Reservoir Heterogeneity Research Articles

Stochastic Risk Assessment Framework of Deep Shale Reservoirs by a Deep Learning Method and Random Field Theory

Risk assessment of deep shale reservoirs is very important for subsurface energy development. However, due to complex geological environments and physicochemical interactions, shale reservoir fabric parameters exhibit variability. Moreover, the actual investigation and testing information is very limited, which is a typical small-sample problem. In this paper, the heterogeneity and statistical characteristics of deep shale reservoirs are clarified by the measured mechanical parameters. A deep learning method for deep shale reservoirs with limited survey data information is proposed. The variability of deep shale reservoirs is characterized by random field theory. A variable stiffness method and stochastic analysis method are developed to evaluate the risk of deep shale reservoirs. The detailed workflow of the stochastic risk assessment framework is presented. The frequency distribution and failure risk of deep shale reservoirs are calculated and analyzed. The risk assessment of deep shale reservoirs under different model parameters is discussed. The results show that a stochastic risk assessment framework of deep shale reservoirs, using a deep learning method and random field theory, is scientifically reasonable. The scatter plots of the elasticity modulus (EM), cohesive force (CF), and Poisson ratio (PR) distribute along the 45-degree line. The different distributed variables of EM, CF, and PR have a positive correlation. The statistical properties of the measurement data and deep learning data are approximately the same. The principal stress of deep shale follows the normal distribution with significance level 0.1. Under positive copula conditions, the maximum failure probability is 5.99%. Under negative copula conditions, the maximum failure probability is 4.58%. Different copula functions under positive and negative copula conditions have different failure probabilities. For the exponential correlation structure, the minimum failure probability is 3.46%, while the maximum failure probability is 6.19%. The mean failure probability of the EM, CF, and PR of deep shale reservoirs is 4.85%. Different random field-related structures and parameters have different influences on the failure risk.

Gulf of Mexico Case Study Highlights Smart-Completion Application

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 218794, “Multifaceted Smart Completion Application: A Gulf of Mexico Case Study,” by Jeffrey P. Manning, Ben J. Carter, SPE, and Leila Bannister, Occidental Petroleum, et al. The paper has not been peer reviewed. _ Reservoir heterogeneity, stacked pay systems, reservoir compartmentalization, and high temperature and pressure are common challenges seen in Gulf of Mexico (GOM) fields. The use of conventional completion design in vertical wells with commingled production from multiple zones has led to multiple operational challenges. In the complete paper, the authors explore the use of intelligent completions in deepwater GOM as a solution. Benefits of Intelligent Completions Intelligent completions allow collection, transmission, and validation of real-time, high-frequency data. This enables analysis of completion, production, and reservoir data while selected reservoir zones are controlled and monitored remotely in real time. Intelligent completions such as multiphase flowmeters, bottomhole pressure gauges, and cased-hole logging tools enable real-time data gathering and monitoring. Additionally, downhole flow-control systems, pinpoint stimulation, intelligent injection, and multizonal isolation technology are examples of intelligent completions that enable production optimization. This study concentrates on the application of downhole intelligent choke systems in wells that use radio frequency identification technology. Combining this technology with smart sensors or Global Positioning System technology enables sensor data to be transmitted wirelessly. Operations such as opening and closing of valves, indexing and release mechanisms, and a variety of other functions can be implemented using this technology. Multiple tools can be run at the same time without compromising the tool’s internal diameter. Each tool has a unique address; therefore, tags can be coded with data that identifies an individual tool to be actuated. Intelligent completions can be broadly categorized into the following: - Downhole flow-control systems: These are passive or active control systems to control flow from each zone. They are remotely controlled by hydraulic lines from the surface facility. - Permanent downhole sensors: These are discrete or distributed systems that monitor downhole parameters. - Permanent downhole chemical injection: These systems allow for treatment at different points in the wellbore and flowlines to prevent issues caused by corrosion, hydrates, asphaltenes, paraffin, and scaling. Case Study: Practical Use of Intelligent Completions in the GOM The operated field in the northern GOM has successfully used intelligent completions in multiple vertical producing wells to help significantly reduce operating costs associated with topsides water processing, mitigate the risk of sand production, and allow for better subsurface understanding using individual zone-flow tests (this synopsis concentrates on the first two of these three aspects). Intelligent completions enabled these operations without the use of an intervention vessel.

