Casing damage resulting from sand production in unconsolidated sandstone reservoirs can significantly impact the average production of oil wells. However, the prediction task remains challenging due to the complex damage mechanism caused by sand production. This paper presents an innovative approach that combines feature selection (FS) with boosting algorithms to accurately predict casing damage in unconsolidated sandstone reservoirs. A novel TriScore FS technique is developed, combining mRMR, Random Forest, and F-test. The approach integrates three distinct feature selection approaches—TriScore, wrapper, and hybrid TriScore-wrapper and four interpretable Boosting models (AdaBoost, XGBoost, LightGBM, CatBoost). Moreover, shapley additive explanations (SHAP) was used to identify the most significant features across engineering, geological, and production features. The CatBoost model, using the Hybrid TriScore-rapper G₁G₂ FS method, showed exceptional performance in analyzing data from the Gangxi Oilfield. It achieved the highestaccuracy (95.5%) and recall rate (89.7%) compared to other tested models. Casing service time, casing wall thickness, and perforation density were selected as the top three most important features. This framework enhances predictive robustness and is an effective tool for policymakers and energy analysts, confirming its capability to deliver reliable casing damage forecasts.

There are significant differences in the structural characteristics of conglomerate reservoirs, among which the presence of gravel not only leads to strong heterogeneity of conglomerate, but also has a significant impact on the mechanical properties and fracture development of conglomerate. At present, the influence of the composition of different particle sizes of gravel on the mechanical properties and crack development of conglomerate is still unclear. This article cites the concept of intermediate gradation in soil mechanics to characterize the composition of different particle sizes of gravel, and quantifies the grading characteristics using the curvature coefficient Cc and non-uniformity coefficient C_u as grading parameters. A numerical model of conglomerate based on particle flow PFC2D simulation software is established. The influence of gravel gradation parameters on the mechanical properties and crack propagation of conglomerate is investigated by analyzing the stress-strain curve, compressive strength, elastic modulus, and crack development status. The findings reveal that the compressive strength of conglomerates is primarily influenced by the non-uniformity coefficient C_u and the average particle size of gravel, exhibiting a negative correlation with C_u. Similarly, the elastic modulus of conglomerates correlates with the average particle size of gravel. The number of cracks is affected by both the curvature coefficient C_c and the average particle size of gravel, demonstrating a negative correlation with C_c. Conversely, the number of cracks positively correlates with the complexity of the fractures. These research findings offer novel insights into exploring the mechanical properties and fracture propagation patterns of conglomerate reservoirs.

Sustained casing pressure (SCP) is a crucial issue in the oil and gas production lifecycle. Epoxy resins, exhibiting exceptional compressive strength, ductility, and shear bonding strength, have the potential to form reliable barriers. The injectivity and sealing capacity of the epoxy resin is crucial parameters for the success of shallow remediation operations. This study aimed to develop and assess a novel solid-free resin sealant as an alternative to Portland cement for mitigating fluid leakage. The investigation evaluated the viscosity, compressive strength, and brittleness index of the epoxy resin sealant, as well as its tangential and normal shear strengths in conjunction with casing steel. The flow characteristics and sealing abilities of conventional cement and epoxy resin were comparatively analyzed in cracks. The results showed that the application of a viscosity reducer facilitated control over the curing time of the epoxy resin, ranging from 1.5 to 6 h, and reduced the initial viscosity from 865.53 to 118.71 mPa⋅s. The mechanical properties of the epoxy resin initially increased with a rise in curing agent content before experiencing a minor decrease. The epoxy resin containing 30% curing agent exhibited optimal mechanical properties. After a 14-day curing period, the epoxy resin's compressive strength reached 81.37 MPa, 2.12 times higher than that of cement, whereas the elastic modulus of cement was 2.99 times greater than that of the epoxy resin. The brittleness index of epoxy resin is only 3.42, demonstrating high flexibility and toughness. The tangential and normal shear strengths of the epoxy resin exceeded those of cement by 3.17 and 2.82 times, respectively. In a 0.5 mm-wide crack, the injection pressure of the epoxy resin remained below 0.075 MPa, indicating superior injection and flow capabilities. Conversely, the injection pressure of cement surged dramatically to 2.61 MPa within 5 min. The breakthrough pressure of 0.5 PV epoxy resin reached 7.53 MPa, decreasing the crack's permeability to 0.02 D, a mere 9.49% of the permeability observed following cement plugging. Upon sealing a 2 mm-wide crack using epoxy resin, the maximum breakthrough pressure attained 5.47 MPa, 3.48 times of cement. These results suggest that epoxy resin sealant can be employed safely and effectively to seal cracks in the cement.

The increasing energy demand has pushed oil and gas exploration and development limits to extremely challenging and harsher HTHP (High Temperature and High Pressure) environments. Maintaining wellbore integrity in these environments, particularly in HPHT reservoirs with corrosive gases, presents a significant challenge. Robust risk evaluation and mitigation strategies are required to address these reservoirs' safety, economic, and environmental uncertainties. This study investigates chemo-mechanical properties degradations of class G oil well cement blended with silica fume, liquid silica, and latex when exposed to high temperature (150 ℃) and high partial pressure of CO₂ saturated brine. The result shows that these admixtures surround the cement grains and fill the interstitial spaces between the cement particles to form a dense crystal system of C–S–H. Consequently, the cement's percentage of pore voids, permeability, and the content of alkali compounds reduce, resulting in increased resistance to CO₂ corrosion. Liquid silica, a specially prepared silica suspension, is a more effective alternative to silica fume in protecting oil well cement against CO₂ chemical degradation. Micro-indentation analysis shows a significant deterioration in the mechanical properties of the cement, including average elastic modulus and hardness, particularly in the outer zones in direct contact with corrosive fluids. This study highlights the significance of incorporating admixtures to mitigate the effects of CO₂ corrosion in HPHT environments and provides a valuable technique for quantitatively evaluating the mechanical-chemical degradation of cement sheath.