Since shale gas is a valuable energy resource, effective planning for its extraction and utilization depends on precise forecasting of gas well production. Conventional models need long computation time, a wide range of geological and fluid data, and suffer from unstable predictions. To develop a low-cost, intelligent, and reliable forecast system for shale gas production, a hybrid Temporal Convolutional Network-Policy Gradient Informer (TCN-PgInformer) model was constructed for multivariate production prediction research. This model is based on the Informer model of its own unique self-attention mechanism, which lowers the temporal complexity of conventional self-attention technique while increasing the model's accuracy. Meanwhile, to completely avoid the gradient vanishing problem, the dilated convolutions of TCN structure are employed to extract the long-term dependency relationships. Ultimately, a policy gradient (Pg) algorithm is introduced to enhance the parameter training speed. The results indicate that the daily gas production may be accurately predicted by TCN-PgInformer model. A detailed performance comparison was carried out among TCN-PgInformer, CNN, GRU and CNN-LSTM models in the literature. The comparison demonstrates that the suggested TCN-PgInformer model outperforms existing techniques. For four different gas production stages, the MAPE/RMSE error of other models is 2–12 times higher than that of the TCN-PgInformer model, while the R² accuracy of TCN-PgInformer model can be as high as 1 time higher than other models. Therefore, the designed model has excellent applicability, which offers reference and guidance for shale gas development.

As CO₂ injection can enhance the efficiency of shale oil extraction and reduce CO₂ emissions, it has been utilized widely in the development of shale oil resources. Minimum miscible pressure is an important parameter describing the miscibility of CO₂ and shale oil, which is of great importance for determination of CO₂ injection strategy. However, due to the unclear phase boundary caused by the confinement effect in shale nanopores, it is difficult to determine the minimum miscible pressure of CO₂ and shale oil. In this study, a new minimum miscible pressure estimation method is constructed, that is suitable for nanopores based on the significant co-evolution of pore wall adsorption and confined-bulk phase interactions. This method can mitigate the limitations of traditional minimum miscible pressure calculation methods relying on fluid interfaces. Furthermore, the confinement effects on the miscibility process are analyzed using a theoretical method and molecular dynamics simulation on the microscopic scale. The results demonstrate that the minimum miscible pressure of CO₂ and shale oil initially decreases as the pore size decreases. When the pore size decreases to a certain extent, the minimum miscible pressure increases with the thickness of the adsorbent layer rising and the CO₂ diffusion coefficient decreasing. Temperature elevation raises the minimum miscible pressure as it intensifies molecular thermal motion, weakens fluid adsorption, and reduces interaction energy, which are not conducive to miscibility. This study can provide an essential basis for the optimization of CO₂ injection pressure in shale oil reservoir development.

As the main factor influencing the flow and preservation of underground fluids, wettability has a profound impact on CO₂ sequestration (CS). However, the influencing factors and internal interaction mechanisms of shale kerogen wettability remain unclear. In this study, we used molecular dynamics to simulate the influence of temperature, pressure, and salinity on wettability. Furthermore, the results were validated through various methods such as mean square displacement, interaction energy, electrostatic potential energy, hydrogen bonding, van der Waals forces, and electrostatic forces, thereby confirming the reliability of our findings. As temperature increases, water wettability on the surface of kerogen increases. At CO₂ pressures of 10 and 20 MPa, as the temperature increases, the kerogen wettability changes from CO₂ wetting to neutral wetting. As the CO₂ pressure increases, the water wettability on the surface of kerogen weakens. When the pressure is below 7.375 MPa and the temperature is 298 or 313 K, kerogen undergoes a wettability reversal from neutral wetting to CO₂ wetting. As salinity increases, water wettability weakens. Divalent cations (Mg²⁺ and Ca²⁺) have a greater impact on wettability than monovalent cations (Na⁺). Water preferentially adsorbs on N atom positions in kerogen. CO₂ is more likely to form hydrogen bonds and adsorb on the surface of kerogen than H₂O. As the temperature increases, the number of hydrogen bonds between H₂O and kerogen gradually increases, while the increase in pressure reduces the number of hydrogen bonds. Although high pressure helps to increase an amount of CS, it increases the permeability of a cap rock, which is not conducive to CS. Therefore, when determining CO₂ pressure, not only a storage amount but also the storage safety should be considered. This research method and results help optimize the design of CS technology, and have important significance for achieving sustainable development.