Facies-related reservoir heterogeneity of grain-supported limestones: insight from the Early Cretaceous Yamama Formation, southern Iraq

Sedimentological investigation of 150 m drill cores and well log analyses, including gamma-ray, resistivity, sonic, neutron, density logs, were conducted to constrain the impact of depositional facies on reservoir quality distribution in limestone succession of the Yamama Formation (Early Cretaceous), Nasiriya Oilfield, southern Iraq. Understanding the factors controlling reservoir heterogeneity in carbonate reservoirs is crucial for developing geological and reservoir models. Nine microfacies were identified: peloidal oncoidal grainstones-rudstones, skeletal cortoids packstones, skeletal dasyclads wackestones, pelletal packstones-grainstones, cortoidal peloidal grainstones, ooidal peloidal grainstones, skeletal grainstones, bioturbated dolomitic wackestones, and spiculitic skeletal mudstones-wackestones. The formation was deposited in open-marine shallow-water carbonate ramp, ranging from the intertidal to outer-ramp during the Berriasian-Valanginian. The depositional ramp was characterized by grainstones shoal barriers in the distal inner-ramp. Sea level fluctuations significantly influenced the vertical facies and reservoir quality distribution. The grain-supported, distal inner-ramp shoal facies formed the reservoir units, while the mud-supported, middle-outer-ramp facies are impervious units. Diagenetic processes, including dissolution of skeletal allochems, physical and chemical compaction, dolomitization, and cementation, have variably affected reservoir quality. Dissolution enhanced porosity by creating vuggs, while compaction and cementation often reduced porosity. Nevertheless, early diagenetic circumgranular calcite and small amount of scattered equant and syntaxial calcite overgrowths helped protecting the grain-supported limestones from physical compaction and thus preserved interparticle pores (≤ 22%) at depth (>3100 m). Conversely, equant calcite cement, which occurs in substantial amounts, has reduced porosity by filling the interparticle and moldic pores. Reservoir heterogeneity of the formation is attributed to depositional facies, which control the texture of the sediments, and to various types of diagenetic alterations.

Heterogeneity Modeling and Heterogeneity-Based Upscaling for Reservoir Characterization and Simulation

Summary Reservoir heterogeneity is a key factor in modeling reservoir performance. Heterogeneity measures can be calculated for a given permeability field but are not sufficiently straightforward to reverse the process. Detailed heterogeneity can be built into a fine-scale model but can be lost during upscaling to a coarse scale, no matter which method is chosen from simple averaging to flow-based (FB). In this paper, we propose a method of heterogeneity modeling and heterogeneity-based upscaling with the aim of solving these problems. Unlike the traditional geostatistical method used to generate a permeability field that is not directly linked to a desired heterogeneity coefficient, the proposed method creates a heterogeneous permeability field directly using a Lorenz coefficient. Using a given Lorenz coefficient as an input, the expected heterogeneous permeability field can be generated via the proposed steps and equations. Using the proposed heterogeneity-based upscaling method, the Lorenz coefficient and curve, defined from the fine-scale model, can be preserved with minimum heterogeneity loss after upscaling such that gas/oil ratio (GOR) can be matched between the fine- and coarse-scale models without using pseudofunctions. The proposed method has been applied successfully in modeling a giant carbonate oil field in the Caspian Sea consisting of a matrix-dominated platform and a fracture/karst-dominated rim. Due to the field’s complex geology and high H2S content, a dual-porosity/dual-permeability compositional model has been created to model compositional flow within/between matrix and fracture/karst initialized with an abnormally high reservoir pressure. The field surveillance data show that reservoir heterogeneity is a key component for the field reservoir characterization and simulation. The Lorenz coefficient and curves can be estimated from the cores and logs, but the challenge is how to preserve the characteristics of the Lorenz coefficient and curves during modeling, upscaling, history matching, and uncertainty analysis (UA). Application of the new method has demonstrated its ability to overcome this challenge and has significantly improved the quality of the field’s reservoir modeling, upscaling, history matching, and UA. The fine-scale model Lorenz coefficient and curves were directly applied to calculate the coarse-scale permeability. The range of the Lorenz coefficient and curves estimated from cores and logs were used to generate a range of heterogeneous permeability fields for UA. Regional Lorenz coefficients were adjusted based on the mismatches of GOR, and new heterogeneous permeability fields were generated to improve the history-matching quality.

