SPE Journal最新文献

The logging-while-drilling (LWD) ultradeep azimuthal electromagnetic tool plays a pivotal role in real-time drilling optimization operations. Established tool designs include arrays of coaxial and tilted coils that, during drilling operations, can be processed to a multicomponent magnetic induction data. These data can then be combined into different detection modes, which accentuate sensitivity to particular geological features. Leveraging the established coil design and definitions of various detection modes for an electromagnetic look-ahead (EMLA) tool, this study undertakes a comprehensive exploration of the disparities in detection performance and characterization of subsurface parameters. Through sensitivity analysis, the varying degrees of sensitivity exhibited by these detection modes concerning parameters such as subsurface formation resistivity, formation inclination, and electrical anisotropy have been investigated. The ensuing conclusions derived from an in-depth analysis are as follows: Detection Mode I exhibits remarkable prowess in delineating subsurface boundaries. Optimal exploration distances can be achieved through the judicious selection of source-receiver distances and frequencies. Detection Mode II displays heightened sensitivity to wellbore inclination and anisotropy, effectively elucidating subsurface resistivity anisotropy. This sensitivity is particularly pronounced at wellbore inclinations approaching 60°. Detection Mode III, while lacking directional capability, nonetheless furnishes fundamental insights into subsurface resistivity. Detection Mode IV demonstrates exceptional sensitivity to electrical anisotropy, particularly at higher wellbore inclinations, manifesting a conspicuous response to subsurface resistivity anisotropy. In summary, the diverse detection modes within the realm of ultradeep azimuthal electromagnetic technology each offer distinctive attributes, facilitating optimal mode selection to attain superior outcomes as per specific requisites. This research contributes significantly to an enhanced comprehension of the performance and applicability of the ultradeep azimuthal electromagnetic tool in the field of optimal drilling.

The grain size distribution along the well depth is of great significance for the prediction of the physical properties and the staged sand control design of the unconsolidated or weakly consolidated sandstone reservoir. In this paper, a new method for predicting the formation median grain size profile based on the combination model of extreme gradient boosting (XGBoost) and artificial neural network (ANN) is proposed. The machine learning algorithm and weighted combination model are applied to the prediction and analysis of reservoir grain size. The prediction model is improved from two aspects: First, the feature engineering of the XGBoost-ANN model is constructed by using the data of multiple sampling points on the logging curve. Second, the prediction accuracy is improved by increasing the dimension of the prediction model, that is, the XGBoost and ANN single-prediction models are weighted by the error reciprocal method and a combined prediction model containing multidimensional information is established. The research results show that compared with the single-point mapping model, the prediction accuracy of the multipoint mapping model considering the vertical geological continuity of the reservoir is higher than that of the single-point prediction and the coefficient of determination in the testing set can be improved up to 14.5%. The influence of different weighting methods on prediction performance is studied, and the prediction performance of original XGBoost, ANN, and XGBoost-ANN combined models is compared. The combined prediction model has a higher prediction accuracy than the single XGBoost and ANN models with the same number of sampling points and the coefficient of determination can be improved by up to 16.5%. The prediction accuracy and generalization ability of the XGBoost-ANN combined model are evaluated comprehensively. The combined model is used to design layered sand control of a well in an adjacent block, and good results have been achieved in production practice. This study provides a new method with high accuracy and efficiency for the prediction of unconsolidated sand median grain size profile.

To address the significant scaling challenges within the near-wellbore formation of ultradeep natural gas reservoirs characterized by high temperature and high salinity, we developed a dynamic scaling prediction model. This model is specifically designed for the prediction of scaling in gas-water two-phase seepage within fractured-matrix dual-porosity reservoirs. It accounts for the concentration effects resulting from the evaporation of water on formation water ions. Our scaling model is discretely solved using the finite volume method. We also conducted on-site dynamic scaling simulations for gas wells, allowing us to precisely predict the distribution of ion concentrations in the reservoir, as well as changes in porosity and permeability properties, and the scaling law dynamics. The simulation results reveal a significant drop in formation pressure, decreasing from 105 MPa to 76.7 MPa after 7.5 years of production. The near-wellbore formation is particularly affected by severe scaling, mainly attributed to the radial pressure drop funneling effect, leading to a reduction in scaling ion concentrations in the vicinity of the wellbore. Calcium carbonate is identified as the predominant scaling component within the reservoir, while calcium sulfate serves as a secondary contributor, together accounting for roughly 85.2% of the total scaling deposits. In contrast, the scaling impact on the matrix system within the reservoir remains minimal. However, the central fracture system exhibits notable damage, with reductions of 71.2% in porosity and 59.8% in permeability. The fracture system within a 5-m radius around the wellbore is recognized as the primary area of scaling damage in the reservoir. The use of the simulation approach proposed in this study can offer valuable support for analyzing the dynamic scaling patterns in gasfield reservoirs and optimizing scaling mitigation processes.

