Heterogeneous Carbonate Reservoir Research Articles

Permeability modelling in a highly heterogeneous tight carbonate reservoir using comparative evaluating learning-based and fitting-based approaches

Permeability modelling is considered a complex task in reservoir characterization and a key component of reservoir simulation. A common method for permeability modelling involves performing static rock typing (SRT) using routine core analysis data and developing simple fitting-based mathematical relations that link permeability to reservoir rock porosity. In the case of carbonate reservoirs, which are associated with high heterogeneities, fitting-based approaches may fail due to porosity–permeability data scattering. Accurate modelling of permeability using petrophysical well log data seems more promising since they comprise a vast array of information about the intrinsic properties of the geological formations. Furthermore, well log data exhibit continuity throughout the entire reservoir interval, whereas core data are discrete and limited in availability and coverage. In this research work, porosity, permeability and log data of two oil wells from a tight carbonate reservoir were used to predict permeability at un-cored intervals. Machine learning (ML) and fitting models were used to develop predictive models. Then, the developed ML models were compared to exponential and statistical fitting modelling approaches. The integrated ML permeability model based on Random Forest method performed significantly superior to exponential and statistical fitting-based methods. Accordingly, for horizontal and vertical permeability of test samples, the Root Mean Squared Error (RMSE) values were 3.7 and 4.5 for well 2, and 1.7 and 0.86 for well 4, respectively. Hence, using log data, permeability modelling was improved as it incorporates more comprehensive reservoir rock physics. The outcomes of this reach work can be used to improve the distribution of both horizontal and vertical permeability in the 3D model for future dynamic reservoir simulations in such a complex and heterogeneous reservoir system.

New insights into estimating the cementation exponent of the tight and deep carbonate pore systems via rigorous numerical strategies

AbstractOne of the main constituents of any reservoir characterization is an accurate forecast of water saturation, which is highly dependent upon the cementation exponent. Even though there have been a lot of studies, the most common correlations depend on total porosity. This means that they do not work as well in heterogeneous carbonate reservoirs, especially tight formations with total porosities less than 10%. This study aims to develop accurate and universal models for estimating the cementation exponent in deep and tight carbonate pore systems located in West Asia. Two heuristic algorithms, including the radial basis function neural network optimized by ant colony optimization (RBFNN-ACO) and gene expression programming (GEP), were employed to calculate the cementation exponent. To do this, we prepared a databank incorporating cementation exponents, total porosity, and various pore types. Then, the databank is classified into the test subset (for model prediction checking) and the train subset (for model construction). The reliability of the new recommended models is inspected by applying several statistical quality measures associated with graphical analyses. So, the consequences of the modeling disclose the large precision of the above-mentioned RBFNN-ACO, GEP Model-I, and GEP Model-II by average absolute percentage relative deviations (AAPRD%) of 6.28%, 6.39%, and 7.45%, respectively. Based on the outliers analysis, nearly 95% of the databank and model estimations are, respectively, valid and reliable. Additionally, the three input variables, including moldic porosity (with a + 70% impact value), non-fabric-selective dissolution (connected) porosity (with a -30% impact value), and interparticle porosity (with a -23% impact value), exhibit the main affecting parameters on the cementation exponent. Comparing current results with traditional literature correlations demonstrates the supremacy of the RBFNN-ACO model (AAPRD = 6.28% and root mean squared error (RMSE) = 0.17) over the examined literature correlations such as Borai’s equation (AAPRD = 12.30% and RMSE = 0.41). In addition, RBFNN-ACO can give better results than Borai’s Eqn. for tight (porosity less than 10%) and deep carbonate samples.