Rheological Properties of Weak Gel System Cross-Linked from Chromium Acetate and Polyacrylamide and Its Application in Enhanced Oil Recovery After Polymer Flooding for Heterogeneous Reservoir

Long-term polymer flooding exacerbates reservoir heterogeneity, intensifying intra- and inter-layer conflicts, which makes it difficult to recover the remaining oil. Therefore, further improvement in oil recovery after polymer flooding is essential. In this study, a weak gel system was successfully synthesized, and possesses a distinct network structure that becomes more compact as the concentration of partially hydrolyzed polyacrylamide increases. The network structure of the weak gel system provides excellent shear resistance, with its apparent viscosity significantly higher than that of partially hydrolyzed polyacrylamide solution. The weak gel system exhibits typical pseudo-plastic behavior, which is a non-Newtonian fluid as well as a viscoelastic fluid. Additionally, the weak gel system’s elasticities exceed its viscosities, and longer crosslinking time further enhances the viscoelasticity. The weak gel system achieves superior conformance control and enhanced oil recovery in highly heterogeneous reservoirs compared to partially hydrolyzed polyacrylamide solutions. The weak gel system is more suited to low-permeability reservoirs with strong heterogeneity, as its effectiveness in conformance control and oil recovery increases with greater reservoir heterogeneity. Enhanced oil recoveries of the weak gel system in low-permeability sandpacks increase from 22% to 48% with a rise in permeability ratios from 14.39 to 35.64 after polymer flooding.

A Data-Driven Approach for Stylolite Detection

Summary Stylolite is a specific geopattern that can occur in both sedimentary rocks and deformed zones, which could change the porosity of the reservoir, modify the permeability, and even result in horizontal permeability barriers. Though there are many related studies to characterize this geopattern, most of them focus on experimental methods. In this work, we investigated a new approach for recovering geometrical information of the stylolite zone (including its size and location) based on neural network architectures including convolutional neural network (CNN), long short-term memory (LSTM), and transformer encoder, which could serve as a data-driven solution to the problem. To our knowledge, this paper is the first to exclusively use well testing data for deducing field geopatterns, whereas other studies have relied on additional data sources such as well logging data. To simplify the problem, we first conducted simulation by building 3D multilayer reservoir models with one stylolite zone. We considered both simplified cases with only a few homogeneous layers and cases with heterogeneous layers to generalize our work. For the heterogeneous case, we extracted the permeability from SPE10 Model 2, a commonly used public resource (SPE 10 Benchmark, Model 2 2008). Producing and observing wells in the model are at different locations and provide pressure and production rate data as inputs for the deep learning models, in the form of multivariant time series data. For homogeneous cases, after zero-padding and standardizing our inputs to tackle different-length data and features with different scales, we applied a CNN-LSTM model to our data set, leveraging the CNN’s ability to capture short-term local features and the LSTM’s capacity to extract long-term dependencies. This combination improves the extraction of important information from time series data related to stylolites. The two subnetworks are connected in parallel to combine the advantages of CNN in extracting local temporal features and the strengths of LSTM in capturing long-time dependency via self-loops. Our work also separately covers the two subnetworks of the CNN-LSTM model as baseline models. For heterogeneous cases, a CNN-based model U-Net and an attention-based model set function for time series (SeFT) were introduced to make the predictions. In a more realistic scenario featuring fluid pathways with irregular shapes within a heterogeneous reservoir, we employed a transformer encoder model to predict both the shape and location of the fluid pathways. On the homogeneous data set, our CNN-LSTM model achieved a satisfactory performance and could predict the locations and sizes of the stylolite zone and outperformed the two baseline models. On the more challenging heterogeneous data set, our baseline and CNN-LSTM models failed to deliver meaningful results. In contrast, SeFT and U-Net performed better in the sense that we could successfully predict the locations of the stylolite zones. In a more realistic scenario involving irregularly-shaped fluid pathways, our transformer encoder model achieved accurate predictions of their locations.

