Neutron Porosity Logs Research Articles

Advancing Reservoir Evaluation: Machine Learning Approaches for Predicting Porosity Curves

Porosity assessment is a vital component for reservoir evaluation in the oil and gas sector, and with technological advancement, reliance on conventional methods has decreased. In this regard, this research aims to reduce reliance on well logging, purposing successive machine learning (ML) techniques for precise porosity measurement. So, this research examines the prediction of the porosity curves in the Sui main and Sui upper limestone reservoir, utilizing ML approaches such as an artificial neural networks (ANN) and fuzzy logic (FL). Thus, the input dataset of this research includes gamma ray (GR), neutron porosity (NPHI), density (RHOB), and sonic (DT) logs amongst five drilled wells located in the Qadirpur gas field. The ANN model was trained using the backpropagation algorithm. For the FL model, ten bins were utilized, and Gaussian-shaped membership functions were chosen for ideal correspondence with the geophysical log dataset. The closeness of fit (C-fit) values for the ANN ranged from 91% to 98%, while the FL model exhibited variability from 90% to 95% throughout the wells. In addition, a similar dataset was used to evaluate multiple linear regression (MLR) for comparative analysis. The ANN and FL models achieved robust performance as compared to MLR, with R2 values of 0.955 (FL) and 0.988 (ANN) compared to 0.94 (MLR). The outcomes indicate that FL and ANN exceed MLR in predicting the porosity curve. Moreover, the significant R2 values and lowest root mean square error (RMSE) values support the potency of these advanced approaches. This research emphasizes the authenticity of FL and ANN in predicting the porosity curve. Thus, these techniques not only enhance natural resource exploitation within the region but also hold broader potential for worldwide applications in reservoir assessment.

Hyperparameter inversion of well logs for the simultaneous estimation of volumetric and zone parameters

We develop a new alternative for the joint inversion of well logs to predict the volumetric and zone parameters in hydrocarbon reservoirs. Porosity, water saturation, shale content, kerogen, and matrix volumes are simultaneously estimated with the tool response function constants with a hyperparameter estimation assisted inversion of the total and spectral natural gamma-ray intensity, neutron porosity, and resistivity logs. We treat the zone parameters, i.e., the physical properties of rock matrix constituents, shale, kerogen, and porefluids, as well as some textural parameters, as hyperparameters and estimate them in a metaheuristic inversion procedure for the entire processing interval. The selection of inversion unknowns is based on parameter sensitivity tests, which indicate that the automated estimation of several zone parameters is favorable, and their possible range can also be specified in advance. In the outer loop of the inversion procedure, we use a real-coded genetic algorithm for the prediction of zone parameters, and we update the volumetric parameters in the inner loop in addition to the fixed values of the zone parameters estimated in the previous step. We apply a linearized inversion process in the inner loop, which allows for the quick prediction of volumetric parameters along with their estimation errors from point to point along a borehole. The derived parameters, such as hydrocarbon saturation and total organic content, indicate good agreement with core laboratory data. The significance of the inversion method is that the zone parameters are extracted directly from the wireline logs, which improves the solution of the forward problem and reduces the cost of core sampling and laboratory measurements. In a field study, we demonstrate the feasibility of the inversion method using real well logs collected from a Miocene tight gas formation situated in the Derecske Trough, Pannonian Basin, East Hungary.

