Multi-resolution Graph-based Clustering Research Articles

Data-driven machine learning approaches for precise lithofacies identification in complex geological environments

ABSTRACT Reservoir characterization is a vital task within the oil and gas industry, with the identification of lithofacies in subsurface formations being a fundamental aspect of this process. However, lithofacies identification in complex geological environments with high dimensions, such as the Lower Indus Basin in Pakistan, poses a notable challenge, especially when dealing with limited data. To address this issue, we propose four common data-driven machine learning approaches: multi-resolution graph-based clustering (MRGC), artificial neural networks (ANN), K-nearest neighbors (KNN), and self-organizing map (SOM). We utilized these proposed approaches to assess their performance in scenarios with varying core sample availability, specifically evaluating their effectiveness in identifying lithofacies within the Lower Goru formation of the middle Indus Basin. The study reveals that in scenarios with a limited number of core samples, MRGC is the preferred choice, while KNN or MRGC is more suitable for larger datasets. The results demonstrate the superior performance of MRGC and KNN in lithofacies identification within the specified geological environment, with SOM following closely behind, and ANN exhibiting comparatively lower efficacy. The accurate identification of lithofacies from the selected model is complemented by the application of the truncated Gaussian simulation method for facies modeling. Comparative results confirm the excellent agreement between the model identification of lithofacies from well logs and electro-facies obtained from the truncated Gaussian simulation electro-facies volume. This study highlights the crucial role of selecting the right machine learning approach for precise lithofacies identification and modeling in complex geological environments. The comparative analysis provides practitioners in the petroleum industry with insights into the strengths and limitations of each method, enhancing existing knowledge. In conclusion, this research emphasizes the significance of comprehensive research and method selection for advancing lithofacies identification in diverse formations or study areas, ultimately benefiting the broader field of subsurface characterization in the petroleum industry.

Pore Size Classification and Prediction Based on Distribution of Reservoir Fluid Volumes Utilizing Well Logs and Deep Learning Algorithm in a Complex Lithology

Pore size analysis plays a pivotal role in unraveling reservoir behavior and its intricate relationship with confined fluids. Traditional methods for predicting pore size distribution (PSD), relying on drilling cores or thin sections, face limitations associated with depth specificity. In this study, we introduce an innovative framework that leverages nuclear magnetic resonance (NMR) log data, encompassing clay-bound water (CBW), bound volume irreducible (BVI), and free fluid volume (FFV), to determine three PSDs (micropores, mesopores, and macropores). Moreover, we establish a robust pore size classification (PSC) system utilizing ternary plots, derived from the PSDs.Within the three studied wells, NMR log data is exclusive to one well (well-A), while conventional well logs are accessible for all three wells (well-A, well-B, and well-C). This distinction enables PSD predictions for the remaining two wells (B and C). To prognosticate NMR outputs (CBW, BVI, FFV) for these wells, a two-step deep learning (DL) algorithm is implemented. Initially, three feature selection algorithms (f-classif, f-regression, and mutual-info-regression) identify the conventional well logs most correlated to NMR outputs in well-A. The three feature selection algorithms utilize statistical computations. These algorithms are utilized to systematically identify and optimize pertinent input features, thereby augmenting model interpretability and predictive efficacy within intricate data-driven endeavors. So, all three feature selection algorithms introduced the number of 4 logs as the most optimal number of inputs to the DL algorithm with different combinations of logs for each of the three desired outputs. Subsequently, the CUDA Deep Neural Network Long Short-Term Memory algorithm(CUDNNLSTM), belonging to the category of DL algorithms and harnessing the computational power of GPUs, is employed for the prediction of CBW, BVI, and FFV logs. This prediction leverages the optimal logs identified in the preceding step. Estimation of NMR outputs was done first in well-A (80% of data as training and 20% as testing). The correlation coefficient (CC) between the actual and estimated data for the three outputs CBW, BVI and FFV are 95%, 94%, and 97%, respectively, as well as root mean square error (RMSE) was obtained 0.0081, 0.098, and 0.0089, respectively. To assess the effectiveness of the proposed algorithm, we compared it with two traditional methods for log estimation: multiple regression and multi-resolution graph-based clustering methods. The results demonstrate the superior accuracy of our algorithm in comparison to these conventional approaches. This DL-driven approach facilitates PSD prediction grounded in fluid saturation for wells B and C.Ternary plots are then employed for PSCs. Seven distinct PSCs within well-A employing actual NMR logs (CBW, BVI, FFV), in conjunction with an equivalent count within wells B and C utilizing three predicted logs, are harmoniously categorized leading to the identification of seven distinct pore size classification facies (PSCF). this research introduces an advanced approach to pore size classification and prediction, fusing NMR logs with deep learning techniques and extending their application to nearby wells without NMR log. The resulting PSCFs offer valuable insights into generating precise and detailed reservoir 3D models.

