Conventional Log Data Research Articles

Automated lithofacies classification: A comprehensive machine learning approach in Shushan Basin reservoirs, Western Desert, Egypt

The application of machine learning serves as a pivotal tool for petroleum geologists in facies classification. This new workflow distinguishes itself from existing classifiers by leveraging hidden statistical patterns in logging data to present a few recognizable clustering options for geologists. These choices are guided by other geological data sources, allowing geologists to retain the dimensional locations of chosen clusters for identification in other wells lacking these additional sources. The classification technique maximizes the value of conventional logging data (gamma ray, resistivity, density, neutron and sonic) for discerning rock typing, porosity ranking, fluid content, highlighting similar petrographic characteristics and elements composition, facilitating the inference of porosity and permeability degrees with high confidence.The workflow is designed in this study to predict siltstone, shale, limestone, basaltic intrusions, and coal, accurately identifies various sandstone sub-facies, differentiates between tight and hydrocarbon-bearing sandstone across four wells, with blind validation on a separate well. The classification is validated using Litho Scanner tool, petrography thin sections, and laboratory analysis.This comprehensive approach demonstrates the efficiency and applicability of the methodology, marking significant advancements in facies classification within petroleum geology.

Enhanced Carbon/Oxygen Ratio Logging Interpretation Methods and Applications in Offshore Oilfields

As the development of most domestic and international oilfields progresses, many fields have entered a mature phase characterized by high water cut and high recovery, with water cut levels often exceeding 90%. Carbon/oxygen ratio logging has proven to be an indispensable tool for distinguishing oil layers from water layers in complex environments, especially where salinity is low, unknown, or highly variable. This logging method has become one of the most effective techniques for determining residual oil saturation in cased wells, providing critical insights into the oil–water interface. In this study, we evaluate two key interpretation models for carbon/oxygen ratio logging: the fan chart method and the ratio chart method. We optimize the interpretation parameters in the ratio chart model using an improved genetic algorithm, which significantly enhances interpretation precision. The optimized parameters enable a more seamless integration of logging results with reservoir and conventional logging data, reducing the influence of lithological variations and physical property differences on the measurements. This research establishes a robust theoretical foundation for enhancing the interpretation accuracy of carbon/oxygen ratio logging, which is crucial for effectively identifying water-flooded layers. These advancements provide vital technical support for monitoring oil–water dynamics, optimizing reservoir management, and improving production efficiency in oilfield development.

The invention relates to a method for simulating phase permeability curve based on conventional logging data

Abstract Phase permeability curve is the curve of relative permeability (effective permeability/oil phase permeability in bound water) and water saturation change obtained through core phase permeability experiment. It is an indispensable data in oilfield development calculation, dynamic analysis, determination of oil-water saturation distribution in reservoirs and other water-driven oil related calculations, and an important index for reservoir evaluation. The study of effective permeability and relative permeability is of great significance for controlling the uniform advance of oil-water or oil-gas boundary and rationally developing oil and gas fields. The only effective way to study the phase infiltration is the core phase infiltration experiment, but the number of lithology samples is very limited, so the phase infiltration in oilfield development and application is still in the estimation stage, the accuracy is low, and the practical application is not strong. In order to improve the applicability of the phase permeability curve, it is necessary to establish a set of electrical interpretation methods associated with logging curves to realize electrical simulation instead of core measurement. In this paper, an electrical interpretation model of oil-water phase permeability is established based on flowable porosity and pore radius, and the effective seepage dynamic prediction and inversion of oil-water phase in each development stage of each reservoir is realized, which lays a solid foundation for the fine development of the reservoir.