Integration of Seismic and Well Log Data for Reservoir Characterization in Offshore Niger Delta, Nigeria

The demand for fossil fuels is continually increasing due to the growing global population and industrialization. It is thus necessary to find hydrocarbon prospects in new fields and those that have previously been classified as marginal fields in order to maximize production. An integrated approach is required to effectively characterize hydrocarbon reservoirs and assess their ability to store and produce hydrocarbons. 3D seismic reflection data and well log data were combined to quantitatively estimate hydrocarbon reserves in a field, offshore Niger delta, Nigeria. The reservoirs were delineated on the gamma ray and resistivity logs which penetrate four wells selected from the field. The petrophysical analysis provided information about the net-to-gross thickness ratio, porosity, shale volume and water saturation of the reservoirs, deduced from well log suites comprising gamma, resistivity, density and neutron-porosity logs through four wells. The seismic interpretation involved mapping of horizons and faults across the wells on the seismic section. Check shot data were used to tie the seismic data to the well log data to generate the synthetic seismogram. The time structural and depth structural maps were generated. The volumetric analysis entailed derivation of the Gross rock volume from the depth structural map, and estimation of the Stock Tank Oil Initially In Place (STOIIP). Four reservoirs were delineated in each of the four wells. The average thickness of the reservoirs ranges from 56.6 m to 232.7 m while the water saturation varies from 0.29% to 0.57%. The average porosities of the reservoirs ranges from 0.18% to 0.22%. The structural interpretation of the eight faults mapped reveals synthetic and antithetic faults, and rollover anticlines. The time- and depth structural maps generated from the mapped horizons show that the reservoirs are penetrated by the different faults. The Stock Tank Oil Initially In Place (STOIIP) of the reservoirs range from 25 MMstb to 468 MMstb. The study shows that the integration of 3D reflection data and well log data can be used effectively to estimate hydrocarbon reserves. The results of the study are expected to contribute to the development of new exploration and production strategies in the Niger delta basin and similar sedimentary environments.

Volve Oil Field S-Wave Log Data Prediction Using GBR and MLPR

The shear wave sonic (S-wave) log data is essential for identifying the reservoir's geomechanical properties, which is an important factor for the drilling, completion, and optimization processes, where obtaining S-wave requires capital investment and reduces the cost. In this paper, we used the multi-layer perceptron (MLPR) and the gradient boosting regression (GBR) to predict the missing S-Wave log data for the Volve Oil Field in the North Sea. Prescriptive and predictive analysis were carried out in series to achieve a high blind data accuracy rate of 0.943 and 0.982 for the multi-layer perceptron regression and the gradient boosting regression, respectively. It was observed that the gradient-boosting algorithm achieved higher accuracy than the MLPR algorithm for the limited dataset. It was also found that hierarchical clustering can reveal information regarding the feature's importance similar to the relevant AI tools, which makes hierarchical clustering a faster tool in eliminating the nonimportant inputs from the dataset while the use of artificial intelligence tools showed a significant effect in predicting the missing values of the sonic wave log and the neutron porosity log in an efficient way by selecting the relevant important features.

Reservoir Characterization of KD Field Located Onshore of Akwa-Ibom State, Niger Delta, Nigeria Using Well Logs Property Comparison to Seismic Amplitude Analysis

Characterization of reservoirs defines how well they can generate and store hydrocarbon, reservoir parameters are used to determine the behavior of reservoir fluids under various conditions and to identify the best production practices that can optimize hydrocarbon production, the objective is to identify and map reservoir sand zones using basic logs, provide information about the reservoir depth and thickness using well log data, interpret faults and sealing system using the seismic data and generate amplitude maps and model basic reservoir facies, the method applied was sectioned into three distinct parts that encompasses qualitative well log analysis, seismic interpretation and cross plotting of basic well logs versus amplitude for reservoir characterization, the well data analyzed shows well number 5, 6 and 8 have incomplete GR logs resulting in missing lithology while well 10 have some missing logs, such as, density (DEN) log, neutron porosity (NEU) log and porosity (POR) log, five logs needed for this study were available in the remaining complete wells. These logs include gamma ray (GR) log, resistivity (RES) log, density (DEN) log, neutron porosity (NEU) log, and porosity (POR) log. This study looks at a possible change in log signal due to an increase or decrease in seismic amplitude in characterizing a reservoir sand and how logs are influence with Amplitude peaks and trough.