Operators often require real-time measurement of fluid flow rates in each well of their fields, which allows better control of production. However, petroleum is a complex multiphase mixture composed of water, gas, oil, and other sediments, which makes its flow challenging to measure and monitor. A critical issue is how the liquid component interacts with the gaseous phase, also known as the flow pattern. For example, sometimes liquids can accumulate in the lower part of the pipeline and block the flow completely, causing a gas pressure buildup that can lead to unstable flow regimes or even accidents (blowouts). On the other hand, some flow patterns can also facilitate sediment deposition, leading to obstructions and reduced production. Thus, this work aims to show that deep neural networks can act as a virtual flowmeter (VFM) using only a history of production, pressure, and temperature telemetry, accurately estimating the flow of all fluids in real time. In addition, these networks can also use the same input data to detect and recognize flow patterns that can harm the regular operation of the wells, allowing greater control without requiring additional costs or the installation of any new equipment. To demonstrate the feasibility of this approach and provide data to train the neural networks, a water-air loop was constructed to resemble an oil well. This setup featured inclined and vertical transparent pipes to generate and observe different flow patterns and sensors to record temperature, pressure, and volumetric flow rates. The results show that deep neural networks achieved up to 98% accuracy in flow pattern prediction and 1% mean absolute prediction error (MAPE) in flow rates, highlighting the capability of this technique to provide crucial insights into the behavior of multiphase flow in risers and pipelines.

Sustained injection of industrial-scale volumes of cold CO₂ into warmer subsurface rock will result in extensive cooling which can alter rock mass mechanical behavior and fluid migration characteristics. Advanced simulation tools are available to assess and characterize such phenomena; however, the effective use of these tools requires appropriate injection temperatures and rock thermophysical parameters (in addition to geomechanical and hydraulic properties). The primary objective of this study was to demonstrate the sensitivity of injection-induced tensile fracturing and fault reactivation to injection temperature and reservoir thermophysical properties during CO₂ injection operations. This was achieved by (1) compiling and reviewing thermophysical parameter data available for formations in the province of Alberta, Canada, and CO₂ injection temperature records for CO₂ injection projects in western Canada and (2) using a 3D, physics-based, fully integrated hydraulic fracturing and reservoir simulation numerical model to examine the geomechanical response of several potential CO₂ reservoirs in the Alberta Basin as a function of injection temperature, thermal conductivity (TC), and coefficient of linear thermal expansion (CLTE) values. The simulation results indicate that reducing the fluid injection temperature from 15°C (assumed in previous work) to 2°C (conservative value selected based on temperature data reviewed in this work) could trigger extensive vertical (20–130 m high, 100–600 m long) tensile fractures with rapid fracture initiation and full vertical growth within short periods (weeks to months) and continued horizontal length increase. When low values for thermophysical properties are used, the results show that thermally-induced tensile fracturing is unlikely, whereas the use of high values results in extensive tensile fracturing in all simulations. A similar conclusion was reached for the thermally-induced reactivation (unclamping) of proximal, critically-stressed faults. Notably, slip is predicted for all simulations where high thermophysical property values are used. This confirms that accurate determination of minimum fluid injection temperature and thermophysical parameters is important for containment risk assessment for commercial-scale CO₂ storage projects. Another significant outcome of this work is the observation that most thermophysical parameters in the available data were measured using experimental conditions and/or temperature paths that are not representative of CO₂ injection projects. As such, the development and validation of best practice approaches for accurate assessment of these parameters seem necessary.

Surfactants and low-salinity brines have been shown to be effective for enhanced oil recovery in carbonate rocks through wettability alteration (WA). Oil wettability of carbonates is ascribed to the adsorbed organic acid components in oil. The removal of the adsorbed acids leads to WA. Previous experiments with wettability-altering surfactants have shown the following: WA is a slow process; acid removal is irreversible in most cases; surfactants can access the rock surface in water-wet regions and at three-phase contact lines rather than the entire rock surface; surfactant molecules become inactive after interactions with acids. Existing models/simulators do not incorporate the aforementioned observations. In this work, a multiphase, multicomponent, finite-difference reservoir simulator incorporating a new mechanistic model for WA was developed. The model captures the key physicochemical reactions between adsorbed acids and surfactant molecules and honors the four experimental evidences. The model was first tested at the core scale. The simulation results demonstrated that the model can accurately predict waterflood performance in rocks with various wettability. It can also effectively account for the influence of injection rates in surfactant flood experiments. The effectiveness of the surfactant, controlled by an interaction constant in the model, was found to be a dominant factor. The model was also tested for field-scale pilot tests. The results revealed that total quantity of chemicals injected and the injection rate have a more pronounced effect on oil recovery compared to the timing of surfactant treatment and the concentration of surfactant slug.