Triassic–Jurassic alkaline magmatism in the Ivrea-Verbano Zone, Southern Alps: A zircon perspective on mantle sources and geodynamic significance

The Ivrea-Verbano Zone (IVZ; western Southern Alps) consists of a distinctive sequence of the lower continental crust of the Adriatic plate, extending down to the subcontinental lithospheric mantle. It is characterized by a large variety of intrusive bodies of variable geochemical composition and age, offering a unique insight into the evolution of mantle-derived magmatism in post-collisional and extensional geodynamic settings. In this study, we characterize a suite of alkaline dykes intruding a mantle massif – the Finero Phlogopite Peridotite in northern IVZ. These dykes include zircon-bearing diorites and anorthosites, mainly composed by HFSE-rich amphibole and phlogopite, albite (>90 vol% in anorthosites) and apatite. Zircon, monazite, ilmenite, titanite, Nb-rich oxides, and carbonates are common accessory minerals. Additionally, a “composite” diorite dyke containing both HFSE-rich and HFSE-poor amphiboles was investigated. The study is aimed at providing new trace element, UPb geochronological and LuHf isotopes dataset on zircon from these alkaline dykes, to refine the understanding of their mantle source characteristics, emplacement age and geodynamic implications. The trace element composition of zircons from the studied dykes points to segregation from melts with alkaline to ultra-alkaline affinity. Concordia UPb ages of zircon from the alkaline diorite dykes span from 221 to 191 Ma, which are interpreted as the result of multiple crystallization/recrystallization stages related to different magmatic pulses. Conversely, zircon from anorthosite and composite diorite dykes yield a narrow time range of 198–202 Ma, highlighting the occurrence of a magmatic pulse around ca. 200 Ma. The εHf(t) values (+13.4 to +5.7) of zircon from the alkaline diorite dykes are significantly more positive compared to the values in those from anorthosite and composite diorite dykes (+4.2 to −0.4), suggesting that the parental melts were derived from heterogeneous asthenospheric mantle sources with low to moderate amount of recycled continental crust components. Our data and reappraisal of the literature indicate that the IVZ experienced a protracted period of alkaline magmatism from ca. 235 Ma to 185 Ma. The melts migrated along mantle shear zones during the Triassic-Jurassic lithospheric extension. Different pulses of alkaline magmatism were associated with relevant extensional tectonic stages recorded by the continental crust, presumably triggering a passive uplift of heterogenous asthenospheric reservoirs over a front of at least 500 km. This tectono-magmatic cycle is a precursor to the focused rifting stage which caused the opening of the Jurassic Alpine Tethys, enhancing Pangea fragmentation.

Damage mechanism analysis of polymer gel to petroleum reservoirs and development of new protective methods based on NMR technique

Polymer gels are widely used in water control and enhanced oil recovery in oil fields. However, the damage mechanism of polymer gels to layers with remaining oil and not requiring plugging and corresponding protective measures are unclear. In this paper, we investigated polymer gels' damage and protection performance through static gel-breaking experiments and dynamic plugging and oil recovery evaluations on rock cores. Moreover, nuclear magnetic resonance (NMR) technology was combined to analyze the damage performance of polymer gels on cores from the pore scale. In addition, a protective technique based on gel breakers for layers with remaining oil and not requiring plugging was proposed. Results showed that when polymer gels were injected into heterogeneous cores, they plugged high-permeability layers while also penetrating low-permeability layers. When the damage to the low-permeability layers was not alleviated, the conformance and oil displacement efficiency were significantly reduced. When the concentration of ammonium persulfate was 2%–5%, the gel-breaking time was shortest and the residue was very minimal. Ammonium persulfate could be used as a gel breaker and reservoir protective material. Furthermore, after injecting ammonium persulfate into heterogeneous reservoir cores, the gel damage on the face of low-permeability layers was relieved. Consequently, the improvement in sweep efficiency was achieved, showing the re-activation of the remaining oil in medium-low permeability layers. Therefore, the low-permeability layer protection process and core experiment study based on gel-breaking agents proposed in this study were suggested to provide a new technique for the field application of conformance modification agents, aiming to achieve higher recovery degrees.