Reservoir Characterization of the Paleogene Khurmala Formation in Tawke and Shaqlawa Areas, Kurdistan Region of Iraq

Well logs were utilized to investigate petrophysical properties of the Khurmala Formation’s surface outcrops in Shaqlawa Subdistrict and Tawke Oilfield, e.g., lithology, shale volume, porosity, and fracture. The thickness of the formation is about 15 m in the Shaqlawa section and 42 m in the Tawke Oilfield. Porosity logs were used to estimate porosity; where the porosity values reached a maximum of 52% from the sonic log, 48% from the density log, and 35% from the neutron porosity log. The reservoir quality of the Khurmala Formation, as determined through the analysis of thin sections, which were obtained from outcrop samples, is deemed to be of low quality. The determined shale volume within the examined interval exhibits a moderate level of clay constituents, with the highest gamma-ray measurement indicating a shale content of 29% at some locations within the reservoir. This integrated method using various conventional well logs suggests a great probability of petrophysical properties in the Khurmala Formation to be considered as the reservoir.

Friction Angle Prediction of Carbonate Rocks: A Case Study, Rumaila Oil Field

Friction angle (𝜑) and Cohesion (𝑐) are the most important factors to depict rock's shear strength. The friction angle (φ) expresses a unit of rock's capacity to endure shear stress. For the optimization of drilling operations, monitoring of the reservoir, and production of hydrocarbons, the prediction of friction angle is essential. From laboratory measurements or wireline logging data, this parameter can be empirically predicted. The main goal of this study is to develop a new correlation for predicting friction angle for carbonate formations from well logs using the typically accessible well log data (i.e. neutron porosity, gamma ray, bulk density, and sonic logs) and core data. A total of 5197 well log data points were collected from carbonate formation with depth interval of (1920 m to 2711 m) from Rumaila oil field. For all 5197 data points neutron porosity, and gamma ray logs were recorded as a function of depth, and the corresponding shale volume and total porosity were estimated. In addition to these well log data, 20 data core points with 9 different values of friction angle were collected. The developed correlation's estimated friction angle has been contrasted with measured ones. The results show that the new correlation is able to predict the friction angle of carbonate rocks with high accuracy (i.e. R coefficient of the new correlation was 90% and average absolute error of 1.6%).Thus, we conclude that the new correltion can be used to estimate the friction angle for carbonate formation. The new correlation helps in providing continues profile for friction angle with depth and leads to reduce the cost of estimating the rock strength.

Generating 3D lithology probability volumes using poststack inversion, probabilistic neural networks, and Bayesian classification — A case study from the mixed carbonate and siliciclastic deposits of the Cisco Group of the Eastern Shelf of the Permian Basin, north-central Texas

The deposition and mixing of carbonates and siliciclastics in the Cisco Group of the Eastern Shelf of the Permian Basin are complicated by the temporal overlap between icehouse eustatic sea-level oscillations and fluctuations in sediment influx due to the rejuvenation of the Ouachita fold belt. Previous investigators have used well-log correlation as the primary tool in their interpretations of the area’s reciprocal depositional model, but well-log correlation alone cannot explain the full range of spatial lithology variations in the system. To better understand the lithology variation in the area, we use an integrated technique that combines wireline log information from 17 wells with 625 km2 3D seismic data through poststack seismic inversion, probabilistic neural networks (PNNs), and Bayesian classification. We use deterministic matrix inversion to derive lithology classes from well logs. Crossplot analyses reveal that the acoustic impedance and neutron porosity log pair can be used to differentiate lithologies. We perform model-based poststack inversion to generate a P-impedance volume and use PNNs to generate a neutron porosity volume. We combine these volumes through supervised Bayesian classification to generate lithology probability volumes for each lithology and a most probable lithology volume throughout the seismic data. The lithology volumes highlight the dominant lithologies (carbonate, shale, sand, and mixed) that allowed the interpretation of major carbonate platforms, sand-to-shale ratio variations, carbonate buildups between wells, and channel fill lithologies. Our semiautomated lithology detection workflow applies to regional studies and is also valid for reservoir-scale studies to determine variations in lithologies.

