Waterflooding persists as the primary secondary oil recovery method, although its efficiency evaluation is often hindered by limitations from reservoir heterogeneities alongside non-uniform sweep efficiency and uncertainty in predicting well-to-well connectivity. To improve the evaluation of waterflooding, this research develops and validates nine advanced machine learning models, covering key types such as neural networks, linear models, tree-based models, ensemble learning, and support vector machines. This research utilized a dataset of 6592 observations collected across four wells over 2000 days to train and evaluate these machine learning models for fluid production rate forecasting and injector-producer connectivity assessments. The most successful model was AdaBoost since it delivered both a minimum root mean squared error (RMSE) of 0.004 and a coefficient of determination (R²) of 0.99, surpassing traditional methods. Furthermore, SHapley Additive Explanations (SHAP) analysis provided interpretability by quantifying injection wells' relative influence on production rates, facilitating the construction of a connectivity-driven contour map. Traditional evaluation methods of empirical and analytical approaches, as well as capacitance-resistance models and numerical simulators, function under restrictive assumptions, extensive computational requirements, and significant data demands. This study is novel in uniting multiple ML algorithms with SHAP interpretability and a connectivity-driven contour map for injector-producer dynamics, providing a practical, real-time workflow for field-scale waterflood optimization. The results of this study demonstrate that ML models surpass traditional methods in terms of predictive accuracy, computer processing speed, and scalability when applied for real-time waterflood optimization. This study illustrates the transformative potential of ML in reservoir engineering by enabling data-driven decision-making and improving the injection optimization process.

CO₂ flooding is an efficient technique for enhancing oil recovery and carbon sequestration, where accurate estimation of the minimum miscibility pressure (MMP) is crucial. Conventional experiment and numerical simulations are time-cost and often inaccurate. Therefore, it is essential to develop an efficient and accurate model for MMP prediction. In this work, we introduced a genetic algorithm-optimized random forest (GA-RF) model using 260 samples with eight input features, including reservoir temperature (T_R), molecular weight of C₅₊ pseudo-components (MWC₅₊), ratio of the mole percent of volatile to intermediate fractions (vol./int.), and mole fractions of injected gas (CO₂, H₂S, CH₄, C₂-C₄, and N₂). Four algorithms-random forest (RF), multilayer perceptron (MLP), support vector regression (SVR), and gradient boosting decision tree (GBDT) are compared. RF performs best and is further optimized by GA. The resulting GA-RF model outperforms traditional methods, with prediction accuracy of 0.99 and a root mean square error of 0.74. The model was applied to CO₂-hydrocarbons system from different low-permeability reservoirs in China. Results show that the GA-RF model demonstrated an average relative error of 3.43%. These findings highlight that the GA-RF model has great potentials for predicting MMP during CO₂ flooding and improving CO₂ miscible flooding efficiency.

The global depletion of conventional oil and gas resources has led to the exploration of alternative energy resources. Among these resources, oil shale is a fine-grained sedimentary rock with a large amount of kerogen that has attracted increasing attention from the hydrocarbon industry and researchers due to the enormous global reserves of oil shale, which are larger than those of conventional crude oil. However, in situ conversion technologies (ICTs) constitute a practical and environmentally friendly technology for extracting hydrocarbons from oil shale formations without traditional surface mining. This study provides a comprehensive analysis of the primary ICTs developed and evaluated in this field to date. Convection-based approaches, such as steam injections and modified thermal injection (MTI), conduction-based heating approaches, such as Shell's in situ conversion process (ICP) and ExxonMobil's Electrofrac™, and radiant heating techniques that utilize radio frequency and microwave energy, are critically evaluated in this study. Advanced innovations, such as near-critical water extraction, nitrogen-assisted in situ conversion process (NAICP), and terahertz spectroscopy (a diagnostic tool for reservoir evaluation), highlight the advanced direction of this field. Mechanistic insights into pore-fracture network formation, kerogen pyrolysis, and their function in improving permeability and hydrocarbon flow have also been analyzed. Challenges related to energy efficiency, operating costs, and environmental impacts still remain despite these technological advancements. Addressing these limitations by maximizing thermal delivery, catalytic enhancement, and integrated modeling will be crucial for realizing the large-scale viability of ICTs. Future research should focus on increasing the efficiency of energy recovery while maintaining environmental sustainability and economic competitiveness.