CO₂ injection into oil reservoirs is expected to achieve enhanced oil recovery along with the benefit of carbon storage, while the application potential of this strategy for shale reservoirs is unclear. In this work, a numerical model for multiphase flow in shale oil reservoirs is developed to investigate the impacts of nano-confinement and oil composition on shale oil recovery and CO₂ storage efficiency. Two shale oils with different maturity levels are selected, with the higher-maturity shale oil containing lighter components. The results indicate that the saturation pressure of the lower-maturity shale oil continues to increase with increasing CO₂ injection, while that of the higher-maturity shale oil continues to decrease. The recovery factor and CO₂ storage rate for higher-maturity shale oil after CO₂ huff-n-puff are 12.02% and 44.76%, respectively, while for lower-maturity shale oil, these are 4.41% and 69.33%, respectively. These data confirm the potential of enhanced oil recovery in conjunction with carbon storage in shale oil reservoirs. Under the nano-confinement impact, a decrease in the oil saturation in the matrix during production is reduced, which leads to a significant increase in oil production and a significant decrease in gas production. The oil production of the two kinds of shale oil is comparable, but the gas production of higher-maturity shale oil is significantly higher. Nano-confinement shows a greater impact on the bubble point pressure of higher-maturity shale oil and a more pronounced impact on the production of lower-maturity shale oil.

Methane adsorption is a critical assessment of the gas storage capacity (GSC) of shales with geological conditions. Although the related research of marine shales has been well-illustrated, the methane adsorption of marine-continental transitional (MCT) shales is still ambiguous. In this study, a method of combining experimental data with analytical models was used to investigate the methane adsorption characteristics and GSC of MCT shales collected from the Qinshui Basin, China. The Ono-Kondo model was used to fit the adsorption data to obtain the adsorption parameters. Subsequently, the geological model of GSC based on pore evolution was constructed using a representative shale sample with a total organic carbon (TOC) content of 1.71%, and the effects of reservoir pressure coefficient and water saturation on GSC were explored. In experimental results, compared to the composition of the MCT shale, the pore structure dominates the methane adsorption, and meanwhile, the maturity mainly governs the pore structure. Besides, maturity in the middle-eastern region of the Qinshui Basin shows a strong positive correlation with burial depth. The two parameters, micropore pore volume and non-micropore surface area, induce a good fit for the adsorption capacity data of the shale. In simulation results, the depth, pressure coefficient, and water saturation of the shale all affect the GSC. It demonstrates a promising shale gas potential of the MCT shale in a deeper block, especially with low water saturation. Specifically, the economic feasibility of shale gas could be a major consideration for the shale with a depth of <800 m and/or water saturation >60% in the Yushe-Wuxiang area. This study provides a valuable reference for the reservoir evaluation and favorable block search of MCT shale gas.

Thermal recovery techniques for producing oil sands have substantial environmental impacts. Surfactants can efficiently improve thermal bitumen recovery and reduce the required amount of steam. Such a technique requires solid knowledge about the interaction mechanism between surfactants, bitumen, water, and rock at the nanoscale level. In particular, oil sands ores have extremely complex mineralogy as they contain many clay minerals (montmorillonite, illite, kaolinite). In this study, molecular dynamics simulation is carried out to elucidate the unclear mechanisms of clay minerals contributing to the bitumen recovery under a steam–anionic surfactant co-injection process. We found that the clay content significantly influenced an oil detachment process from hydrophobic quartz surfaces. Results reveal that the presence of montmorillonite, illite, and the siloxane surface of kaolinite in nanopores can enhance the oil detachment process from the hydrophobic surfaces because surfactant molecules have a stronger tendency to interact with bitumen and quartz. Conversely, the gibbsite surfaces of kaolinite curb the oil detachment process. Through interaction energy analysis, the siloxane surfaces of kaolinite result in the most straightforward oil detachment process. In addition, we found that the clay type presented in nanopores affected the wettability of the quartz surfaces. The quartz surfaces associated with the gibbsite surfaces of kaolinite show the strongest hydrophilicity. By comparing previous experimental findings with the results of molecular dynamics (MD) simulations, we observed consistent wetting characteristics. This alignment serves to validate the reliability of the simulation outcomes. The outcome of this paper makes up for the lack of knowledge of a surfactant-assisted bitumen recovery process and provides insights for further in-situ bitumen production engineering designs.