Analyzing the impact of geological features on reservoir heterogeneity using heterogeneity logs: A case study of Permian reservoirs in the Persian Gulf

Carbonate reservoir heterogeneity has a major impact on fluid flow and hydrocarbon recovery. It is challenging to characterize due to the complex and heterogeneous nature of carbonate reservoirs. It is important to express heterogeneity as a quantitative parameter for comparing various reservoirs. It can also be used for investigating the influence of different factors on the degree of heterogeneity. Heterogeneity log is one of the effective methods for quantifying heterogeneity. The aim of the present study is to investigate the application of heterogeneity log to determine the geological features affecting heterogeneity in Upper Dalan Member (Late Permian) in the central Persian Gulf. The gamma ray, bulk density, neutron porosity, acoustic, and deep resistivity well logs data were used to calculate the heterogeneity logs. A total of 1093 thin sections were studied and primary depositional as well as secondary diagenetic changes were recorded. Porosity and permeability of the same number of samples were measured on the plugs. Heterogeneity logs were calculated by Lorenz coefficient method. The changes in heterogeneity logs were classified into different ranges and compared with frequency of geological features in corresponding intervals. Findings show that heterogeneity logs can help identify areas of the reservoir with distinct reservoir behavior (permeability and/or porosity). These areas can be targeted for increased oil and gas production. This can help reduce exploration and development costs. Dolomite samples are observed in both low and high heterogeneity values. Anhydrite lithology reduces porosity and increases heterogeneity. High heterogeneity values correspond with mudstone texture from supratidal, peritidal, and lagoon subenvironments with low porosity. Grainstone textures from shoal subenvironment and crystalline carbonates are associated with low values of heterogeneity. These microfacies have high porosity and permeability as well as a high rate of marine cementation. Also, interparticle and moldic pores along with stylolites and fractures are observed in low values of heterogeneity. The increase in the ratio of pore space to the solid component of samples has resulted in a reduction in the value of heterogeneity. By creating different subenvironments with different microfacies and average porosities, environmental energy controls the heterogeneity of the studied member. Therefore, the amount of heterogeneity is mainly controlled by two factors namely, porosity and pore type.

Improving permeability prediction via Machine Learning in a heterogeneous carbonate reservoir: application to Middle Miocene Nullipore, Ras Fanar field, Gulf of Suez, Egypt

Predicting and interpolating the permeability between wells to obtain the 3D distribution is a challenging mission in reservoir simulation. The high degree of heterogeneity and diagenesis in the Nullipore carbonate reservoir provide a significant obstacle to accurate prediction. Moreover, intricate relationships between core and well logging data exist in the reservoir. This study presents a novel approach based on Machine Learning (ML) to overcome such difficulties and build a robust permeability predictive model. The main objective of this study is to develop an ML-based permeability prediction approach to predict permeability logs and populate the predicted logs to obtain the 3D permeability distribution of the reservoir. The methodology involves grouping the reservoir cored intervals into flow units (FUs), each of which has distinct petrophysical characteristics. The probability density function is used to investigate the relationships between the well logs and FUs to select high-weighted input features for reliable model prediction. Five ML algorithms, including Linear Regression (LR), Polynomial Regression (PR), Support Vector Regression (SVR), Decision Trees (DeT), and Random Forests (RF), have been implemented to integrate the core permeability with the influential well logs to predict permeability. The dataset is randomly split into training and testing sets to evaluate the performance of the developed models. The models’ hyperparameters were tuned to improve the model’s prediction performance. To predict permeability logs, two key wells containing the whole reservoir FUs are used to train the most accurate ML model, and other wells to test the performance. Results indicate that the RF model outperforms all other ML models and offers the most accurate results, where the adjusted coefficient of determination (R2adj) between the predicted permeability and core permeability is 0.87 for the training set and 0.82 for the testing set, mean absolute error and mean squared error (MSE) are 0.32 and 0.19, respectively, for both sets. It was observed that the RF model exhibits high prediction performance when it is trained on wells containing the whole reservoir FUs. This approach aids in detecting patterns between the well logs and permeability along the profile of wells and capturing the wide permeability distribution of the reservoir. Ultimately, the predicted permeability logs were populated via the Gaussian Random Function Simulation geostatistical method to build a 3D permeability distribution for the reservoir. The study outcomes will aid users of ML to make informed choices on the appropriate ML algorithms to use in carbonate reservoir characterization for more accurate permeability predictions and better decision-making with limited available data.