Controls on mechanical properties of a carbonate mudstone: Insights from non-destructive techniques and geochemical data

This study quantifies geomechanical properties of carbonate mudrocks in the Arabian intra-shelf basin, focusing on different sedimentologic zones (proximal to distal). Core-based mechanical properties were obtained through unconfined conditions. Micro Rebound Testing (MRT) was conducted on 700′ ft of core at 0.5 ft intervals to capture high-resolution rock hardness (HLD), unconfined compressive strength (UCS), and static Young's Modulus (E). Additional tests validated log-based parameters, including impulse hammer, acoustic velocity testing for (E), P-wave and S-wave velocities (Vp and Vs), scratch testing, and XRF/XRD analyses for mineralogical and elemental variability impact on mechanical properties.Multivariate statistical analysis correlates mechanical properties with elemental data, showing direct proportionality to calcium percentage, (E), Vp, and Vs, while inversely correlated to aluminum, iron, sulfur, and total organic carbon (TOC), especially in distal zones. Porosity, as measured by neutron porosity log, is a significant control on mechanical properties, with higher porosity resulting in lower mechanical properties at any given calcium percentage. Variations in geomechanical properties within each zone are principally controlled by depositional environments, sediment supply, diagenesis, and a pycnocline dividing the water column into anoxic, dysoxic (or suboxic), and oxic conditions.Zone 1A and 4 exhibit high brittleness characterized by high Ca-calcite, HLD, (E), Vp, Vs., low terrigenous elements/minerals, and porosity. Zone 1B and 2A are moderately brittle, showing a slight decrease in brittleness and a slight increase in terrigenous elements/minerals, porosity, and organic matter, particularly in the distal zone (Zone 1B). Zone 2B, 3A, and 3B represent moderate to highly ductile zones with low levels of Ca-calcite, HLD, (E), Vp, Vs, and an increase in terrigenous elements/minerals, porosity, and TOC, especially in the distal zone (Zone 2B and 3B). This study highlights the value of high-resolution mechanical data from cores with applications for improved stress estimation and hydraulic fracture stimulation in unconventional reservoirs.

Integration of Rock Properties Crossplot and Petrophysics for Enhancement of Litho-Fluid Discrimination and Compartmentalizationto Reduce Uncertainty and Mitigate Hydrocarbon Exploration Risk. ‘A Case of ‘NICK' Field, Onshore Niger Delta’.

The main components of reservoir rocks are hydrocarbons and immiscible water in varying ratios. It is essential to precisely identify, characterize, and divide the fluids in these reservoirs into distinct groups according to the characteristics of their rock properties to conduct a successful hydrocarbon exploration. For this reason, petrophysics and rock physics analysis were combined on the "NICK" field in the onshore Niger Delta. Through precise litho-fluid discrimination in the field, this study seeks to improve field hydrocarbon production, lower uncertainty, and mitigate risks related to hydrocarbon exploration. The suites of well logs (Gamma-ray neutron, bulk density, sonic, and resistivity logs) from three wells—NICK-1, NICK-3, and NICK-6—make up the data used. Gamma-ray log signatures were used to identify and correlate lithologies throughout the field. Potential reservoirs and fluid content were identified and delineated by high resistivity and adequate neutron-porosity log signatures. Hydrocarbon-bearing sands were recorded at low values of elastic attributes (acoustic impedance, rigidity, incompressibility, and others), which were computed to aid in the characterizations. Two potentialreservoirs’ Sands A and B delineated, constituted the correlated pay zones observed in three wells across the field at depths ranging from 1986.24 to 2599.82m. Petrophysics results generally revealed fair to good porosities of (11-25%) for easy accumulation of hydrocarbon. Permeability ranged from 210- 809mD for Sand A and 27 – 887mD for Sand B, showing that there are suitable permeabilitiesfor fluid movement/migration within the reservoirs. Cross plot of Lambdarho versus Murho, Lambdarho versus Velocity Ratio, and Velocity Ratio versus Acoustic Impedance gave four distinct clusters for litho-fluid zones identification given as gas-sand, oil-sand, brine-sand, and shale. This study has assisted in better characterization and distinguishing of the litho-fluid details for enhancement of hydrocarbon production in the field.