Identification of the low resistivity-low contrast (LRLC) gas-bearing pay zones in Shaly sand reservoirs using acoustic data: a case study from the Messinian Abu Madi formation, onshore Nile Delta, Egypt

Detection of the low resistivity-low contrast (LRLC) reservoirs is among the main challenges in the oil industry. In this concern, the LRLC pay zones of the Upper Messinian Abu Madi clastic reservoirs in the onshore Nile Delta Gas fields became a main challenge for significant exploration. This type of reservoirs, including low resistivity-low contrast zones and thin-bedded intervals, are often overlooked using the conventional petrophysical evaluation techniques, especially in the wildcat exploratory wells or highly agitated shoreline depositional environments like the Nile Delta of Egypt. These hidden low contrast reservoirs are generally challenging due to the presence of many shale intercalations/laminations and/or due to increasing the shale volume represented in the form of dispersed distribution, and the dominance of conductive clay minerals. Therefore, in this study, the expected high resistivity values of the gas-bearing reservoir intervals of the Abu Madi Formation in the onshore Begonia gas Field, as a typical case study of the LRLC reservoirs, are masked due to the relatively high shale conductivity, particularly when the thickness of these intervals is less than the vertical resolution of the utilized conventional resistivity log. To verify the LRLC phenomena of the Begonia gas Field, the obtained data was compared to the South Abu El Naga gas Field as a normal case study with a relatively high resistivity gas-bearing pay zone. To overcome the impact of the conductive clay mineral content and identify these hidden low resistivity reservoir intervals, it is necessary to integrate the conventional logging data (gamma-ray, shallow and deep resistivity, density, and neutron) with the acoustic log data including shear and compressional sonic data. In this way, a useful relationship can be established enabling the detection of these hidden LRLC reservoir intervals. This integration is based on the principle that shear waves are not influenced by the fluids types, whereas the compressional sonic waves are influenced by the reservoir fluids. However, to effectively investigate these concealed LRLC reservoir intervals, which can boost production and increase the potential reserves, it is essential to have a low water cut value. The present study represents introduces an efficient workflow, which can be extended to other similar LRLC pay zones in the Nile Delta and northeast Africa. It is also extendible to the LRLC reservoirs in similar deltaic systems having conductive minerals-bearing reservoirs or thin beds.

An integrated convolutional neural network prediction framework for in situ shale oil content based on conventional logging data

The quantification of total organic carbon (TOC) and the free hydrocarbon content (S 1 ) is crucial for evaluating the shale oil generation and bearing properties of source rocks. This study aimed to enhance the accuracy of TOC and S 1 quantification in shale oil evaluations. The scope encompassed the development of a novel deep learning framework to overcome the limitations of traditional physical and machine learning or deep learning methods. This paper proposes an integrated swarm optimization algorithm–convolutional neural network/machine learning framework. This framework uses a swarm optimization algorithm for the hyperparameter optimization of a convolutional neural network or machine learning framework, utilizing experimental data from core samples preserved in liquid nitrogen alongside well logging data. The application of the proposed framework to the H11 well in the Subei Basin, China, using 110 core samples, demonstrated a superior performance. The results validate the framework's effectiveness in predicting the TOC and S 1 contents at various depths. The proposed framework stands out for its convenient methodology, wide application range and high precision in prediction. These attributes contribute significantly to the field of petroleum engineering and development, offering a novel approach that promises to enhance both the efficiency and accuracy of well logging evaluation.

Identification of Multi-Parameter Fluid in Igneous Rock Reservoir Logging—A Case Study of PL9-1 in Bohai Oilfield