Permeability Prediction and Facies Distribution for Yamama Reservoir in Faihaa Oil Field: Role of Machine Learning and Cluster Analysis Approach

Empirical and statistical methodologies have been established to acquire accurate permeability identification and reservoir characterization, based on the rock type and reservoir performance. The identification of rock facies is usually done by either using core analysis to visually interpret lithofacies or indirectly based on well-log data. The use of well-log data for traditional facies prediction is characterized by uncertainties and can be time-consuming, particularly when working with large datasets. Thus, Machine Learning can be used to predict patterns more efficiently when applied to large data. Taking into account the electrofacies distribution, this work was conducted to predict permeability for the four wells, FH1, FH2, FH3, and FH19 from the Yamama reservoir in the Faihaa Oil Field, southern Iraq. The framework includes: calculating permeability for uncored wells using the classical method and FZI method. Topological mapping of input space into clusters is achieved using the self-organizing map (SOM), as an unsupervised machine-learning technique. By leveraging data obtained from the four wells, the SOM is effectively employed to forecast the count of electrofacies present within the reservoir. According to the findings, the permeability calculated using the classical method that relies exclusively on porosity is not close enough to the actual values because of the heterogeneity of carbonate reservoirs. Using the FZI method, in contrast, displays more real values and offers the best correlation coefficient. Then, the SOM model and cluster analysis reveal the existence of five distinct groups.

Simulation of reservoir dynamic damage and influencing factors of horizontal wells in heterogeneous carbonate reservoirs

AbstractAs the development of well types for low permeability reservoirs shifts from vertical to highly inclined/horizontal wells, targeted studies of oil and gas reservoir damage must be conducted for such well types to improve the science and effectiveness of acidification conversion. Taking into account the formation of mud cake and the dynamic damage factors caused by drilling fluid on the reservoir during horizontal well drilling, a refined dynamic description model for nonuniform damage in horizontal wells was established. The coupled simulation accounts for the dynamical destruction of the skin factor of the reservoir and the dynamical generation of the slurry. The coupled model consists of a radial mud cake dynamic formation model, a reservoir damage radius model, a permeability distribution model, and a horizontal well inhomogeneous skin factor model. It captures the dynamic evolution of reservoir damage due to drilling fluid contamination during horizontal well drilling and provides a real‐time representation of changes in the reservoir skin factor of horizontal wells, measuring the depth and extent of radial reservoir damage. Clarifying the extent of reservoir damage and its impact in horizontal well sections can help select appropriate acidification methods, improve acid distribution processes, and ultimately improve the effectiveness of mitigating reservoir damage and increasing oil and gas well productivity.

Editorial for the Special Issue “Mineralogical, Geological and Geochemical Heterogeneities of Carbonate Reservoirs”

Carbonate reservoir heterogeneity is a defining characteristic that fundamentally shapes the behavior and properties of subsurface formations [...]

Integration of electromagnetic, resistivity-based and production logging data for validating lithofacies and permeability predictive models with tree ensemble algorithms in heterogeneous carbonate reservoirs

This study develops an innovative workflow to identify discrete lithofacies distributions with respect to the well-log records exploiting two tree-based ensemble learning algorithms: extreme gradient boosting (XGBoost) and adaptive boosting (AdaBoost). In the next step, the predicted discrete lithofacies distribution is further assessed with well-log data using an XGBoost regression to predict reservoir permeability. The input well-logging records are gamma ray, neutron porosity, bulk density, compressional slowness, and deep and shallow resistivity. These data originate from a carbonate reservoir in the Mishrif Basin of southern Iraq's oilfield. To achieve a solid prediction of lithofacies permeability, random subsampling cross-validation was applied to the original dataset to formulate two subsets: training for model tuning and testing for the prediction of subsets that are not observed during the model training. The values for the total correct percentage (TCP) of lithofacies predictions for the entire dataset and testing subset were 98 and 93% using the XGBoost algorithm, and 97 and 89% using the AdaBoost classifier, respectively. The XGBoost predictive models led in attaining the least uncertain lithofacies and permeability records for the cored data. For further validation, the predicted lithofacies and reservoir permeability were then compared with porosity–permeability values derived from the nuclear magnetic resonance (NMR) log, the secondary porosity of the full-bore micro imager (FMI) and the production contribution from the production–logging tool (PLT). Therefore, it is believed that the XGBoost model is capable of making accurate predictions of lithofacies and permeability for the same well's non-cored intervals and other non-cored wells in the investigated reservoir.