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.

Uncertainty and explainable analysis of machine learning model for reconstruction of sonic slowness logs

Logs are valuable information for oil and gas fields as they help to determine the lithology of the formations surrounding the borehole and the location and reserves of subsurface oil and gas reservoirs. However, important logs are often missing in horizontal or old wells, which poses a challenge in field applications. To address this issue, conventional methods involve supplementing the missing logs by either combining geological experience and referring data from nearby boreholes or reconstructing them directly using the remaining logs in the same borehole. Nevertheless, there is currently no quantitative evaluation for the quality and rationality of the constructed log. In this paper, we utilize data from the 2020 machine learning competition of the Society of Petrophysicists and Logging Analysts (SPWLA), which aims to predict the missing compressional wave slowness (DTC) and shear wave slowness (DTS) logs using other logs in the same borehole. We employ the natural gradient boosting (NGBoost) algorithm to construct an Ensemble Learning model that can predicate the results as well as their uncertainty. Furthermore, we combine the SHAP (SHapley Additive exPlanations) method to investigate the interpretability of the machine learning model. We compare the performance of the NGBosst model with four other commonly used Ensemble Learning methods, including Random Forest, GBDT, XGBoost, LightGBM. The results show that the NGBoost model performs well in the testing set and can provide a probability distribution for the prediction results. This distribution allows petrophysicists to quantitatively analyze the confidence interval of the constructed log. In addition, the variance of the probability distribution of the predicted log can be used to justify the quality of the constructed log. Using the SHAP explainable machine learning model, we calculate the importance of each input log to the predicted results as well as the coupling relationship among input logs. Our findings reveal that the NGBoost model tends to provide greater slowness prediction results when the neutron porosity (CNC) and gamma ray (GR) are large, which is consistent with the cognition of petrophysical models. Furthermore, the machine learning model can capture the influence of the changing borehole caliper on slowness, where the influence of borehole caliper on slowness is complex and not easy to establish a direct relationship. These findings are in line with the physical principle of borehole acoustics. Finally, by using the explainable machine learning model, we observe that although we did not correct the effect of borehole caliper on the neutron porosity log through preprocessing, the machine learning model assigned a greater importance to the influence of the caliper, achieving the same effect as caliper correction.

A multidisciplinary approach to facies evaluation at regional level using well log analysis, machine learning, and statistical methods

Geological facies evaluation is crucial for the exploration and development of hydrocarbon reservoirs. To achieve accurate predictions of litho-facies in wells, a multidisciplinary approach using well log analysis, machine learning, and statistical methods was proposed for the Lower Indus Basin. The study utilized five supervised machine learning techniques, including Random Forest (FR), Support Vector Machine (SVM), Artificial Neural Network (ANN), Extreme Gradient Boosting (XGB), and Multilayer Perceptron (MLP), to analyse gamma ray, resistivity, density, neutron porosity, acoustic, and photoelectric factor logs. The Concentration-Number (C-N) fractal model approach and log–log plots were also used to define geothermal features. In a study on machine learning models for classifying different rock types in the Sawan field of the Southern Indus Basin, it was discovered that sand (fine, medium and coarse) facies were most accurately classified (87–94%), followed by shale (70–85%) and siltstone facies (65–79%). The accuracy of the machine learning models was assessed using various statistical metrics, such as precision, recall, F1 score, and ROC curve. The study found that all five machine learning methods successfully predicted different litho-facies in the Lower Indus Basin. In particular, sand facies were most accurately classified, followed by shale and siltstone facies. The multilayer perceptron method performed the best overall. This multidisciplinary approach has the potential to save time and costs associated with traditional core analysis methods and enhance the efficiency of hydrocarbon exploration and development.

Thermal conductivity of sediments from well-logs and its application to determine heat flow density in the Pannonian Basin, Hungary

The thermal conductivity of rocks can be deduced from available data of exploration wells such as core samples, cuttings, lithological descriptions and geophysical well logs. We present a methodology for determining the thermal conductivity of the Neogene and Quaternary clastic sediments in the Pannonian Basin using geophysical well logs and thermal conductivity data measured in laboratory. The lithological composition consisting of shale, sand and water was identified and the volumetric fractions of these components were derived from wireline logging data that included natural gamma ray, resistivity, bulk density and neutron porosity logs. The bulk thermal conductivity was computed by applying an appropriate mixing law using the thermal conductivity values of the lithological components. The thermal conductivities of the lithological components were determined by the best fit of the calculated bulk thermal conductivity values to the laboratory data. Different mixing models are recommended to calculate thermal conductivity of shales and sandstones. The thermal conductivity of shales was calculated with the weighted harmonic mean of the constituents, where the volumetric fractions and the porosity were the weights. The thermal conductivity of sandstones can be equally well estimated by the weighted geometric mean and the three component lower Hashin-Shtrikman bound. The geometric mean was used in calculations due its simplicity. Shale to sand ratio was utilized to separate shaly and sandstone samples. Heat flow density determination was carried out using the Bullard-plot technique with thermal conductivities calculated by our method. Heat flow density calculated by our method is in the range of heat flow density values previously conducted but with substantially lower uncertainties. The significance of the method is that it allows the harmonization of the heat flow density in the countries lying in the Pannonian Basin, because the Neogene and Quaternary sedimentary cycle resulting in the filling of the basin was uniform throughout the basin.

Hydrocarbon Formation Evaluation Using Well Log Data Of Well Tmg-02, Opolo Field, Niger Delta

Well TMG-02 with the depth interval of 5058.77 to 9389.43ft of Opolo field located in the Niger delta was assessed for hydrocarbon using suite of geophysical well logs. Suite includes gamma ray (GR), formation density (RHOB), neutron porosity (NPHI), and resistivity logs. The analysis was carried out to estimate the field’s hydrocarbon prospect by identifying hydrocarbon bearing reservoirs and their properties. The quantitative and qualitative results, identified thicker units of sand than shale lithology, three reservoirs A, B, C within the depth ranges from 5058.77ft to 9389.43ft, capable of accumulating hydrocarbon based on the petrophysical parameters calculated were delineated. The effective porosity for each of the reservoir are: 27%, 24% and 19% respectively. It was observed that reservoir A, B had excellent permeability while reservoir C was low as a result of thicker shale sequence within the reservoir. The result obtained shows presence of hydrocarbon bearing gas water contact in Reservoir A at depth of 5119.70ft, gas oil contact and oil water contact at depths 7310.00ft and 7438.69ft in Reservoir B and Gas water contact at depth 9032.00ft at Reservoir C.&#x0D;

Research on instrument structure scheme for improving the sensitivity of wireline D-T source porosity logging