Since the “13th Five-Year Plan”, the exploration of large-scale structural oil and gas reservoirs in the Bohai oilfield has become more complex, and the exploration of igneous oil and gas reservoirs has become the focus of current attention. At present, igneous rock reservoir fluid identification methods are mainly based on the evaluation method of logging single parameter construction, which is primarily a qualitative identification due to lithology, physical property, and engineering factors. Accurate acquisition of interference logging data, and multi-parameter coupling and recording coupling methods are few, lacking systematic and comprehensive evaluation and analysis of logging data. Since conventional logging data in the study area have difficulty accurately and quickly identifying reservoir fluid properties, a systematic analysis was conducted of three factors: lithology, physical properties, and engineering, as well as a variety of logging parameters (gas measurement, three-dimensional quantitative fluorescence, geochemical, FLAIR, etc.) that can reflect fluid properties were integrated. Based on parameter sensitivity analysis, the quantitative characterization index FI of multi-parameter coupling fluid identification was established using the data from testing, sampling, and laboratory testing to develop the identification standard. The sensitivity analysis and optimization of characteristic parameters were carried out by integrating the data reflecting fluid properties such as gas surveys, geochemical data, and related logging data. Combined with gas logging-derived parameters and improved engineering parameters (the value of alkanes released by rock cracking per unit volume Cadjust, C1 abnormal multiple values, three-dimensional quantitative fluorescence correlation factor N), the fluid properties were identified, evaluation factors were constructed based on factor analysis, and fluid identification interactive charts were established. By analyzing test wells in the PL9-1 well area, the results of comparison test data are more reliable. Compared with conventional methods, this method reduces the dependence of a single parameter by synthesizing multiple parameters and reduces the influence of lithology, physical properties, and engineering parameters on fluid identification. It is more reasonable and practical. It can accurately and quickly identify the fluid properties of igneous rock reservoirs in the study area. It has a guiding significance for improving the accurate evaluation of logging data and increasing exploration benefits.

Bedding-parallel fracture density prediction using graph convolutional network in continental shale oil reservoirs: A case study in Mahu Sag, Junggar basin, China

The natural fractures, e.g., the bedding-parallel fractures (BPF), have a significant impact on the storage space and horizontal permeability of continental shale oil reservoirs. However, the conventional logging response of BPF is complex. To solve the problem of BPF density prediction, we propose the Edge Generation Weighted Graph Convolutional Network (EG-WGCN) method, which is an improved Graph Convolutional Network (GCN). The new method integrates the relationship information between the fractures and lithology into the edge generation method, and adds the generated edge weight information into the message passing process, thereby effectively improving the accuracy of fracture density prediction. The prediction process is divided into three steps: first, using conventional logging sampling points as vertices, establish vertex sets for each lithology and single lithologic layer. Second, based on the depth sequence of a single lithologic layer and the Euclidean distance of the same lithologic vertex sets, the connection between vertices is established and overlapping edges generated by the above methods are given higher weights. Third, the vertices in the constructed graph are classified through graph convolutional layers, which are integrated into the edge weight information. This method is applied to the fracture density prediction of continental shale oil reservoirs in the Mahu Sag of the Junggar Basin, western China. The results show that the accuracy of our proposed EG-WGCN method in testing data is 95.13%. Compared to the GCN (boundless) without using the edge generation method and EG-GCN with the edge generation method but without the weighted aggregation mechanism, model accuracy is improved by 14.48% and 1.22%, respectively. In summary, this method proves that if a proper ML model is used, conventional logging data can become valuable for predicting BPF density with high accuracy even when cores are missing.

Shear-Wave Velocity Prediction Based on the CNN-BiGRU Integrated Network with Spatiotemporal Attention Mechanism

Shear wave velocity is one of the important parameters reflecting the lithological and physical properties of reservoirs, and it is widely used in the fields of lithology and fluid property identification, reservoir evaluation, seismic data processing, and interpretation. However, due to the high cost and challenge of obtaining shear wave velocity, only a few key wells are measured. Considering the intricate nonlinear mapping relationship between shear wave velocity and conventional logging data, an integrated network incorporating an attention mechanism, a convolutional neural network, and a bidirectional gated recurrent unit (STACBiN) is proposed for predicting shear wave velocity. The impact of conventional logging data on shear wave velocity is analyzed, thus employing the attention mechanism to focus on data correlated with shear wave velocity, which can enable the prediction results of the method proposed superior to those of conventional methods. Additionally, the prediction results of this method are compared with the prediction results of the two-dimensional convolutional neural network (2DCNN) and bidirectional gated recurrent unit (BiGRU). It is verified that the network proposed can effectively predict the shear wave velocity, with minimal error between predicted and true values.