CO2 Injectivity Test Proves Concept of CCUS Field Development

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 216673, “CO2 Injectivity Test Proves the Concept of CCUS Field Development,” by Yermek Kaipov, SPE, and Bertrand Theuveny, SLB, and Ajay Maurya, Saudi Aramco, et al. The paper has not been peer reviewed. _ The complete paper presents a unique case study on injectivity tests done in Saudi Arabia to prove the concept of carbon capture, use, and storage (CCUS) capability. It describes the design of surface and downhole testing systems, lessons learned, and recommendations. The injectivity tests were effective in identifying and confirming the best reservoir for CO2 injection and defining the best completion strategy. Creating injection conditions close to CCUS is vital, especially in heterogeneous carbonate reservoirs where the petrophysical correlations for the reservoir model require calibration with dynamic data. Introduction The energy company has conducted an extensive evaluation campaign by drilling appraisal wells through multizone saline aquifer reservoirs on different sites close to potential sources of CO2 at the surface. The evaluation program included coring, openhole logging, formation testing for stress-test and water sampling, and injectivity testing in the cased hole. Apart from reservoir characterization, different completion strategies were evaluated by performing injectivity tests in both vertical and horizontal wells. The lower completion was represented by perforated casing and an open hole. Injectivity Testing Injection tests are a commonly used method in waterflood projects to evaluate the injectivity capacity of the well and reservoir. The test involves an injection period with one or more injection rates, followed by a falloff period (Fig. 1). During the injection period, the liquid is injected at a stable rate to reduce the risk of near-wellbore formation damage caused by fluid incompatibility or exceeding the fracture gradient and inducing formation fracturing. The bottomhole-pressure data acquired during the test is analyzed using the pressure transient analysis method to estimate the permeability thickness, skin factor, and lateral heterogeneities. Additionally, the injection logging profile can be conducted along the sandface to assess completion efficiency and formation heterogeneity. By interpreting the results of the injection test, engineers can optimize the injection rate and improve the performance of the well and reservoir, ultimately leading to more-efficient oil recovery. Injectivity Test: Case Study The injectivity tests were conducted on virgin reservoirs using vertical appraisal wells that were sidetracked horizontally into the reservoirs with the greatest potential for storage. The reservoirs’ depths varied from 4,000 to 8,000 ft, with a normal gradient of reservoir pressure and temperature. The injectivity test design used reservoir properties estimated from the openhole evaluation, such as porosity, permeability, reservoir pressure, temperature, reservoir fluid sample, and fracture gradient. These data were used to set injectivity-test objectives, calculate expected well parameters, select equipment, and plan operations. The primary goal of the tests was to assess reservoir injectivity by injecting water, nitrogen, and CO2 to prove the concept for a CO2-injection project. While water and nitrogen injections are well-known in the industry, the CO2 injectivity test was new and required more attention during the design phase to evaluate all possible risks.

Assessment of World’s First Two Polymer Injectivity Tests Performed in Two Giant High-Temperature/High-Salinity Carbonate Reservoirs Using Single-Well Simulation Models and Pressure Falloff Tests Analysis

Summary Polymer flooding is a mature enhanced oil recovery (EOR) technology that has been widely implemented around the world for more than 60 years. Polymer flooding mostly targets medium- to high-permeability sandstone reservoirs with moderate salinity, hardness, and temperatures. However, in the last few years, the envelope of polymer flooding has been expanded to harsher reservoir conditions of high-temperature and high-salinity mixed-wet to oil-wet heterogeneous carbonate reservoirs. Development of novel polymers and innovative field application concepts has allowed for the reconsideration of polymer-based EOR as a promising technology to improve sweep efficiency for these challenging reservoirs. Polymer injectivity is one of the key challenges for polymer flooding projects and requires a rigorous derisking program that includes laboratory and field testing. A comprehensive laboratory program was designed to assess and investigate polymer thermal stability, polymer rheology in porous media, adsorption, and injectivity using reservoir core samples. In addition, two polymer injectivity tests (PITs) were performed in two giant light oil (0.3 cp) carbonate reservoirs in onshore Abu Dhabi under harsh conditions of high salinity (>200 g/L), high divalent ions (>20 g/L), high temperature (>250°F), and H2S concentration of up to 40 ppm. The polymer used during the two PITs (PIT 1 2019 and PIT 2 2021) is a new generation of EOR polymer (SAV 10) with high 2-acrylamido-tertiary-butyl sulfonic acid content that was specifically developed to tolerate such harsh conditions. This paper is focused on the interpretation of the PITs and lessons learned for future polymer-based EOR projects. The detailed data acquired in both tests were used to evaluate the polymer injectivity at representative field conditions and in-depth mobility reduction. The PITs together with the extensive laboratory studies are part of a thorough derisking program for the upcoming world’s first innovative hybrid EOR multiwell pilots—simultaneous injection of miscible gas and polymer (SIMGAP) and simultaneous injection of water and polymer (SIWAP). Both PITs are composed of three stages that include a multirate waterflood baseline, polymer injection using different rates, and polymer concentrations followed by extended chase waterflooding. In addition, a sequence of multiple pressure falloff (PFO) tests was acquired during the PIT executions and analyzed to obtain the required uncertainty parameters for the history-matching exercise. Polymer preshearing was considered as part of both PIT programs with the aim to homogenize the polymer molecular weight distribution and reduce possible shear-thickening effects near the wellbore as per laboratory measurements. Two single-well 3D simulation models were built to incorporate the information from polymer laboratory studies and to interpret the large field data sets acquired during the PITs. Lessons learned from PIT 1 allowed us to optimize the PIT 2 design program and achieve better understanding of polymer characteristics. The interpretation of the pressure transient analysis (PTA) of the PFO tests and the 3D simulation models of the two PITs confirmed the generation of polymer banks and demonstrated effective propagation of the polymer into the reservoirs at target concentrations and representative rates of the future SIWAP and SIMGAP interwell pilots.

Fracture-vuggy carbonate reservoir characterization based on multiple geological information fusion

The complexity and strong heterogeneity of carbonate reservoirs with fracture-vuggy structures present significant challenges in reservoir characterization. To address these challenges, we propose a novel multi-element information fusion modeling approach. This approach is designed to integrate multiple methods and incorporate multi-probability fusion at various facies and scales, thereby bridging the gap between geological information and reservoir modeling. Our methodology involves four key steps. First, the statistics between frequency of karst and geological information are acquired, and we quantify the statistics to regression equations. Second, these regression equations are transferred to probability bodies. The probability bodies can be applied in modeling as a soft control. But just one single body can be input in modeling process. Third, multiple probability bodies are fused into a fusion probability body by a probability fusion algorithm, which can keep the potential information of probability bodies. Finally, we apply the probability body in modeling workflow. By this way, the fusion method bridges the gap between geological information and modeling. The model established through our proposed method showed a significant level of consistency with reservoir re-evaluation, achieving an impressive 90% degree of alignment. Furthermore, the history match analysis revealed a high correlation, indicating the model's reliability. The method effectively integrates various scales and types of geological information, offering an accurate approach to complex carbonate reservoir modeling.

An Approach to Define and Quantify Porosity System in Heterogeneous Carbonate Reservoirs through Textural Analysis using High-Resolution Resistivity Image in the Bassein Formation of Mumbai Offshore Basin, India

Summary In evaluating the porosity system of the Bassein formation, high-resolution resistivity image (FMI) data have been used. Major features observed from the FMI data are the presence of low-angle fractures and solution channels of varying length and thickness. Against the layers with very-low-density neutron porosity, solution channels are seen as conductive features. These mud-filled solution channels can actively participate in production. Being very small-scale features, these types of porosities are not identifiable from conventional log data. To define and quantify porosity types such as vugs, solution channel, and matrix porosity, the resistivity image data are processed and image-based porosity is calculated. A core thin-section study is carried out and is used for calibration of the image-derived porosity system. Using resistivity image data, we have computed porosity types, for example, vugs, connected matrix and resistive, and their contribution to total porosity. Identification of hydrocarbons in reservoir layers with their porosity contribution (porosity partitioning) gives better insight into hydrocarbon production in these low-porosity layers which are producing after acid job.

