Determination of electrical facies using MRGC and ANN clustering methods

doi:10.21203/rs.3.rs-2987078/v1

Download PDF

Research Article

Determination of electrical facies using MRGC and ANN clustering methods

https://doi.org/10.21203/rs.3.rs-2987078/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Mar, 2024

Read the published version in Scientific Reports →

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

The present research was performed on Iran's most crucial reservoir formation (Asmari Formation) in the Mansouri oil field in southwest Iran. This Formation is generally composed of carbonate lithologic, but it has sandstone layers in some zones. It is also the youngest reservoir rock in Iran. In the first step, the lithology of the Asmari Formation was determined by cross-lithology diagrams, and a combination of lime, dolomite, anhydrite, and sandstone with shaley layers was estimated. Also, in the sequence of this geological/reservoir zoning formation, five zones were identified with different geological/reservoir conditions, and the average of each of these petrophysical parameters was calculated in each zone, with the best reservoir quality in zones 3 and 5 with sandstone/shale lithology. Also, this study determines facies clustering methods MRGC and ANN and reading of gamma, neutron, density, acoustic and resistivity diagrams, facies having common geological/reservoir conditions in different subdivisions. For this purpose, the combined log readings of DT, RHOB, GR, Sw, NPHI, and PHIE were used. After applying the software (Geolog) method, an optimal model with 8 clusters (facies) was obtained. The reservoir features and plots in the software for better separation of rock species studied. Finally, five facies with different lithological compositions and reservoir conditions were identified in this study. The information obtained from determining electrical facies by clustering method determines the reservoir zone from non-reservoir and qualitatively (good, medium, and poor) of Mansouri oilfield were achieved.

Petroleum Geology

Petroleum Engineering

Artificial Intelligence and Machine Learning

Asmari Reservoir

Mansouri Field

Electrofacies

Clustering MRGC

ANN

Electrofacies specify reservoir rock properties, especially permeability, to simulate fluid flow in porous media. These are determined based on the classification of similar logs among different groups of logging data. Data classification is accomplished by statistical analyses such as principal component analysis, cluster analysis, and differential analysis (Amraei and Falahat 2021; Bhattacharya et al. 2016; Davis 2018; Ding et al. 2022; Kianoush et al. 2022; Kianoush et al. 2023b; Rastegarnia et al. 2016; Ye and Rabiller 2005). A model of the reservoir's geological structure is also provided to create a three-dimensional model of a reservoir's properties, in addition to estimating porosity, permeability, water saturation, and other petrophysical properties. For geological modeling, lithofacies describe rock or sedimentary units with texture, structure, mineralogy, and rock properties, and these units can be employed to match important reservoir characteristics such as permeability and porosity (Gualda and Vlach 2007; Hosseini et al. 2023; Ibrahim et al. 2016; Kadkhodaie et al. 2006; Kianoush et al. 2023a; Manshad et al. 2021).

Therefore, identifying different rock facies of reservoir rock is a fundamental task in describing the characteristics of oil reservoirs. Detection of lithofacies in the repository is complex because the type and distribution of the facies are determined by the deposition system and are affected or altered by diagenesis and tectonics (Esfandyari et al. 2023; Mathis et al. 2003; Mirkamali et al. 2016; Shakiba et al. 2015; Shoghi et al. 2020). The most common method for determining lithofacies is the use of coring. The core data directly observes the lithofacies and can accurately distinguish the different facies. However, despite all the positive features, due to the high cost of core-extraction operations and the lack of 100% recovery, this method can only be used in a small section of the field around the well (Antelo and Aguirre 2001; Heydari et al. 2012; Ismail et al. 2021; Jouini et al. 2008; Madani et al. 2019).

On the other hand, core description is time-consuming and depends on the geologist's experience. Therefore, solving this problem requires a method cheaper than coring and capable of providing precision and separation of the rock facies to a suitable size similar to that of cores without wells (Kelkar 2005; Shirneshan et al. 2018). An excellent way to respond to this need is to use well-log data. Data loggers obtain indirect information from underground data and are much cheaper than kernels. Well-log calculations can be categorized into facies or electrical facies. Log facies can display rock and reservoir fluid properties and allow users to separate sedimentary units and layers. If the lithofacies match the lithofacies, lithofacies can be substituted for lithofacies and applied. Created facies can be later used to predict rock facies in core-free or core-free intervals (Abrar 2011; Frew 2004; Kianoush et al. 2022; Kianoush et al. 2023b). A practical method for facies analysis is to create a facies classification model that splits the log data into a set of log responses that describes sediment and provides sediment separation and identification from other sediments. This set of responses is called a cluster. These clusters have introduced various methods in which there is a significant problem, namely, the dependence on the dimensionality, which is different from the geological distance and the two points that are similar in interpreting the cluster. They may not be geologically similar. The reason for this problem is the different views of geologists and loggers. This problem creates a nonlinear solid relationship between log and lithology and causes the problem of determining any partitioning of the log data into the associated sedimentary unit (Kianoush et al. 2023a; Kianoush et al. 2023b; Kianoush et al. 2023c; Lai et al. 2023; Luthi 2001).