Injecting CO₂ into shale oil reservoirs has the dual purpose of enhancing oil recovery rate and geological storage of CO₂. The phase behavior and mass transfer laws of the CO₂-shale oil are crucial for achieving highly efficient production shale oil and gas. Based on an improved vapor-liquid equilibrium (VLE) model in nanopores, which achieves a relative error of less than 3.16%, this study investigates the phase behavior and CO₂-oil mass transfer laws of two representative shale oils under CO₂ injection. Research results show that after CO₂ injection, the saturation pressure tends first to increase and then reduce. The CO₂ dissolution results in a significant reduction in the molecular weight and viscosity of Wolfcamp shale oil. During the multi-contact shale oil with CO₂ injection, both density and the interfacial tension (IFT) reduce at the forward contact edge, indicating forward-contact miscibility. Furthermore, the nanoconfinement effect reduces the density difference between the liquid and gas phases, leading to a reduction in the IFT, which is beneficial for gas-oil miscibility. The study results can provide a theoretical basis for understanding compositional variations in shale oils under CO₂ injection, offering promising improvements in miscibility and recovery efficiency.

To enable the precise prediction and control of erosion in hydraulic fracturing pipelines, this study employs a CFD-DEM method coupled with a wear model to quantitatively analyze the influence of shear-thinning fracturing fluids on proppant transport and elbow erosion. The study reveals a critical, non-monotonic effect of fluid concentration on erosion. Compared to the water baseline, the low-concentration (0.2%) fluid, by enhancing particle impact kinetic energy, counter-intuitively caused the maximum elbow erosion rate to surge by approximately 40%. However, as the concentration was further increased to 0.4%, the fluid's effective viscosity became dominant, causing the maximum erosion rate to plummet relative to the 0.2% case, ultimately falling to a level ~8% below the water baseline. More importantly, the high-concentration fluid altered particle trajectories, leading to a significant shift of the peak erosion zone from the elbow's midpoint (~77°) to its outlet (90°). This finding quantifies the complex impact of fracturing fluid rheology on erosion and provides critical data and a new perspective for proactively managing pipeline integrity by optimizing the fluid's non-Newtonian properties.

The drilling of non-vertical wellbores requires effective real-time monitoring techniques to improve drilling efficiency. This study innovatively develops nine robust hybrid machine learning (HML) models to develop a real-time surrogate model for the drilling cuttings transport index (DCTI). The DCTI, a comprehensive indicator of hole-cleaning efficiency, is typically calculated with a 2-4 h delay due to its dependency on laboratory-measured fluid properties. Our hybrid ML models overcome this limitation by accurately emulating the DCTI value using only real-time, surface-recorded drilling data. A 6818-record dataset was collected from three oil and gas fields (Eastern Province of Saudi Arabia), comprising three offshore and two onshore wells. Following a meticulous dataset preprocessing, a nondominated sorting genetic algorithm (NSGA-II) was applied with a multi-layer perceptron neural network (MLPNN), identifying seven key features influencing DCTI - inclination angle (ANG), rate of penetration, flow in, depth, torque, stand-pipe pressure (SPP), and revolutions per minute. Extreme learning machine (ELM), MLPNN, and radial basis function neural networks (RBFNN) were hybridized with particle swarm optimization (PSO) and a growth optimizer in two stages: (1) to determine the optimal model structure, and (2) to fine-tune the model hyperparameters to precisely estimate DCTI based on the selected features. The HML models achieved better estimation performance and generalizability than the standalone models. Achieving the lowest root mean square error (RMSE = 0.00344), the RBFNN-PSO model exhibited the best DCTI estimation performance. Analytical scoring, robustness, and Monte Carlo simulation confirmed this model's superior performance, robustness, and high generalization. Shapley additive explanations (SHAP) feature influence analysis revealed that ANG exerted the most influence on the RBFNN-PSO estimations, while SPP had the least impact. During drilling operations, the proposed data-driven approach practically enables drilling crews to diagnose and monitor hole-cleaning performance in real time and improve the rate of penetration. Furthermore, SHAP analysis transforms the model into a prescriptive tool, identifying the primary real-time parameters affecting hole cleaning. This enables drillers to make targeted operational decisions, such as adjusting the rate of penetration or flow rate, to mitigate developing issues.

Offshore extended-reach drilling is widely employed for the efficient development of deepwater hydrocarbon fields. However, the complex subsea formation environment and downhole frictional losses lead to significant inaccuracies and limitations in bottomhole weight on bit (WOB) calculations, severely impacting drilling operations. Therefore, this study proposes an improved bottomhole WOB prediction model that combines CNN-BiLSTM networks with Transformer encoders. By introducing a loss function (named DRE) based on the rate of penetration equation into the intelligent algorithm, four models were constructed: CNN-BiLSTM, CNN-Encoder-BiLSTM, CNN-BiLSTM (DRE), and CNN-Encoder-BiLSTM (DRE). The eight models were compared and evaluated, including both the established models and conventional models. Experimental results show that the newly developed CNN-Encoder-BiLSTM (DRE) model parameters' influence degree basically conforms to the DRE with high generalization capability. On the test well, it achieved MAE of 0.859, MRE of 0.1, MAPE of 10%, MSE of 1.201, and R² of 0.867, demonstrating good performance. The research results prove that the new model is suitable for bottomhole WOB prediction in offshore extended-reach wells, which is expected to improve drilling efficiency and enable subsequent complex situation analysis, providing new ideas for intelligent drilling research in the marine sector.