A fresh look at the Lucia classification using mud- and grain-dominated reservoirs of the Persian Gulf

Carbonate reservoir heterogeneity, a multifaceted phenomenon in the domain of subsurface exploration, plays a pivotal role in the assessment of hydrocarbon resources. The upper Dalan and Dariyan formations are significant gas and oil reservoirs in the Persian Gulf. Both of these formations are strongly heterogenous. Different methods are used to evaluate reservoir heterogeneities and determine rock types. In this research, Lucia's standard petrophysical method was used to manage and compare the heterogeneity of the upper Dalan and Dariyan formations. To investigate the factors influencing the distribution of samples within Lucia's petrophysical classes, microfacies and diagenetic processes were studied in both formations using 1483 thin sections. To determine reservoir rock types, 2080 porosity and permeability data were plotted on Lucia's petrophysical chart. The charts were plotted at multiple scales, including the whole well, reservoir units, and microfacies levels. All Lucia's classes were observed in the upper Dalan Formation. Dissolution and dolomitization in the mud-dominated microfacies have led to the increase in both porosity and permeability, resulting in the distribution of samples primarily within Lucia classes 1 and 2. On the other hand, calcite and anhydrite cements in the grain-dominated microfacies have reduced the reservoir quality and increased the number of samples in the lower parts of the Lucia's chart. In contrast to the upper Dalan, the samples from the Dariyan Formation were concentrated within class 3 and the lower part of the Lucia's chart. This is due to the high abundance of mud-dominated microfacies and the presence of high microporosity and low permeability. Finally, by comparing the results obtained from the two formations, it can be concluded that the Lucia's method is more suitable for grain-dominated formations compared to mud-dominated reservoirs. The findings of this study can help for better understanding of the relationships between carbonate textures and porosity-permeability. Results also shed light on evaluating the heterogeneity of different carbonate reservoirs.

Secondary porosity classification in heterogeneous carbonate reservoirs based on the sonic log data

Many carbonate oil and gas reservoirs are characterized by the presence of secondary pores: vugs and fractures. To estimate the volume of oil and gas in carbonate rocks with double porosity, the information about secondary pore structure is required. In the present work we propose a simple and easy to code methodology to determine secondary pore types in carbonate sediments using acoustic well log data. The methodology is based on the comparison of the normalized differences between the calculated and measured in a borehole velocity of the compressional and shear waves. The results were obtained for the rocks with polymineral skeleton (limestone and dolomite). The mineral concentration is estimated by the solution of the inverse problem using the data obtained by a neutron, density, and PEF logging. The classification results were adjusted by using the micromechanical methods (the Kuster-Toksöz method and the Effective Medium Approximation (EMA)). Examples of the developed methodology application in the carbonate reservoirs are presented.

An advanced ensemble modeling approach for predicting carbonate reservoir porosity from seismic attributes

This study uses a machine learning (ML) ensemble modeling approach to predict porosity from multiple seismic attributes in one of the most promising Main Dolomite hydrocarbon reservoirs in NW Poland. The presented workflow tests five different model types of varying complexity: K-nearest neighbors (KNN), random forests (RF), extreme gradient boosting (XGB), support vector machine (SVM), single layer neural network with multilayer perceptron (MLP). The selected models are additionally run with different configurations originating from the pre-processing stage, including Yeo–Johnson transformation (YJ) and principal component analysis (PCA). The race ANOVA method across resample data is used to tune the best hyperparameters for each model. The model candidates and the role of different pre-processors are evaluated based on standard ML metrics – coefficient of determination (R2), root mean squared error (RMSE), and mean absolute error (MAE). The model stacking is performed on five model candidates: two KNN, two XGB, and one SVM PCA with a marginal role. The results of the ensemble model showed superior accuracy over single learners, with all metrics (R2 0.890, RMSE 0.0252, MAE 0.168). It also turned out to be almost three times better than the neural net (NN) results obtained from commercial software on the same testing set (R2 0.318, RMSE 0.0628, MAE 0.0487). The spatial distribution of porosity from the ensemble model indicated areas of good reservoir properties that overlap with hydrocarbon production fields. This observation completes the evaluation of the ensemble technique results from model metrics. Overall, the proposed solution is a promising tool for better porosity prediction and understanding of heterogeneous carbonate reservoirs from multiple seismic attributes.

