Summary The characterization of ultradeep shale reservoirs is hindered by the scarcity and limited size of available samples. With this study, we develop a constrained big data nanoindentation analytics framework for the upscaling characterization of shales, integrating mineralogical and microstructural constraints into grid-based nanoindentation data sets. The three dominant shale minerals are further classified into five representative microstructural types based on their distinct microscopic mechanical responses. Significant variations in elastic modulus and fracture toughness are observed among microstructures of the same mineral composition. Compared with conventional composition-based upscaling, the microstructure-constrained model yields results that are in closer agreement with macroscopic laboratory measurements. The error between the upscaling modulus and the static elastic modulus under 30-MPa confining pressure does not exceed 5.38% when indentation depth exceeds 350 nm. The homogenized modulus from the surround effect model differs by merely 2.1% from the dynamic modulus derived from P-wave velocity. Microcrack accumulation around mineral debris clusters (MDCs) is identified as the dominant mechanism controlling fracture propagation in three-point bending (TPB) tests. Moreover, the microscale fracture toughness of MDCs obtained from nanoindentation agrees well with the effective fracture toughness estimated from the upscaling size effect model. These findings establish an upscaling method for ultradeep shale reservoirs and provide a mechanistic basis for optimizing hydraulic fracturing design and shale gas production efficiency.

Summary Cementing plays a critical role in ensuring zonal isolation and long-term well performance, motivating continued evaluation of supplementary cementitious materials (SCMs) to partially replace Portland cement while maintaining oilwell operability and sealing-relevant properties. With this study, we examine the effect of partial replacement of Class G oilwell cement with fly ash (FA) across a range of blend compositions under controlled laboratory conditions. Five cement systems containing 0%, 20%, 40%, 60%, and 80% FA by weight of binder (BWOB) were prepared at a constant water content of 44% BWOB and conditioned at 150°F. An integrated experimental framework was adopted in which all blends were first screened using unconfined compressive strength and splitting tensile strength measurements after 24 hours curing, followed by a detailed characterization of the selected formulation using fresh-state performance, elastic response, and microstructural indicators. Fresh-state testing included rheology, thickening time, free water, and high-pressure, high-temperature (HPHT) fluid loss to assess placement-related behavior. The 60:40 ordinary Portland cement (OPC)/FA blend exhibited mechanical performance comparable to neat cement, with compressive and tensile strengths of 4,873 psi and 587 psi, respectively, while providing improved slurry stability, extended thickening time, and reduced fluid loss. Density profiling and X-ray computed tomography (CT) indicated reduced segregation and improved internal homogeneity, accompanied by a decrease in porosity from 32.3% to 29.8%. Elastic wave measurements showed modest increases in dynamic Young’s modulus and Poisson’s ratio relative to neat cement. The results demonstrate that moderate FA substitution can maintain key operational and mechanical criteria while enhancing slurry stability and microstructural uniformity, supporting its consideration in laboratory-scale formulation screening for oilwell cementing.

Summary Gas drilling is a promising technique for developing tight sandstone reservoirs. Cuttings accumulation in horizontal and deviated gas drilling impairs wellbore cleaning, reducing drilling efficiency and elevating operational risks. A novel model for predicting cuttings bed height in horizontal and deviated gas drilling is proposed. Based on particle dynamics theory, mechanical equilibrium equations are derived for three distinct cuttings initiation mechanisms—sliding, rolling, and lifting—to solve for the critical frictional velocity. Combined with wellbore geometric relationships, the cuttings bed height is predicted using an iterative method. This model is validated by computing the critical annular gas velocity and comparing the results with previous experimental data, yielding average prediction errors of 12.1% and 11.1%. Based on the model, the effects of key parameters, including wellbore inclination, equivalent particle diameter, particle sphericity, exposure level, and gas injection rate, on cuttings bed height are analyzed. The results indicate that wellbore inclination exerts a threshold effect on cuttings bed height. Specifically, cuttings bed height rises as inclination increases up to the 60–65° range, where the bed height peaks and declines as inclination continues to rise beyond this interval. This trend holds across different equivalent particle diameters. Cuttings initiation mode is also consistent across different equivalent particle diameters, adopting the lifting mode at inclinations below 60° and shifting to the rolling mode at inclinations around 60° and above. Equivalent particle diameter also affects cuttings bed height, with particularly pronounced effects in the range of 1–5 mm. Higher particle sphericity promotes accumulation, while higher exposure level and gas injection rate reduce it. This model provides a reliable tool for optimizing drilling parameters to mitigate cuttings bed accumulation, thereby supporting the efficient development of tight sandstone reservoirs.

