Next Article in Journal
Intelligent Systems Supporting the Use of Energy Devices and Other Complex Technical Objects: Modeling, Testing, and Analysis of Their Reliability in the Operating Process
Next Article in Special Issue
Global Genetic Algorithm for Automating and Optimizing Petroleum Well Deployment in Complex Reservoirs
Previous Article in Journal
An Open Digital Platform to Support Interdisciplinary Energy Research and Practice—Conceptualization
Previous Article in Special Issue
Strike-Slip Fault Effects on Diversity of the Ediacaran Mound-Shoal Distribution in the Central Sichuan Intracratonic Basin, China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Energies
Volume 15
Issue 17
10.3390/en15176415
energies-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

3D Modeling of Fracture-Cave Reservoir from a Strike-Slip Fault-Controlled Carbonate Oilfield in Northwestern China

by
Rujun Wang
1,
Jianping Yang
1,
Lunjie Chang
1,
Yintao Zhang
1,
Chong Sun
1,*,
Xiaoguo Wan
1,
Guanghui Wu
2,* and
Bingchen Bai
2
1
PetroChina Tarim Oilfield Company, Korla 841000, China
2
School of Geoscience and Technology, Southwest Petroleum University, Chengdu 610500, China
*
Authors to whom correspondence should be addressed.
Energies 2022, 15(17), 6415; https://doi.org/10.3390/en15176415
Submission received: 17 May 2022 / Revised: 22 July 2022 / Accepted: 6 August 2022 / Published: 2 September 2022
(This article belongs to the Special Issue New Insights into Reservoir 3D Modeling and Simulation)
Browse Figures
Versions Notes

Abstract

:
A giant strike-slip fault-controlled Fuman Oilfield has been found in the Ordovician fractured carbonates of the Tarim Basin. However, conventional seismic methods are hardly able to distinguish the fractured reservoir and its connectivity in the ultra-depth (>7000 m) carbonate fault zones. We propose thin-likelihood and tensor-thickness process methods to describe the fracture network and large cave reservoir, respectively. Together with the two methods for 3D visualization of fracture-cave reservoirs, we had an application in the ultra-deep well deployment in Fuman Oilfield. The results show that the fracture network and cave reservoir can be 3D-imaged more clearly than conventional methods. The fracture network and cave reservoir show distinct segmentation by the fault assemblage in Fuman Oilfield. Furthermore, 3D modeling is favorable for the reservoir connectivity description along the carbonate fault zones. There are three distinct reservoir models: fault core-, fault damage zone- and overlap zone-controlling fractured reservoirs along the fault zones. This revealed variable fractured reservoirs that are related to fault maturity and segmentation. The method has been widely used in fracture-cave reservoir description and subsequent well optimization, suggesting a favorable method for economic oil exploitation in the ultra-depth reservoirs. This case study is not only useful for the complicated reservoir 3D description and modeling but also helpful for well employment to provide support for the target evaluation and optimization in ultra-depth fractured reservoirs.

1. Introduction

With the rapid decrease in shallow hydrocarbon resources, the deep (>4500 m), buried carbonate oil/gas resource has become a major exploitation domain [1,2]. Due to tight matrix reservoirs in the deep carbonates, fractured reservoirs are becoming significant in hydrocarbon exploitation in the ultra-depth (>6000 m) tight carbonates [1,2,3,4,5]. Natural fracture can increase the permeability by more than 1–2 orders of magnitude in the deep, tight carbonate reservoirs [4,5,6,7,8]. In addition, the tight carbonate reservoir porosity can be increased more than several times by the secondary dissolution porosity along the fracture network [8,9]. Hence, fractured reservoirs make a significant contribution to petroleum reserves and economic production in the deep subsurface. Modeling of fractured reservoirs has been studied by geological, geophysical and engineering methods [4,5,6,7,8]. Due to intense heterogeneity of fractured reservoirs, the seismic method is of great importance for the “sweet spot” description and well deployment in the deep subsurface [10,11,12,13,14]. By low-resolution seismic data, however, seismic identification of fractured reservoirs in the deep subsurface is still challenging [10]. In this context, there is still little understanding of the fracture network and its relation to the fractured reservoirs, which in turn constrains hydrocarbon exploitation of the deep fractured reservoirs.
Large amounts of hydrocarbon reserves have been found in the ultra-depth Ordovician limestones in the Tarim Basin, NW China [15]. However, there is generally low economical exploitation benefit from the tight matrix reservoir (low porosity < 4%, low permeability < 1 mD) in the deep subsurface. Recently, the largest strike-slip, fault-controlled oilfield in China, Fuman Oilfield, has been found in the central Tarim Basin [16,17,18]. The “sweet spot” of a fracture-cave reservoir is the major drilling target for high oil production along the strike-slip fault zones in the ultra-depth (>7000 m) depression. Due to the complicated fracture network and intense heterogeneous caves in the deep fault damage zones, conventional seismic methods are difficult to effectively describe the fracture network and its relationship with the cave reservoirs [13,16,19]. Seismic description of the fractured reservoir and subsequent well optimization in the fault zones are major issues for economical exploitation in Fuman Oilfield [13,16]. There are still more than 60% of wells that have low production, which calls for fine 3D modeling of the heterogeneous fracture-cave reservoir along the deep fault zones.
For this study, we proposed an integrated 3D model of fracture and the fracture-cave reservoir in Fuman Oilfield. Furthermore, we analyzed the typical fracture-cave models along the strike-slip fault zones. Finally, we investigated applications to take advantage of the 3D description of the fracture-cave reservoir for high production well optimization.