Study on Main Controlling Factors and Fluid Identification Methods of Low-resistivity Reservoirs based on BP Nerve Network Verification in X Area of Ordos Basin

The low-resistivity oil layers refer to a formation whose resistivity response is opposite to the typical oil layers and similar to water layer. Distinguishing it from water layer in logging interpretation is very difficult, which brings great difficulties to oil and gas exploration and development. The exploration and development research shows that there is obvious resistivity anomaly in X area of Ordos Basin, and typical low-resistivity oil layers are developed in formation A in this area, so it is difficult to accurately divide oil layers and water layers only from logging response. In order to identify the fluid properties in low-resistivity oil layer areas, based on conventional logging data of formation A in X area, through petrophysical experimental analysis, combined with formation water analysis data, logging, oil test and other data, the main control factors of low-resistivity oil layers in X area are deeply studied in many aspects, and a more effective fluid identification method is put forward by combining the secondary evaluation of porosity and permeability and the flow unit method. The BP nerve network simulation cross plot is used to verify the identification results. The research shows that the main control factors of low resistivity oil layers in the study area are related to the salinity of water and pore connectivity, and also to the lithologic properties of different formations. Based on the study of the main control factors of low-resistivity oil layers, a multi-factor fluid identification method is proposed. The coincidence rate between the evaluation results of this identification method and oil test is 85.7%, which can identify the oil-water reservoirs in X area more effectively and accurately, finally identify the location of low-resistivity oil layers and boost the growth rate of oil and gas production.

Formation mechanical mechanism of low-angle fractures in pre-salt carbonate reservoirs: a case study from Carboniferous KT-I formation of the NT oilfield in the eastern margin of the Precaspian Basin, Kazakhstan

Fracture is an important factor that affects the oil and gas productivity of carbonate reservoirs. Much researches have been done on the origin of high-angle fractures in carbonate reservoirs, but few efforts have been made on the genetic mechanical mechanism of low-angle fractures. Based on the seismic data, core data, conventional logging data and rock mechanics experimental data, combined with three-dimensional in situ stress field simulation methods, the features, formation geological conditions and genetic mechanical mechanism of low-angle fractures (LAFs) were analyzed by applying Coulomb–Moore criterion, Griffith criterion and non-coordination criterion. The proportion of the number of shear fractures is as high as 90.2%, while that of tensile fractures is only 9.8% in the study area. Shear fractures are mainly unfilled fractures, and tensile fractures are mainly partially filled fractures. The LAFs were formed in the second tectonic movement, in which the knee-fold structure with high in the west and low in the east developed in the study area. The buried depth of most parts of the KT-I formation is 800 m when the study area develops the knee-fold structure, with a maximum depth of 1800 m and a minimum buried depth of 70 m, and the dip angle of the steepest part of the stratum is about 20°. A large number of LAFs were formed in the study area under the joint influence of tectonics and abnormally high pressure of water, including near-horizontal LAFs in the non-weak fabrics section (type I low-angle shear fractures), the LAFs having a certain angle with bedding in the non-weak fabrics section (type II low-angle shear fractures) and near-horizontal LAFs in the weak fabrics section (type III low-angle shear fractures). The formation of type I and type II low-angle shear fractures follows the Coulomb–Moore criterion. Type I low-angle shear fractures are formed in strata with a certain dip angle, while type II low-angle shear fractures are formed in near-horizontal strata. Type III low-angle shear fractures are formed under the comprehensive influence of pre-existing weak fabrics and strong horizontal extrusion, which follows the non-coordination criterion. Low-angle tensile fractures are mainly caused by abnormally high pressure and reverse faults in the study area, following Griffith’s criterion. The research in this paper not only reveals the formation mechanical mechanism of LAFs in pre-salt carbonate reservoirs but also provides guidance for the prediction of LAFs and solving the problem of water channeling caused by LAFs in oil fields.