Summary Fracture-cavity carbonate reservoirs are key targets for ultradeep oil exploration but feature complex structures and highly heterogeneous fluid migration. Fractures and dissolution cavities exert cascading transport effects that significantly influence hydrocarbon movement. In this study, we develop a mesoscale phase-field model based on 3D digital rock to simulate two-phase flow and systematically evaluate the coupled effects of structure, pressure, and temperature on oil migration and accumulation. Key findings include the following: (1) Fluid pressure distribution and pore/fracture connectivity critically affect hydrocarbon migration efficiency. In fractures, pressure decreases linearly along the flow direction, enabling efficient oil transport. In contrast, narrow throats in porous media cause exponential pressure drops, reducing flow velocity and injection efficiency. (2) Increasing pressure gradients activate more reservoir space, but the effect diminishes with scale. In a fractured and porous-type model, increasing the pressure differential from 50 Pa to 200 Pa raises average flow velocity by 194.8% in fractures and 118.1% in pores. (3) Temperature rise has a greater impact in the porous domain. From 80°C to 140°C, average flow velocity and oil phase volume fraction in pores increase by 61.3% and 99.5%, respectively, while in fractures, velocity rises by only 15.2% with minimal change in saturation. These results reveal the nonlinear, synergistic influence of porous structure, pressure, and temperature on oil migration in fracture-cavity systems, providing theoretical insights to support the efficient development of ultradeep carbonate reservoirs.

Summary Extended reach drilling (ERD) is an advanced technique that achieves broad reservoir coverage by significantly extending the well’s horizontal reach. However, drilling these long, high-angle sections faces a significant challenge—wellbore cleaning. Traditional wellbore cleaning strategies form the foundation of hole cleaning by optimizing parameters such as rate of penetration (ROP), drillstring rotation, and fluid rheology. In the demanding context of ERD, however, the cleaning capacity provided by these methods alone may be inadequate to manage cuttings in long horizontal sections, calling for enhanced cleaning techniques. The deployment of cuttings bed remover (CBR) to mechanically disrupt persistent accumulations has proved to be an effective solution. The CBR disrupts the structure of the cuttings bed by generating a high-intensity spiral flow field and alleviates the cuttings accumulation caused by gravitational settling in low-flow velocity zones. However, the underlying mechanisms governing the structure-performance relationships require further systematic investigation. To this end, we develop a 3D flow model by coupling computational fluid dynamics (CFD) with the discrete element method (DEM). Meanwhile, we simulate the migration and sedimentation behaviors of cuttings in the annulus by considering the coupling effect between the drilling fluid and cutting particles, thereby systematically analyzing the influence of the CBR on the morphology, thickness, and removal efficiency of the cuttings bed. Based on the simulation results, the action zone of CBR can be divided into the intense agitation zone and the intense transport zone, with swirl strength and transport intensity used as quantitative evaluation indicators, respectively. Moreover, we discuss the effects of drillpipe rotational speed, CBR structural parameters, and annular flow velocity. Compared with rectangular-type CBR configurations, the deflection-angled CBR design demonstrates marked superiority in performance of generating swirl, exhibiting enhanced generation capacity, although potentially incurring more pressure loss. With this study, we introduce an innovative method for delineating the operational zones of the CBR and quantitatively assessing its effectiveness, thereby establishing a robust theoretical framework for evaluating the impact of CBR. By analyzing swirl intensity and effective action distance in the high-agitation zone, the key parameters of CBR are optimized as blade height (25–27 mm), length (1.5 m), count (4–5), and inclination angle (30–40°). Additionally, a multiunit short-CBR configuration outperforms a single long-unit design.