2. Geological Background

The Tarim Basin covers an area of 56 × 104 km2, has multiple tectonic–sedimentary cycles and deposited a more than 15,000 m thick Cryogenian–Quaternary sedimentary strata [20]. The multi-stage tectonic events have resulted in multiple unconformities and faults, recording the supercontinent assembly and breakup in the late Neoproterozoic, the opening and closure of the Tethys in the Palaeo-Mesozoic and the Indo-Asian collision in the Cenozoic [21].
There is a large Cambrian–Ordovician carbonate platform in the central and western basin, with a more than 30 × 104 km2 area and 3000 m thick carbonate [15,21]. More than 70 large strike-slip fault zones have been identified by 3D seismic data [22]. Connected with the major source rock at the base of the Cambrian and multiple carbonate reservoirs in the Cambrian–Ordovician [16], a strike-slip, fault-controlled petroleum system occurred with an area of 9 × 104 km2 in the central Tarim Basin (Figure 1a) [16,21]. Large amounts of hydrocarbon resources have been found in the Ordovician limestones (Figure 1). The oil and gas geological reserves exceed 1 billion tons of oil equivalent, forming a fault-controlled petroleum system of more than 5 billion tons of resources in the central Tarim Basin [16].
The largest ultra-deep (>7000 m) strike-slip fault-controlled oilfield, Fuman Oilfield, has an oil production of more than 2 × 106 t/y in the central depression (Figure 1) [16,17,18]. The strike-slip faults mainly presented transpressional faults in the Cambrian–Ordovician carbonates but are inherited by transtensional faults upward to the overlying strata [21,22,23,24,25]. The carbonate reservoir is composed of a secondary triple porosity assemblage of the fractures, pores-vug and cave, but the oil production is mainly from the localized large-scale, fracture-cave reservoir along the strike-slip fault zones [16,17,18,25,26,27]. The fracture-cave reservoirs have irregular 3D shapes and complicated connectivity, and in turn complicated oil/gas distribution and production [16,17,18,19]. The oilfield presented complicated and unstable production with ~30% high-production wells that support more than 70% of production. As a result, it is still hard to describe the complicated boundaries of the reservoir along the strike-slip fault zones, which led to complicated oil/water contact and production along the fault zones.

3. Data and Methods

More than 10,000 km2 3-D seismic surveys have been carried out in Fuman Oilfield. The prestack depth migration seismic dataset is favorable for detailed strata and major fault mapping [16]. Constrained by low resolution of the seismic data, it is difficult to identify small fault and fault linkage along a fault zone (Figure 1) [16,19]. First, we carry out fault modeling to show the different fault models of each segment [28], particularly of the complex fault networks. Furthermore, we use a maximum-likelihood attribute to map the small fault. The maximum-likelihood attribute is used to get the most probable location and probability of fault in the study area by scanning the similarity between the fracture points in the whole seismic volume and improving the accuracy of fault identification [29,30]. Compared with the conventional seismic attributes of coherence and curvature, a maximum-likelihood attribute can enhance the seismic imaging effect of small fault and fracture assemblages and help to describe the internal structures in fault damage zones [16,19,30,31]. In this context, we have seismic fault interpretation and modeling of each fault segment.
Generally, the large cave reservoirs showing a “bead−shape” seismic reflection in the seismic section can be identified by seismic methods in the complicated ultra-depth oilfield [15]. The seismic methods of amplitude attribute, reservoir inversion, etc., have been used to describe the 3D distribution of the large cave reservoir [13,16,19,31]. However, the information of the small vug-cave reservoir and connectivity of the fractured reservoir is hard to describe with conventional methods [16,19,27]. In this study, we use a structure tensor attribute for modeling of the fault damage zone and the fracture-cave reservoirs. The method is based on a tensor matrix that uses a seismic data process to identify different structural features or texture units (such as layered textures, messy textures, etc.) in seismic images [31,32]. The structure tensor attribute can ensure that the vertical and lateral resolution and signal-to-noise ratio of the seismic data are not affected during the processing and can restore and strengthen the fault damage zone to the greatest extent. Furthermore, it can also be used to identify the cave reservoirs and the fault damage zone in the Tarim Basin [16].
Utilizing fault segment modeling, the structure tensor attribute is processed to constrain the boundary of the fault damage zone. Constrained by the structure tensor attribute, we use a maximum-likelihood attribute to model the major fracture network in the fault damage zone. Through well calibration, the structure tensor is used to distinguish the cave reservoirs from the surrounding rocks. Finally, we use the 3D modeling module in Geoeast to obtain the 3D models of fractures and cave reservoirs.

4. Results

4.1. Fault Characteristics

In Fuman Oilfield, the main target layers of the Middle Ordovician carbonates are easily traced, interpreted and mapped by the seismic data (Figure 2). The fault with a vertical displacement of more than 20 m could be identified from the seismic sections. On the other hand, the fault styles vary along the fault strike and some show relatively continuous reflection or a kink shape in seismic sections. Through seismic attributes such as coherence and curvature, the major strike-slip fault can be identified, but with ambiguous fault linkage. In this way, maximum-likelihood attributes are used to divide the fault segment. Further, fault throw along the fault strike is much more helpful for distinguishing the fault segment [28]. On this basis, the fault modeling and interpretation present a detailed fault model of each segment (Figure 2). The isolate fault segment presents a fault throw that decreases from the center to zero at its tip. In the soft-linked zone, each fault throw decreases respectively. In the hard-linked zone, each fault segment has increased fault throw to show transpressional horst or transtensional graben.
The strike-slip faults usually penetrate to the basement in steep fault planes and spread upward in a flower structure. Most faults terminate upward at the top of the Middle Ordovician carbonates, but some secondary faults spread upward to the Devonian and even extend to the Permian. Together with seismic attribute analysis, there are different fault models along different segments (Figure 2). As opposed to previous interpretations (Figure 1), the large fault zones also display distinct segmentation rather than a through-going fault zone at the top of the Middle Ordovician carbonates (Figure 2a). The strike-slip fault segments showed oblique assemblage, overlapping zone and fault tip splay to the north. In the hard-linked overlap zones, secondary faults developed and formed faulted horsts. Some small grabens formed in localized trantensional segments. They are mostly of inherited transtensional faults in the Silurian–Devonian, and some developed into Carboniferous–Permian. They show slightly fluctuating morphology by small-fault displacement and weak deformation. The secondary faults are mostly small faults formed along the NW-trending R shear plane, which form positive geomorphic blocks with slight undulations with the major faults. At the strike-slip fault tips, a horsetail structure can be formed in divergent assemblage. The major fault activity is weakened and the width of the fault damage zone is increased. Some fault tips are dominated by oblique fault combination. These suggest that there is distinct fault segmentation, consistent with the varied displacement along the strike-slip fault strike.

4.2. Fracture Network along the Fault Damage Zone

By the maximum-likelihood process, the small fractures can be identified from the seismic section (Figure 3). It improved the accuracy of the fracture image by scanning the whole seismic data volume and calculating the similarity between data samples. In this study, we had artificial participation for the analysis of seismic reflection characteristics of faults, and we carried out fracture imaging strengthening under dip-angle control. In order to characterize the fracture network, a discontinuous attribute body in the reservoir interval can reveal the plane and longitudinal distribution of the fracture network. In this way, the discontinuous attribute volume based on maximum likelihood can clearly reflect the planar distribution characteristics of the fracture network. The integration of this attribute volume with conventional 3D seismic interpretation can show the planar segmentation of the fracture network.
In order to improve the identified fracture accuracy, a thin-likelihood attribute is used to obtain a variance attribute body to show the fracture network (Figure 3b). It shows the distinct feature of the fracture network from the seismic section. Considering many false imagines by the lithology and seismic data, we excluded unreasonable “fractures” by human–computer interaction from the plane coherence attribute. In addition, it showed that the variance attribute body based on the thin-likelihood attribute is better to interpret the fractures. The plane distribution of the fracture network is consistent with the coherence plane attribute and seismic interpretation. Furthermore, a 3D visualization map has been created to show the fracture network (Figure 4). It should be noted that the fracture point should be corrected by the seismic section and plane attribute to obtain the 3D closure of point, line and plane.
The results show that the thin-likelihood attribute has a high precision in detailed description of the fracture network (major fault segment, small fault and fracture set) along the fault damage zone. The fracture length, width and intensity can be visualized in 3D (Figure 4). In the thin-likelihood section, there is clear internal structure and distinct fracture segmentation along the NE-trending fractures. The major fault is composed of multiple small fault branches in overlapping and oblique combination. It also characterizes the intensive fracture network. It is concentrated at the intersection between NE- and NW-trending faults, which is consistent with the joint orientation from the logging data interpretation. In addition, the fracture network is variable to show different internal structures in the fault zone, which cannot be obtained by the coherent attribute.
In the M1 well area, there is a fault overlap zone in the southern segment and a secondary splay in the northern segment. The vertical displacement is more than 100 m in the major fault segment, but lower than 50 m within the secondary fault segments. Multiple overlapped major fault segments present a “braided structure”. The fault splays led to a complicated and interweaved fault assemblage. There is stronger transpressional and weaker transtensional deformation in the narrow overlap zone, which resulted in complicated localized deformation. A secondary fault generally displays a linear fault plane with unclear vertical displacement. The displacement and width of different segments vary along the fault strike to show a complicated fracture network and linkage. The fault segments and linkage can be distinguished in a 3D diagram, although there is some ambiguous linkage in the overlap zone. The complicated fracture network suggests strong linkage and interaction in the overlapping zone. These led to a wider fault zone by the interaction of multiple fault damage zones. In the northern area, there are two soft–hard linked oblique segments. The relatively small segments are an almost linearly oblique assemblage. They show weak deformation by relatively smaller length in plane and vertical displacement (<10 m) in seismic section. The straight fault surface is coupled with a narrow fault damage zone and a lack of obvious deformation to show weak fracture interaction. In this way, the 3D model presents detailed fracture segmentation and linkage in the spatial assemblage (Figure 4).

4.3. Cave Reservoir along the Fault Damage Zone

On the basis of strike-slip fault interpretation, we carried out a variety of seismic attributes and well control inversion for fractured cave reservoir identification and characterization. Well data indicate that fault damage zone and fractured reservoirs are consistent with the stronger structure tensor (tensor-thickness) attribute. Due to intense dissolution along the fault damage zone, the cave reservoir shows varied tensor-thickness value (Figure 5). There is generally low tensor-thickness attribute value—less than 120—in the country rock, but much higher value—more than 200—in the fracture-cave reservoir along the fault damage zone. Generally, there is a distinct cave reservoir boundary from the tensor-thickness attribute. The relative location is consistent with drilling results from well data (Figure 5). In addition, the fracture network has a seismic response on the tensor-thickness section, suggesting the internal structures in the fractured reservoirs.
By well data calibration, the large cave reservoir presents clear boundaries by the tensor-thickness attribute. Through the correlation analysis, there is a distinct porosity logging interpretation and tensor-thickness value for the cave reservoir other than the country rocks. After the elimination of the country rocks, the tensor-thickness attribute can show a 3D image of the large cave reservoirs in the Middle Ordovician carbonates along the fault damage zone (Figure 6). Most linear fracture networks less than the lower limit of the effective cave reservoir also can be eliminated from the dataset and can in turn be used to show the visualized cave reservoir body. In this way, the cave reservoir of more than 5 × 104 m3 has been 3D-visualized from the tensor-thickness attribute data. In the 3D dataset, the cave reservoirs are generally isolated cylindrical bodies in the Middle Ordovician carbonate (Figure 6). It should be noted that there are some more connected caves in the southern segment, and the larger caves have better connectivity.
The integrated 3D model presents the relationship between the fracture network and cave reservoirs in 3D visualization (Figure 7). In this study area, the cave reservoirs developed and showed segmentation along the fault segments. In the northeastern segment, the cave developed in larger scale along the fault cores (Figure 7a). This is consistent with the “bead-shape” seismic reflection along fault cores and higher fluid production in this area. These caves show linear distribution along fault cores within 200 m width rather than the wide-fault damage zones. The narrow-fault damage zone generally constrains the cave development along the fault cores. These suggest that fault core is preferable for cave localization rather than the wide-fault damage zones. The western splay presents scattered cave reservoirs. It should be noted that most caves are isolated in small scale along the fault damage zone. This segment shows a wider fault damage zone than the eastern segment. The dissolution caves are mainly developed along the fault damage zone rather than the major fault cores. In this context, the cave distribution is related to the width of the fault damage zone. However, the caves number is much less than the eastern segment. This suggests that there are more cave reservoirs developed along the fault cores than along the damage zone.
In the southern segment, more cave reservoirs developed and are constrained in the fault overlap zones (Figure 7b). Most reservoirs developed in the fractured overlap zones, but few occur in the fault cores to show a much wider fractured reservoir. The cave assemblage shows a block shape among the braid structure. This suggests that the overlap zone is favorable for cave localization rather than the fault core or damage zones. In addition, there is better connectivity among the caves and the fractured reservoirs, as well as much higher oil production in this segment. In addition, some caves are connected by the fracture network to show a large oil pool with uniform oil–water contact. These are inconsistent with the northern segments, which suggest that an intense deformation and wide superimposed fracture network plays an important role in the reservoir connectivity.
As a consequence, there are segmentation models of the fracture network and cave reservoirs along a strike-slip fault zone in the Tarim Basin.

5. Applications and Discussions

5.1. Applications in Fuman Oilfield

In the strike-slip fault-controlled Fuman Oilfield, wells generally target the fracture-cave reservoirs along the carbonate strike-slip fault zones. The fracture-related “sweet spots” of large cave reservoirs are favorable targets for hydrocarbon exploitation [16,17,18,19].
Based on the 3D fracture-cave modeling and description, the large fracture network and cave reservoir and their connectivity can be visualized for well optimization along the fault zone. In addition, the boundary of the fractured reservoirs could be identified to distinguish the traps along the fault damage zone. The applications showed much higher resolution in the identification of “sweet spot” along the fault zone by this method. Before the application, drilling break and leakage ratios are lower than 60%, and the success ratio of the high production wells is lower than 75%. Through the optimized application of 3D fracture-cave reservoirs description in more than 100 wells along strike-slip fault zone, the drilling penetration rate of the reservoir is increased to more than 95%, and the success rate of drilling is also improved greatly in Fuman Oilfield. According to the calibration in 9 wells (Figure 8), the drilling break ratio in the fracture-cave reservoir is up to 75%, and the drilling fluid loss in the fault damage zone is up to 83.3%. These suggest that the 3D modeling and description of fracture-cave reservoirs are very helpful for well trajectory design in the complicated strike-slip, fault-controlled oilfield.
In the ultra-depth fractured reservoirs, there is generally high initial oil production but rapid decline during oil well production in the Tarim Basin [16,17,18,19]. Highly productive wells are generally the ones targeted on connected cave reservoirs along fault zones. The stable-production wells had better reservoir connectivity by the fracture network. This clearly indicates that oil production is driven by an understanding of the connectivity between the fracture-cave bodies along the strike-slip fault zones. Calibrated by well data, 3D fracture-cave reservoir modeling can be used to evaluate the connectivity between the fractures and caves (Figure 9). For example, there are similar production features in the connected reservoirs between well 3-H2 and 302H, whereas the adjacent well 3-H6 in the unconnected cave showed a different oil production and oil pressure. In this way, 3D reservoir modeling is of significance in enhancing oil recovery from the complicated fractured reservoirs.
In the circum-Aman petroleum system, we had applications on the well optimization in more than 100 wells. The results show that there is a higher reservoir penetration rate and drilling leakage along the strike-slip fault zone. 3D fracture-cave reservoir modeling is also helpful for well trajectory design and drilling monitoring. In this way, it important in enhancing oil recovery from the complicated ultra−depth oilfield for economical exploitation. As a result, there is a more than 20% increase ratio in high and stable oil production wells in Fuman Oilfield. Furthermore, it has been the largest ultra−deep, strike-slip, fault-controlled oilfield, with oil production up to 1500 × 104 bbl/a.

5.2. Cave Reservoir along the Fault Damage Zone

Seismic method has been widely used in modeling of fractured reservoirs in the deep subsurface [10]. Constrained by seismic resolution, 3D seismic description of a fractured reservoir and its connectivity has become a major issue in deep oil/gas exploitation. Seismic attributes such as conventional coherence and curve have been widely used in the strike-slip fault identification in the Tarim basin [16,20]. The major strike-slip fault zones have been identified in the central Tarim Basin by new 3D seismic modeling and interpretation [16,21,22]. Furthermore, the methods of seismic amplitude and inversion have been used to describe the fracture-cave reservoirs [13,16,19,31]. However, these methods are ambiguous in the fracture network (small fault and joint set) and its linkage and connectivity with the cave reservoir. By the thin-likelihood process of the 3D seismic data, small fault segments and the fracture network can be fine-mapped in 3D visualization (Figure 4). This makes it possible for small-fracture identification and provides a favorable tool for the fault architecture description and modeling. Furthermore, the tensor-thickness attribute from the seismic data is very helpful in describing the cave reservoirs from the fault damage zone (Figure 6). With the thin-likelihood and tensor-thickness methods, the detailed fractures, caves and their connectivity can be mapped in 3D visualization (Figure 7). Compared with conventional methods, such as seismic attribute and reservoir inversion, these methods provide better images of the fractured reservoirs and their connectivity in the deep fault damage zone (Figure 7, Figure 8 and Figure 9). Of course, further study of the detailed fracture network and small caves in the deep fractured reservoirs is still needed.
This case presents three distinct fractured reservoirs along the strike-slip fault zone. Generally, the small-fault zone is favorable for fractured-reservoir development that is controlled by the immature fault core. The fault core has controlled the fracture network and in turn the fracture-related dissolution porosity. The width of the fault damage zone is generally correlated with fault displacement by a power–law relationship [33,34]. However, segment 2 has a wider damage zone but a smaller scale than segment 1 (Figure 1 and Figure 2). Particularly, smaller segment 2 has a damage-zone-controlled cave reservoir (Figure 7). According to fault damage zone analysis (Figure 4), segment 2 has an oblique fault assemblage that led to a relatively wide fault damage zone and subsequently formed a wider fracture network for cave development. In the southern segment, the cave reservoir is prone to initiate the overlapped fault blocks. The high oil production wells occurred in the higher faulted blocks. This suggests that the overlap zone has intense fracture networks for cave localization that are inconsistent with the northern segments. In addition, there is better reservoir connectivity in the southern segment to show a larger block-shaped oil pool rather than the small, isolated reservoirs in the northern segments. These suggest that the maturity of the fault damage zone plays an important role in reservoir development and in turn results in fault segmentation and varied fractured reservoirs along a fault zone.

6. Conclusions

In ultra-deep, strike-slip, fault-controlled fractured reservoirs, 3D reservoir modeling provides a powerful tool for the “sweet spot” description of heterogeneous fractured reservoirs and well optimization in the Tarim Basin. The integrated 3D description and modeling of the Ordovician fracture-cave reservoirs presented in this case study is summarized in the following conclusions.
(1) 3D modeling of fracture-cave reservoirs is feasible by thin-likelihood and tensor-thickness processes for optimal imaging of fracture networks and cave reservoirs along the ultra-deep, strike-slip fault zone.
(2) The fracture network and large cave reservoir have distinct segmentation by the fault segments and are subsequently available for reservoir connectivity description along the carbonate fault zones.
(3) 3D modeling presents three distinct reservoir models: fault core-, fault damage zone- and overlap zone-controlling fractured reservoirs along the fault zones, which are related to fault maturity and segmentation.
(4) 3D modeling of fracture-cave reservoirs is very helpful in well optimization in ultra-depth fractured reservoirs.

Author Contributions

Conceptualization, R.W. and J.Y.; methodology, Y.Z. and L.C.; software, C.S. and X.W.; investigation, Y.Z. and L.C.; data curation, L.C. and G.W.; writing, G.W. and C.S.; visualization, C.S., X.W. and B.B.; supervision, R.W. and J.Y.; funding acquisition, R.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Prospective Basic Research Project of CNPC (2021DJ1502) and the National Natural Science Foundation of China (Grant No. 91955204 and 41472103).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

The authors thank the editor and reviewers for their comments regarding manuscript improvement. We also thank Shiyin Li, Yongfeng Zhu, Xinwei Chen, Yawen Zhao and Jingyi Yuan for their helps in data and visualization.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Sun, L.D.; Zou, C.N.; Zhu, R.K.; Zhang, Y.H.; Zhang, S.C.; Zhang, B.M.; Zhu, G.Y.; Gao, Z.Y. Formation, distribution and potential of deep hydrocarbon resources in China. Pet. Explor. Dev. 2013, 40, 687–695. [Google Scholar] [CrossRef]
  2. Pang, X.Q.; Jia, C.Z.; Wang, W.Y. Petroleum geology features and research developments of hydrocarbon accumulation in deep petroliferous basins. Pet. Sci. 2015, 12, 1–53. [Google Scholar] [CrossRef]
  3. Zhu, G.Y.; Alexei, V.M.; Zhang, Z.Y.; Sun, C.H.; Zhou, X.X.; Chen, F.R.; Han, J.F.; Zhu, Y.F. Formation and preservation of a giant petroleum accumulation in superdeep carbonate reservoirs in the southern Halahatang oil field area, Tarim Basin, China. AAPG Bull. 2019, 103, 1703–1743. [Google Scholar] [CrossRef]
  4. Luca, S.; Marco, M.; Stefano, T.; Antonio, P.; Michael, K.; Andrea, B.; Eugenio, C. 3D Discrete Fracture Network (DFN) models of damage zone fluid corridors within a reservoir-scale normal fault in carbonates: Multiscale approach using field data and UAV imagery. Mar. Pet. Geol. 2021, 126, 104902. [Google Scholar]
  5. Yasin, Q.; Ding, Y.; Baklouti, S.; Boateng, C.D.; Du, Q.Z.; Golsanami, N. An integrated fracture parameter prediction and characterization method in deeply-buried carbonate reservoirs based on deep neural network. J. Pet. Sci. Eng. 2022, 208, 109346. [Google Scholar] [CrossRef]
  6. Wu, G.H.; Zhao, K.Z.; Qu, H.Z.; Nicola, S.; Zhang, Y.T.; Han, J.F.; Xu, Y.F. Permeability distribution and scaling in multi-stages carbonate damage zones: Insight from strike-slip fault zones in the Tarim Basin, NW China. Mar. Pet. Geol. 2020, 114, 104208. [Google Scholar] [CrossRef]
  7. Liu, G.P.; Zeng, L.B.; Han, C.Y.; Ostadhassan, M.; Lyu, W.Y.; Wang, Q.Q.; Zhu, J.W.; Hou, F.X. Natural fractures in carbonate basement reservoirs of the Jizhong Sub-Basin, Bohai Bay Basin, China: Key aspects favoring oil production. Energies 2020, 13, 4635. [Google Scholar] [CrossRef]
  8. Moore, P.J.; Fernández-Ibáñez, F. Non-Matrix Quick Pass: A rapid evaluation method for natural fractures and karst features in core. Energies 2022, 15, 4347. [Google Scholar] [CrossRef]
  9. Yang, H.J.; Wu, G.H.; Scarselli, N.; Sun, C.H.; Qing, H.R.; Han, J.F.; Zhang, G.W. Characterization of reservoirs, fluids, and productions from the Ordovician carbonate condensate field in the Tarim Basin, northwestern China. AAPG Bull. 2020, 104, 1567–1592. [Google Scholar] [CrossRef]
  10. Ou, C.H.; Li, C.C.; Huang, S.Y.; Lu, W.T.; Sheng, J.J.; Xiong, H.L. Three-dimensional discrete network modeling of structural fractures based on the geometric restoration of structure surface: Methodology and its application. J. Pet. Sci. Eng. 2018, 161, 417–426. [Google Scholar] [CrossRef]
  11. Ding, Y.; Du, Q.; Yasin, Q.; Zhang, Q.; Liu, L. Fracture prediction based on deep learning: Application to a buried hill carbonate reservoir in the S area. Geophys. Prospect. Pet. 2020, 59, 267–275. [Google Scholar]
  12. Benmadi, M.; Sayantan, G.; Roger, S.; Kurt, M.; Mashhad, F. Practical aspects of upscaling geocellular geological models for reservoir fluid flow simulations: A case study in integrating geology, geophysics, and petroleum engineering multiscale data from the Hunton Group. Energies 2020, 13, 1604. [Google Scholar]
  13. Li, X.; Li, J.; Li, L.; Wan, Z.; Liu, Y.; Ma, P.; Zhang, M. Seismic wave field anomaly identification of ultra-deep heterogeneous fractured-vuggy reservoirs: A case study in Tarim Basin, China. Appl. Sci. 2021, 11, 11802. [Google Scholar] [CrossRef]
  14. Babasafari, A.A.; Chinelatto, G.F.; Vidal, A.C. Fault and fracture study by incorporating borehole image logs and supervised neural network applied to the 3D seismic attributes: A case study of pre-salt carbonate reservoir, Santos Basin, Brazil. Pet. Sci. Techn. 2022, 40, 1492–1511. [Google Scholar] [CrossRef]
  15. Du, J.H. Oil and Gas Exploration of Cambrian-Ordovician Carbonate in Tarim Basin; Petroleum Industry Press: Beijing, China, 2010. (In Chinese) [Google Scholar]
  16. Wang, Q.H.; Yang, H.J.; Wang, R.J.; Li, S.Y.; Deng, X.L.; Li, Y.; Cang, L.J.; Wan, X.G.; Zhang, Y.T. Discovery and exploration technology of fault-controlled large oil and gas fields of ultra-deep formation in strike slip fault zone in Tarim Basin. China Pet. Explor. 2021, 26, 58–71, (In Chinese with English abstract). [Google Scholar]
  17. Yang, X.W.; Wang, R.J.; Deng, X.L.; Li, S.Y.; Zhang, H.; Yao, C. Theoretical exploration and practice of water injection gravity flooding oil in ultra-deep fault-controlled fractured-cavity carbonate reservoirs. Pet. Explor. Dev. 2022, 49, 133–143. [Google Scholar] [CrossRef]
  18. He, X.; Wang, R.; Yang, J.; Li, S.; Yan, C.; Wu, G. Optimization of oil productivity from the ultra−depth strike−slip fault−controlled carbonate reservoirs in northwestern China. Energies 2022, 15, 3472. [Google Scholar] [CrossRef]
  19. Jiang, T.W.; Chang, L.J.; Deng, X.L.; Li, S.Y.; Wu, G.H.; Wan, X.G.; Guan, B.Z. Geological understanding and evaluation technology of fault controlled carbonate reservoir development: A case study of the Tarim Basin. Nat. Gas Ind. 2021, 41, 1–9, (In Chinese with English abstract). [Google Scholar]
  20. Jia, C.Z. Tectonic Characteristics and Petroleum, Tarim Basin, China; Petroleum Industry Press: Beijing, China, 1997. (In Chinese) [Google Scholar]
  21. Wu, G.H.; Pang, X.Q.; Li, Q.M.; Yang, H.J. The Structural Characteristics of Carbonate Rocks and their Effects on Hydrocarbon Exploration in Craton Basin: A Case Study of the Tarim Basin; Science Press: Beijing, China, 2016. (In Chinese) [Google Scholar]
  22. Yang, H.J.; Wu, G.H.; Han, J.F.; Su, Z. Structural analysis of strike-slip faults in the Tarim intracratonic Basin. Chin. J. Geol. 2020, 55, 1–16, (In Chinese with English abstract). [Google Scholar]
  23. Deng, S.; Li, H.L.; Zhang, Z.P.; Zhang, J.B.; Yang, X. Structural characterization of intracratonic strike-slip faults in the central Tarim Basin. AAPG Bull. 2019, 103, 109–137. [Google Scholar] [CrossRef]
  24. Wu, G.H.; Ma, B.S.; Han, J.F.; Guan, B.Z.; Chen, X.; Yang, P.; Xie, Z. Origin and growth mechanisms of strike-slip faults in the central Tarim cratonic basin, NW China. Pet. Explor. Dev. 2021, 48, 595–607. [Google Scholar] [CrossRef]
  25. Ding, Z.W.; Wang, R.J.; Chen, F.F.; Yang, J.P.; Zhu, Z.Q.; Yang, Z.M.; Sun, X.H.; Xian, B.; Li, E.P.; Shi, T.; et al. Origin, hydrocarbon accumulation and oil-gas enrichment of fault-karst carbonate reservoirs: A case study of Ordovician carbonate reservoirs in in South Tahe area of Halahatang oilfield, Tarim Basin. Pet. Explor. Dev. 2020, 47, 306–317. [Google Scholar] [CrossRef]
  26. Lu, X.B.; Wang, Y.; Tian, F.; Li, X.H.; Yang, D.B.; Li, T.; Lv, Y.P.; He, X.M. New insights into the carbonate karstic fault system and reservoir formation in the Southern Tahe area of the Tarim Basin. Mar. Pet. Geol. 2017, 86, 587–605. [Google Scholar] [CrossRef]
  27. Wu, G.H.; Gao, L.H.; Zhang, Y.T.; Ning, C.Z.; Xie, E. Fracture attributes in reservoir-scale carbonate fault damage zones and implications for damage zone width and growth in the deep subsurface. J. Struct. Geol. 2019, 118, 181–193. [Google Scholar] [CrossRef]
  28. Wu, G.J.; Kim, Y.S.; Su, Z.; Yang, P.; Ma, D.B.; Zheng, D.M. Segment interaction and linkage evolution inconjugate strike-slip fault system from the Tarim Basin, NW China. Mar. Pet. Geol. 2020, 112, 104054. [Google Scholar] [CrossRef]
  29. Hale, D. Fault Surfaces and Fault Throws from 3D Seismic Images; SEG Technical Program Expanded Abstracts; Society of Exploration Geophysicists (SEG): Houston, TX, USA, 2012; pp. 1–6. [Google Scholar]
  30. Ma, D.B.; Zhao, Y.M.; Zhang, Y.T.; Yang, P.F.; Yang, M.; Li, L. Application of maximum likelihood attribute to fault identification: A case study of Rewapu block in Halahatang area, Tarim Basin, NW China. Nat. Gas Geosci. 2018, 29, 817–825, (In Chinese with English abstract). [Google Scholar]
  31. Li, Z.; Yang, Z.; Li, H.; Liu, J.; Gong, W.; Ma, X.; Yang, L.; Huang, C. Three-dimensional seismic exploration method for ultra-deep fault-related dissolution reservoirs in the Shunbei desert area. Geophys. Prospec. Pet. 2020, 59, 283–294, (In Chinese with English abstract). [Google Scholar]
  32. Wu, X.M. Directional structure-tensor-based coherence to detect seismic faults and channels. Geophysics 2017, 82, 13–17. [Google Scholar] [CrossRef] [Green Version]
  33. Torabi, A.; Berg, S.S. Scaling of fault attributes: A review. Mar. Pet. