Well logging plays a critical role in characterizing reservoir petrophysical properties and is essential for hydrocarbon exploration and development. Nevertheless, the absence of well logs is a common issue during development, which can lead to a decrease in identification accuracy. A novel method for synthesizing missing well logs based on SBOA-XGBoost and Shapley Additive Explanations (SHAP) was proposed, where the missing well logs were reconstructed using available conventional well logs. The method can efficiently learn subsurface patterns and temporal trends from conventional well logs. SBOA-XGBoost was applied to synthesize natural gamma ray (GR), gamma ray without uranium (KTH), photoelectric (PE), compensated density (DEN), compensated neutron (CNCF), and compressional wave velocity (DTC) logs. For comparison, SBOA-XGBoost was compared with Fully Connected Neural Networks (FCNN), Bidirectional Long Short-Term Memory Networks (BiLSTM), and Convolutional Bidirectional Long Short-Term Memory Networks (CNN-BiLSTM) to evaluate its performance. Mean absolute error, root mean square error, and the coefficient of determination (R2) were utilized to objectively evaluate the predictive capability of the models. Synthesis accuracy was the highest for GR, KTH, DEN, and DTC, as indicated by their R2 values of 0.95, 0.92, 0.93, and 0.90. Reasonable accuracy was achieved for PE and CNCF, with R2 values of 0.82 and 0.83, respectively. SBOA-XGBoost surpassed the other methods in synthesizing all types of well logs. An interpretability analysis of the SBOA-XGBoost model was performed using SHAP, demonstrating that the model's decision-making mechanism aligns with domain knowledge.

Precise identification of water-flooded intervals is critical for optimizing reservoir development, particularly during the mid-to-late stages of water injection in oilfields. This study uses the Guantao Formation in Block X as a case study, where the prevalence of clay-rich lithologies and complex pore structures has led to the widespread occurrence of low-resistivity oil reservoirs. These conditions pose significant challenges to conventional logging interpretation techniques. To address this, the study investigated the underlying mechanisms responsible for low resistivity through integrated rock physics experiments. Building on these insights, an enhanced logging-based evaluation workflow was proposed to more accurately identify water-flooded zones. Firstly, a lithology-based correction was used to the Simandoux equation to improve the estimation of mixed water resistivity under low-resistivity conditions. Secondly, the extreme gradient boosting (XGBoost) machine learning algorithm was employed to predict effective porosity (EPOR), thereby reducing dependence on nuclear magnetic resonance (NMR) logging and improving evaluation efficiency. Finally, a cross-plot technique combining ΔRD and Swz-Swir was introduced to quantitatively classify water-flooded intervals. Validation using production well data from Block X yielded an interpretation accuracy of 90.6%, demonstrating significant improvement in identifying water-flooded zones in low-resistivity formations. Furthermore, successful application in adjacent blocks confirms the method’s transferability. Overall, this research offers a robust and scalable workflow for log-based reservoir evaluation in complex resistivity environments, providing critical insights for remaining oil prediction and improved reservoir management in similar geological settings.

The Archie equation is a fundamental model for describing the relationships between formation resistivity, porosity and water saturation, and plays an important role in log-based saturation evaluation. However, in complex reservoirs, its original structure and constant-parameter assumptions are often insufficient for refined reservoir evaluation because of complex mineral compositions, diverse pore structures and other controlling factors. Meanwhile, numerous Archie-extended equations have been developed, each relying on different applicability conditions, which introduces considerable uncertainty into the selection of the optimal model structure. To address this issue, this study proposes a physics-constrained symbolic-regression method for Archie-extended relationships based on network electrical conduction theory. The method consists of two stages. In the first stage, Archie-extended structures are systematically enumerated under the constraint of a general saturation interpretation model to identify formula structures that better conform to reservoir characteristics. In the second stage, a parallel symbolic regression network (PSRN) is used to construct explicit functional relationships between residuals or structural parameters and reservoir petrophysical properties, including porosity, permeability and shale content, for the formation factor–porosity relationship ( F–φ ) and the resistivity index–water saturation relationship ( I–Sw ), respectively. The proposed method is applied to a complex-lithology carbonate reservoir in the Persian Gulf Basin, yielding a weighted dual-power-law formation-factor model and a variable saturation-exponent resistivity-index model controlled by porosity and permeability. Compared with the classical Archie equation, the coefficients of determination,R2 , of the two models increase from 0.569 to 0.772 and from 0.856 to 0.986, respectively. The log-application results further indicate that the proposed method improves the applicability and accuracy of saturation evaluation in complex reservoirs while preserving the interpretability of Archie-type conduction behavior.

Accurate prediction of reservoir permeability constitutes a central task in hydrocarbon field development, but it remains challenging because core measurements are sparse and heterogeneous reservoirs commonly exhibit nonunique logging responses. Conventional machine learning methods typically treat each core measurement as an independent sample for end-to-end permeability prediction, thereby overlooking the stratigraphic context embedded in depth-ordered core sequences. In this study, we propose a depth-aware Transformer-based hybrid modeling framework, termed DATS, for geological parameter prediction from sparsely sampled core data. DATS constructs sequential samples using sliding windows of core measurements and incorporates depth information and stratigraphic priors through three key components: a continuous Fourier-based depth encoding with learnable frequency parameters, an attention mechanism incorporating depth-distance bias penalties, and a two-stage training strategy that bridges continuous well-logging data and discrete core measurements. Applied as a feature extractor to carbonate reservoirs in the study area, DATS was evaluated by using 2856 core samples. Through comparison among three feature combination strategies and 14 downstream models, the best-performing model, DATS-XGBoost_Combined, achieved an R 2 of 0.9404. Ablation experiments indicate that each constituent module of DATS contributes positively to the overall model performance. SHAP values and attention matrix analyses further suggest that the performance improvement is associated with the ability of DATS to capture depth-dependent stratigraphic patterns consistent with petrophysical principles. These results suggest that DATS can serve as a supplementary tool for permeability prediction in sparsely cored reservoirs and assist parameter evaluation in uncored or undersampled wells.

Abstract Electrofacies analysis is a key component of reservoir characterization because it links well-log responses to lithological and petrophysical properties; however its effectiveness is often limited by an overreliance on macroscopic logging signals. In heterogeneous reservoirs, microscopic pore structures exert a decisive influence on fluid storage and flow, and overlooking this control can lead to ambiguous or geologically inconsistent facies classifications. To overcome this limitation, this study integrates nuclear magnetic resonance while drilling (NMR-WD) T2 spectrum features with conventional logging data, allowing pore architecture to be quantified and incorporated directly into electrofacies analysis. Multidimensional pore-structure parameters are first extracted from the T2 spectrum and subsequently reduced through principal component analysis (PCA) to retain the dominant microstructural information while minimizing redundancy. Clustering is then carried out using an improved K-means++ algorithm, which enhances stability and reduces sensitivity to initial conditions, a common issue in unsupervised classification. The resulting electrofacies are validated through correlation with core-derived lithological data, ensuring geological consistency, and Fisher discriminant analysis is further applied to establish a predictive model for uncored intervals. The results indicate that the inclusion of NMR-WD-derived microstructural parameters significantly improves electrofacies discrimination, particularly in reservoirs where lithological differences are subtle but pore-scale heterogeneity is pronounced. Overall, this pore-aware framework advances traditional electrofacies analysis by bridging macroscale logging responses and microscale reservoir characteristics, providing a reliable and transferable tool for evaluating complex reservoirs and supporting informed exploration and development decisions.