This study addresses some of the ambiguities of using drilling cuttings and logs data in analyzing petroleum formations by combining the results of logs and describing log facies. It is mainly done by defining the log facies and the reservoir zoning of the study well.

The easy determination and application of this zoning and its reliance on log data make it possible to quickly and accurately distinguish reservoir segments from non-reservoir segments. This debate is recent in Iran, and due to past computers' inability to perform complex analyses of log facies processes, the world is also overgrowing, and the cluster analysis method is used as the principal direction analysis of the log facies identification. It is used as an essential tool for logging in advanced petroleum software. Since these facies are determined only by pure mathematical processes in the cluster analysis method and no training or fitting function is used, they are very accurate, and their application to a particular well or field is not limited. After identifying the log facies by assigning geological features to them, it can be applied to other wells on the surface of adjacent fields to match and predict the desired features (Kianoush et al. 2023a; Kolbikova et al. 2021; Rabbani et al. 2018; Roslin and Esterle 2016).

Serra and Abbott (1982), and Serra (1984) defined electric facies as "a set of responders that, in addition to characterizing the sediments, allows them to be separated." Wolf and Pelissier-Combescure (1982), and Selley (1995) introduced the first automated method for classifying "facsimiles" into electrical facies. Tavakkoli and Amini (2006) used a set of logs instead of one log to assign more features to a particular façade simultaneously. In 2015, some studies of identifying and interpreting electrical facies using the self-organized neural network (SOM) method and its prediction for sedimentary facies in the Asmari reservoir of an SW oil field of Iran have been carried out (Farazani et al. 2015; Kiaei et al. 2015; Salehi et al. 2015; Zahmatkesh et al. 2015). Kiaei et al. (2015) Modeled three-dimensional reservoir electro-physics using integrated clustering and geostatistical methods in the Persian Gulf. Their approach included hierarchical cluster analysis (HCA), Multi-resolution graph-based clustering (MRGC), and self-organized maps (SOM). Based on their results, hierarchical clustering as a robust and practical approach to data clustering was performed based on profile validation and geological information.

El Sharawy and Gaafar (2016) zoned the reservoir based on statistical analyzes in the Nubian sandstone Gulf of Suez, Egypt. They showed that the entire reservoir could be divided into at least four electro physics with significant changes in reservoir quality, which correlated with permeability changes. Tian et al. (2016) used the multi-resolution graph-based clustering (MRGC) method for determining the electrofacies (EF) and lithofacies (LF) from well-log data obtained from the intraplatform bank gas fields located in the Amu Darya Basin. Rastegarnia et al. (2016) predicted 3D flow zone index (FZI) and electrified (EFACT) volumes from a large volume of 3D seismic data. Rafik and Kamel (2017) evaluated the permeability of the Formation for a Sandstone Reservoir in the South Ramsey reservoir formations from log well data using multivariate methods. Their results showed that permeability prediction using variable selection to non-parametric alternating conditional expectations (ACE) regression is the best way to predict permeability. Kadkhodaie and Kadkhodaie (2018) reviewed different reservoir rock typing approaches from geology to seismic and dynamic and proposed an integrated rock type workflow for worldwide carbonate reservoirs. Jafarzadeh et al. (2019) investigated the distribution of reservoir facies by using the available data to identify areas that are prone to be considered in the production and development plans of the field in terms of their storage and fluid flow capacity. Eventually, the experimental results illustrated that the Adaptive multi-resolution graph-based clustering algorithm for electrifying analysis also outperformed the original MRGC method on clustering and propagation prediction with higher efficiency and stability (Kianoush et al. 2023c; mohammadinia et al. 2023; Wu et al. 2020). Recently, the electrifies predicted using the MRGC approach to generate rock mechanical properties such as Young's modulus, Poisson's ratio, unconfirmed compressive strength, and internal friction coefficient (Alameedy et al. 2023; Mahadasu and Singh 2022; Masroor et al. 2023).

Finally, Okhovvat et al. (2023) used Kernel principal component analysis (KPCA) to improve the performance of electrical facies classification.