Simultaneously improving ROP and maintaining wellbore stability in shale gas well: A case study of Luzhou shale gas reservoirs

The ROP (rate of penetration) within the horizontal section of shale gas wells in the Luzhou oil field is low, seriously delaying the exploration and development process. It is proved that reducing mud density mitigates the bottom-hole differential pressure (ΔP) and increases the ROP during overbalanced drilling. However, wellbore collapse may occur when wellbore pressure is excessively low. It is urgent to ascertain the optimal equilibrium point between improving ROP and maintaining wellbore stability. The safe mud weight window and the lower limit of mud density in the horizontal section of the Luzhou block are predicted using the piecewise fitting method based on conventional logging data. Then, the accuracy of the collapse pressure prediction was verified using the distinct element method (DEM), and the effect of wellbore pressure, in-situ stress, rock cohesion, and natural fracture density on borehole collapse was investigated. Finally, a fitting model of ΔP and ROP of the horizontal section in the Luzhou block is established to predict ROP promotion potential after mud density reduction. The field application of this approach, demonstrated in 8 horizontal wells in the Luzhou block, effectively validates the efficiency of reducing mud density for ROP improvement. This study provides a useful method for simultaneously improving ROP and maintaining wellbore stability and offers significant insights for petroleum engineers in the design of drilling parameters.

Advancing subsurface analysis: Integrating computer vision and deep learning for the near real-time interpretation of borehole image logs in the Illinois Basin-Decatur Project

The accurate quantification and mapping of subsurface natural fracture systems using borehole imaging logs are critical for the success of CO2 sequestration in geologic formations, optimization of engineered geothermal systems, and hydrocarbon production enhancement. However, traditional interpretation processes suffer from time-consuming procedures and human bias. To address these challenges and expedite fracture analysis, we investigated the application of integrated computer vision and DL workflows to automate image log analysis. Specifically, the design of our workflow was crafted to swiftly detect fractures and baffles by using actual electrical resistivity of borehole wall from microresistivity imaging device alongside their binary representation. This novel approach significantly reduces computational time while providing invaluable insights. By incorporating conventional logging and microseismic data, we present a regional subsurface natural fracture mapping technique. Through the minimization of human bias in image log analysis, our automated workflow achieves reduced fracture interpretation time and costs while ensuring robust and reproducible results. We demonstrated the efficacy of our approach by applying the workflow to the Illinois Basin-Decatur Project site. The automated workflow successfully identified major fractured zones, multiple baffles, and an interbedded layer with a high resolution of 0.01 ft or 0.12 in. (0.3 cm) and can be upscaled to any desired resolution. Validation through microseismic and image log interpretations allows for accurate and near-real-time mapping of fractures and baffles, significantly enhancing CO2 pressure forecasting and postinjection site care. Our approach stands out due to its robustness, consistency, and reduced computational cost compared with alternative feature extraction technologies. It presents exciting possibilities for advancing CO2 sequestration and engineered geothermal efforts by offering comprehensive and efficient fracture mapping solutions. This technology can contribute significantly to the optimization of CO2 sequestration projects, facilitating sustainable environmental practices, and combating climate change.

Prediction and effectiveness study of fracture distribution based on an artificial neural network technology and production performance data detection method: a case study of Ordovician carbonate formation in the Tahe oilfield, Tarim Basin

To solve the difficulty of predicting the spatial distribution of fracture linear density and clarify the controlling influence of tectonic factors on the effectiveness of fractures. This article combines core, conventional logging, imaging logging, and 3D seismic volume data and uses artificial intelligence neural network technology combined with the sequential Gaussian method to establish a linear density volume prediction model for fractures in the study area. In addition, production dynamic data are applied to verify the effectiveness of the fractures. The results show that the artificial intelligence neural network technique is highly applicable with an accuracy of up to 76.8% in predicting the fracture linear density. The fracture linear density is controlled by tectonic activity, and the fracture linear density increases sharply in the areas adjacent to the main faults, the fracture superposition site, and the intersection site of multiple faults, with the fracture linear density reaching 8.65 fractures per meter in study areas. The effective distance of fracture connectivity is mostly concentrated between 550 and 670 m. When the difference in fracture density along the structural ridge is between 2 and 4 fractures/meter, the effectiveness of the fractures is strong.