In the process of directional and horizontal well drilling, cuttings tend to settle and form a bed at the low side of the annulus due to gravity, which decreases the drilling rate and even causes accidents in severe cases. This paper analyzes the performance of a new tool, the vortex cuttings cleaner, which can be effective without rotation of the drillpipe. Based on the computational fluid dynamics (CFD) approach, together with the discrete phase, Euler, and dynamic mesh models, the vortex cuttings cleaner is investigated with respect to the turbine torque, turbine velocity, pressure drop, and cuttings transport in the annulus. The working mechanism of the vortex cuttings cleaner is clarified. Finally, field tests are conducted on the tool to evaluate its application in terms of service life, wellbore friction, and rate of penetration (ROP). The results show that the turbine can rotate continuously under hydraulic drive. The turbine torque/velocity and the tool’s pressure drop increase with increasing displacement. The cuttings transport in the annulus is jointly affected by factors such as turbine velocity, fluid velocity, and particle size. A too low or high turbine velocity is unfavorable for cuttings transport. Through the analysis of the number of particles and particle concentration, the optimal velocity is determined to be 125 rev/min. The swirling flow intensity in the annulus flow field increases with the increase in turbine velocity. Field applications suggest a service life longer than 200 hours, a notable decrease in wellbore friction, and an average increase in ROP by more than 20%. This study provides a theoretical basis for the research on wellbore cleaning tools.

Drilling-fluid loss caused by millimeter-scale fractures is a notoriously difficult problem in drilling engineering, and both rigid and flexible plugging materials are commonly used to address this issue. This investigation aims to comprehensively explore the plugging efficacy and underlying mechanisms of rigid, flexible, and fiber materials when used individually and in combination. The findings of our investigations into macroscopic high-temperature and high-pressure plugging experiments divulge a revelation: Under conditions of enhanced concentration, rigid particles evince the remarkable ability to engender a pressure-enduring plugging stratum; in contrast, independent attempts by flexible and fiber materials to yield a stable plugging layer are challenging. In this context, the optimal ratio of rigid, flexible, and fiber materials has been determined through composite plugging experiments. Calcite particles with a concentration of 5–8%, rubber particles with a concentration of 2–3%, and polypropylene fibers with a concentration of 1–2% were compounded for fracture plugging with widths of 1 mm, 3 mm, and 5 mm, respectively. The resulting plugging strengths were 10 MPa, 9 MPa, and 7 MPa. The microscopic visualized plugging experiments showed that the rigid particles form an I-shaped plugging layer with high strength but are difficult to transport to the deep part of the fracture. Flexible particles can be transported into the deep part of the fracture to form a plugging layer, but the “V”-shaped formation is unstable and has low strength. Based on the experimental results of “rigid-flexible synergistic” composite bridging-plugging formulations for different scales of fractured strata, the preferred template for bridging-plugging material formulations in the field is investigated to provide a reference for the bridging-plugging material formulations in the field.

Molten metal jet cutting, based on the transient superexothermic characteristics of aluminum thermal reaction, presents a novel technology for swiftly cutting and disposing of stuck drilling columns in downhole oil and gas wells. The key to achieving efficient cutting in drilling columns lies in the jetting mechanism, which guides the high-speed radial ejection of aluminum thermal reaction products that act upon the metal pipe wall. This study uses computational fluid dynamics (CFD) simulation to establish a fluid domain model for the process of cutting molten drilling columns. The optimization of the jetting mechanism is conducted to improve the circumferential coverage by the molten metal by analyzing the impact of molten metal yield and jetting mechanism parameters (cone angle of the conical conductor, diameter, number and length of nozzles, and shape of the diverter). Finally, an ejection test is carried out to verify the optimized jetting mechanism. Research results show that increasing the cone angle of the conical conductor can increase the flow rate of the molten metal at the upper end of the axial nozzle assembly to smoothly discharge the molten metal. Increasing the number of nozzles with equal diameters can increase the circumferential distribution range of molten metal ejected into the cutting area. However, the molten metal circumferential coverage will be impacted by increasing cutting distance. Increasing the nozzle size can reduce the divergence of the molten metal, thereby improving the coverage of the molten metal in the cutting area. When the nozzle arc length L = 8 mm, the molten metal can cover almost the entire cutting area. Adding a 2-mm horizontal draining table at the end of the diverter can assist the molten metal in changing its flow direction, allowing the molten metal to be ejected in a radial direction. The research results provide a theoretical basis for optimizing fusion cutting tools and formulating cutting processes.