In this study, as an innovation, the wells were zoned using well-logging diagrams, and then, by using neural network clustering methods, MRGC, and ANN, electrical facies from log data were determined.

2.1. Zoning

In the sequence of oil/gas wells, zoning is one operation that segregates the sequence studied into zones with common conditions (geological or reservoir conditions, etc.). This section used log data to represent the Asmari Formation in the Mansouri Field accurately. Zones with typical reservoir geology (lithology) were studied in the well sequences using read diagrams. Figure 1 shows the sequence of the study well after zoning.

2.2. Determination of production zone

After evaluating the sequence of the Asmari Formation, the petrophysical parameters were averaged to determine the best production intervals in terms of oil accumulation in the well, and the best intervals with the highest hydrocarbon volume were determined. The net production zone is an essential indicator in evaluating reservoir quality. This thickness comprises a part of the reservoir with acceptable reservoir and petrophysical conditions. Shear limits were determined to achieve this goal for some of the calculated petrophysical parameters. These intervals are the zones where the focus of the reservoir studies is concentrated. Based on zonation in wells, the cost and technology for cutting boundaries differed.

The shear boundaries for the carbonate and sandstone sequences of the Asmari Formation are presented in Table 1.

Table 1

Cutting limits for carbonate and sandstone sections.
Parameter	Type	Cut off _(%) _CARBONATE	Cut off _(%) _SANDSTONE
PHIE	≥	4.5	8
SWE	≤	50	50
Vsh	≤	20	30

2.3. Stratigraphy of the studied field

Based on recent studies and using petrophysical information, the Asmari Formation has been divided into five zones.

Zone 1 (3538.5-3472.5 m): This zone exists in all drilled wells. The central part of the zone consists of dolomite plus a thick layer of limestone. Limestones are mostly cream to light brown and cream to gray, semi-hard to hard, fine-grained, micro-crystalline with anhydrite, chert, mudstone argillic to packstone.

Zone 2 (3582.5-3538.5 m): This zone is present in all wells. The lithology of this zone mainly consists of dolomite and sandstone.

Zone 3 (3605-3582.5 m): Its dominant lithology includes shale, limestone, and sandstone.

Zone 4 (3715 − 3605 m): This zone exists in all wells, and most of it contains a barrier/ beach ridge and is likely to be associated with the Ahvaz sand dunes. Much of the lithology of this zone is sandstone and shale.

Zone 5 (3772 − 3715 m): This zone cannot be identified in all wells due to a lack of logging data. The main lithologies in this zone are shale, sandstone, and limestone.

In the data preparation stage, after loading the data into the software, their quality in terms of readings, well control conditions, and logs used was checked. Afterward, the pre-computational stage was performed using well information, which finally obtained information about the well conditions and drilling mud properties in the evaluated sequence. Then environmental correction of the data was performed, and the effects of wells, drilling mud, etc., were removed from the logs readings, and consequently, zoning was performed using charts. In the next step, the lithology was evaluated and estimated in each sequence using corrected and edited logs and lithology cross-sections (neutron-density, Rho-U plot, MID plot, and MN plot). Finally, using the probabilistic method, the petrophysical parameters were calculated in the whole sequence, and the average of these parameters was calculated in the whole well and each zone. The production zones were also identified in the wells studied using the average petrophysical parameters and determining the cutoff for some of these parameters.

In the sequence of oil/gas wells, zoning is one procedure that segregates the sequence studied into zones with common conditions (geological or reservoir conditions, etc.). This section used log data to represent the Asmari Formation in the Mansouri Field accurately. Zones with typical reservoir geology (lithology) were studied in the well sequences using read diagrams.

According to Fig. 2, the segmentation 1 zone is 14.3 m thick in the well sequence. The dominant lithology of this zone is mainly anhydrite. A small percentage of shale and dolomite is also found in this zone. Due to the petrophysical property of anhydrite, this zone plays no role in the Asmari Formation in the well-studied. It has very low porosity and almost no hydrocarbon accumulation in this zone.

According to Fig. 3, this zone, which has the most sequence in the studied well, is 116 m thick and has the dominant lithology of calcite and dolomite. At some intervals, there are pores in the zone with little oil accumulation. The average porosity in this zone is also low and has high water saturation and, consequently, low oil volume.

The third segment in Fig. 4 has a minimum thickness of about 10 m from the study well. It has good oil accumulation capacity with sand and shale lithology. The high porosity, and low water saturation with good hydrocarbon volume, make this zone the main reservoir zones in the studied well sequenced.