Degradation of biomaterials under high temperatures and saline conditions leads to deterioration and loss of performance in deep well drilling applications. This research focuses on examining the impact of colloidal and dually modified tapioca starch (CMITS) on filtration characteristics. Mud filtration properties under various conditions of temperatures were evaluated through API and HPHT filter presses. Different rock samples, including sandstone and limestone, each characterized by unique pore size distributions, porosity, and permeability, are utilized to simulate filtration processes. The thermal stability of drilling fluid additives and the interactions between rocks and filtrate are investigated in hot roll oven and aging cells. The findings evaluate the interaction between temperature and starch additives on mud filtration properties. Notably, in sandstone samples, at 200 ℉, Bandera Brown exhibits the lowest filtrate volume (6.2 mL) with DS 0.66 starch, while Boise sandstone produced a higher volume (38.6 mL) due to its inherent permeability. As the temperature rises to 250 ℉, the effects become more pronounced. At 250 ℉, Indiana limestone with DS 0.66 starch shows minimal filtrate loss (5.4 mL), while Boise sandstone faces pore plugging issues resulting in higher filtrate volumes. At 300 ℉, temperature continues to exert change, with Indiana limestone and Bandera Brown displaying minimal filtrate volumes.

Casing window sidetracking is a key technique for tapping remaining oil in small-scale enriched zones, such as minor anticlines and thin interbeds. While reaming-while-drilling (RWD) technology can mitigate issues like wellbore necking and stuck pipe caused by narrow sidetracked holes, conventional RWD tools are often excessively long, failing to meet the demands of high-curvature window side-tracking operations. To address this challenge, this study systematically investigates the structural principles and mechanical performance of hydraulic RWD tools. Through theoretical analysis, structural design, and numerical simulation, a novel micro hydraulic RWD tool activated by differential pressure was developed, followed by prototype fabrication and laboratory testing. The research results demonstrate that: Cutter stress analysis reveals that axial stress on arm is proportional to the cutting arc angle, while cutting resistance inversely correlates with it. An optimal 35° cutting arc angle minimizes resistance and prevents tool failure caused by locking ball bearing damage. The locking mechanism employs a line-line contact design between ball, housing, and sliding sleeve, effectively avoiding stress concentration and plastic strain. Based on hydrodynamic theory, the influence of the sliding sleeve's inner diameter on flow velocity, pressure differential, and total head loss was investigated, establishing a design methodology to ensure stable tool activation while minimizing energy dissipation. Applying thick-walled cylinder elastic theory, an interference-fit length calculation model was developed for the sleeve-housing interface. This methodology effectively prevents premature tool triggering while guaranteeing reliable arm retraction performance. Laboratory tests confirm successful tool activation/deactivation at theoretical flow rates, validating the design methodology. This study provides theoretical foundations for slimhole RWD tool design, enabling customized reaming applications. Its technical solutions hold significant economic and practical value for precisely exploiting residual reserves in mature wells and enhancing reservoir recovery efficiency.

Pipelines are an indispensable part of the global infrastructure for energy transportation, and their integrity poses a significant risk as the dynamic loads generated throughout their life cycle might result in catastrophic failures if not properly managed. A systematic review is presented, including the sources, effects, and mitigation measures of dynamic loads classified into four categories: environmental, mechanical, accidental, and human-induced. A thorough literature review highlights computational methods, particularly finite element method and computational fluid dynamics, contribute to estimating failure modes and developing preventive measures. Primary mitigation measures comprise flexible joints for the seismic response of pipelines, vortex-induced vibration suppressors for offshore applications, and protective barriers against impact loads. There are gaps in modeling coupled multiaxial loads and aging infrastructure under seismic loading conditions, with limitations in real-time monitoring. Adaptive computational frameworks integrating FSI with probabilistic risk assessments should be developed, with innovative technologies such as self-healing coatings and IoT-enabled monitoring requiring rigorous attention. This study aims to serve as a guide to enhancing pipeline resilience in extreme environments, thus directing future endeavors toward safe and sustainable transportation of energy through resilient infrastructures.