Porosity prediction from prestack seismic data via deep learning: incorporating a low-frequency porosity model

Abstract Porosity prediction from seismic data is of considerable importance in reservoir quality assessment, geological model building, and flow unit delineation. Deep learning approaches have demonstrated great potential in reservoir characterization due to their strong feature extraction and nonlinear relationship mapping abilities. However, the reliability of porosity prediction is often compromised by the lack of low-frequency information in bandlimited seismic data. To address this issue, we propose incorporating a low-frequency porosity model based on geostatistical methodology, into the supervised convolutional neural network to predict porosity from prestack seismic angle gather and seismic inversion results. Our study demonstrates that the inclusion of the low-frequency porosity model significantly improves the reliability of porosity predictions in a heterogeneous carbonate reservoir. The low-frequency information can be compensated to enhance the network's capabilities of capturing the background porosity trend. Additionally, the blind well tests validate that considering the low-frequency constraint leads to stronger model prediction and generalization abilities, with the root mean square error of the two blind wells reduced by up to 34%. The incorporation of the low-frequency reservoir model in network training also remarkably enhances the geological continuity of seismic porosity prediction, providing more geologically reasonable results for reservoir characterization.

Technology Focus: Field Development (September 2023)

The integration of machine learning and physics-based simulation modeling offers a comprehensive approach to optimizing unconventional development and addressing parent-/child-well depletion issues. By combining the predictive capabilities of machine-learning models with the insights gained from physics-based modeling, operators can develop customized solutions tailored to the specific characteristics of the reservoir. Machine-learning models can optimize well-spacing, -placement, and -completion designs, while physics-based simulation modeling provides a deeper understanding of reservoir behavior and performance. Machine learning has emerged as a powerful tool, leveraging large volumes of data to provide data-driven insights and predictive capabilities. By training machine-learning models on geologic, completion, and spacing data, operators can predict production performance and make informed decisions to enhance production and cost-efficiency. Parent/child interactions, where production from a parent well negatively affects subsequent child wells, pose challenges in unconventional development. Benchmarking, through the comparison and analysis of performance data, plays a crucial role in understanding and mitigating these issues. By benchmarking production data, completion designs, and reservoir properties, operators can identify best practices and develop strategies to minimize the interference between parent and child wells. By harnessing the power of data-driven insights and the knowledge gained from physics-based simulation modeling, operators can make informed decisions regarding well spacing, completion designs, and operational strategies. This integrated approach enhances decision-making, reduces uncertainties, and increases overall efficiency in unconventional development projects. As technology continues to advance, the use of machine learning and physics-based simulation modeling will further improve the optimization of unconventional development, driving the industry toward greater efficiency and sustainability. The selected papers and suggested additional reading provide valuable insights into the evolving field of development optimization using physics-based simulation models and machine learning. Recommended additional reading at OnePetro: www.onepetro.org. SPE 205651 Digital Field Development Planning: A Collaboration Between Technology and Process To Enable Fast and Efficient Field Development Planning by Mohd Aminuddin Bin Md Karim, Petronas SPE 207899 Artificial Intelligence for Infill Well Placement and Design Optimization in Multilayered/Stacked Reservoirs Under Subsurface Uncertainty by Shi Su, SLB, et al. SPE 208529 A Comprehensive Machine-Learning Approach for Quantitatively Analyzing Development Performance and Optimization for a Heterogeneous Carbonate Reservoir in the Middle East by Ruijie Huang, Research Institute of Petroleum Exploration and Development, et al.