The lithological conditions in of fourth segment and zone (Fig. 5) are approximately the same as those of the second zone, with the dominant lithology being calcite and dolomite containing shale and anhydrite. Its thickness is 15 meters. This zone has little porosity, and the water saturation is somewhat high, resulting in small quantities of oil.

This zone occupies the last part of the Asmari Formation sequence in the studied well (Fig. 6) and is the main reservoir zone in the studied sequence. The lithology of this zone is similar to that of the 3rd zone, sand and shale zones, which have high porosity, low water saturation, and consequently higher oil accumulations than other zones. The zone is 42 meters thick.

The average calculated petrophysical parameters of studied zones are shown in Table 2. According to the information presented in this table, it can be seen that the average petrophysical parameters in zones 3 and 5 have the best reservoir conditions compared to other zones. Due to the consequence of these two zones, the main reservoir zones in the Asmari Formation sequence are studied in shaley sandstone lithology.

Table 2

Average petrophysical parameters calculated per weight.
Interval	Vsh (%)	PHIE (%)	SWE (%)	Uoil (%)
Zone-1	2.7	2.4	99	0.01
Zone-2	18.1	3.9	91.8	0.055
Zone-3	27.4	13.1	46.5	7.4
Zone-4	11.5	8.9	76.7	2.9
Zone-5	29.5	12.9	63	6.3

3.1. Determination of electrical facies

The facies were categorized into wells by combining appropriate electrical diagram readings and proper clustering methods. These facies are categorized using electrical diagrams readings of electricity. The total sedimentary units with lithological features, sedimentary structure, geometrical form, fossil appendages, longitudinal flow patterns, and similar layering surfaces are distinguishable from the characteristics mentioned above by facies (Selley 1995). Well logs are one of the primary sources of subsurface information in oil/gas fields. These tools exhibit features such as mineralogical composition, texture, sedimentary structures, and reservoir properties (such as porosity and permeability) directly or indirectly (Serra 1984; Serra and Abbott 1982). Nowadays, determining electrical facies in reservoir formations is one of the standard studies in the characterization of reservoirs. The wide use of these facies and their flexibility to determine specific reservoir parameters due to the type of input data has today made this method one of the most potent tools in reservoir studies. Facies data are used in cases such as separating reservoir zones from non-reservoir and in field-scale and large-scale structural adaptation. This data's importance is called the virtual core (Ye and Rabiller 2005)

In this study, facies with common geological/reservoir properties were classified into different categories using readings of gamma, neutron, density, acoustic, and resistivity diagrams. Figure 7 and Fig. 8 show the frequency diagrams of the input logs of the model and cross diagrams of these logs, respectively.

3.2. MRGC method

Data clustering is the basis of modeling and classification algorithms. This process aims to define small natural and fundamental groups of large datasets (Kadkhodaie-Ilkhchi et al. 2013). The MRGC (Multi-resolution graph-based clustering) method is one of the few non-parametric and very suitable methods for studying and analyzing data clusters from the logs with the characteristics mentioned. In this method, the log data is indexed by two NI (Neighboring Index Parameter) that determine the position of the proximity of each point in a dataset to the summit or probability of the density function of the data and the KRI (Kernel Representative Index) is an indicator. To determine the prone points to represent the core or center of the cluster.

In this study, based on the MRGC clustering method, the upper and lower limit of optimal data and models were determined. Finally, applying this model produced an optimal model with eight facies. The results of categorized facies are presented in Fig. 9.

Figure 9 shows the readings of each model input log in the separated facies with their weight. Figure 10 shows the facies classification throughout the sequence of the Asmari Formation in the studied wells in the Mansouri Field. Furthermore, Fig. 10 compares the lithofacies produced by the lithological column and the saturated and hydrocarbon columns. Comparing lithological columns in sandstone and carbonate lithologies and facies columns is clearly illustrated.

3.3. ANN method

An artificial neural network is a computerized model that attempts to emulate biological learning processes and simulate specific tasks of the human nervous system. Neurons are composed of neurons as microprocessors, each of which has a simple task. Based on the ANN clustering method, the present study assumed eight optimal facies in the previous stage to estimate an entire facies in the well by constructing an ANN model between petrophysical logs and the pre-facies logs. In constructing the neural network model, the Levenberg-Marquardt algorithm trained the data. This network has three layers (input, hidden, and output). The number of neurons was calculated through trial and error and response optimization. The facies classification throughout the sequence of the Asmari Formation is shown in Fig. 11.