Conceptual framework for data-driven reservoir characterization: Integrating machine learning in petrophysical analysis

Reservoir characterization plays a critical role in optimizing oil and gas production by accurately determining subsurface properties such as porosity, permeability, and fluid saturation. Traditionally, this process has relied on the interpretation of well logs, seismic data, and core samples using manual or physics-based methods. However, these approaches often face limitations when dealing with complex datasets and heterogeneous reservoir properties. This review develops a conceptual framework for integrating machine learning (ML) into petrophysical analysis to revolutionize reservoir characterization. By utilizing advanced ML algorithms such as supervised learning, unsupervised learning, and deep learning, this model demonstrates the potential of data-driven methods to improve the accuracy and efficiency of reservoir analysis. The framework focuses on how ML techniques can automate the interpretation of well logs, enhance seismic data processing, and predict rock properties from core samples, leading to more accurate identification of reservoir boundaries and productive zones. This approach not only reduces interpretation errors but also accelerates decision-making in reservoir management. Moreover, machine learning can handle the vast amounts of data generated in reservoir studies, enabling real-time analysis and adaptive decision support. The study also addresses challenges such as data quality, model interpretability, and computational scalability, offering insights into future applications of ML in reservoir characterization. By combining petrophysical expertise with machine learning capabilities, this framework aims to significantly advance the field of reservoir characterization and contribute to more efficient resource management in the oil and gas industry.

Controls on microbially-induced carbonate precipitation in geologic porous media

Microbially-induced carbonate precipitation (MICP) provides a natural biomineralization approach to secure the geologic storage of gases (e.g., carbon dioxide, hydrogen and methane). Cracks in embrittled wellbore cement, for example, provide a pathway for atmospheric gas leakage, while permeability heterogeneities in the storage reservoir leads to fingering effects that diminish the storage capacity. The design of MICP processes, however, remains a challenge due to limited understanding of the coupled nonlinear reaction kinetics and multiphase transport involved. Specifically, previous attempts at MICP through porous media have been encumbered by carbonate precipitation localized to the first ∼ cm of the bulk injection surface. In this study, we investigate the reactive transport controls on MICP necessary to enable deep MICP penetration into the formation. We use a micromodel with pore geometry and geochemistry representative of real geologic media to image direct pore- and pore-ensemble-level mineral, fluid, and microbial distributions. An approach to adsorb microbes uniformly across the micromodel, rather than local accumulation near the inlet, is developed that enables deep MICP penetration into the porous medium. A sensitivity analysis was performed to investigate the impact of injection conditions (e.g., rates, concentrations) required to maximize CaCO3 precipitation away from the injection site. With multiple cycles of MICP, a ∼ 78 % reduction in permeability was achieved with ∼8 % carbonate pore volume occupation. Overall, this study establishes the possibility of MICP as an effective and controllable method to enhance the security of gas storage in geologic media.

Unsupervised machine learning-based multi-attributes analysis for enhancing gas channel detection and facies classification in the serpent field, offshore Nile Delta, Egypt

The prediction of highly heterogeneous reservoir parameters from seismic amplitude data is a major challenge. Seismic attribute analysis can enhance the tracking of subtle stratigraphic features. It is challenging to investigate these subtle features, including channel systems, with conventional-amplitude seismic data. Over the past few years, the use of machine learning (ML) to analyze multiple seismic attributes has enhanced the facies analysis by mapping patterns in seismic data. The purpose of this research was to assess the efficiency of an unsupervised self-organizing map (SOM) approach supported by multi-attribute analysis that could improve gas channel detection and seismic facies classification in Serpent Field, offshore Nile Delta, Egypt. As well as evaluates the importance of several available seismic attributes in reservoir characterization rather than analyzing individual attribute volumes. In this study, the single attribute (spectral decomposition attribute) highlighted the gas channel spatial distribution using three distinct frequency magnitude values. Subsequently, we employ principal component analysis (PCA) as an attribute selection method, discovering that combining seismic attributes such as sweetness, envelope, spectral magnitude, and spectral voice as input for SOM reflects an effective method to determine facies. The clustering results distinguish between shale, shaly sand, wet sand, and gas-saturated sand and identify gas–water contact on a 2D topological map (SOM), where each pattern indicates certain facies. This is achieved by associating the SOM outputs with lithofacies determined from petrophysical logs. Reducing exploration and development risk and empowering the geoscientist to generate a more precise interpretation are the ultimate objectives of this multi-attribute analysis.

