Abstract
Point bar reservoirs are critical development targets in the late-stage recovery efforts of the Daqing Oilfield. This study presents a numerical simulation framework based on high-resolution point bar modeling. By integrating dynamic production data to generate simulated well histories and acquiring X-unit splitting factors, the researchers performed rigorous history matching. The simulation results reveal that oil saturation follows a distinct belt-like distribution influenced by lateral accretion interlayers during oil-water flooding. Specifically, the remaining oil is primarily trapped in the middle and upper sections of these lateral accretion bodies due to updip pinchout control. Based on these findings, the study demonstrates that horizontal wells effectively target this residual oil, a conclusion validated by further numerical simulation results.
Introduction
As mature oil fields like the Daqing Oilfield enter an extra-high water-cut phase, identifying and extracting remaining oil becomes increasingly complex. The primary challenge lies in the internal heterogeneity of thick reservoirs, particularly within point bar deposits formed by meandering rivers. These reservoirs are characterized by lateral accretion layers—inclined silt or mudstone baffles—that disrupt fluid flow. These layers create "shelter effects," trapping significant oil volumes while water bypasses them. Understanding the spatial distribution of this residual oil is essential for maintaining production stability in late-stage field development.
Methodology
The research utilized a multi-stage workflow to model and simulate the D Block of the Daqing Oilfield.
Geological Modeling and Upscaling
Using PETREL software, the team constructed a fine-scale 3D geological model focusing on the "∫-type" (sigmoidal) geometry of lateral accretion layers. To balance computational efficiency with geological accuracy, the initial 5m × 5m grid was upscaled to a 15m × 15m resolution while preserving the critical architectural features of the point bar.
Simulation and History Matching
The model incorporated production history from nine active wells in Target Layer X. To accurately represent fluid dynamics, 55 distinct relative permeability curves were applied, accounting for the varying porosity and permeability across the reservoir. The researchers performed history matching to align simulated production with historical field data, achieving a high degree of accuracy with only a 1.2% error in moisture content and a 4.25% error in cumulative oil production.
Results and Discussion
The numerical simulations provided deep insights into the mechanisms of oil entrapment and the effectiveness of targeted recovery strategies.
Oil Saturation Distribution
A key finding was the "belt" or "strip" distribution of oil saturation, which aligns with the strike of the lateral accretion layers. Due to gravity and the slanted orientation of these layers, injected water tends to migrate downward, flooding the lower sections of the sand body while leaving the upper sections saturated with oil.
Horizontal Well Performance
To exploit the identified remaining oil in the upper sections of the point bar, the team simulated the deployment of a horizontal well. The results were promising for a mature field: the well achieved an initial production rate of 8.8 m³/d. Although the water cut was high (92.31%), it remained viable for late-stage extraction, maintaining a production rate of 3.8 m³/d after one year.
Conclusion
This study confirms that high-resolution numerical simulation is a powerful tool for identifying "hidden" oil pockets in complex point bar reservoirs. The discovery that lateral accretion layers trap oil in the upper reservoir sections provides a clear geological basis for deploying horizontal wells. While this study focuses on spatial distribution and immediate viability, future work should include detailed economic analyses and optimization of well trajectories to further enhance sweep efficiency in mature fields.
Summary updated