Well-log Data Research Articles

Strategic Geosteering Workflow with Uncertainty Quantification and Deep Learning: Initial Test on the Goliat Field Data

Continuous integration of real-time logging-while-drilling data into a subsurface model with relevant geological uncertainties enables strategic geosteering: a field-level optimization of the well-placement strategy. Model errors arising from oversimplified conceptual geological models and imperfect simulation of measurements result in unreliable subsurface-model updates. The model errors are particularly pronounced when synthetic measurements are approximated with a fast but imperfect model, such as a deep neural network (DNN).#xD;We present a practical data-assimilation workflow consisting of offline and online phases. The offline phase involves DNN training and building an uncertain prior near-well geo-model. The online phase utilizes the flexible iterative ensemble smoother (FlexIES) to perform real-time assimilation of extra-deep electromagnetic data while accounting for the model errors in the approximate DNN model. We demonstrate the proposed workflow on historic well-log data from the Goliat Field (Barents Sea). #xD;The median of our probabilistic estimation is on par with proprietary inversion, regardless of the number of layers in the chosen prior or the approximate DNN model. By estimating model errors, FlexIES automatically quantifies the uncertainty in the boundaries and resistivities of layers, which is not standard in proprietary inversion. #xD;This capability allows us to capture uncertainties more efficiently, thus providing input for future quantitative decision support methods. We demonstrate the potential of quantitive decision support by visually estimating the ahead-of-bit risk of reservoir exit that has occurred during the considered operation.

Seismic coloured inversion to explore the hydrocarbon prospectivity of a field in the Upper Assam basin

This study demonstrates an integrated approach using a seismic-coloured inversion to derive the hydrocarbon potential of a field in the Upper Assam basin. The well-log analysis indicated the presence of hydrocarbon in the form of gas, oil, and water in sandstones with thin streaks of limestone and shale layers. The Acoustic Impedance (AI) model was generated from post-stack seismic data using a coloured inversion. Shale volume (Vshale), density, porosity and water saturation are the petrophysical parameters spatially populated using a multilayered feed-forward neural network on the inversion-based AI model. The estimated range of model-derived parameters varies as Velocity: 2136–5034 m/s, Vshale: 0–48%, Density: 2.14–2.72 gm/cc, Porosity: 6–29%, Water saturation: 9–28%, and demonstrates a reasonable correlation to well-log data. Seismic attribute for thin-fault likelihood was used to interpret the major faults. From the combined analysis based on interpreted faults and modelled parameters, the northwestern part of the study area displays a thick hydrocarbon-bearing zone (due to increasing thickness and reservoir quality). This interpretation is based on the consideration that faults are open. Thus, assuming continuity in the sequence, this northwestern region can be considered as one of the promising candidates for further exploration.

Electrofacies-driven 3D-static reservoir modeling of the Late-Cenomanian AbuRoash’G lithounit (Abu-Gharadig Basin, Egypt): Sequence stratigraphic and geomodel constraints for a gas-bearing estuarine system

Estuarine-systems, developed upon transgressive-phases, feature high-quality reservoir-facies, e.g. tidal-bars, that are important stratigraphic-plays critical for hydrocarbon exploration-development. However, capturing their intricate architectural elements (heterogeneity and quality) is still challenging due to the complex stacking-nature and limited-examples. Moreover, defining reservoir-boundaries upon static-modeling of reservoirs cannot be efficient unless it is controlled by stratal-geometries and established depositional-models. To this end, in this study, we performed 3D-static geocellular reservoir-modeling process for the Late-Cenomanian AbuRoash“G” Member (Abu-Gharadig Basin, Egypt) with sequence-stratigraphic and geomodel, relative-geological-time (RGT) model and horizon-stacks, constraints. In this investigation, as an effective-workflow, not only facies-analysis, integrating seismic-stratigraphy and GR-log motifs, was applied for paleo-environment reconstruction, but also machine learning-based electrofacies were applied, through self-organized-maps (SOM), to accurately recognize complex facies-assemblages present. Object-based and pixel-based stochastic-simulation processes were applied upon geocellularly modeling rock and fluid properties, utilizing key-information scales of seismic and well-log data. The results show that three third-order depositional sequences dominate the succession, resting on the Late-Cretaceous unconformity, of which sequence-1 encloses the lowstand and transgressive systems-tracts of the fluvio-estuarine Bahariya and Abu Roash“G” units, respectively. The transgressive phase built AbuRoash“G” lithounit features an estuarine depositional-system encompassing four facies-associations, of which tidal-sand-bars represent significant gas-bearing reservoir-quality facies. The tidal-bar facies’ efficient reservoir quality calls for attention and testing in future development plans and investigation in similar settings. Furthermore, the facies-constrained workflow established in this study, for reservoir modeling, can worldwide help identify the ultimate reservoir-configuration, as long as the 3D-static modeling process is controlled by the stratal and geomodel restraints.

