threshold Pressure Gradient: The study highlights that the threshold pressure gradient isn't static but dynamically changes with water saturation. It initially increases to a peak value before decreasing, a crucial insight for understanding non-Darcy flow.
- Non-Darcy Flow Dominance: In low-permeability reservoirs, especially during two-phase flow, the flow significantly deviates from Darcy's law, necessitating models that account for this non-linearity.
- Impact on Relative Permeability: The proposed model, by incorporating the threshold pressure gradient, leads to lower calculated oil-phase relative permeability and higher water-phase relative permeability compared to the JBN method. This suggests that traditional methods might overestimate oil flow potential.
- Permeability Influence: Higher permeability in these reservoirs generally leads to more favorable oil-water flow characteristics, including expanded two-phase flow regions and higher equipermeability points.
Key Results
- The developed model successfully characterized the dynamic quasi-threshold pressure gradient in low-permeability sandstone cores.
- The new relative permeability calculation model, which integrates the threshold pressure gradient, showed improved accuracy in representing non-Darcy flow behavior.
- Validation against the JBN method indicated that the new model yields lower oil-phase relative permeability and higher water-phase relative permeability. This discrepancy generally diminishes as reservoir permeability increases.
- The study categorized water-phase relative permeability curves into concave-downward and linear types, offering insights into reservoir heterogeneity.
Limitations and Future Work
- Assumptions: The theoretical model relies on several simplifying assumptions, including incompressible fluids, neglect of capillary and gravitational forces, and the assumption that the threshold pressure gradient primarily affects the flow initiation. These might not fully capture complex multi-physics coupling effects like rock-fluid interactions or wettability alteration.
- Experimental Constraints: The experiments were limited to cores from a single reservoir block and a specific range of permeabilities. The applicability of the derived empirical relationships to reservoirs with significantly different lithologies, higher clay content, or broader permeability ranges needs further validation. Additionally, experiments were conducted at a single temperature and oil-water viscosity ratio.
- Model Scalability: The characterization model was calibrated at the core scale. Its direct extrapolation to field-scale simulations requires careful consideration of heterogeneity and upscaling effects. The model's performance in oil-wet or mixed-wet systems also warrants further investigation.
- Future Work: The authors suggest extending the model's applicability to more complex conditions, such as fractured reservoirs, oil-wet systems, and varying temperature and salinity environments. They also propose integrating the model into digital twin frameworks for smart field management.
Summary updated