Based on a petrophysical study of 280 core samples from one of the wells drilled in the Asmari reservoir located in the Mansouri field, the following results are outlined:

Determination of the production zones in the Asmari Formation Sequence was determined using the method. Accordingly, shear limits were considered based on the combination of different lithologies in designated zones for three parameters of effective porosity, effective water saturation, and shale volume. After applying these shear zones, the potential zones with high oil accumulation were identified.
Regarding estimating electrical facies by MRGC and ANN methods, despite the differences in concepts and methods, these techniques are all exclusively powerful tools for predicting electrical facies.
In addition, the results show that intelligent systems can successfully model patterns in test data when there is a logical relationship between input data or between input and output data. The model can be used for Recognition and making reasonable conclusions between actual and estimated data. Suppose the parameters necessary for making these models are adjusted, regardless of the lithology and formation characteristics in the relevant oilfield. In that case, these methods can also be applied to other field wells that lack specific facies.
On the other hand, by comparing the performance of the parameters measured and estimated from different methods, it can be concluded that the MRGC method is more accurate and successful than other methods for determining these parameters in the wells.
In conclusion, it is worth noting that MRGC is an efficient method for clustering and homogenizing data and determining different reservoir parameters, which, in addition to the high speed of operation, is not limited to data size and numbers.

ACE

Alternating conditional expectations

ANN

Artificial Neural Network

Electrofacies

FZI

Flow zone index

HCA

Hierarchical cluster analysis

KPCA

Kernel principal component analysis

KRI

Kernel Representative Index

Lithofacies

MRGC

Multi resolution graph based clustering

SOM

Self-organized neural network

Acknowledgment

The authors consider it necessary to express their sincere gratitude to the esteemed experts of the RIPI and Exploration Directorates of the National Iranian Oil Company (NIOC). The authors would like to sincerely thank Dr. Nasser Kwshavarz Faraj Khah for constructive comments to improve the article's scientific level

