Unconventional natural gas development requires a deeper insight into how CH₄ and CO₂ adsorb and diffuse in the pores of coal seams. Graphene (GRA) is frequently employed in microscopic mechanism simulations on coal surfaces because its structure closely resembles that of the coal seam matrix. In this study, molecular dynamics simulations were conducted to systematically investigate the diffusion, adsorption, and desorption behaviors of CH₄ and CO₂ within the pore system of hydrated graphene under three representative temperature and pressure conditions: 190 K-6 MPa, 298 K-0.1 MPa, and 320 K-8 MPa. The results show that heatinfg and depressurization significantly enhance the diffusion ability of gas molecules and promote their desorption from the graphene surface. Low temperature and high pressure are conducive to the formation of a stable adsorption layer, and more hydrogen bond structures are formed between CO₂ and water. However, under high-temperature conditions, this ordered structure is significantly weakened. The density distribution further reveals the spatial distribution characteristics of water molecules and gases and their evolution trends with changes in temperature and pressure. This research is conducive to a deeper understanding of the multiphase behavior of coalbed methane and its regulatory mechanism, providing theoretical support for the gas storage and displacement processes.

Accurate prediction of electricity consumption in crude oil pipeline transportation is of significant importance for optimizing energy utilization, and controlling pipeline transportation costs. Currently, traditional machine learning algorithms exhibit several limitations in predicting electricity consumption. For example, these traditional algorithms have insufficient consideration of the factors affecting the electricity consumption of crude oil pipelines, limited ability to extract the nonlinear features of the electricity consumption-related factors, insufficient prediction accuracy, lack of deployment in real pipeline settings, and lack of interpretability of the prediction model. To address these issues, this study proposes a novel electricity consumption prediction model based on the integration of Grid Search (GS) and Extreme Gradient Boosting (XGBoost). Compared to other hyperparameter optimization methods, the GS approach enables exploration of a globally optimal solution by exhaustively evaluating all hyperparameter combinations, thereby enhancing the model's predictive performance. To improve the interpretability of the model's predictions, SHapley Additive exPlanations (SHAP) is employed to quantify the contribution of input parameters to the model's predictive outcomes. Experimental results demonstrate that the GS-XGBoost hybrid model outperforms Multilayer Perceptron, Support Vector Machine, Extreme Learning Machine, Random Forest, and Gradient Boosting Regression Tree, achieving the best predictive performance with mean absolute percentage error (MAPE) and coefficient of determination (R²) values of 4.1% and 0.98, respectively. SHAP analysis identifies the key determinants of electricity consumption in descending order of importance: daily transport volume, average pump-out pressure, next station inlet pressure, average converging pressure, ground temperature, next station inlet temperature, and average output temperature. The proposed model holds substantial practical significance for improving energy utilization efficiency, optimizing cost management, and accelerating digital transformation in pipeline operations.

Natural gas hydrates are undoubtedly an important clean energy source in the future. However, its temperature sensitivity poses a significant challenge to the stability of the wellhead and wellbore during the drilling process of hydrate-bearing sediments. In this study, the evolution characteristics of key parameters, including hydrate dissociation, bearing capacity and wellhead displacement, were analyzed using a thermo-hydro-mechanical-chemical coupled model. The results show that hydrate dissociation leads to a ∼40% reduction in bearing capacity, resulting in a final wellhead displacement of 4.1 cm under the base case. Meanwhile, the effects of factors such as hydrate saturation and drilling fluid temperature on wellhead stability were then analyzed. It was found that higher hydrate saturation in sediment and lower drilling fluid temperature significantly enhances wellhead stability. Specifically, increasing hydrate saturation from 20% to 60% reduces the final wellhead displacement from 5.4 cm to 1.7 cm, while raising drilling fluid temperature from 15 °C to 30 °C increases displacement from <0.2 cm to 7.5 cm. Furthermore, the effectiveness of vacuum insulation casing in maintaining wellbore stability was numerically evaluated through numerical simulation. The simulation results indicate that, hydrate dissociation is almost completely suppressed, bearing capacity remains stable, and wellhead displacement is nearly 0 cm when vacuum insulation casing is employed. This study provides valuable guidance for ensuring the safe and efficient drilling operation in hydrate-bearing sediment.

Molecular dynamics simulations were conducted to examine how methylated and hydroxylated coal molecules interact with SiO ₂ surfaces bearing corresponding functional groups. Coal adsorption and wetting strongly depended on both coal functional groups and substrate chemistry. Methylated coal (coalCH ₃ ) formed dense, planar clusters on hydrophobic SiO ₂ –CH ₃ , achieving nearly complete surface coverage (contact angles: 58.5°–67.7° at the surface; 81.9o–94.4° at the tetrahedral bottom), whereas on hydrophilic SiO2–OH it formed sparse spherical or rod-like aggregates (surface: 88.4°–130.6°; bottom: 64.4°–114.4°), indicating poor wettability. Hydroxylated coal (coalOH) exhibited chain-like clusters on SiO ₂ –CH ₃ evolving into continuous adsorption layers, while on SiO ₂ –OH it formed interconnected strip-like structures with limited spreading (surface: 62.7°–113.2°; bottom: 69.9°–71°). The energy decomposition results indicated that the interaction in coalCH ₃ systems is primarily governed by van der Waals forces, whereas electrostatic contributions play a dominant role in stabilizing coalOH structures, highlighting the influence of surface functionalization on adsorption and wetting behaviors.