Mathematical model and its solution for water-altering-gas (WAG) injection process incorporating the effect of miscibility, gravity, viscous fingering and permeability heterogeneity

AbstractThe oil recovery from the water-alternating-gas (WAG) injection process is significantly impacted by gravity, viscous fingering, and the permeability heterogeneity of the reservoir. Therefore, the combined effect of these parameters cannot be neglected in the WAG injection process. This article presents the development of a mathematical model of oil recovery and its solution for the WAG injection process that takes into account the combined effects of miscibility change, viscous fingering, gravity, and permeability heterogeneity in an inclined stratified reservoir. First, the governing equations and fractional flow functions were explained in relation to the effects of gravity, permeability heterogeneity, viscous fingering, and miscibility change in an inclined stratified porous medium. Then, a mathematical model was developed using fractional flow functions and conservation equations for both injected water and solvent. The model was generated in the form of a quasi-linear first-order partial differential equation, which was solved analytically in two dimensions (2-D) utilizing vector calculus. Next, this model was solved analytically by applying wave theory to practical constant pressure boundary conditions, which generate distinct waves at different times to provide pressure and saturation at the displacement front location. The total volumetric flux and breakthrough time are calculated from the analytical solution at various times. The presented analytical solutions can be used to predict different parameters for stratified porous media in a fast and efficient way. Finally, the results of analytical solution validated with high-resolution numerical simulation for a wide range of permeability heterogeneity, which shows excellent agreement for breakthrough time, saturation, and pressure versus displacement location of different waves at different times. This analytical solution will save time and money by offering guidance to engineers for analyzing the saturation and pressure distribution at different times and predicting oil recovery. It will also improve the understanding of the physics underlying the multiphase flow WAG injection process in heterogeneous reservoirs.

Geological model calibration based on gradual deformation and connectivity function.

The connectivity is an important feature of the reservoir geological heterogeneity that effects fluid flow responses. In geostatistical modeling, random realizations are generated to describe reservoir heterogeneities. But these realizations do not necessarily honor connectivity data. Especially in multiple-point geostatistical modeling, ensuring the connectivity data of these realizations consistent with the training image is a challenge. The connectivity data of the training image is an important factor in assessing the quality of simulation results in modeling a fluvial hydrocarbon reservoir with meandering channels. The calibration of geological/geostatistical model realizations by measured data is generally performed through history matching, which is an inversion process. This method requires a parameterization of the geostatistical model to allow the updating of an initial model realization. The gradual deformation method has been used to parameterize geostatistical realizations. This method uses a perturbation mechanism to smoothly modify model realizations generated by sequential (not necessarily Gaussian) simulations while preserving its spatial variability. In this paper, a workflow is presented to calibrate the model realizations to the connectivity data. This workflow ensures that the connectivity of model realizations generated by SNESIM is consistent with the training image, and it is applicable to both multiple-point and two-point geostatistical modeling methods based on sequential simulation. This workflow incorporates the computation of the model connectivity function and its calibration to connectivity data using the gradual deformation method. This involves defining and minimizing an objective function, which quantifies the mismatch between the model connectivity function and the connectivity data. In the case study, multiple initial realizations are utilized to construct a final realization of the reservoir model that honors the connectivity data.

Effects of Fines Migration and Reservoir Heterogeneity on Well Productivity: Analytical Model and Field Cases

Abstract Formation damage due to fines migration after water breakthrough during oil and gas production results in significant well productivity decline. A recent study derived an analytical model for fines migration during commingled water–oil production in homogeneous reservoirs. Yet, reservoir heterogeneity highly affects formation damage and well productivity. This article develops an analytical model for layer-cake reservoirs. We develop a novel methodology for characterizing productivity decline by considering the impedance as a function of water-cut, two quantities that are commonly measured throughout the production life of the well. The methodology is based on a new analytical model for inflow performance in layer-cake reservoirs under fines migration. The new model integrates pseudo-phase-permeability functions for commingled water–oil production with equations for fines release and induced permeability damage. The analytical model reveals linear well impedance growth versus water-cut increase, where the slope is determined by a modified form of the mobility ratio which includes the extent of formation damage. This linear form is shown to arise when the formation damage factor is constant, regardless of the distribution of reservoir permeabilities. The model is validated by comparison with production histories of five wells from three fields, which exhibit good agreement with the linear trend predicted by the new model. The explicit formulae allows for the prediction of productivity at abandonment, determining the optimal well stimulation time, as well as reconstructing skin values during the early stages of production to better estimate the influences of other formation damage factors, like those induced during drilling and completion.