Summary As part of the energy transition and the aim to reduce greenhouse gas emissions, more carbon in the form of carbon dioxide (CO2) will be captured and stored underground in wells intersecting suitable reservoirs for storage in future. The long-term integrity of such wells is a considerable concern, given that CO2 can react with Portland cement and steel, which can erode well barriers over time. Moreover, low temperatures and temperature cycling in injection and storage wells can lead to cement cracking and debonding from casing, creating annular flow paths for CO2 to surface and allowing for CO2 to attack cement more severely. In this paper, we report on an investigation into using creeping shale formations as alternative annular barriers providing integrity during CO2 injection and long-term storage. Building upon previous work done as part of our ongoing shale-as-a-barrier (SAAB) investigation, rock mechanical laboratory investigations were conducted into the behavior of shale creep in wells experiencing CO2 injection. A special experimental setup was constructed to be able to establish an annular shale barrier at simulated field conditions (using either in-situ formation temperature or thermal stimulation) and then testing this barrier during simulated CO2 injection conditions. During CO2 injection, the well will experience significant reductions and increases in temperature, which can lead to, e.g., the debonding of cement from the casing and the formation of a microannulus that compromises the annular barrier. Note that temperature cycling in wells is a leading cause of the loss of annular isolation and the associated flow to surface of formation fluids and gases in oil and gas wells. In the experiments, shale barriers were first generated and verified at a variety of in-situ and elevated temperatures (which affect shale creep rate). These barriers were then subjected to a significant temperature reduction, as well as temperature cycling with wellbore temperatures reaching a low value of −14°C (7°F), followed by periods where temperature was allowed to rise back up to in-situ conditions. In all cases, the shale barrier continued to function and maintain annular pressure integrity, indicating that well temperature reduction and cycling associated with CO2 injection will not negatively affect it. This is a promising result and insight because the same cannot be guaranteed for a Portland cement barrier. In addition, shale barriers are impervious to any chemical attack by CO2 and are expected to last for a very long time period, given that we are dealing with actual caprock material. Carbon storage wells present new challenges to well construction. These include the low absolute temperatures and large cyclic temperature cycles during CO2 injection, which could lead to cement debonding and microannulus formation providing a pathway for CO2 migration to the surface, as well as chemical attack of cement and casing by CO2 during long-term storage. This work shows that creeping shale formations can deal with both challenges and provide a superior annular isolation solution when compared with conventional Portland cement. The work could have large positive implications for how (barriers in) carbon storage wells will be constructed in the future, and how permanent storage of CO2 underground can be achieved.

Summary Accurate assessment of reservoir health is essential for efficient and sustainable oilfield development. In this study, we develop a data-driven approach to evaluate and classify health status in conventional sandstone reservoirs using machine learning and interpretable analysis. The categorical boosting (CatBoost) ensemble model predicts cumulative oil production based on 15 key development indicators, trained and validated on a data set comprising 42 real blocks and 158 simulated cases from a mature oil field. K-fold cross-validation yields high predictive accuracy [coefficient of determination (R2) = 0.964; root mean squared error (RMSE) = 0.3096]. Shapely additive explanations (SHAP) analysis quantifies feature importance, identifying remaining recoverable reserves per well as the dominant positive driver and water-cut increase rate as the least influential. An enhanced radar chart, integrated with a quantified risk coefficient, effectively visualizes reservoir health status and categorizes warning levels as green (safe), yellow (light warning), and red (heavy warning) in accordance with established industry standards for conventional sandstone reservoirs. The framework effectively detects anomalous development trends, enabling proactive adjustment of injection/production strategies. This interpretable data-driven framework supports rapid decision-making and near-real-time health assessment based on periodically updated field data.