Utilizing a D-T source to replace the chemical source is the development trend of nuclear logging technology. However, the wireline porosity logging with a D-T source has a lower sensitivity. To solve this problem, the key factors on the sensitivity of neutron porosity logging are analyzed. Combining the numerical simulation method, special structures such as neutron-source deceleration block and detector shielding sleeve are designed to improve the sensitivity of D-T source wireline porosity logging tool, and the detector-spacing selection of the improved instrument was discussed. Study indicates that the sensitivity of neutron porosity logging mainly depends on the fast-neutron deceleration length, the proportion of formation information in the detector count and the detector-spacing. In the instrument design, the tungsten neutron-source deceleration block is adopted to reduce the energy of neutrons emitted by source, thereby shortening the fast-neutron deceleration length and improving the logging sensitivity. Additionally, the detector shielding sleeve made of boron carbide can absorb thermal neutrons from the borehole to increase the proportion of formation information in the detector count, which can also effectively improve the sensitivity. Based on the above structure design, it was found that when the near detector-spacing was set at 30 cm and the far detector-spacing was set at about 70 cm, the performance of the instrument can reach the best. This study can provide theoretical guidance for the application of D-T source in the wireline porosity logging tool in the future.

Feature engineering process on well log data for machine learning-based SAGD performance prediction

The conventional numerical simulation requires a reliable three-dimensional reservoir model to evaluate the steam-assisted gravity drainage (SAGD) performance, but it may be time-consuming and need a significant amount of static data. This study aims to develop a machine learning (ML) model that predicts the SAGD performance based on well log data from a near single well. Although the ML model is useful in capturing underlying non-linear relationships between input features and target features, it requires a large number of training data. However, the number of data is usually small in commercial fields, limiting the availability of all log responses as input features to consider reservoir heterogeneity. In case of using small dataset, the overfitting can occur with an excessive number of model parameters. Therefore, this study introduces a feature engineering process to extract and select key features from well log data to alleviate the overfitting problem.In this study, the feature engineering process was optimized by sensitivity analysis on feature extraction and feature selection methods. For data preparation, the 63 SAGD well pairs and their matched logged wells were screened from the six commercial oil sand fields in Alberta, Canada. The SAGD process mechanism and petrophysics were considered to build data screening criteria. Each depth of the log data was categorized into pay-zone or shale barrier. For feature extraction, sampling intervals were placed on log data. This calculates the net pay thickness, averaged gamma ray, deep resistivity, and neutron porosity logs in the pay-zone. In addition, the elevation of two shale barriers and the distance between the SAGD well and logged well are extracted. Sensitivity analysis of the sampling interval length was conducted to determine the optimal conditions considering heterogeneity and feature reduction. For the feature selection, various combinations of the extracted features were used to examine their inherent meaning for predicting the SAGD performance. The multi-layer perceptron model was trained based on various selected features to predict the SAGD performance indicators of a seven-year operation. The optimal sampling interval length was determined as 10m and the combination of gamma ray, deep resistivity, shale barrier elevation, and the distance information was selected as key input features. The model showed reasonable prediction results for SAGD productivity and economic indicator with R2 of 0.67 and 0.72, respectively.

Emplacement dynamics of a complex thick mafic intrusion revealed by borehole image log facies analyses: Implications for fluid migration in the Parnaíba Basin petroleum system, Brazil

Sill intrusions may play a significant role in the development of petroleum systems in sedimentary basins by promoting maturation of source rocks, influencing fluid migration and acting as reservoirs, traps, and seals for hydrocarbon accumulations. Within this study, we present a novel approach to interpreting geophysical borehole data for a ca. 174 m thick sill from the Parnaíba Basin revealing new insights into both the emplacement mechanisms and impacts of the sill on hydrocarbon migration and trapping. High-resolution ultrasonic and microresistivity borehole image logs were integrated with conventional log data, petrography of eight side wall cores and total gas data recorded during drilling. The petrographic analysis revealed finer crystalline dolerite at the top and bottom layers of the sill while the upper central portion revealed coarser crystalline gabbroic facies, associated with anomalies in gamma ray, neutron porosity, sonic and geochemical log responses. From borehole image logs we interpreted key facies units including massive, fractured, cooling jointed and layered dolerite along with clear evidence for layers containing diverse inclusions interpreted as enclave structures. The integration and correlation of these data with analogue studies allowed the interpretation of gradational cooling with magmatic differentiation, partially melted host rock linked to sulphide production and organic matter maturation, mafic microgranular enclaves related to mingling processes of magma recharge, banding layers related to cumulate zones, and intense cooling fractures in the central sill body. An interval of ca. 44 m at the top of the sill with notably less jointing appears to be responsible for sealing the gas underneath, whilst the fractured central portion of the sill has enabled gas migration evidenced by increased gas signatures. In addition, this work also reveals for the first-time image log evidence for magma mingling and indications of assimilation and anatexis processes within the Parnaíba Basin. This study has important implications for improving the understanding of emplacement mechanisms of thick parallel layer intrusions and their implications for fluid flow in sedimentary basins worldwide.