Estimating petrophysical properties using Geostatistical inversion and data-driven extreme gradient boosting: A case study of late Eocene McKee formation, Taranaki Basin, New Zealand

Estimation of petrophysical properties holds significant importance in evaluating the feasibility of hydrocarbon extraction and enhancing production efficiency. Therefore, theintegration of seismic data with well data for reservoir characterization has garnered significant attention in the oil and gas industry. This approach provides a comprehensive view of the lateral variations across the entire reservoir. In this study, our goal is to establish a relationship between seismic attributes and the petrophysical properties of the reservoir, enabling the prediction of property distribution across the entire reservoir. We employed geostatistical inversion to generate detailed P-impedance maps of the reservoir, capturing its spatial heterogeneity. A total of 100 realizations were produced and ranked into P10, P50, and P90 scenarios, representing different probabilistic outcomes. The P90 scenario, considered the most representative of the reservoir's average properties, was selected for interpretation, enabling a comprehensive analysis of reservoir heterogeneities and their distribution. Subsequently, we estimated porosity using various seismic attributes, employing a Data-driven Extreme Gradient Boosting (Xgboost) approach within the reservoir analysis. This method enabled us to estimate a high-resolution acoustic impedance and porosity with minimal root mean square error of 3.8 and r2score of 66 % across the spatial gap which ranges from 8468.84 to 9367.85(g/cc) *(m/s) and 10 to 45 % between wells, revealing the lateral variation of acoustic impedance and porosity within the reservoir interval of interest. The information provided was found to be of significant value in the identification of potential sweet spots and plays a pivotal role in the development of a reservoir management strategy for future hydrocarbon exploration.

On the feasibility of an ensemble multi-fidelity neural network for fast data assimilation for subsurface flow in porous media

Uncertainty quantification (UQ) of the reservoir heterogeneity is essential to predict fluid flow behavior in subsurface formations accurately, and the task is often accomplished by integrating high-fidelity forward physics simulators with iterative data assimilation methods, and such workflows are usually computationally expensive due to the iterative nature and the prohibitive cost of physics simulations.In this work, we develop a new Ensemble Multi-Fidelity Neural Network (EMF-Net) to mitigate the efficiency bottleneck of UQ. EMF-Net directly infers the uncertain variables (e.g., permeability) via the sparse observation of the state variables (e.g., pressure). By leveraging the regression capability of deep neural networks, EMF-Net directly learns the nonlinear mapping from the innovation vector to the inferred update vector without the hypothesis of a linear mapping. At the training phase, high-fidelity data computed by a physics simulator (fh) and low-fidelity data computed by a forward proxy model (fl) are used to train the EMF-Net at two different stages, respectively. At the inference phase, we first adopt the 1-stage EMF-Net to infer the update vector for the initial ensembles with a global tuning and then efficiently update the innovation vector by fl. As an option, we further refine the inference, via a 2-stage EMF-Net, which captures the locality and enhances consistency with the observation data. We demonstrate that EMF-Net can reach equivalent inference accuracy compared to classic data assimilation algorithms like ES-MDA, in addition to decreasing the CPU time. Therefore, the accuracy and efficiency make it an attractive alternative for scalable real-time history-matching tasks in field-scale subsurface engineering applications.

Physical and experimental simulation of unconventional reservoir formation for carboniferous in hudson oilfield, Tarim Basin, China

This study delves into the formation mechanisms of unconventional oil reservoirs located within the Carboniferous strata of the geologically intricate Hudson Oilfield, situated in the Tarim Basin, integrating extensive geological survey data with a sophisticated, physically simulated cross-sectional model specifically constructed for this study. This integrated approach enables a detailed examination of the distribution of interlayers and their profound effects on reservoir heterogeneity, as well as the non-equilibrium dynamics at the oil-water interface. Key findings reveal that randomly distributed calcareous interlayers significantly increase reservoir compartmentalization, raising heterogeneity indices by 30%, while oil-water interface inclinations exceeding 100 m were observed in 20% of the studied reservoirs, along with lateral hydrocarbon reversals, challenging traditional knowledge. Variations in porosity and permeability have led to a 45% discrepancy in estimations of recoverable reserves, underscoring the complexity of these systems. Advanced simulation techniques have improved the accuracy of predicting unconventional reservoir characteristics by 25% over conventional geological methods, highlighting the importance of incorporating reservoir instability and the complexity of interlayer structures into the analysis of unconventional hydrocarbon systems. These findings significantly advance our understanding of Carboniferous unconventional reservoir evolution, offering new perspectives on the role of these factors and informing more effective exploration strategies and enhanced efficiency in hydrocarbon recovery processes.