Summary Gas-liquid flow is one of the most observed in nature and in many industrial applications. The presence of dense gas is a special case (i.e., a flow with a liquid/gas viscosity ratio of 10 or even less). Dense-gas/liquid flow is found in the Brazilian offshore production scenario, which is characterized by subsea lines operating under reservoir pressures typically ranging from 100 bar to 500 bar, temperatures around 60°C, and high CO2 content, while upstream of the choke valve pressures commonly range between 30 bar and 160 bar. Despite its relevance, the technical literature shows significant gaps. Although some studies employ dynamic pressure signals to analyze two-phase flows, most rely on numerical simulations or high-frequency experimental systems. Consequently, the analysis and classification of dense-gas/liquid flow using low-frequency differential pressure signals that are precisely the most widely available measurements in offshore operations have not been addressed. Likewise, there are no approaches that integrate multiple signal-to-image encoding techniques with advanced hybrid deep learning architectures for flow-pattern characterization from pressure-gradient signals. To address these gaps, we introduce a methodology that transforms low-frequency pressure-gradient time series into multivariate images using four complementary encoding techniques. Gramian angular fields (GAFs) [Gramian angular summation field (GASF)/Gramian angular difference field (GADF)] capture temporal correlations and angular relationships; recurrence plots (RPs) reveal recurrence patterns and nonlinear dynamic structures; and Markov transition fields (MTF) model state-transition probabilities while preserving the statistical structure of the signal. These representations are integrated into a unified 2 × 2 mosaic, generating a final image of size 2m × 2n. In addition, a hybrid deep learning model convolutional neural network (CNN)-transformer with coordinated attention is proposed, designed to leverage both local features and global dependencies. The convolutional modules extract fine spatial patterns, the Swin Transformer blocks learn long-range relationships through hierarchical self-attention, and the coordinated attention module preserves directional information by recalibrating feature responses along the horizontal and vertical axes. This combination enables the model to capture rapid local variations in the pressure gradient as well as global hydrodynamic trends, both essential in complex multiphase systems. By relying exclusively on low-frequency pressure-gradient signals, which are typically available in offshore monitoring systems, the proposed method offers strong industrial applicability. The results show that the hybrid model achieves 99% accuracy in classifying dense-gas/liquid flow patterns, significantly outperforming traditional approaches based on physical correlations or statistical analysis. These advancements position the methodology as a significant contribution to multiphase-flow characterization in the oil and gas industry.

Summary The effects and mechanisms of three nanomaterials with different reactivities—nano-silica sol (NSS), nano-alumina (NA), and nano-calcium carbonate (NC)—on sand-enhanced oilwell cement cured at 200°C and 20 MPa for 1–28 days were investigated. Incorporation of NSS markedly increased the amorphous fraction, reaching approximately 60% at 5 days and 55% at 12 days of curing. The addition of 1% NC induced the formation of more xonotlite (C6S6H) crystalline phases, with the mass fraction increasing from 18% (5 days) to 28% (12 days). Both additives improved the compactness of the cement after early-stage curing and significantly enhanced the early compressive strength of the sand-added cement. Incorporation of 2.5% NA facilitated the formation of tobermorite (C5S6H), peaking at 23% at 5 days of curing; nevertheless, it produced negligible changes in the mechanical properties of the sand-added cement. Excess NSS (20%) severely inhibited cement hydration. The addition of excessive NSS and NA led to the formation of large numbers of ~500-nm spherical calcium-silicate-hydrate (C–S–H) precipitates (Ca:Si = 1:1) that covered the surfaces of other hydration products. This phenomenon correlates with lower compressive strength.

Summary Accurate outbound volume (OV) prediction of refined oil depots is vital for efficient energy distribution and logistics management. However, refined oil depots typically store multiple types of oil products, the OVs of which exhibit strong correlations and significant time-varying characteristics, with sparse data availability in some depots. Existing methods fail to deliver accurate predictions as they overlook inherent interdependencies and latent temporal dynamics among product volumes, and struggle with effective knowledge transfer under sparse data conditions. To address these challenges, we propose an innovative multitask learning (MTL) framework. First, a multivariate data decomposition module incorporating fast Fourier transform (FFT) and wavelet decomposition (WD) to optimize multivariate variational mode decomposition (MVMD) is designed, enhancing the capability of extracting multidimensional time-varying properties from OV data. Next, an improved multigate mixture-of-experts (IMMOE) network is introduced to capture dynamic patterns, refine critical temporal features, and effectively model complex dependencies across various refined products, thus achieving accurate OV predictions. Finally, a progressive unfreezing transfer learning strategy is proposed to mitigate prediction accuracy degradation under limited data conditions. Experimental results on real-world data sets demonstrate that the proposed framework achieves more than 50% reduction in error across all evaluation metrics compared with advanced baseline models. Additionally, the transfer learning strategy achieves R2 and Nash-Sutcliffe efficiency values exceeding 0.9 in data-scarce depots. Sensitivity analysis confirms the contributions of each component and emphasizes the more significant importance of multidimensional time-varying feature extraction in performance improvement of OV prediction compared with the temporal feature extraction module.