Fracture parameters play a crucial role in productivity prediction, reservoir evaluation and fracturing production of buried-hill reservoirs and carbonate reservoirs. The main method for calculating fracture parameters is electrical logging based on rock-electric experiments. However, due to significant differences in observation systems between rock-electrical experiments and various electrical logging methods, directly calibrating electrical logging data with cores in different fractured formations will lead to large errors. Based on the finite element method calibrated with core samples, we established micro-fractured formation models and conducted fracture parameter simulation experiments for plunger core samples, full-diameter core samples, electrical imaging logging, micro-spherical focusing logging, shallow lateral logging, and deep lateral logging, respectively, aiming at the influence of multiple fracture parameters. A comparison of the measurement results of the six models for the same fractured formation showed that the fracture-induced resistivity reduction rates were ranked in descending order as follows: electrical imaging logging, plunger core testing, micro-spherical focusing logging, full-diameter core testing, shallow lateral logging, and deep lateral logging, with the maximum discrepancy in resistivity reduction rates across these models reaching a factor of 45. Specifically, the resistivity reduction rate of plunger cores was 2.7 times higher than that of full-diameter cores, and the rate of electrical imaging logging was 11.8 times higher than that of micro-spherical focusing logging, whereas the values for shallow lateral logging and deep lateral logging were identical. Finally, this study proposed a correction rate required for fracture core calibration, which could comprehensively optimize the interpretation range of various resistivity logging methods and effectively improve the interpretation accuracy of reservoirs.

The genetic mechanism of low-resistivity oil reservoirs in the Guantao Formation of the Bohai Sea is complex, and understanding the role of clay minerals and pore structures in forming low resistivity remains unclear. This study employed multiscale digital core technology to integrate multisource digital core data. Combined with flow characteristic-based upscaling technology, microscale and nanoscale digital cores were reconstructed to calculate parameters such as porosity, permeability, and formation factor. At the same time, mercury intrusion and seepage experiments were simulated. The research reveals two key mechanisms underlying the influence of clay minerals and pore structures on low-resistivity oil reservoirs: first, the additional conductive effect of clay. Among clay minerals, illite-smectite mixed-layer minerals exhibit the strongest conductivity, which is the key factor contributing to reduced resistivity, followed by illite, kaolinite, and chlorite in decreasing order of additional conductive capacity. The second is the regulatory role of pore structures. Kaolinite fills intergranular pores to form a complex micropore system, resulting in high irreducible water saturation in the reservoir, which is the primary cause of low-resistivity oil reservoirs. This study clarifies the influence mechanisms of clay minerals and pore structures on resistivity, provides key technical support for fine geological modeling and optimization of oil and gas reservoir development schemes, and holds significant practical value for reducing exploration risks and improving oil recovery efficiency.

Tight sandstone reservoirs are characterized by highly heterogeneous pore structures, in which multiscale pore-throat systems jointly control the shapes of capillary pressure curves and relative permeability, thereby exerting a fundamental influence on water production behavior and the overall development performance of gas reservoirs. The Ordos Basin is generally characterized by the development of tight sandstone. The tight sandstones exhibit porosities of 2-13% and permeabilities of 0.01-10 & times; 10-3 mu m2. To quantitatively elucidate the controlling mechanisms of multiscale pore structure on capillary pressure curve morphology and relative permeability, this study systematically investigates the fractal and multifractal characteristics of pore structures in tight sandstones based on high-pressure mercury intrusion (MICP) and nuclear magnetic resonance (NMR) experimental data, and establishes a quantitative relationship between fractal parameters and the capillary pressure curve shape parameter lambda. First, capillary pressure curves were fitted using the Brooks-Corey model within the effective saturation interval to extract the shape parameter lambda, which characterizes the concentration degree of pore-size distribution and the drainage behavior. Subsequently, based on NMR T2 spectra, the small-pore fractal dimension D1, large-pore fractal dimension D2, and the multifractal singularity spectrum width Delta alpha were extracted to quantitatively describe the geometric complexity of pore structures at different scales. On this basis, the correlations between lambda and D1, D2, and Delta alpha were systematically analyzed, and the predictive performance of lambda under different parameter combinations was compared. The results indicate that: (1) the pore structures of tight sandstones exhibit pronounced fractal and multifractal characteristics at the NMR T2 scale, with significant differences among samples; (2) lambda shows an overall negative correlation with fractal parameters, among which the correlations with the large-pore fractal dimension D2 and the multifractal spectrum width Delta alpha are the most significant; (3) compared with models using a single fractal dimension, the multiparameter model incorporating Delta alpha provides a more comprehensive characterization of multiscale pore heterogeneity, leading to a substantial improvement in the accuracy and stability of lambda prediction; and (4) lambda exerts a clear control on the shape of relative permeability curves, where a larger lambda corresponds to earlier initiation and forward-shifted rising segments of water-phase flow, while a smaller lambda results in overall flatter relative permeability curves. From the perspectives of fractal and multifractal theory, this study establishes an intrinsic linkage among pore structure, capillary pressure curve shape parameters, and relative permeability, providing a novel quantitative framework for constraining relative permeability curve morphology in tight sandstones under conditions where systematic relative permeability experiments are unavailable.