SeisResoDiff: Seismic resolution enhancement based on a diffusion model

High resolution of post-stack seismic data assists in better interpretation of subsurface structures as well as high accuracy of impedance inversion. Therefore, geophysicists consistently strive to acquire higher resolution seismic images in petroleum exploration. Although there have been successful applications of conventional signal processing and machine learning for post-stack seismic resolution enhancement, there is limited reference to the seismic applications of the recent emergence and rapid development of generative artificial intelligence. Hence, we propose to apply diffusion models, among the most popular generative models, to enhance seismic resolution. Specifically, we apply the classic diffusion model—denoising diffusion probabilistic model (DDPM), conditioned on the seismic data in low resolution, to reconstruct corresponding high-resolution images. Herein the entire scheme is referred to as SeisResoDiff. To provide a comprehensive and clear understanding of SeisResoDiff, we introduce the basic theories of diffusion models and detail the optimization objective's derivation with the aid of diagrams and algorithms. For implementation, we first propose a practical workflow to acquire abundant training data based on the generated pseudo-wells. Subsequently, we apply the trained model to both synthetic and field datasets, evaluating the results in three aspects: the appearance of seismic sections and slices in the time domain, frequency spectra, and comparisons with the synthetic data using real well-logging data at the well locations. The results demonstrate not only effective seismic resolution enhancement, but also additional denoising by the diffusion model. Experimental comparisons indicate that training the model on noisy data, which are more realistic, outperforms training on clean data. The proposed scheme demonstrates superiority over some conventional methods in high-resolution reconstruction and denoising ability, yielding more competitive results compared to our previous research.

Seismic Node Arrays for Enhanced Understanding and Monitoring of Geothermal Systems

Abstract Harnessing geothermal energy will likely play a critical role in reducing global CO2 emissions. However, exploration, development, and monitoring of geothermal systems remain challenging. Here, we explore how recent low-cost seismic node instrumentation advances might enhance geothermal exploration and monitoring. We show the results from 450 nodes deployed at a geothermal prospect in Cornwall, United Kingdom. First, we demonstrate how the nodes can be used to monitor the spatiotemporal and size distribution of induced seismicity. Second, we use focal mechanisms, shear-wave source polarities, and anisotropy to indicate how the dense passive seismic observations might provide enhanced insight into the stress state of the geothermal systems. All the methods are fully automated, essential for processing the data from many receivers. In our example case study, we find that the injection-site fracture orientations significantly differ from that of the crust above and the regional stress state. These observations agree well with fracture orientations inferred from independent well-log data, exemplifying how the nodes can provide new insight for understanding the geothermal systems. Finally, we discuss the limitations of nodes and the role they might play in hybrid seismic monitoring going forward. Overall, our results emphasize the important role that low-cost, easy-to-deploy dense nodal arrays can play in geothermal exploration and operation.