Prediction of igneous lithology and lithofacies based on ensemble learning with data optimization

Igneous rocks are widely developed in various Mesozoic and Cenozoic continental and marine basins. Igneous reservoirs are the key reservoirs for current oil and gas development. Accurate prediction of lithology and lithofacies is a prerequisite for the effective exploration of igneous reservoirs. Igneous lithology and lithofacies are complex and correlated. The existing single-label igneous rock identification methods only consider the prediction of individual properties, and less consideration is given to the correlation of reservoir properties. Therefore, lithology and lithofacies prediction based on conventional logging data is regarded as a typical class-imbalanced multilabel classification problem when considering both attribute correlation in algorithms and evaluation metrics. To solve this problem, an ensemble method of data optimization combined with multigrained cascade forest (CF) is used in this study to develop a new multilabel lithology and lithofacies prediction model based on data from nine conventional logs in the eastern depressional reservoirs of the Liaohe Basin, and satisfactory results are obtained. The imbalance problem of the conventional logging data sets is first solved by using K-means and synthetic minority oversampling technique methods; then, the model is trained by scenario transformation and stripping with multigrained CF; and next, a multilabel classification evaluation index with multiple perspectives is introduced. The differences between the model and typical intelligent algorithms such as CF, adaptive boosting, random forest, and support vector machine are compared in simulated wells, and the new model is found to have obvious advantages. The model is finally applied to an actual well, and the accurate prediction results illustrate that the new model designed and trained for the class imbalance multilabel classification problem in this paper has application value in the lithology and lithofacies multilabel prediction of igneous rocks and also provides a theoretical basis for more complex multilabel reservoir evaluation using machine learning in the future.

Prediction of total gas content in low-resistance shale reservoirs via models fusion — Taking the Changning shale gas field in the Sichuan Basin as an example

The low-resistivity phenomenon appeared during the development of shale gas in the Changning area of southern Sichuan. The accurate prediction of GAST is crucial in formulating optimal development plans for low-resistivity shale reservoirs. A series of formulas such as Archie in traditional electrical logging is no longer applicable to the characterization of gas-bearing properties of low-resistivity shale reservoirs. In this paper, we propose a fusion model for predicting the total gas content of low-resistivity shale reservoirs using conventional logging and total gas content data of core analysis. The model combines three popular machine learning models: random forest, extreme gradient boosting tree, and deep neural network. First, rock physics logging knowledge and the Spearman coefficient method are used to select suitable logging data types as the input of three models. Second, three models are trained by using the selected logging and core gas content data to achieve optimal performance. Finally, based on the stacking model fusion strategy, the three are fused to obtain a fusion model with better generalization performance. The experimental results show that the average absolute errors of the fusion model in predicting the total gas content of low-resistance wells and ultra-low resistance wells are 0.025 and 0.023 respectively, and the R2 coefficients reach 0.952 and 0.964. The fusion model effectively predicts the gas content of low-resistivity shale reservoirs and has high application value.

Research on Inversion Log Evaluation Method of Special Mineral in Alkali Lake Shale Oil Reservoir—A Case Study of the Fengcheng Formation in the Mahu Sag, China