Visualization and Simulation of Foam-Assisted Gas Drive Mechanism in Surface Karst Slit-Hole Type Reservoirs

Nitrogen injection technology has become an important production technology after water injection development in the karst fracture-vuggy reservoir in Tahe Oilfield. However, due to the influence of reservoir heterogeneity and the high mobility of gas fluid, nitrogen easily forms a dominant channel and gas channeling occurs, and the recovery effect time is short. Based on this, a visual surface karst model is designed and created to study nitrogen foam-assisted gas drive. The results show that after gas channeling occurs in the dominant channel of nitrogen flooding, foam injection-assisted gas flooding can improve oil recovery. In the longitudinal direction, foam-assisted gas drive mainly displaces the remaining oil because of gravity differentiation and the reduction of oil–water interfacial tension. In the horizontal direction, foam-assisted gas drive is mainly used to block the large pore cracks and dominant channels, promote the gas to turn into large tortuous and small cracks, and expand the swept efficiency of the gas. After forming the dominant channel, injecting 0.3 pv salt-sensitive foam with a gas–liquid ratio of 2:1 in the middle of the gas channel can improve the recovery rate of the model from 4% to about 25%, and the recovery rate can be increased by about 20%, which improves the effect of gas flushing and improves the development efficiency of the oil field at the same time.

Intelligent Prestack Multi-trace Seismic Inversion Constrained by Probabilistic Geological Information

Seismic inversion uses observed data including well-logging and seismic data to infer rock parameters. However, it still suffers from non-unique solutions. The non-uniqueness increase when extrapolating away from well location. To mitigate this effect, we incorporate geological information into a deep learning (DL) framework in the form of generated probabilistic labels. We transform geological information into reservoir heterogeneity (RH) weights and geological pattern constraint (GPC) weights to generate probabilistic labels. The RH weights are used to capture the variability in accuracy during the spatial extrapolation of well data. The RH weights are used to balance the constraint weights between well and seismic data in the DL model. The GPC weights are used to characterize the changes of geological patterns from well locations to non-well locations. Based on variations in local prestack seismic waveforms, the DL model learns extrapolation patterns of well data with similar geological patterns. By combining the rock parameters derived from well logs with these two types of weights, we can generate a series of probability labels. Two types of geological information and well-logging data are deeply integrated into a supervised 2D-ResUNet framework. In this way, we develop a novel DL-based prestack multi-trace seismic inversion method. We validate our method on a 3D synthetic model and a 3D field dataset. The results show that our method exhibits great potential for characterizing the spatial variation of subsurface rocks and outperforms traditional deep learning methods.

Well-Test Interpretation Model of Water-Injection Well in a Low-Permeability Reservoir and Its Application

For low-permeability reservoirs, water-flooding development is usually adopted, which leads to induced fractures near the wellbore, increasing reservoir heterogeneity, and making water-flooding development more complex. This paper focuses on low-permeability reservoirs, considering the characteristics of induced fractures and elliptic-flow composite, and the well-test model for injection wells is established. The mathematical model in Laplace space is obtained through dimensionless transformation and Laplace transformation. Subsequently, the Mathieu function is introduced to obtain the bottom hole pressure, and the pressure response curve is drawn. The six flow stages of the curve are defined, and the sensitivity of parameters such as half-length of induced fractures, range of lateral-swept area, permeability in unswept area, and outer boundary distance at constant pressure are analyzed. The results show that the half-length of the fracture mainly affects the linear flow of the fracture, the range of the lateral wave-affected area mainly affects the radial flow of the swept area, the permeability of the unswept area mainly affects the radial flow of the unswept area, and the outer boundary distance at constant pressure mainly affects the boundary flow. Based on the production performance of a certain injection well in J Oilfield, a series of key parameters are obtained through analytical solution model inversion, including the induced-fracture half-length of 10.32 m, the lateral-swept range of elliptic partition flow of 128.95 m, the permeability of the swept area of 6.87 mD, and the mobility ratio of 119.92, which show the superiority of the analytical solution model.