Accurate inversion of original formation resistivity is critical for reliable evaluation of water-flooded reservoirs; however, conventional methods often fail in strongly heterogeneous formations due to the neglect of lithology-controlled nonlinear responses. To overcome this limitation, this study proposes a lithology-constrained adaptive inversion framework for original resistivity in complex reservoirs. Unlike conventional approaches where lithology classification is only used for data grouping, lithological information is explicitly incorporated as a constraint to guide model selection and inversion strategy, thereby linking geological characteristics with data-driven modeling. Based on core experiments and logging responses, reservoirs in Oilfield A of the Pearl River Mouth Basin are classified into three types: clean sandstone, argillaceous sandstone, and calcareous reservoirs. For each reservoir type, an adaptive modeling scheme is developed, in which machine learning models (Random Forest, XGBoost, and Support Vector Machine) are optimally matched according to data scale, nonlinearity, and electrical response characteristics. The inverted original resistivity is further integrated with rock-electrical experiments and the Indonesian model to calculate water saturation, bound water saturation, oil displacement efficiency, and water production rate, enabling a physically consistent and quantitative evaluation of water-flooded layers. Compared with non-classified unified inversion models, the proposed lithology-constrained adaptive framework achieves lower prediction errors, demonstrating that lithological constraints can effectively improve the accuracy and robustness of original resistivity inversion in heterogeneous reservoirs. Application to 19 wells yields an overall agreement rate of 89.5% with actual production data, demonstrating strong reliability and practical applicability. This study highlights that incorporating lithological constraints into data-driven modeling is essential for capturing reservoir heterogeneity, and provides a generalizable and physically consistent methodology for resistivity inversion and water-flooded layer evaluation in complex geological settings.

The tight sandstone reservoirs in the Linxingdong area of the Ordos Basin are characterized by low porosity, high irreducible water saturation, low gas saturation, and strong heterogeneity, with conventional logging methods showing low accuracy in identifying reservoir fluid properties. This paper proposes a method based on principal component analysis (PCA) for identifying post-fracturing productivity classification in complex tight gas reservoirs. Initially, the fluid productivity classification type of the test layers was determined by integrating gas and liquid production data from the fracturing and production stages. Subsequently, logging data were utilized to select curves sensitive to reservoir fluid properties, and a calculation formula for the fluid property identification factor D was established based on PCA. The classification of reservoir fluid productivity in the study area was then identified based on the D value. The results indicate that the fluid property identification factor method based on PCA performed well in the study area, with an accuracy of 83%. Compared with other conventional methods, this approach fully exploits geophysical logging data, has strong generalizability, and provides data support for the formulation and adjustment of subsequent development plans.

Accurate production capacity prediction plays a crucial role in formulating efficient development plans for tight gas reservoirs. Due to the complexity of tight gas reservoir characteristics and significant reservoir heterogeneity, traditional prediction methods often fail to meet the accuracy and stability requirements in practical applications. This paper proposes an innovative production capacity prediction model, ADASVRLGBM, which integrates AdaBoost (Adaptive Boosting), SVR (Support Vector Regression), and LGBM (Light Gradient Boosting Machine) algorithms. The model utilizes GridSearchCV (Grid Search Cross-Validation) to fine-tune the hyperparameters of each algorithm and applies a genetic algorithm to optimize the weight combinations of the sub-models. The integrated model systematically analyzes the correlation of factors influencing tight gas production, extracts key feature parameters, and builds a predictive model with gas well production capacity as the output label. The study demonstrates that the integrated model significantly outperforms single algorithms in terms of prediction accuracy, achieving an average agreement rate of 93.33% after training. Furthermore, the paper provides an in-depth discussion of the contributions of different sub-models to overall prediction performance and highlights their advantages in handling complex data. The findings offer theoretical and practical support for the efficient development of tight gas reservoirs and valuable insights for extending the application of similar models.

