Tectonic Fractures Modeling Based on Multi-Source Data Fusion in Deep Carbonate Reservoirs, Northeast Sichuan Basin, China
Deep carbonate reservoirs exhibit pronounced multiscale heterogeneity, driven by polyphase fault activation and matrix heterogeneity, which collectively control the stochastic hierarchical spatial distribution of fracture networks. The construction of high-fidelity discrete fracture network (DFN) mo...
| Published in: | Lithosphere |
|---|---|
| Main Authors: | , , , , , |
| Format: | Article |
| Language: | English |
| Published: |
GeoScienceWorld
2025-06-01
|
| Online Access: | https://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/doi/10.2113/2025/lithosphere_2024_245/659390/lithosphere_2024_245.pdf |
| Summary: | Deep carbonate reservoirs exhibit pronounced multiscale heterogeneity, driven by polyphase fault activation and matrix heterogeneity, which collectively control the stochastic hierarchical spatial distribution of fracture networks. The construction of high-fidelity discrete fracture network (DFN) models proves critical for deciphering fracture propagation mechanisms and mitigating characterization uncertainties in tectonically overprinted reservoirs. This study proposes a novel multi-constraint fusion framework integrating geological, geomechanical, petrophysical, and 3D seismic datasets, validated through its application to the Feixianguan Formation carbonate reservoir in the Puguang Gas Field, northeastern Sichuan Basin. It employs a multi-data fusion approach, combining geological, geomechanical, well-based, and seismic data constraints for DFN modeling. The geomechanical-based data integrate quantitative predictions of fractures based on the tectonic stress field, considering multi-period faults and different matrix types. The geological-based data utilizes quantitative predictions based on proximity to faults. The well-based data uses image logging in conjunction with array sonic logging to predict fractures. The seismic-based data uses several fracture-sensitive attributes for fracture prediction. The modeling results indicate that fracture development is influenced by the interaction of fault periods, distance from faults and matrix types. Numerical simulation of gas reservoirs using this integrated model shows a 5% improvement in the historical fit rate, providing further validation of its accuracy. |
|---|---|
| ISSN: | 1941-8264 1947-4253 |