Pilot Area Formation Evaluation: Upper Shale Member/ Rumaila Oil Field

This study aims to conduct a comprehensive formation evaluation of a pilot area within the Upper Shale Member of the Rumaila Oil Field. This evaluation is an essential step in the full development of the field. The application of well-log data and core analyses can help in obtaining the desired information about the geological characteristics of the formation. The process begins with measuring the formation temperature and water resistance utilizing Schlumberger’s charts and equations. The volume of shale was determined by two different methods, which were then used together to obtain the final shale volume. The porosity was determined using the conventional porosity equations from the porosity logs and the saturation was estimated based on Archie's equation. In core data analysis, an unconventional technique was utilized to determine rock type and permeability. The core porosity and the permeability were classified into four groups mainly using a self-organizing map and an unsupervised machine-learning method, and selected regression equations of each group were applied to estimate permeability in the core. The method depicted a good agreement between the core and estimated permeabilities, proving it as an effective tool. A complicated training data set was constructed based on the use of a multilayer perceptron neural network on coreless wells to identify rock types and permeability. Analyzing the petrophysical properties of the study area showed evidence that this area is characterized by heterogeneity. The heterogeneity of this formation is due to the presence of a considerable amount of shale, in addition to the significant characteristic differences in the layers and the same layer in different locations. The abundance of shale rock poses challenges during drilling operations, particularly due to shale washout which can lead to mechanical issues with the drilling string. Therefore, caution is advised when drilling new wells in the area to mitigate shale washout risks. Furthermore, the analysis identified layers with high hydrocarbon saturation that are viable for production. Conversely, some layers, characterized by shale presence and poor rock quality, are deemed unsuitable for production, and should not be considered as reservoir rock.

Stochastic inversion based on spectral simulation of non-stationary statistical characteristic parameters

Abstract The traditional method of constructing the prior model for stochastic inversion mainly uses well-logging data to obtain the stationary range from the variogram to characterize the complex non-stationary spatial tectonic features of the subsurface layers. Unfortunately, it is hard for the stationary range to precisely describe the complex non-stationary spatial tectonic features of the subsurface layers. The established prior model is also difficult to describe the real spatial tectonic features of the subsurface media; thus, it cannot provide reliable prior information for the inversion. Therefore, according to the theory of random media, we extract the non-stationary statistical characteristic parameters that change with spatial coordinates from the known information such as seismic and well-logging data, to accurately and reliably characterize the non-stationary spatial tectonic of the subsurface media. Then, combined with FFT-MA spectral simulation, a spectral simulation method based on non-stationary statistical characteristic parameters is built to construct the spectral simulation results that can more accurately conform to the spatial tectonic features of the subsurface media, to provide reliable and reasonable geostatistical prior information for a better inversion result.

Quantitative Elimination of Seismic Pseudofaults and Fine Analysis of True Faults Underlying Igneous Rocks of No-Well Areas: A Case Study of Shuntuoguole Uplift in Tarim Basin

Abstract After multistage tectonic movement and evolution, large superimposed oil and gas basins generally developed many igneous rocks in the early rifting stages. The lithology and lithofacies of igneous rocks are complex, which is easy to lead to the distortion of the underlying migration velocity field and thus the response of seismic pseudofaults. Also, because of the obvious shielding and absorption effect of igneous rocks on seismic waves, the waveform quality of underlying strata is poor and the seismic response characteristics of faults are fuzzy. Currently, relevant studies have shown that the influence of igneous rock can be eliminated by the prestack depth migration with an accurate igneous rock velocity model. However, improving the accuracy of the velocity model needs to be corrected by well-logging data, resulting in poor applicability of the existing velocity modeling technology underlying igneous rocks without well, which is an obvious technical bottleneck. In this paper, the secondary strike-slip fault in Shuntuoguole low uplift of Tarim Basin, which has great oil and gas exploration potential but a very low degree of drilling, is selected as the research object. Aiming at difficult fault detection underlying igneous rocks caused by lack of drilling, the accuracy of fault seismic identification is improved by “interpretative fault preprocessing” and “fault sensitive attribute optimization.” In addition, through the “extreme hypothesis method” to maximize the complex migration velocity and simulate the underlying target layer distortion maximization, we realize the quantitative elimination of seismic pseudofaults. The practical application shows that this technology can determine the true and fake underlying faults quantitatively without establishing an accurate igneous rock velocity model. It is crucial not only for exploring oil and gas in the Tarim Basin’s secondary strike-slip faults but also for offering a method and technical guide for identifying faults in other basins affected by igneous rocks.