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Abrar B (2011) Integration of Log Data Analysis and Facies Core to Define Electrofacies Using Multi Resolution Graph-based Clustering. In: 73rd EAGE Conference and Exhibition incorporating SPE EUROPEC 2011. European Association of Geoscientists & Engineers, pp cp-238-00580. https://doi.org/00510.03997/02214-04609.20149524
Alameedy US, Almomen AT, Abd N (2023) Evaluating Machine Learning Techniques for Carbonate Formation Permeability Prediction Using Well Log Data. The Iraqi Geological Journal:175-187
Amraei H, Falahat R (2021) Improved ST-FZI method for permeability estimation to include the impact of porosity type and lithology. Journal of Petroleum Exploration and Production 11(1):109-115. https://doi.org/110.1007/s13202-13020-01061-13206
Antelo R, Aguirre O (2001) Permeability Calculations from Clustering Electrofacies Technique for the Petrophysical Evaluation in La Peña and Tundy Oil Fields. In: SPE Latin American and Caribbean Petroleum Engineering Conference. pp https://doi.org/10.2118/69400-ms .
Bhattacharya S, Carr TR, Pal M (2016) Comparison of supervised and unsupervised approaches for mudstone lithofacies classification: Case studies from the Bakken and Mahantango-Marcellus Shale, USA. Journal of Natural Gas Science and Engineering 33:1119-1133. https://doi.org/1110.1016/j.jngse.2016.1104.1055
Davis JC (2018) Electrofacies in Reservoir Characterization. In: Daya Sagar BS, Cheng Q, Agterberg F (eds) Handbook of Mathematical Geosciences: Fifty Years of IAMG. Springer International Publishing, Cham, pp 211-223. https://doi.org/210.1007/1978-1003-1319-78999-78996_78911
Ding J-w, Ma H-y, Yang Q-s, Lu Y, Yin S-j (2022) Study on Logging Identification Method of Complex Lithology in X Oilfield. In: Lin Je (ed) Proceedings of the International Field Exploration and Development Conference 2021. Springer Nature Singapore, Singapore, pp 38-45
El Sharawy MS, Gaafar GR (2016) Reservoir zonation based on statistical analyses: A case study of the Nubian sandstone, Gulf of Suez, Egypt. Journal of African Earth Sciences 124:199-210. https://doi.org/110.1016/j.jafrearsci.2016.1009.1021
Esfandyari M, Mohseni H, Heidari M (2023) Facies analysis, depositional sequences and platform evolution of the Sarvak Formation (late Albian-Turonian) in the Zagros Basin, West of Iran. Journal of African Earth Sciences 198:104811. https://doi.org/104810.101016/j.jafrearsci.102022.104811
Farazani N, Khoshbakht F, Tavakoli V (2015) Improving Electrofacies Analysis by Integrating Wireline and Image Logs in Asmari Formation Using MRGC. Journal of Petroleum Research 25(83):186-194. https://doi.org/110.22078/pr.22015.22543
Frew K (2004) Litho ToolKit: Lithofacies estimation with the most comprehensive suite of lithofacies tools. In: Schlumberger Information. Slumberger Geology Office, p.
Gualda GAR, Vlach SRF (2007) The Serra da Graciosa A-type Granites and Syenites, southern Brazil: Part 2: Petrographic and mineralogical evolution of the alkaline and aluminous associations. Lithos 93(3):310-327. https://doi.org/310.1016/j.lithos.2006.1006.1002
Heydari M, Maddahi I, Moradpour M, Esmaeilpour S, Keshavarz Faraj Khah N, Naderi A (2012) Application of core study, seismic inversion and petrophysical formation evaluation in seismic facies analysis in Dezful Embayment-Iran. In: International Geophysical Conference and Oil & Gas Exhibition, Istanbul, Turkey, 17-19 September 2012. pp 1-4. https://doi.org/10.1190/ist092012-092001.092045
Hosseini SA, Keshavarz Faraj Khah N, Kianoush P, Afzal P, Ebrahimabadi A, Shirinabadi R (2023) Integration of Fractal modeling and Correspondence Analysis Reconnaissance for Geochemically High-Potential Promising Areas, NE Iran. Results in Geochemistry 11:100026. https://doi.org/100010.101016/j.ringeo.102023.100026
Ibrahim WS, Watanabe K, Yonezu K (2016) Structural and litho-tectonic controls on Neoproterozoic base metal sulfide and gold mineralization in North Hamisana shear zone, South Eastern Desert, Egypt: The integrated field, structural, Landsat 7 ETM+ and ASTER data approach. Ore Geology Reviews 79:62-77. https://doi.org/10.1016/j.oregeorev.2016.1005.1012
Ismail MJ, Ettensohn FR, Handhal AM, Al-Abadi A (2021) Facies analysis of the Middle Cretaceous Mishrif Formation in southern Iraq borehole image logs and core thin-sections as a tool. Marine and Petroleum Geology 133:105324. https://doi.org/105310.101016/j.marpetgeo.102021.105324
Jafarzadeh N, Kadkhodaie A, Ahmad BJ, Kadkhodaie R, Karimi M (2019) Identification of electrical and petrophysical rock types based on core and well logs: Utilizing the results to delineate prolific zones in deep water sandy packages from the Shah Deniz gas field in the south Caspian Sea basin. Journal of Natural Gas Science and Engineering 69:102923. https://doi.org/102910.101016/j.jngse.102019.102923