Tight gas is an unconventional resource abundantly found in low-porosity, low-permeability sandstone reservoirs. Production can be significantly reduced due to water production during the development process. Therefore, it is necessary to predict water production during the logging phase to formulate development strategies for tight gas wells. This study analyzes the water production mechanism in tight sandstone reservoirs and identifies that the core of water production evaluation in the Shihezi Formation of the Linxing block is to clarify the pore permeability structure of tight sandstone and the type of intra-layer water. The primary challenge lies in the accurate characterization of bound water saturation. By integrating logging data with core experiments, a bound water saturation evaluation model based on grain size diameter and pore structure index was established, achieving a calculation accuracy of 92% for the multi-parameter-fitted bound water saturation. Then, based on the high-precision bound water saturation, a gas–water ratio prediction model for the first month of production, considering water saturation, grain size diameter, and fluid type, was established, improving the prediction accuracy to 87.7%. The bound water saturation evaluation and water production evaluation models in this study can achieve effective water production prediction in the early stage of production, providing theoretical support for the scientific development of tight gas in the Linxing block.

As the exploration and development of fracture-type hydrocarbon reservoirs progress, accurately identifying reservoir fracture development is crucial for reservoir evaluation. This method is founded on array acoustic logging data, integrating fractal theory and singular spectrum analysis. It computes and evaluates the full-wave signals and dipole shear wave signals to assess their self-similarity and chaotic characteristics in various stratigraphic conditions. A comprehensive fracture identification method suitable for complex formations has been established, with its application to carbonate fracture-type reservoir characterization demonstrated through a case study of the Asmari Formation in the M oilfield, southeastern Iraq, as a case study. This method demonstrates an accuracy of approximately 90 % in identifying both low-to-moderate and high-angle fractures. Compared to traditional mode wave attenuation methods, this approach significantly reduces errors in complex wellbore environments, providing a new technical pathway for exploring and developing fracture-type hydrocarbon reservoirs.

The strong heterogeneity of carbonate reservoirs poses significant technical challenges in reservoir classification and permeability evaluation. This study proposes a new method for reservoir classification based on nuclear magnetic resonance (NMR) logging data for the Asmari formation of the Middle East M Oilfield, a carbonate reservoir. By integrating NMR T2 spectrum characteristic parameters (such as T2 geometric mean, T2R35/R50/R65, and pore volume fraction) with principal component analysis (PCA) for dimensionality reduction and an improved slope method, this study achieves fine reservoir type classification. The results are compared with core pressure curves and petrographic pore types. This study reveals that the Asmari reservoir can be divided into four categories (RT1 to RT4). RT1 reservoirs are characterized by large pore throats (maximum pore throat radius > 3.8 μm), low displacement pressure (<0.2 MPa), and high permeability (average 22.16 mD), corresponding to a pore structure dominated by intergranular dissolution pores. RT4 reservoirs, on the other hand, exhibit small pore throats (<1 μm), high displacement pressure (>0.7 MPa), and low permeability (0.66 mD) and are primarily composed of dense dolostone or limestone. The classification results show good consistency with capillary pressure curves and petrographic pore types, and the pore–permeability relationships of each reservoir type have significantly higher fitting goodness (R2 = 0.48~0.68) compared with the unclassified model (R2 = 0.24). In the new well application, the root mean square error (RMSE) of permeability prediction decreased from 0.34 mD using traditional methods to 0.21 mD, demonstrating the method’s effectiveness. This approach does not rely on a large number of mercury injection experiments and can achieve reservoir classification solely through NMR logging. It provides a scalable technological paradigm for permeability prediction and development scheme optimization of highly heterogeneous carbonate reservoirs, offering valuable references for similar reservoirs worldwide.