Seismo-stratigraphic and morpho-bathymetric analysis revealing recent fluid-rising phenomena on the Adventure Plateau (northwestern Sicily Channel)

The northwestern region of the Sicily Channel hosts a great number of morphological highs, the widest of which is the Adventure Plateau that is part of the Sicilian Maghrebian Fold and Thrust Belt system, formed since the Neogene. The Adventure Plateau was shaped in the Early Pliocene by an extensional phase that produced high-angle normal faults mostly WNW-ESE to N-S oriented. Through these faults, magmatic fluids ascended and produced widespread volcanic manifestations often associated to fluid flow processes. The interpretation of multibeam echosounder, seismic reflection (sparker, airgun) and well-log data allow us to identify several features related to the presence of fluids in the study area. The morpho-structural analysis showed a NW–SE oriented fault system and a string of pockmarks that follow the same trend. A detailed well-log analysis confirmed the presence of oil traces, at a depth of ~ 250 m, and gas (i.e., CO2) at a depth of ~ 450 m. The seismo-stratigraphic analysis highlighted seismic signals located below the pockmarks, (e.g. seismic chimneys, bright spots) which suggest the presence of fluids that would rise to a few meters’ depth. Based on the observations, two sources and two corresponding rising mechanisms have been identified. Morphometric analysis of pockmarks has been performed to delineate their possible interaction with the bottom currents. A fluids pathway model has been reconstructed, revealing the source of fluids emissions at depth in the Adventure Plateau, and providing new insights into the identification of fluid leakage pathways.

Rock-physics modeling and pre-stack seismic inversion for the Cambrian superdeep dolomite reservoirs in Tarim Basin, Northwest China

The Cambrian superdeep dolomite in Tarim Basin has great resource potentialities. However, the superdeep dolomite reservoirs have the seismogeological characteristics of deep burial, low porosity and permeability, complex fracture networks and low signal-to-noise, which cannot support the efficient exploration and development of superdeep dolomite reservoir due to restricting the precision of fracture prediction. The fractures and pores compose the main storage spaces of superdeep dolomite reservoir. To better characterize the superior reservoir of superdeep dolomite with complex fracture, we first describe the properties of dolomite using core data, which illustrates that the low-porosity and low-permeability dolomite reservoirs have three tendencies: relatively higher porosity and lower permeability, relatively lower porosity and lower permeability, and relatively lower porosity and higher permeability. The development of porosity is decided by the size of dolomite particle, and when the porosity is greater, the pore type controls the permeability of the reservoir. It is difficult to effectively characterize dolomite with multiple types fractures and pores with the traditional rock physical analysis. Considering the influence of complex pore structures on the elastic parameters of superdeep dolomites, we propose a fracture characterization indicator (FCI) based on Biot theory. We test the FCI using cores and well-logging data, which can quantitatively describe the change of pores in dolomite reservoirs, much better than conventional rock-physics modeling. Meanwhile, the pre-stack seismic gather optimization process is established to employ the combination of denoising and flattening. The density indicator (DI) is proposed due to the poor result of density inversion. The priority distributions of superdeep dolomite reservoirs are revealed through the FCI and DI applied to 3D pre-stack seismic inversions. The results are consistent with those by actual drilling, which confirms that the proposed method can serve as a new way to predict superdeep dolomite reservoirs.

Three-dimensional shear wave velocity prediction by integrating post-stack seismic attributes and well logs: application on Asmari formation in Iran

Understanding the distribution of shear wave velocity (VS) in hydrocarbon reservoirs is a crucial concern in reservoir geophysics. This geophysical parameter is utilized for reservoir characterization, calculating elastic properties, assessing fractures, and evaluating reservoir quality. Unfortunately, not all wells have available VS data due to the expensive nature of its measurements. Hence, it is crucial to calculate this parameter using other relevant features. Therefore, over the past few decades, numerous techniques have been introduced to calculate the VS data using petrophysical logs in wells with limited information. Unfortunately, the majority of these methods have a drawback they only offer insight into the location of the wells and do not provide any details regarding the distribution of VS in the space between the wells. In this article, we employed three-dimensional post-stack seismic attributes and well-logging data integration to predict the distribution of VS in the Asmari formation in an Iranian oil field. To accomplish this objective, the model-based seismic inversion algorithm was utilized to convert the seismic section into the acoustic impedance (AI) section. Then, AI and seismic data were utilized in the cross-validation method to determine the relevant attributes for predicting the spatial distribution of VS throughout the entire reservoir area, using an artificial neural network. The proposed method was shown to provide 94% correlation and 109 m/s error between the actual and estimated VS. Also, the calculated VS section has a high correlation with the actual logs at the location of the wells.

Prism-shaped rock-physics template

Despite the emergence of statistical and intelligent approaches for quantitative seismic interpretation in recent years, rock-physics templates (RPTs) are still preferred because of their simplicity and easy implementation. RPTs have been introduced as a fundamental tool for lithology and fluid discrimination based on well-log and seismic data, which have already been proven in terms of accuracy and reliability. Considering the demand for comprehensive RPTs and improving their efficiency, I develop a novel 3D prism-shaped template for lithology and fluid discrimination called a prism-shaped RPT (P-RPT). For this, a theoretical methodology for designing conventional RPT is introduced with the aim of bounds and hybrid rock-physics models. Next, some novel triangular and rectangular ternary and binary templates are designed and connected to one another to build a prism for fluid discrimination and lithology identification. Two ternary charts are used for fluid and lithology, incorporating the P- and S-wave velocity ratios and lambda-mu parameters. Furthermore, binary charts, including acoustic impedance, Lamé parameters, and porosity, are designed and modified from the literature for fluid and lithology identification. Next, the obtained templates are successfully validated on blind data sets (ultrasonic, well logging, and seismic data) in different reservoirs with various lithologies and fluid types. The results indicate that the P-RPT could integrate the available RPTs into a 3D diagram and give a reliable framework for applications in seismic interpretation. User friendliness and generalizability are the most prominent advantages of this template for detecting fluid type and lithology based on well logs and seismic inversion, the results of which can be interpreted on a single chart rather than analyzing the data with different templates. The methodology and framework for implementing and generating templates are thoroughly explained in theory and practice to localize P-RPT or update the new RPT for another region.

An integrated approach to the substantiation of the boundary values and the assessment of the character of saturation in the deposits of the Tyumen formation

Background. The Tyumen formation is characterized by the complexity of the structure of deposits and consists of the sediments of different facies groups. To build hydrodynamic models, improve the development system and identify prospective zones, the conditions of sedimentation must be taken into account. Objective. Increasing the reliability of reservoir permeability assessment and clarifying critical resistivities when assessing the nature of saturation, which requires new approaches to petrophysical modeling. Materials and methods. The work uses the results of core studies and interpretation of well-logging data for the JV2 formation considering the facies. Results. The wettability of extracted and nonextracted samples is studied, the dependence of the saturation parameter on the water saturation coefficient is corrected taking into account the wettability of reservoirs, and the relative phase permeabilities are calculated for drainage conditions in the “oil–water” system. Conclusions. An integrated approach makes it possible to clarify the values of permeability, critical resistivities and oil and gas saturation of reservoirs, which can be used to identify prospecting zones when drilling new wells.

A Three-Zone Identification Method for Coal Mine Area Based on DS Evidence Theory

As coal ore and other resources are continuously mined, a three-zone structure is formed underground consisting of a sagging zone, fault zone, and caving zone. The use of well-logging data to identify the three zones is important for production safety and environmental management. Owing to the scarcity of data that can reflect three zones in normal coal mining, conventional identification and prediction methods face challenges when extracting data features, incurring a degree of uncertainty within prediction results. Accordingly, the accurate identification of the three zones has become a critical objective in daily production. To address this issue, we developed a method called a method called backpropagation neural networks with Dempster–Shafer (DS) evidence theory. Initially, we preprocessed the training data and deployed two backpropagation neural networks (BPNNs) to predict the three zones according to two parameters. According to these prediction results, the local and global credibility of each prediction is calculated and used to obtain the basic probability assignment function required for the DS evidence theory. Finally, the DS evidence theory is used to fuse the two BPNNs prediction results, thereby producing the final prediction results. The proposed method was demonstrated to improve prediction accuracy by 6.4% compared to a conventional neural network.

Geophysical characterization of Saraswati River palaeochannel in parts of Yamuna Nagar and Kurukshetra districts of Haryana, India.

Palaeochannels are remnants of rivers or stream channels filled with younger sediments over the period of time. In ancient times, these rivers/channels were thriving in phenomenal conditions, but due to frequent tectonic activities, they lost the direction of their original path and were gradually either lost or buried under thick beds of younger alluvium. Palaeochannels act as reservoirs for fresh groundwater since they are made up of coarser sediments and were formerly flowing rivers. Depending on the groundwater regime and local topography, these could either be saturated or dry. The palaeochannels have high groundwater potential if saturated. These are ideal sites for artificial groundwater recharge, if dry. The identification of palaeochannels becomes quite challenging if they are buried under thick deposits of finer younger sediments. In the present study, an attempt has been made to characterize the Saraswati River Palaeochannel in parts of Yamuna Nagar and Kurukshetra districts of Haryana by using surface and subsurface geophysical methods. Till date, the palaeochannels in this area were mainly discerned on the basis of remote sensing only; therefore, geophysical characterization of these palaeochannels has been attempted in this study. In surface geophysical methods, electrical resistivity surveys, especially gradient resistivity profiling (GRP) and vertical electrical sounding (VES), were conducted in the study area, while electrical and natural gamma logging was used as subsurface geophysical approaches to identify the coarser sands of buried palaeochannels. The main objective of the study was to characterize the Saraswati River palaeochannel and analyze the quality of the groundwater stored in the palaeochannel in the study area. The findings were compared with the well-log data and were found in good agreement.

Post-Fracture Production Prediction with Production Segmentation and Well Logging: Harnessing Pipelines and Hyperparameter Tuning with GridSearchCV

As the petroleum industry increasingly exploits unconventional reservoirs with low permeability and porosity, accurate predictions of post-fracture production are becoming critical for investment decisions, energy policy development, and environmental impact assessments. However, despite extensive research, accurately forecasting post-fracture production using well-log data continues to be a complex challenge. This study introduces a new method of data volume expansion, which is to subdivide the gas production of each well on the first day according to the depth of logging data, and to rely on the correlation model between petrophysical parameters and gas production to accurately combine the gas production data while matching the accuracy of the well-log data. Twelve pipelines were constructed utilizing a range of techniques to fit the regression relationship between logging parameters and post-fracture gas production These included data preprocessing methods (StandardScaler and RobustScaler), feature extraction approaches (PCA and PolynomialFeatures), and advanced machine learning models (XGBoost, Random Forest, and neural networks). Hyperparameter optimization was executed via GridSearchCV. To assess the efficacy of diverse models, metrics including the coefficient of determination (R2), standard deviation (SD), Pearson correlation coefficient (PCC), mean absolute error (MAE), mean squared error (MSE), and root-mean-square error (RMSE) were invoked. Among the several pipelines explored, the PFS-NN exhibited excellent predictive capability in specific reservoir contexts. In essence, integrating machine learning with logging parameters can be used to effectively assess reservoir productivity at multi-meter formation scales. This strategy not only mitigates uncertainties endemic to reservoir exploration but also equips petroleum engineers with the ability to monitor reservoir dynamics, thereby facilitating reservoir development. Additionally, this approach provides reservoir engineers with an efficient means of reservoir performance oversight.

Mechanisms for Microseismicity Occurrence Due to CO2 Injection at Decatur, Illinois: A Coupled Multiphase Flow and Geomechanics Perspective

ABSTRACT We numerically investigate the mechanisms that resulted in induced seismicity occurrence associated with CO2 injection at the Illinois Basin–Decatur Project (IBDP). We build a geologically consistent model that honors key stratigraphic horizons and 3D fault surfaces interpreted using surface seismic data and microseismicity locations. We populate our model with reservoir and geomechanical properties estimated using well-log and core data. We then performed coupled multiphase flow and geomechanics modeling to investigate the impact of CO2 injection on fault stability using the Coulomb failure criteria. We calibrate our flow model using measured reservoir pressure during the CO2 injection phase. Our model results show that pore-pressure diffusion along faults connecting the injection interval to the basement is essential to explain the destabilization of the regions where microseismicity occurred, and that poroelastic stresses alone would result in stabilization of those regions. Slip tendency analysis indicates that, due to their orientations with respect to the maximum horizontal stress direction, the faults where the microseismicity occurred were very close to failure prior to injection. These model results highlight the importance of accurate subsurface fault characterization for CO2 sequestration operations.