Jouini S, Umbhauer F, Leduc J-P, Keskes N (2008) Petrophysical Properties Prediction Using 3D Core Scanner Imagery. In: SPE Annual Technical Conference and Exhibition. pp https://doi.org/10.2118/116393-ms
Kadkhodaie-Ilkhchi R, Rezaee R, Moussavi-Harami R, Kadkhodaie -Ilkhchi A (2013) Analysis of the reservoir electrofacies in the framework of hydraulic flow units in the Whicher Range Field, Perth Basin, Western Australia. Journal of Petroleum Science and Engineering 111:106-120. https://doi.org/110.1016/j.petrol.2013.1010.1014
Kadkhodaie A, Kadkhodaie R (2018) A Review of Reservoir Rock Typing Methods in Carbonate Reservoirs: Relation between Geological, Seismic, and Reservoir Rock Types. Iranian Journal of Oil and Gas Science and Technology 7(4):13-35. https://doi.org/10.22050/ijogst.22019.136243.131461
Kadkhodaie A, Rezaee R, Moallemi SA (2006) A fuzzy logic approach for estimation of permeability and rock type from conventional well log data: An example from the Kangan reservoir in the Iran Offshore Gas Field. Journal of Geophysics and Engineering 3:https://doi.org/10.1088/1742-2132/1083/1084/1007
Kelkar M (2005) Exploitation and optimization of reservoir performance in Hunton formation, Oklahoma. U.S. Department of Energy, Assistant Secretary for Fossil Energy:190. https://doi.org/110.2172/815450
Kiaei H, Sharghi Y, Ilkhchi AK, Naderi M (2015) 3D modeling of reservoir electrofacies using integration clustering and geostatistic method in central field of Persian Gulf. Journal of Petroleum Science and Engineering 135:152-160. https://doi.org/110.1016/j.petrol.2015.1008.1019
Kianoush P, Mohammadi G, Hosseini SA, Keshavarz Faraj Khah N, Afzal P (2022) Compressional and Shear Interval Velocity Modeling to Determine Formation Pressures in an Oilfield of SW Iran. Journal of Mining and Environment 13(3):851-873. https://doi.org/810.22044/jme.22022.12048.22201
Kianoush P, Mohammadi G, Hosseini SA, Keshavarz Faraj Khah N, Afzal P (2023a) Inversion of seismic data to modeling the Interval Velocity in an Oilfield of SW Iran. Results in Geophysical Sciences 13:100051. https://doi.org/100010.101016/j.ringps.102023.100051
Kianoush P, Mohammadi G, Hosseini SA, Keshavarz Faraj Khah N, Afzal P (2023b) Determining the drilling mud window by integration of geostatistics, intelligent, and conditional programming models in an oilfield of SW Iran. Journal of Petroleum Exploration and Production Technology 13(6):1391-1418. https://doi.org/1310.1007/s13202-13023-01613-13206
Kianoush P, Mohammadi G, Hosseini SA, Keshavarz Faraj Khah N, Afzal P (2023c) ANN-based estimation of pore pressure of hydrocarbon reservoirs—a case study. Arabian Journal of Geosciences 16(5):302. https://doi.org/310.1007/s12517-12023-11373-12516
Kolbikova E, Gusev S, Malinovskaya O, Garaev A, Valiev R (2021) Lithofacies analysis of Devonian carbonate deposits based on geological and geophysical information analysis by using machine learning methods. European Association of Geoscientists & Engineers 2021(1):1-6. https://doi.org/10.3997/2214-4609.202159047
Lai J, Pang X, Zhao X, Zhao Y, Wang G, Huang Y, Li H, Li Y (2023) Typical misinterpretations and scientific concepts in well-logging geologic studies. Natural Gas Industry B 10(2):198-211. https://doi.org/110.1016/j.ngib.2023.1002.1003
Luthi S (2001) Geological Well Logs: Their Use in Reservoir Modeling. In: Springer Berlin, Heidelberg, pp XII, 373. https://doi.org/310.1007/1978-1003-1662-04627-04622
Madani N, Biranvand B, Naderi A, Keshavarz Faraj Khah N (2019) Lithofacies uncertainty modeling in a siliciclastic reservoir setting by incorporating geological contacts and seismic information. Journal of Petroleum Exploration and Production Technology 9(1):1-16. https://doi.org/10.1007/s13202-13018-10531-13207
Mahadasu P, Singh KH (2022) 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. Arabian Journal for Science and Engineering:https://doi.org/10.1007/s13369-13022-07521-x .
Manshad AK, Pashaki RS, Ali JA, Iglauer S, Memariani M, Akbari M, Keshavarz A (2021) Geochemical study of the early cretaceous Fahliyan oil reservoir in the northwest Persian Gulf. Journal of Petroleum Exploration and Production Technology 11(6):2435-2447. https://doi.org/2410.1007/s13202-13021-01178-13202
Masroor M, Emami Niri M, Sharifinasab MH (2023) A multiple-input deep residual convolutional neural network for reservoir permeability prediction. Geoenergy Science and Engineering 222:211420. https://doi.org/211410.211016/j.geoen.212023.211420
Mathis B, Leduc JP, Vandenabeele T (2003) From the Geologists’ Eyes to Synthetic Core Descriptions: Geological Log Modeling Using Well-Log Data. In: AAPG Annual Meeting. Salt Lake City, Utah
Mirkamali MS, Keshavarz Faraj Khah N, Bakhtiari MR (2016) Evolution analysis of miocene channels and faults in offshore area of Strait of Hormuz (Eastern part of Persian Gulf) using seismic meta-attributes. Journal of Petroleum Science and Engineering 147:116-128. https://doi.org/110.1016/j.petrol.2016.1005.1012