In this study, we propose a self-adaptive weighted multi-mineral inversion model (SQP_AW) based on Sequential Quadratic Programming (SQP) and the Adam optimization algorithm for the accurate evaluation of mineral content in carbonate reservoir rocks, addressing the high costs of traditional experimental methods and the strong parameter dependence in geophysical inversion. The model integrates porosity curves (compensated density, compensated neutron, and acoustic time difference), elastic modulus parameters (shear and bulk moduli), and nuclear magnetic porosity data for the construction of a multi-dimensional linear equation system, with calibration coefficients derived from core X-ray diffraction (XRD) data. The Adam algorithm dynamically optimizes the weights, solving the overdetermined equation system. We applied the method to the Asmari Formation in the M oilfield in the Middle East with 40 core samples for calibration, achieving a 0.91 fit with the XRD data. For eight additional uncalibrated samples from Well A, the fit reaches 0.87. With the introduction of the elastic modulus and nuclear magnetic porosity, the average relative error in mineral content decreases from 9.45% to 6.59%, and that in porosity estimation decreases from 8.1% to 7.1%. The approach is also scalable to elemental logging data, yielding inversion precision comparable to that of commercial software. Although the method requires a complete set of logging data and further validation of regional applicability for weight parameters, in future research, transfer learning and missing curve prediction could be incorporated to enhance its practical utility.

Geophysical logging curves are crucial for oil and gas field exploration and development, and curve reconstruction techniques are a key focus of research in this field. This study proposes an inversion model for deep resistivity curves in marine carbonate reservoirs, specifically the Mishrif Formation of the Halfaya Field, by integrating a deep learning model called CNN-GRU-ATT, which combines Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and the Attention Mechanism (ATT). Using logging data from the marine carbonate oil layers, the reconstructed deep resistivity curve is compared with actual measurements to determine reservoir fluid properties. The results demonstrate the effectiveness of the CNN-GRU-ATT model in accurately reconstructing deep resistivity curves for carbonate reservoirs within the Mishrif Formation. Notably, the model outperforms alternative methods such as CNN-GRU, GRU, Long Short-Term Memory (LSTM), Multiple Regression, and Random Forest in new wells, exhibiting high accuracy and robust generalization capabilities. In practical applications, the response of the inverted deep resistivity curve can be utilized to identify the reservoir water cut. Specifically, when the model-inverted curve exhibits a higher response compared to the measured curve, it indicates the presence of reservoir water. Additionally, a stable relative position between the two curves suggests the presence of a water layer. Utilizing this method, the oil–water transition zone can be accurately delineated, achieving a fluid property identification accuracy of 93.14%. This study not only introduces a novel curve reconstruction method but also presents a precise approach to identifying reservoir fluid properties. These findings establish a solid technical foundation for decision-making support in oilfield development.

The tight sandstone gas reservoirs in the LX area of the Ordos Basin are characterized by low porosity, poor permeability, and strong heterogeneity, which significantly complicate fluid type identification. Conventional methods based on petrophysical logging and core analysis have shown limited effectiveness in this region, often resulting in low accuracy of fluid identification. To improve the precision of fluid property identification in such complex tight gas reservoirs, this study proposes a hybrid deep learning model named ResViTNet, which integrates ResNet (residual neural network) with ViT (vision transformer). The proposed method transforms multi-dimensional logging data into thermal maps and utilizes a sliding window sampling strategy combined with data augmentation techniques to generate high-dimensional image inputs. This enables automatic classification of different reservoir fluid types, including water zones, gas zones, and gas–water coexisting zones. Application of the method to a logging dataset from 80 wells in the LX block demonstrates a fluid identification accuracy of 97.4%, outperforming conventional statistical methods and standalone machine learning algorithms. The ResViTNet model exhibits strong robustness and generalization capability, providing technical support for fluid identification and productivity evaluation in the exploration and development of tight gas reservoirs.