Fault and fracture network characterization using soft computing techniques: application to geologically complex and deeply-buried geothermal reservoirs

Geothermal energy is a sustainable energy source that meets the needs of the climate crisis and global warming caused by fossil fuel burning. Geothermal resources are found in complex geological settings, with faults and interconnected networks of fractures acting as pathways for fluid circulation. Identifying faults and fractures is an essential component of exploiting geothermal resources. However, accurately predicting fractures without high-resolution geophysical logs (e.g., image logs) and well-core samples is challenging. Soft computing techniques, such as machine learning, make it possible to map fracture networks at a finer resolution. This study employed four supervised machine learning techniques (multilayer perceptron (MLP), random forests (RF), extreme gradient boosting (XGBoost), and support vector regression (SVR)) to identify fractures in geothermal carbonate reservoirs in the sub-basins of East China. The models were trained and tested on a diverse well-logging dataset collected at the field scale. A comparison of the predicted results revealed that XGBoost with optimized hyperparameters and data division achieved the best performance than RF, MLP, and SVR with RMSE = 0.02 and R2 = 0.92. The Q-learning algorithm outperformed grid search, Bayesian, and ant colony optimizations. The blind well test demonstrates that it is possible to accurately identify fractures by applying machine learning algorithms to standard well logs. In addition, the comparative analysis indicates that XGBoost was able to handle the complex relationship between input parameters (e.g., DTP > RD > DEN > GR > CAL > RS > U > CNL) and fracture in geologically complex geothermal carbonate reservoirs. Furthermore, comparing the XGBoost model with previous studies proved superior in training and testing. This study suggests that XGBoost with Q-learning-based optimized hyperparameters and data division is a suitable algorithm for identifying fractures using well-log data to explore complex geothermal systems in carbonate rocks.Graphical abstract

Assessment of Sealing Potential of Fault-Bounded Hydrocarbon Prospect in Gabo Field, Niger Delta, Nigeria

Sealing potentials of faults are an important exploration risk to consider in the assessment of fault-bounded hydrocarbon prospects. This technique has been used for predicting potential hydrocarbon columns for mitigating risk in exploration and appraisals. Hence, the objective of the paper is to assess the sealing potential of the fault-bounded hydrocarbon prospects of the Gabo Field in the Niger Delta basin using seismic, well-log, and X-ray diffraction data. Three faults were interpreted as normal and syn-depositional faults. The well-log correlation showed that the shale sequence was inter-bedded with sands and showed a good spread with a relative thickness ranging between 12-105m from shallow to intermediate depth intervals. Results of the rock property analysis showed that the volume of shale ranged from 18 – 73.72% with an average of 63.59%. The total porosity ranged from 15.9 – 31.7 (good to excellent) with an average of 16.7 (good) in the shales and 27.2 (very good) in the sand. The effective porosity of the field ranged from 3.58 – 22.71 with an average of 6.028 in the shales and 20.13 in the sand. The estimated pore pressure ranged from 42.13 – 47.62psi with an average of 47.0psi in the shales and 42.8psi in the sand. The results of the X-ray diffraction analysis showed that the predominant minerals were kaolinite, rutile, gypsum, albite, microcline and quartz, which constitute the caprock sequence and the faulted rocks. Results of the sealing potential showed poor sealing in the shale sequence while the fault analysis showed good sealing along the fault planes. These results deduced the sealing uncertainty of the field and may be applied to similar siliciclastic trap configurations in deltaic environments.

An anisotropic rock physics modeling for the coalbed methane reservoirs and its applications in anisotropy parameter prediction

The elastic properties and anisotropy of coalbed methane reservoirs are still poorly understood due to the complicated pore structure and the existing state of the coalbed methane. Therefore, an anisotropic rock physics model for the coalbed methane reservoirs is proposed based on the related effective medium theories, which focuses on the equivalent calculation of the elastic properties for adsorbed coalbed methane and the characterization of the multiple pore structure. Many factors related to the elastic properties of coal are considered, including the composition, pore structure, adsorbed gas content and pore fluid. The measured ultrasonic and well-logging data demonstrate that the proposed physics model is effective in predicting the elastic constants and velocity anisotropy parameters. The results suggest that the elastic constants increase with the increase of adsorbed gas content to some extent, and the effect of porosity on the elastic constants is much more obvious than that of adsorbed gas content. The velocity anisotropy parameters increase with the aligned fracture porosity and the fracture aspect ratio, and the P-wave anisotropy is more sensitive to these two factors. With increasing organic matter and clay content and porosity, Young's modulus and brittleness index decrease while the Poisson's ratio increases, and the impact of the porosity on the brittleness index is more significant than that of organic matter and clay content. The results are helpful in establishing the relationship between reservoir physics parameters and seismic response and can provide a basis for coalbed methane reservoir prediction and hydraulic fracturing.