mohammadinia F, Ranjbar A, Kafi M, Keshavarz R (2023) Application of machine learning algorithms in classification the flow units of the Kazhdumi reservoir in one of the oil fields in southwest of Iran. Journal of Petroleum Exploration and Production Technology 13(6):1419-1434. https://doi.org/1410.1007/s13202-13023-01618-13201
Okhovvat HR, Riahi MA, Akbari Dehkharghani A (2023) Kernel principal component analysis (KPCA) in electrical facies classification. Iranian Journal of Oil and Gas Science and Technology:https://doi.org/10.22050/ijogst.22023.360469.361653
Rabbani E, Davarpanah A, Memariani M (2018) An experimental study of acidizing operation performances on the wellbore productivity index enhancement. Journal of Petroleum Exploration and Production Technology 8(4):1243-1253. https://doi.org/1210.1007/s13202-13018-10441-13208
Rafik B, Kamel B (2017) Prediction of permeability and porosity from well log data using the nonparametric regression with multivariate analysis and neural network, Hassi R’Mel Field, Algeria. Egyptian Journal of Petroleum 26(3):763-778. https://doi.org/710.1016/j.ejpe.2016.1010.1013
Rastegarnia M, Sanati A, Javani D (2016) A comparative study of 3D FZI and electrofacies modeling using seismic attribute analysis and neural network technique: A case study of Cheshmeh-Khosh Oil field in Iran. Petroleum 2(3):225-235. https://doi.org/210.1016/j.petlm.2016.1006.1005
Roslin A, Esterle JS (2016) Electrofacies analysis for coal lithotype profiling based on high-resolution wireline log data. Computers & Geosciences 91:1-10. https://doi.org/10.1016/j.cageo.2016.1003.1006
Salehi MA, Kazem Shiroudi S, Moussavi Harami SR, Ghafouri M, Lashkari Pour G (2015) Integrated Petrophysical Rock Typing Approaches for the Upper Surmeh Formation in an Oil Field in the Central Persian Gulf. Journal of Petroleum Research 25(84):72-87. https://doi.org/10.22078/pr.22015.22550
Selley RC (1995) Ancient Sedimentary Environments: And Their Sub-surface Diagnosis. Routledge, London, p 318.
Serra O (1984) Fundamentals of well-log interpretation / O. Serra ; translated from the French by Peter Westaway and Haydn Abbott. Elsevier ; Elf Aquitaine, Amsterdam ; New York : Pau
Serra O, Abbott HT (1982) The Contribution of Logging Data to Sedimentology and Stratigraphy. Society of Petroleum Engineers Journal 22(01):117-131
Shakiba S, Asghari O, Keshavarz Faraj Khah N, Sarallah Zabihi S, Tokhmechi B (2015) Fault and non-fault areas detection based on seismic data through min/max autocorrelation factors and fuzzy classification. Journal of Natural Gas Science and Engineering 26:51-60. https://doi.org/10.1016/j.jngse.2015.1005.1024
Shirneshan G, Bakhtiari A, Memariani M (2018) Identification distribution pattern and origin of petroleum hydrocarbons in core sediment of southwest coast of the Caspian Sea (Kiashahr). Journal of Environmental Studies 43(4):743-755
Shoghi J, Bahramizadeh-Sajjadi H, Nickandish AB, Abbasi M (2020) Facies modeling of synchronous successions - A case study from the mid-cretaceous of NW Zagros, Iran. Journal of African Earth Sciences 162:103696. https://doi.org/103610.101016/j.jafrearsci.102019.103696
Tavakkoli V, Amini A (2006) Application of Multivariate Cluster Analysis in Logfacies Determination and Reservoir Zonation, Case Study of Marun Field, South of Iran. Journal of Science University of Teheran 32(2):69-75
Tian Y, Xu H, Zhang X-Y, Wang H-J, Guo T-C, Zhang L-J, Gong X-L (2016) Multi-resolution graph-based clustering analysis for lithofacies identification from well log data: Case study of intraplatform bank gas fields, Amu Darya Basin. Applied Geophysics 13(4):598-607. https://doi.org/510.1007/s11770-11016-10588-11773
Wolf M, Pelissier-Combescure J (1982) Faciolog - Automatic Electrofacies Determination. In: SPWLA 23rd Annual Logging Symposium.
Wu H, Wang C, Feng Z, Yuan Y, Wang H-F, Xu B-S (2020) Adaptive multi-resolution graph-based clustering algorithm for electrofacies analysis. Applied Geophysics 17(1):13-25. https://doi.org/10.1007/s11770-11020-10806-x
Ye S-J, Rabiller P (2005) Automated Electrofacies Ordering. Petrophysics - The SPWLA Journal of Formation Evaluation and Reservoir Description 46(06):409-423.
Zahmatkesh I, Soleimani B, Ghobeishavi A, Sheikhzadeh H (2015) Prediction of Sedimentary Features in Asmari Formation, SW of Iran: Using Electrofacies Analysis. Advanced Applied Geology 5(2):72-81. https://doi.org/10.22055/aag.22015.11528

Download PDF

Journal Publication

published 01 Mar, 2024

Read the published version in Scientific Reports →

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Determination of electrical facies using MRGC and ANN clustering methods

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Methodology

2.1. Zoning

2.2. Determination of production zone

2.3. Stratigraphy of the studied field

3. Results and Discussion

3.1. Determination of electrical facies

3.2. MRGC method

3.3. ANN method

4. Conclusions

Abbreviations

Declarations

References

Status:

Journal Publication

Version 1