A novel physics-based resolution improvement method for neutron-density log

In the later stages of oilfield development, the accurate identification of thin reservoirs become increasingly crucial for oil and gas exploration. However, conventional logs have limited vertical resolution, making it challenging to identify thin reservoirs. This results in potential hydrocarbon layers getting overlooked during exploration. To address this issue, a novel resolution enhancement method for density and neutron porosity logs based on physical constraint optimization is proposed. The proposed method involves three main steps: (1) developing an objective function that relates the true formation density and porosity to the measurement based on the physics of nuclear, (2) setting density neutron cross-constraints using the volumetric model of logging interpretation, and (3) establishing the objective function matrix equations using sliding window technology and solving the matrix equations using the constrained least square method to obtain high-resolution density and neutron porosity logs. The proposed method is evaluated using both synthetic and field logs obtained from compensated density and compensated neutron tools. The results demonstrate an effective improvement in neutron porosity and density log resolution to 15 cm, which is significantly enhanced compared to the typical vertical resolution of 2 ft. This study offers a new perspective for enhancing the resolution of logs and contributes to the exploration and development of thin oil and gas resources. And the proposed method could be used in future exploration and production projects to identify potential hydrocarbon layers that might have been overlooked by conventional logs.

Robust estimation of shear wave velocity in a carbonate oil reservoir from conventional well logging data using machine learning algorithms

In recent decades, various applications of shear wave velocity have been reported in oil and gas projects. These applications include the determination of lithology, estimation of geomechanical parameters, reservoir fluid detection, etc. The shear wave velocity is usually measured or recorded by DSI (Dipole Shear Sonic imager) instruments. This log is not usually available in all wells of a hydrocarbon field because of the high cost of running this log compared to other conventional logs. However, it is tried to estimate the shear wave velocity from other related logs.The literature contains many empirical rock physics models used to estimate shear wave velocity. Most of them were developed for sandstone lithologies and didn't applicable in carbonate reservoir because of existing some vertical and horizontally complex fracture networks in the carbonate reservoirs that cause a quite different behavior for the shear wave velocity.The main objective of this research is to concentrate on the effect of lithology in the estimation of the target log for a carbonate reservoir. At the earliest step, preprocessing is performed and 455 data points of several conventional well logs, i.g., acoustic, density, neutron porosity, resistivity, gamma-ray, calcite volume, dolomite volume, and water saturation logs have been applied to generate the synthetic shear wave velocity log in a carbonate oil reservoir.First, original data are divided into training, validation, and testing subsets. Different data-driven predictive models were built: Multiple Linear Regression (MLR), Ensemble Learning Method (ELM), and three architectures of Artificial Neural Networks (ANN). The performance of all techniques was compared with each other. Finally, the feed-forward neural network shows the highest accuracy with R-values of 0.99 and 0.96 for the training and testing datasets. Moreover, the network design and tuning parameters of the model are adjusted using the grid search optimization technique to obtain the optimal design of the artificial neural network for estimating the shear wave velocity log. Hence, a deep artificial neural network, i.e., [8–11–15-1] network with bayesian regularization training algorithm and hyperbolic tangent sigmoid transfer function, is proposed to predict the target response in other wells.