The microscopic pore throat structure of shale reservoir rocks directly affects the reservoir seepage capacity. The occurrence and flow channels of shale gas are mainly micron–nanometer pore throats. Therefore, to clarify the microstructural characteristics and influencing factors of the deep organic-rich shales, a study is conducted on the marine shale from the Upper Silurian to Lower Ordovician Wufeng–Longmaxi Formation in the southern Sichuan Basin. Petrographic lithofacies division is carried out in combination with petro-mineralogical characteristics, and a high-resolution scanning electron microscope, low-temperature nitrogen and low-temperature carbon dioxide adsorption, and micron-computed tomography are used to characterize the mineral composition and pore structure qualitatively and quantitatively, upon which the influencing factors of the microstructure are further analyzed. The results show that with the increase in burial depth, the total organic carbon content and siliceous mineral content decrease in the Wufeng formation to Long-1₁ subsection deep shale, while clay mineral content increases, which corresponds to the change in sedimentary environment from anoxic to oxidizing environment. Unexpectedly, the total pore volume of deep shale does not decrease with the increase in burial depth but increases first and then decreases. Using total organic carbon (TOC), siliceous mineral content showed a good correlation with total pore volume and specific surface area, with correlation coefficients greater than 0.7, confirming the predominant role of these two factors in controlling the pore structure of deep shales. This is mainly because the Longmaxi shale is already in the late diagenetic stage, and organic matter pores are generated in large quantities. Clay minerals have a negative correlation with the total pore volume of shale, and the correlation coefficient is 0.7591. It could be that clay minerals are much more flexible and are easily deformed to block the pores under compaction. In addition, the longitudinal heterogeneity of the deep shale reservoir structure in southern Sichuan is also controlled by the thermal effect of the Emei mantle plume on hydrocarbon generation of organic matter and the development of natural microfractures promoted by multistage tectonic movement. Overall, the complex microstructure in the deep shales of the Longmaxi Formation in the southern Sichuan Basin is jointly controlled by multiple effects, and the results of this research provide strong support for the benefit development of deep shale gas in southern Sichuan Basin.

To address the limitations of traditional methods in modeling complex nonlinear relationships in horizontal in situ stress prediction for shale reservoirs, this study proposes an integrated framework that combines well logging interpretation with machine learning to accurately predict horizontal in situ stress in shale reservoirs. Based on the logging data from five wells in the Luzhou Block of the Sichuan Basin (16,000 samples), Recursive Feature Elimination (RF-RFE) was used to identify nine key factors, including Stoneley wave slowness and caliper, from 30 feature parameters. Bayesian optimization was employed to fine-tune the hyperparameters of the XGBoost model globally. Results indicate that the XGBoost model performs optimally in predicting maximum horizontal principal stress (SHmax) and minimum horizontal principal stress (SHmin). It achieves R² values of 0.978 and 0.959, respectively, on the test set. The error metrics (MAE, MSE, RMSE) of the XGBoost model are significantly lower than those of SVM and Random Forest, demonstrating its precise capture of the nonlinear relationships between logging parameters and in situ stress. This framework enhances the model’s adaptability to complex geological conditions through multi-well data training and eliminating redundant features, providing a reliable tool for hydraulic fracturing design and wellbore stability assessment in shale gas development.

The geological conditions of shale gas in the deep layers (burial depth of 3500-4000m) of the northern area of Luzhou block are superior, but the shale gas wells in the work area have frequent casing deformation and poor gas well performance. How to scientifically divide the geological-engineering sweet spots of shale gas reservoirs in the work area determines the effective development of the gas field. Starting from the geological-engineering conditions of shale reservoir, this article conducts an analysis of the factors affecting the geological-engineering sweet spots in shale matrix. The research shows that the geological-engineering conditions in the work area are complex, with fault develop showing characteristics of multiple stages, levels, and types. The lower layer of Type I reservoirs has a large continuous thickness (7-18m), excellent reservoir quality, and developed unidirectional and network fractures; The reservoir has high brittleness content, high Young's modulus (3.685~5.004 × 104MPa), low Poisson's ratio (0.195~0.315), and good compressibility of shale; However, the reservoir stress demonstrates complex characteristics such as large differences in horizontal stress, rapid changes in direction, and a strike-slip stress state, which makes faults prone to slippage. Based on a comprehensive analysis of geological conditions, “continuous thickness of type I reservoirs, fault scale, and natural fracture type characteristics” are selected as evaluation indicators, and the work area is divided into three types of geological-engineering sweet spots of the shale matrix: “Zone A of Type I, Zone B of Type I, and Type II”; Based on a comprehensive analysis of fault slip distribution are chosen as the main indicators for evaluating engineering risks, which are divided into high, medium, and low categories; Coupling geological sweet spots of shale matrix geological-engineering sweet spots with engineering risk areas, the geological-engineering sweet spots of shale gas reservoirs in the work area are divided into four categories: I, II, III, and IV.

The deep formations (burial depth: 3500–4000 m) in the northern Luzhou Block boast favorable geological conditions for shale gas accumulation. However, field development is hindered by the frequent casing deformation of shale gas wells and significant variations in single-well productivity. These issues severely restrict the efficient development of shale gas resources. Existing studies mainly focus on the identification and optimization of geo-engineering dual sweet spots, but few have established a systematic and comprehensive evaluation system from the perspective of engineering risk prevention and control. Based on traditional research on geo-engineering dual sweet spots, this study integrates engineering risk factors. It innovatively establishes a geo-engineering dual sweet spot evaluation system that incorporates engineering risks. Four key evaluation indicators for shale matrix geo-engineering sweet spots are selected: the continuous thickness of a Class I reservoir, the structural location, the fault scale, and natural fracture characteristics. Accordingly, shale matrix geo-engineering sweet spots are classified into three categories: Class I-A Area, Class I-B Area, and Class II Area. Meanwhile, three key indicators affecting fault slip—the angle between fractures and the maximum horizontal in situ stress direction, fracture dip angle, and friction coefficient—are optimized to establish the fault slip risk evaluation criteria. Combined with the distribution characteristics of slip faults, the engineering risks are divided into three levels: high, medium, and low. Finally, by coupling the geo-engineering sweet spots of a shale matrix with engineering risk zones, the geo-engineering sweet spots of shale reservoirs in the study area are classified into four categories (I, II, III, IV).