Abstract
The central Sichuan Basin, located in western China, holds great significance in terms of hydrocarbon production, especially in relation to complex carbonate reservoirs, notably the Qixia Formation of the Middle Permian epoch. However, the comprehensive evaluation of this geological formation presents considerable challenges due to complex lithology, limited availability of reservoir property data at various scales, and the absence of a reliable reservoir ranking system. This study integrates advanced technologies like X-ray diffraction, micro-CT, and scanning electron microscope (SEM) techniques in digital drill cutting analysis at the microscale. A novel scoring system has been developed using a prominent component analysis (PCA) approach and an expert scoring system, which incorporates existing conventional well log data and digital cutting analysis. This scoring system has been validated using actual production data, establishing a robust evaluation methodology for assessing exploration potential and optimizing development strategies for carbonate gas reservoirs.
Introduction
Evaluating carbonate reservoirs, particularly the Qixia Formation in China’s Sichuan Basin, is notoriously difficult. These formations are highly heterogeneous, featuring "dual-pore" systems of fractures and vugs (small cavities). Traditional evaluation relies on macroscale well logs, which often lack the resolution to see micro-pores, or physical core samples, which are expensive and difficult to recover in fractured zones where drilling fluid is frequently lost. This data gap makes it hard for engineers to rank reservoirs or make informed development decisions.
Methodology
The study employs a multi-scale integration approach to fill the microscale information gap using "digital cuttings"—the small rock chips produced during drilling—and combining them with standard logging data.
Microscale and Digital Cutting Analysis
The research team utilized X-ray Diffraction (XRD) to identify mineralogy, Micro-CT to reconstruct 3D pore structures at a 1.5 μm resolution, and Scanning Electron Microscopy (SEM) to capture nanoscale porosity. By applying the Lattice Boltzmann method—a fluid dynamics simulation—to the 3D CT models, they calculated absolute permeability, effectively turning rock chips into "digital cores."
Dimensionality Reduction and Scoring
To handle the "curse of dimensionality" from 14 different input parameters, the researchers used Principal Component Analysis (PCA). This identified the most influential factors, which were then combined with an expert-weighted scoring system to create a final "Reservoir Score."
Study Area and Datasets
The study utilized a dataset of 14 drill cutting samples specifically selected from the Middle Permian Qixia Formation in the Gaoshiti-Moxi area of the Sichuan Basin. These samples were chosen to represent both high-potential reservoirs and non-reservoir zones.
Key Results and Insights
- The "90" Threshold: The scoring system produced a clear demarcation line. Wells with a score above 90 were consistently high-producers (reservoirs), while those below 90 showed little to no gas indication (non-reservoirs).
- High Correlation with Production: The scores showed a strong positive correlation with actual daily production volumes. For example, well GS001-X32 scored a 98 and demonstrated substantial gas flow, while well MX108-b scored an 87 and showed no gas.
- Dominance of Micro-Porosity: Through PCA, the researchers found that micron-scale porosity is the single most important variable, carrying a weight of 38.4% in the final score.
- Efficiency: By reducing 14 parameters down to 7 primary controlling factors (Micron-scale porosity, CNL, average pore throat radius, micron permeability, dolomite content, DEN, and AC), the model becomes computationally efficient for rapid field evaluation.
Conclusion and Future Work
This paper demonstrates that with micro-CT and SEM, drill cuttings can provide high-fidelity petrophysical data that macro-logs cannot capture. The successful integration of micro-parameters with macro-parameters provides a holistic view necessary for predicting gas flow in complex carbonates. Future work will focus on expanding the system to evaluate unconventional shale gas reservoirs and automating the analysis pipeline to provide real-time scoring during the drilling process.
Summary updated