The Fengcheng Formation in the Mahu Sag of the Junggar Basin, China, is characterized by alkaline lake deposits, featuring abundant alkaline minerals. The content of alkaline minerals affects the physical properties and oil-bearing properties of the reservoir, and existing mineral inversion methods cannot calculate the content of alkaline minerals. Based on Litho Scanner Log data, we can calculate the dry weight of elements using the oxide closure model. By improving the rock volume physical model; adding trona, shortite, eitelite, and reedmergnerite to the rock volume physical model; and combining with the least squares method, the mineral content calculation was carried out, using the inversion method of combination models (Shortite Model, Eitelite Model, Reedmergnerite Model, and Trona Model) to achieve mineral inversion of alkali-bearing shale oil reservoirs. Litho Scanner Log is expensive, and its widespread application will increase exploration costs. This article scales the mineral inversion results of Litho Scanner Log into conventional log data, improves the rock volume physical model of conventional log, and uses a combination model to achieve mineral inversion of alkali-bearing shale oil reservoirs in conventional log. Compared with the results of X-ray diffraction experiments, the average absolute error of all minerals except for trona and feldspar is less than 10%, and the inversion results are consistent with the core test results. The research results of this article can provide theoretical and technical support for the log evaluation of alkali-bearing shale oil reservoirs.

High-resolution sequence stratigraphy research based on continuous wavelet transform and INPEFA - A case study in the Kalashayi formation, Tarim Basin, China

The Carboniferous Kalashayi Formation, situated in the Lungu-Sangtamu area within the Tarim Basin, exhibits protracted sedimentation periods, intricate sand-mud depositional sequences, and scant paleontological and core datasets. Precision in delineating high-resolution sequences using conventional core and logging data poses a considerable challenge. To enhance the stratigraphic accuracy of the Kalashayi Formation in the Tarim Basin and facilitate quantitative analysis, this study employs continuous wavelet transform on the gamma ray (GR) curves obtained from core wells within the study area. Subsequently, various sequence boundaries are discerned by integrating the resulting wavelet coefficient curves with time-frequency energy maps. Discrimination and subdivision of base-level cycle structures of varying orders are achieved through temporal trend analyses of integrated prediction error filter analysis (INPEFA) curves. Integration of drilling, logging, lithofacies, and core data enables the identification and subdivision of high-resolution sequence stratigraphy using wavelet analysis and INPEFA techniques. Ultimately, the Carboniferous Kalashayi Formation in the Lungu-Sangtamu region is classified into 2 long-term, 5 medium-term, and 14 short-term base-level cycles, establishing a meticulously delineated isochronous stratigraphic framework. This framework serves as a fundamental basis for subsequent discussions on reservoir prediction within the study area.

Geochemical evaluation and source rock zonation by multi-layer perceptron neural network technique: a case study for Pabdeh and Gurpi Formations-North Dezful Embayment (SW Iran)

In this study, using a multi-layer perceptron neural network (MLPNN) model, total organic carbon (TOC) and hydrogen index (HI) values for Pabdeh and Gurpi Formations in the oil fields of Naft Sefid (NS-13), Kupal (KL-36, KL-38, and KL-48) and Palangan (PL-2) were calculated in the North Dezful Embayment located in the southwest of Iran. To build the MLPNN model, the geochemical data calculated by the Rock–Eval pyrolysis method (TOC and HI) and the conventional petrophysical well log data, including sonic transit time log (DT), formation density log (RHOB), total resistivity log (RT), spectral gamma-ray log, computed gamma-ray log and neutron porosity log from the NS-13 well were used. The log data were the input layer, and the geochemical data were the output layer of the model. Twenty-four datasets were used for MLPNN training, and seven datasets were used for MLPNN testing. Two hidden layers were considered in this technique. Each hidden layer has an activation function (tanh) and a solver parameter (lbfgs). The accuracy of measurement of TOC and HI indices of Pabdeh and Gurpi Formations in terms of R2 was 0.93 and 0.90, respectively. This model has higher accuracy than the ΔlogR technique (R2: 0.28). Considering the relationships between the input data and other wireline logs is an advantage of this technique. These two formations have five source rock zones. Pabdeh Formation has three zones. The middle zone of the Pabdeh Formation (Pz. II) has the highest TOC (2.6 wt%) and source rock potential. Pabdeh Formation has kerogen type II. Gurpi Formation has a weaker source rock potential than Pabdeh Formation due to its low TOC content (< 1%). Both source rock zones of this formation have low TOC, but in some layers of the lower zone of the Gurpi Formation (Gz. II), high values for TOC were predicted. Gurpi Formation has Kerogen types II and III.