Petrophysical Parameters and Electrofacies Models of the Itararé Group Applying Multi-Resolution Graph-Based Clustering Method, Paraná Basin

Volume shale (Vsh) and effective porosity (Φe) were calculated and partially applied as petrophysical input parameters to generate electrofacies models using the Multi-Resolution Graph-Based Clustering method in sedimentary successions of the Itararé Group. The petrophysical parameters and electrofacies models were determined from geophysical data from two wells located in the eastern portion of the Paraná Basin. The stratigraphic intervals in each well (analogues of hydrocarbon reservoirs) allowed petrophysical analysis and the determination of electrofacies models based on gamma ray profiles, apparent density and neutron porosity, together with lithological data. The petrophysical parameters were calculated by different procedures, and the effective porosity results were applied as input parameters for electrofacies modeling. The electrofacies models were correlated to lithological data and patterns of gamma ray profiles and apparent density, as well as different types of porosity. The generated models provide differences related to the type of porosity, the method applied to calculate the effective porosity and present probable relationships with the lithological units of the wells.

An integrated geomechanical and petrophysical multiparameter approach for gas reservoir evaluation

Integrating petrophysical and geomechanical parameters is an efficient approach to evaluating shale gas reservoir potential. The high cost of corings and their limited number, coupled with time-intensive investigation, led researchers to use this alternative combination approach. In the Jiaoshiba area, from single-pilot well core data and log measurements, petrophysical and geomechanical parameters such as shale volume, total organic carbon, gas content, as well as pore pressure, stress components, and mineral brittleness were first estimated using established methods. In the second phase, based on logging curves, the reservoir electro-facies (EF) classification was performed using the unsupervised multi-resolution graph-based clustering method on a series of twenty wells, identifying five EF with different intrinsic characteristics. Unsupervised analyses were developed using the multilayer artificial neural network while incorporating the K-nearest neighbors and graphical classification algorithms. The results from the first and second phases indicate reservoir richness in organic matter, with the best reservoir exhibited by EF2 and EF3. In addition, effective stress components (SV, SH, and Sh) evaluation shows a normal stress regime with hydraulic fracture systems perpendicular to the minimum horizontal stress at each measured depth of the reservoir (Sv > SH > Sh). This research workflow can efficiently evaluate shale reservoirs with a realistic approach for identifying favorable fracturing positions while reducing errors due to human interference.

Determination of Reservoir Rock Type in Sarvak Reservoir of an Iranian Oilfield

Integrated reservoir rock typing in carbonate reservoirs is a significant step in reservoir modelling. The key purpose of this study is the identification of integrated rock types in the Sarvak Formation of an Iranian oilfield. In this study, electrofacies (EFAC) analysis of the Sarvak reservoir was done in detail to determine the reservoir quality and rock types of the Sarvak Formation in the studied field. The core data and conventional petrophysical logs were used for rock typing. Some petrophysical logs such as porosity, sonic, neutron, density, and Photo electric factor were applied as input data for electrofacies analysis. Multi-Resolution Graph-Based Clustering was used among six approaches, resulting in four electrofacies after merging similar clusters. EFAC-1 has the lowest porosity, and EFAC-4 has the highest porosity. In addition, based on Winland and Amaefule approaches, three rock types were determined using core data (porosity and permeability). As a result, three rock types were defined; rock type-1 has the smallest pore-throat size and flow zone indicator while rock type-3 has the highest ones. The correlation coefficient between permeability and porosity in each rock type is more than 80%. Rock type 1 is mainly composed of EFAC-1 and EFAC-2, while rock types 2 and 3 share mostly the EFACs 1, 3 and 4.

A new method based on multiresolution graph-based clustering for lithofacies analysis of well logging

A new method based on multiresolution graph-based clustering for lithofacies analysis of well logging

Electrical facies of the Asmari Formation in the Mansouri oilfield, an application of multi-resolution graph-based and artificial neural network clustering methods

Electrofacies analysis conducted the distribution effects throughout the reservoir despite the difficulty of characterizing stratigraphic relationships. Clustering methods quantitatively define the reservoir zone from non-reservoir considering electrofacies. Asmari Formation is the most significant reservoir of the Mansouri oilfield in SW Iran, generally composed of carbonate and sandstone layers. The stratigraphical study is determined by employing 250 core samples from one exploratory well in the studied field. Five zones with the best reservoir quality in zones 3 and 5 containing sandstone/shale are determined. Moreover, multi-resolution graph-based and artificial neural network clustering involving six logs are employed. Utilizing Geolog software, an optimal model with eight clusters with better rock separation is obtained. Eventually, five electrofacies with different lithological compositions and reservoir conditions are identified and based on lithofacies describing thin sections, sandstone, and shale in zones 3 and 5 show high reservoir quality. According to the depth related to these zones, most of the facies that exist in these depths include sandstone and dolomite facies, and this is affected by the two factors of the primary sedimentary texture and the effect of the diagenesis process on them. Results can compared to the clustering zone determination in other nearby sandstone reservoirs without cores.

Employing Statistical Algorithms and Clustering Techniques to Assess Lithological Facies for Identifying Optimal Reservoir Rocks: A Case Study of the Mansouri Oilfields, SW Iran

The crucial parameters influencing drilling operations, reservoir production behavior, and well completion are lithology and reservoir rock. This study identified optimal reservoir rocks and facies in 280 core samples from a drilled well in the Asmari reservoir of the Mansouri field in SW Iran to determine the number of hydraulic flow units. Reservoir samples were prepared, and their porosity and permeability were determined by measuring devices. The flow zone index (FZI) was calculated for each sample using MATLAB software; then, a histogram analysis was performed on the logarithmic data of the FZI, and the number of hydraulic flow units was determined based on the obtained normal distributions. Electrical facies were determined based on artificial neural network (ANN) and multi-resolution graph-based clustering (MRGC) approaches. Five electrical facies with dissimilar reservoir conditions and lithological compositions were ultimately specified. Based on described lithofacies, shale and sandstone in zones three and five demonstrated elevated reservoir quality. This study aimed to determine the Asmari reservoir’s porous medium’s flowing fluid according to the C-mean fuzzy logic method. Furthermore, the third and fourth flow units in the Asmari Formation have the best flow units with high reservoir quality and permeability due to determining the siliceous–clastic facies of the rock units and log data. Outcomes could be corresponded to the flow unit determination in further nearby wellbores without cores.

Lithofacies identification and prediction of Chang 7<SUB align="right">3 shale oil reservoirs in Ordos Basin: insights from the multi-resolution graph-based clustering analysis

Lithofacies identification and prediction of Chang 7<SUB align="right">3 shale oil reservoirs in Ordos Basin: insights from the multi-resolution graph-based clustering analysis

Lithofacies identification and prediction of Chang 7<SUB align="right">3 shale oil reservoirs in Ordos Basin: insights from the multi-resolution graph-based clustering analysis

Lithofacies identification and prediction of Chang 7<SUB align="right">3 shale oil reservoirs in Ordos Basin: insights from the multi-resolution graph-based clustering analysis

Reservoir Petrofacies Predicted Using Logs Data: A Study of Shale Oil from Seven Members of the Upper Triassic Yanchang Formation, Ordos Basin, China