The accurate prediction of water saturation in reservoir exploration and development remains a significant challenge, particularly in regions like the Middle East with complex carbonate formations such as the Mishrif Formation. While geophysical logging data is widely utilized for this purpose, however, the complex pore structures render Archie's formula unsuitable, leading to non-Archie phenomenon in rock-electrical experiments. Although the electrical efficiency model has been employed in calculating water saturation in carbonate reservoirs, there has been no prior study incorporating electrical porosity for its refinement. This study enhances the conventional electrical efficiency model by introducing the concept of electrical porosity. The improvement aims to mitigate the impact of isolated mold pores and sparse regions of current density distribution on electrical efficiency, focusing on the Mishrif Formation and quantitatively computing water saturation. Initially, the study area is categorized into three distinct rock-physics types of reservoirs using the Winland R35 method, with respective electrical porosities calculated. Subsequently, these results are integrated with the enhanced electrical efficiency model, and trial calculations are performed using geophysical well-logging data, followed by a comparison with core data. The findings reveal that the improved electrical efficiency model yields an average relative error of only 10.36 % compared to core data, whereas the respective errors for Archie's formula and traditional electrical efficiency models are 17.65 % and 20.92 %, indicating enhanced accuracy with the improved approach. Across different reservoir types, a decrease in electrical porosity proportion is observed with diminishing pore-throat radius. Additionally, the consistency of this trend is validated by nuclear magnetic resonance logging data. Lastly, the necessity of reservoir rock-physics type classification for electrical porosity computation is confirmed. For heterogeneous reservoirs, direct calculation of electrical porosity is infeasible, thus underscoring the essential groundwork of reservoir rock-physics type delineation. This study improves water saturation prediction accuracy and applicability by introducing electrical porosity to refine the conventional electrical efficiency model, holding significant implications for the exploration and development of complex reservoirs.

To address the challenge of identifying water-flooded layers in the high-porosity, high-permeability, and strongly heterogeneous reservoirs of the Guantao Formation in the Penglai 19-3 Oilfield, research on water-flooded layer identification methods was systematically conducted. The logging characteristics of oil layers and water-flooded layers at different levels overlap considerably, which limits the accuracy of traditional identification methods. Meanwhile, the Archie equation shows significantly reduced applicability during the moderate and strong water-flooding stages. A water-flooded layer identification model was constructed using HistGBDT, and performance comparison between the base model and the optimized model reveals that the latter achieves a test accuracy of 91.6%. Compared with BPNN and SVM, the optimized HistGBDT model demonstrates substantially higher test accuracy and better generalization performance. Based on six sets of logging data, the optimized HistGBDT model developed enables the accurate identification of oil layers and multi-level water-flooded layers. It provides a reliable technical approach for tapping remaining oil in the high-water-cut stage of the Penglai 19-3 Oilfield and offers a new method and engineering reference for water-flooded layer identification in similar high-porosity, high-permeability heterogeneous reservoirs in the Bohai Bay Basin.

Accurate evaluation of permeability parameters is critical for the exploration and development of oil and gas fields. Among the available techniques, permeability assessment based on nuclear magnetic resonance (NMR) logging data is one of the most widely used and precise methods. However, the rapid biochemical variations in marine environments give rise to complex pore structures and strong reservoir heterogeneity, which diminish the effectiveness of traditional SDR and Timur–Coates models. To address these challenges in complex carbonate reservoirs, this study proposes a high-precision permeability evaluation method that integrates the Gaussian distribution model with the Thomeer model for more accurate permeability calculations using NMR logging data. Multimodal Gaussian distributions more accurately capture the size and distribution of multiscale pores. In this study, we innovatively employ the Gaussian distribution function to construct NMR-derived pseudo-pore size distribution curves. Subsequently, Thomeer model parameters are derived from Gaussian distribution parameters, enabling precise permeability calculation. The application of this method to the marine dolomite intervals of the Asmari Formation, Section A, within Oilfield A in southeastern Iraq, demonstrates its superior performance under both bimodal and unimodal pore size distributions. Compared to traditional models, this approach significantly reduces errors, providing crucial support for the accurate evaluation of complex reservoirs and the development of hydrocarbon resources.