A deep-learning framework for borehole formation properties prediction using heterogeneous well-logging data: A case study of a carbonate reservoir in the Gaoshiti-Moxi area, Sichuan Basin, China

The properties of borehole formations, such as porosity, permeability, and water saturation, play a crucial role in characterizing and evaluating subsurface reservoirs. Although core sample experiments offer precise measurements, they are time consuming and cost intensive. An alternative method is to use the logging data to construct an empirical model that predicts formation properties, which is widely studied due to its speed and affordability. Nevertheless, because the response of a logging point reflects its surrounding formation, conventional logging methods relying on point-to-point (P2P) mapping perform poorly in complex reservoirs. Furthermore, the resolution of conventional logging is lower compared with imaging logging. To address these limitations, this study presents a novel approach to predict formation properties based on a deep-learning framework using heterogeneous well-logging data. Our neural network framework takes short sequences of conventional logging data and windowed imaging logging data as inputs. The neural network applies 1D convolution to extract features from the conventional logging sequences and 2D convolution to extract features from the resistivity imaging data. Then, these two feature vectors are fused and fed into a multilayer fully connected neural network to predict formation properties. A case study of a carbonate reservoir demonstrates that our method delivers more accurate predictions of formation porosity, permeability, and water saturation than the P2P, sequence-to-point, and image-to-point prediction methods. Moreover, it is expected that our paradigm will serve as a source of inspiration for forthcoming research endeavors aimed at enhancing the accuracy of predicting borehole formation properties in complex reservoirs.

Total Organic Carbon Logging Evaluation of Shale Hydrocarbon Source Rocks in the Shan 1 Section of the Sulige Gas Field, Ordos Basin, China

The mass fraction of total organic carbon (TOC) is one of the key indicators for evaluating the hydrocarbon generation potential of shale source rocks. Experimental measurements to evaluate the TOC content require significant cost and time. Furthermore, the experimental data are often fragmented and may not provide an accurate depiction of the source rocks throughout the entire block. To solve the above problems, this paper proposes to use the combination of conventional logging data and experimental data after an in-depth study of the geophysical characteristics of hydrocarbon source rocks in the Ordos Basin. A quantitative model between logging data and source rocks is established, and then the continuous distribution value of the TOC content in the hydrocarbon source rock interval is calculated. Firstly, the mud shale formation of the Permian–Shanxi Formation in the Upper Paleozoic, located in the Jingbian area of the Ordos Basin, is selected as the research target using the “Jinqiang method”. The model is constructed by selecting appropriate logging curves (acoustic time difference logging, resistivity logging, and density logging) and experimental results based on the response relationship between logging data and TOC data. This method provides more accurate and comprehensive data for source rock studies, combining experimental sampling to contribute to a better evaluation of TOC in source rock. The shale hydrocarbon source rock logging data from 10 wells are selected, and the model is used to realize the full-well section of the logging data to find the hydrocarbon source rock TOC, which is compared with the TOC data from the experimental core tested at a sampling point. The results demonstrate that the model is highly effective and accurate, with a mere 2.7% percentage error observed across 185 sample data points. This method greatly improves the accuracy and completeness of TOC evaluation compared with the results of previous studies and provides a guide for subsequent TOC logging evaluation of source rocks in other areas. With the study in this paper, continuous TOC values of source rocks are obtained, discarding the TOC values representing the whole set of hydrocarbon source rocks with a limited number of sample averages. This method can reflect the contribution of the layers with high and low organic matter abundance, and the calculated reserves are more accurate. By utilizing the measured TOC values of the study area to invert the model to find the parameters, this study contributes to the decision-making of hydrocarbon exploration in domestic and international basins.