Utilizing CO₂ to enhance shale oil recovery has a huge potential and thus has gained widespread popularity in recent years. However, the microscopic mechanisms of CO₂ enhancing shale oil recovery remain poorly understood. In this paper, the molecular dynamics simulation method is adopted to investigate the replacement behavior of CO₂ in shale oil reservoirs from a micro perspective. Three kinds of $n$-alkanes are selected as the simulative crude oil in silica nanopores. Molecular dynamics models are established to study the occurrence patterns of different alkanes on the rock surface and the alkane-stripping characteristics of CO₂. The fluid density, mean square displacement and centroid variation are evaluated to reveal the effect of CO₂ on alkanes. The results indicate that different alkanes exhibit varying occurrence characteristics of oil film on the rock surface of the shale reservoir. Specifically, a higher carbon number leads to a thicker oil film. Through the alkane molecular gaps, CO₂ penetrates the alkane molecular system and reaches the rock surface to effectively strip the oil film of different alkane molecules. CO₂ will more readily mix with the stripped oil molecules and displace them from the rock surface when the carbon number is small. The process for CO₂ replacing crude oil on the rock surface can be divided into four typical stages, namely, CO₂ diffusion, competitive adsorption, emulsification and dissolution, and CO₂-alkanes miscible phase (for light alkanes). This study contributes to the improvement of micro-scale enhanced oil recovery mechanisms for shale oil via CO₂ injection and provides a guidance for enhancing shale oil recovery by using CO₂.

Accurate construction of a seepage model for a multifractured horizontal well in a shale gas reservoir is essential to realizing the forecast of gas well production, the pressure transient analysis, and the inversion of the postfracturing parameters. This study introduces a method for determining the fracture control region to characterize the flow area of the matrix within the hydraulic fracture network, distinguishing the differences in the flow range of the matrix system between the internal and external regions caused by the hydraulic fracture network structure. The corresponding derivation and solution methods of the semi-analytical seepage model for fractured shale gas well are provided, followed by the application of case studies, model validation, and sensitivity analysis of parameters. The results indicate that the proposed model yields computational results that closely align with numerical simulations. It is observed that disregarding the differentiation of matrix flow area between the internal and external regions of the fracture network led to an overestimation of the estimated ultimate recovery, and the boundary-controlled flow period in typical well testing curves will appear earlier. Because hydraulic fracture conductivity can be influenced by multiple factors simultaneously, conducting a sensitivity analysis using combined parameters could lead to inaccurate results in the inversion of fracture parameters.

A comprehensive dataset from 594 fracturing wells throughout the Duvernay Formation near Fox Creek, Alberta, is collected to quantify the influences of geological, geomechanical, and operational features on the distribution and magnitude of hydraulic fracturing-induced seismicity. An integrated machine learning-based investigation is conducted to systematically evaluate multiple factors that contribute to induced seismicity. Feature importance indicates that a distance to fault, a distance to basement, minimum principal stress, cumulative fluid injection, initial formation pressure, and the number of fracturing stages are among significant model predictors. Our seismicity prediction map matches the observed spatial seismicity, and the prediction model successfully guides the fracturing job size of a new well to reduce seismicity risks. This study can apply to mitigating potential seismicity risks in other seismicity-frequent regions.

On 2019-03-04, the largest induced earthquake (M_L4.18) occurred in the East Shale Basin, Alberta, and the underlying physical mechanisms have not been fully understood. This paper proposes a synthetical geoengineering methodology to comprehensively characterize this earthquake caused by hydraulic fracturing. Based on 3D structural, petrophysical, and geomechanical models, an unconventional fracture model is constructed by considering the stress shadow between adjacent hydraulic fractures and the interactions between hydraulic and natural fractures. Coupled poroelastic simulations are conducted to reveal the triggering mechanisms of induced seismicity. It is found that four vertical basement-rooted faults were identified via focal mechanisms analysis. The brittleness index (BI) along two horizontal wells has a high magnitude (BI > 0.5), indicating the potential susceptibility of rock brittleness. Due to the presence of overpressure, pre-existing faults in the Duvernay Formation are highly susceptible to fault reactivation. The occurrence of the earthquake clusters has been attributed to the fracturing fluid injection during the west 38^th-39^th stage and east 38^th stage completions. Rock brittleness, formation overpressure, and large fracturing job size account for the nucleation of earthquake clusters, and unconventional natural-hydraulic fracture networks provide fluid flow pathways to cause fault reactivation. This workflow can be used to mitigate potential seismic risks in unconventional reservoirs in other fields.