Geo-Body and Geostatistical Modelling of Carbonate Reservoir Facies Architecture and Characterization

Carbonate reservoirs present significant challenges in characterizing and extracting hydrocarbons due to their low permeability, matrix heterogeneities, fractures, and dissolution patterns. Accurately predicting the facies architecture and reservoir properties in such complex formations has been a persistent challenge for geoscientists. This paper proposes an integrated approach that combines geo-body extraction and geostatistical modeling to accurately predict the facies architecture and reservoir properties in carbonate reservoirs. The methodology begins by generating 3D seismic root mean square amplitude (RMS) attributes, which are then used to extract geo-bodies along the pay sequences. The extracted geo-bodies are then subjected to geostatistical modeling to analyze reservoir properties to facilitate the optimization of drilling and production strategies. To validate the effectiveness of the proposed approach, a small field in the Mumbai offshore basin is chosen as a case study. This field is located on the Mumbai High-Deep Continental Shelf and exhibits westerly dipping structures. Structural mapping confirms the presence of an antiformal structure, with one particular well (D-8) at the crest showing the absence of hydrocarbons. The proposed approach mapped two seismic reflectors within the reservoir zones and generated window-based 3D seismic RMS attributes to extract three geo-bodies within the reservoir. Facies and property modeling revealed the presence of distinct non-reservoir facies with poor reservoir properties near dry wells (D-8, D-4, and D-7), which is in line with the production performance observed in the drilled wells. The proposed integrated approach of geo-body extraction and geostatistical modeling is effective in delineating the facies architecture and reservoir heterogeneity of carbonate reservoirs. It enables the identification of favorable reservoir facies and facilitates a comprehensive assessment of the remaining potential.

Water alternative gas (WAG) optimization for a heterogeneous Brazilian pre-salt carbonate reservoir

Abstract Water alternating gas (WAG) is a cyclical process that involves alternating water and gas injections with the primary goal to improve sweep efficiency by maintaining initial high pressure, slowing water and gas breakthrough, and lowering oil viscosity. The objective of this work is to apply and optimize a WAG strategy on a carbonate field with light oil, compare it to the initially planned water-flooding strategy, and investigate the capability of WAG to improve field production. In this research, a compositional reservoir simulator was used to model a WAG process by injecting produced gas into the reservoir, using the same well structure as an optimized water-flooding strategy. Subsequently, a WAG strategy was created, optimizing the number and locations of wells, to facilitate a comparative analysis of the two recovery methods. The WAG optimization involved a detailed assessment of variables such as bottom hole pressure (BHP), WAG cycle duration, maximum gas oil ratio (GOR), and well positioning, to achieve a high net present value (NPV). The study focuses on the application of WAG optimization modeling in unconventional reservoirs, specifically pre-salt carbonate reservoirs, and investigates its implications on production strategy and forecast, emphasizing its potential for maximizing NPV and oil recovery in a recently producing field. The results showed that WAG improved reservoir performance when compared to water injection and produced a greater amount of oil. This solution showed potential to be tested under uncertainties (reservoir heterogeneity, faults, fractures, karsts, vugs, etc.) as future steps.

Multi-level ultra-deep fault-controlled karst reservoirs characterization methods for the Shunbei field

Ultra-deep carbonate fault-controlled reservoirs are characterized by large burial depths, low dissolution degrees, strong heterogeneity, and limited effective well data. Accurate 3D characterization based on seismic data is essential for efficient development of this type of reservoir. However, traditional seismic prediction methods are insufficient to accurately characterize different reservoir levels in the exploration and development of fault-controlled ultra-deep reservoirs. We propose a set of improved multi-level characterization methods for fault-controlled reservoirs. The improved methods could recover seismic information obscured by strong reflections and reduce the uncertainty of seismic interpretation. This study combined seismic strong reflection suppression and sedimentary strata seismic reflection interference elimination, proposed improved inversion methods, and advanced attribute calculation methods to improve the identification accuracy of reservoirs. In particular, we proposed the karst cave carving method based on improved inversion method and karst cave enhancement algorithm, the dissolved pore zone identification method based on optimization energy envelope algorithm, and the fault-fracture zone characterization method based on optimized atomic decomposition texture contrast. These methods were thoroughly validated by theoretical 3D models and field data. The proposed multi-level characterization methods can effectively improve the identification accuracy of fault-controlled karst reservoirs, provide a benchmark for predicting similar strong heterogeneous carbonate reservoirs, and provide reliable support for further facies modeling.