The identification and prediction of petrofacies plays a crucial role in the study of shale oil and gas “sweet spots”. However, the petrofacies identified through core and core test data are not available for all wells. Therefore, it is essential to establish a petrofacies identification model using conventional well logging data. In this study, we determined the petrofacies of shale oil reservoirs in the Upper Triassic Yanchang Formation, Ordos Basin, China, based on scanning electron microscopy, core porosity and total organic carbon (TOC), and brittleness index calculations from X-ray diffraction (XRD) experiments conducted on seven members of the formation. Furthermore, we compared the interpreted logs with the raw well logs data clustered into electrofacies in order to assess their compliance with the petrofacies, using the Multi-Resolution Graph-Based Clustering (MRGC) method. Through an analysis of pore structure type, core porosity, TOC, and brittleness index, we identified four types of lithofacies with varying reservoir quality: PF A > PF B > PF C > PF D. The compliance of the clustered electrofacies with the petrofacies obtained from the interpreted logs was found to be 85.42%. However, the compliance between the clustered electrofacies and the petrofacies obtained from the raw well logs was only 47.92%. Hence, the interpreted logs exhibit a stronger correlation with petrofacies characterization, and their utilization as input data is more beneficial in accurately predicting petrofacies through machine learning algorithms.

Lithofacies Characteristics and Sweet Spot Distribution of Lacustrine Shale Oil Reservoirs: A Case Study of the Second Member of the Kongdian Formation in the Cangdong Sag, Bohai Bay Basin

In contrast to marine shale oil reservoirs, lacustrine shale exhibits rapid lithofacies changes and strong mineral compositional heterogeneity, posing new challenges for the evaluation and distribution prediction of shale oil sweet spots. The oiliness, reservoir properties, oil fluidity, and fracability of different lithofacies were analyzed using emission-scanning electron microscopy (FE-SEM) observation, low-pressure nitrogen physisorption (LNP) analysis, mercury intrusion porosimetry (MIP), nuclear magnetic resonance (NMR), and triaxial compression testing. Based on the mineral composition obtained from X-ray diffraction (XRD) analysis, total organic carbon (TOC) content, and sedimentary structure, four lithofacies were classified, which are organic-rich laminated calcareous shale (LC), organic-rich laminated siliceous shale (LS), organic-rich laminated mixed shale (LM), and organic-poor massive calcareous shale (MC). Considering the factors of oiliness, reservoir properties, oil fluidity, and fracability, the LC and LS lithofacies were determined as being high-quality sweet spots (type I). Within the stratigraphic sequence divided by GR-INPEFA curves, multi-resolution graph-based clustering (MRGC) analysis of sensitive well logs was used to classify the lithofacies, after which the distribution of sweet spots was predicted. The results reveal that the sweet spots exhibit regular changes in their vertical distribution and a ring-like pattern in their planar distribution, influenced by variations in the sedimentary environment. This finding can offer valuable guidance for the future exploitation of shale oil in the Guandong region.

Electro-facies classification based on core and well-log data

Facies studies represent a key element of reservoir characterization. In practice, this can be done by making use of core and petrophysical data. The high cost and difficulties of drilling and coring operations coupled with the time-intensive nature of core studies have led researchers toward using well-log data as an alternative. In the Teapot Dome Oilfield, where core data are limited to those from only a single well, we used well-log data for reservoir electro-facies (EF) studies via two unsupervised clustering methods, namely multi-resolution graph-based clustering (MRGC) and self-organizing map (SOM). Satisfactory results were obtained with both methods, distinguishing seven electro-facies from one another, where MRGC had the highest discriminatory accuracy. The best reservoir quality was exhibited by electro-facies 1, as per both methods. Our findings can be used to avoid some time-intensive steps of conventional reservoir characterization approaches and are useful for prospect modeling and well location proposal.

Multiparameter Logging Evaluation of Chang 73 Shale Oil in the Jiyuan Area, Ordos Basin

The accurate evaluation of shale oil and gas reservoirs is of great significance to the integrated development of geology and engineering. Based on the core analysis, conventional logging, and array acoustic logging data, the total organic carbon, hydrocarbon generation potential, brittleness, and anisotropy of shale reservoirs were calculated. The p-wave time difference curves calculated by the artificial neural network (ANN) method and the conventional logging curve fitting method were compared. The multiresolution graph-based clustering (MRGC) method was used to classify shale reservoirs into three categories and evaluate the classification results. Brittle minerals such as quartz and feldspar were mainly found to be present in shale reservoirs and clay minerals which mainly consisted of illite. The Chang 73 reservoir is rich in organic matter and has great potential for survival. The p-wave time difference calculated by the fitting formula of the shear wave time difference meter demonstrated high accuracy and did not require a complex ANN model. MRGC method can well classify shale reservoir types. The classification results reduce the interference of human factors and are more scientific and reasonable. This research method is of great significance for the scientific classification and evaluation of shale reservoirs.

Evaluating Machine Learning Techniques for Carbonate Formation Permeability Prediction Using Well Log Data

Machine learning has a significant advantage for many difficulties in the oil and gas industry, especially when it comes to resolving complex challenges in reservoir characterization. Permeability is one of the most difficult petrophysical parameters to predict using conventional logging techniques. Clarifications of the work flow methodology are presented alongside comprehensive models in this study. The purpose of this study is to provide a more robust technique for predicting permeability; previous studies on the Bazirgan field have attempted to do so, but their estimates have been vague, and the methods they give are obsolete and do not make any concessions to the real or rigid in order to solve the permeability computation. To verify the reliability of training data for zone-by-zone modeling, we split the scenario into two scenarios and applied them to seven wells' worth of data. Moreover, all wellbore intervals were processed, for instance, all five units of Mishrif formation. According to the findings, the more information we have, the more accurate our forecasting model becomes. Multi-resolution graph-based clustering has demonstrated its forecasting stability in two instances by comparing it to the other five machine learning models.

Integrating facies analysis and geostatistical methods in an onshore oil field using petrophysical groups

Reservoir facies studies are of great importance in different stages of exploration and development of hydrocarbon fields. We have aimed to generate a reservoir facies model for the Asmari Formation in an onshore oil field located in southwest Iran. Input data for electrofacies (EF) clustering algorithms are used, which include gamma-ray (GR), density (RHOB), porosity, and sonic logs from four wells. We obtain the petrophysical group (PG) and EF class using core drill data (mercury injection capillary pressure) and well-logs analysis. The integration of PGs and log EF significantly decreases the uncertainty in reservoir modeling, which alternatively enhances field development decisions. We compare the multiresolution graph-based clustering (MRGC) and k-means clustering methods. EF clustering results find nine EF classes. We delineate high-quality reservoirs based on lower GR, RHOB, and high-porosity logs. Next, we use the clustering results in the static reservoir modeling process, using the sequential index simulation and indicator kriging methods. The comparison between the facies obtained models and existing drilling core data finds that the absolute percentage error of the MRGC algorithm is less than that of the k-means algorithm. The results obtained by this study can provide useful information for the development of hydrocarbon exploration plans in the studied oil field.

Electrofacies Estimation of Carbonate Reservoir in the Scotian Offshore Basin, Canada Using the Multi-resolution Graph-Based Clustering (MRGC) to Develop the Rock Property Models

Electrofacies Estimation of Carbonate Reservoir in the Scotian Offshore Basin, Canada Using the Multi-resolution Graph-Based Clustering (MRGC) to Develop the Rock Property Models

Lithofacies Characteristics and Sweet Spot Distribution of Lacustrine Shale Oil: A Case Study from the Dongying Depression, Bohai Bay Basin, China

Abstract Lacustrine shale is characterized by rapid lithofacies transformation and compositional heterogeneity, which present challenges in shale oil sweet spot evaluation and distribution prediction and should be systematically studied. Field emission-scanning electron microscopy (FE-SEM), low-pressure adsorption isotherm analysis, mercury intrusion porosimetry (MIP), and triaxial compression testing were employed to comprehensively analyze the oil-bearing capacity, reservoir properties, fluidity, and frackability of different lithofacies. Via analyses of mineral composition, total organic carbon (TOC) content, and sedimentary structure, seven lithofacies were identified: organic-rich calcareous shale (L1), organic-rich laminated calcareous mudstone (L2), organic-rich laminated carbonate-bearing mudstone (L3), intermediate-organic laminated calcareous mudstone (L4), organic-poor laminated calcareous mudstone (L5), organic-poor thin-bedded calcareous mudstone (L6), and organic-rich laminated silty mudstone (L7). Considered together, the oil-bearing capacity, reservoir properties, fluidity, and frackability suggested that the L1 and L7 lithofacies were high-quality sweet spots, with satisfactory oil-bearing capacity (TOC>3.5%; S1>10 mgHC/grock), well-developed pores and microfractures, notable fluidity (as indicated by a high oil saturation index value), and suitable brittleness. The sweet spot distribution was predicted according to multiresolution graph-based clustering analysis of well logs. The results indicate that comprehensive research of the key factors for shale oil and lithofacies prediction can promote sweet spot prediction and enhance shale oil exploration.

Characterization of a heterogeneous carbonate reservoir by integrating electrofacies and hydraulic flow units: a case study of Kangan gas field, Zagros basin

Reservoir characterization is one of the key stages in oil and gas exploration, appraisal, development, and optimal production. During the exploration phase, core analysis and well logging are basic information obtained after the pay zone identification. This study aims to integrate the concept of electrofacies (EFs) with hydraulic flow units (HFUs) to effectively characterize Permo–Triassic carbonate formations at the Kangan giant gas field in southwestern Iran. For this purpose, well log data collected from 10 drilled wellbores and 490 core data have been analyzed based on similar clusters and statistical features in which rock-type approaches were used in order to integrate petrophysical data with facies. In particular, we used a multi-resolution graph-based clustering (MRGC) method to determine EFs and the probability plot method, histogram analysis, and the plot of reservoir quality index versus the normalized porosity to identify HFUs. Based on the data, five electrofacies and six HFUs were identified. To establish a good connection between the electrofacies and the pay zones, we analyzed wells’ cross sections to compare lithology, production potential in each zone, rock type, and amount of shale and eventually to adapt facies to flow units and determine the best quality reservoir zone. The main difference between this work and other studies in the literature is adopting a systematic approach based on integrating the geology (via EFs analysis) and engineering examination (via HFUs) to accurately characterize the hydrocarbon-bearing formations. The results of this study help in rapid and cost-effective carbonate reservoir characterization by combining electrofacies clustering and HFU analysis based on core and log data, which are available and routine information in all oil/gas fields. This, in turn, assists in developing the field, using more appropriate production and EOR scenarios, and even locating proper perforation sites.

Electrofacies modeling as a powerful tool for evaluation of heterogeneities in carbonate reservoirs: A case from the Oligo-Miocene Asmari Formation (Dezful Embayment, southwest of Iran)

The Asmari Formation comprises carbonate successions in the east and southeast of the Dezful Embayment, SW Iran. Besides, this Formation has undergone complicated diagenetic events. The Asmari reservoir shows high variation in quality and hydrocarbon production across the Bibi Hakimeh oilfield. Contrary to core data, petrophysical well-logs are almost available for all drilled wells in the oilfield. Therefore, electrofacies (EF) clustering was applied to divide well-log data into distinct clusters with individual petrophysical and geological attributes using Multi-Resolution Graph-based Clustering (MRGC) method. For the definition of diagenetic rock classes (DR), in which both depositional and diagenetic textures have been contributed, the main diagenetic modifications were analyzed in all sedimentary facies belonging to various depositional environments to know whether depositional environments have played a controlling role in reservoir heterogeneities of the Asmari Formation. The quality of electrofacies clusters was evaluated using the distribution of diagenetic rock classes and porosity-permeability data through these clusters. This study reveals that diagenetic rock classes, in which dolomitization and touched pores have prevailed, are more consistent with high-quality electrofacies, while low-quality electrofacies clusters are characterized by dense and cemented matrix together with some isolated pores. As a result, the reservoir heterogeneities of this reservoir have been widely controlled by diagenetic modifications, while depositional environments provided the required circumstances for domination of these modifications within each sedimentary facies. Finally, geostatistical modeling of electrofacies clusters reveals that domination of dolomitization and touched pores, as a consequence of dolomitizing fluids, together with microfracture systems improved reservoir quality of western/northwestern and central parts of the oilfield.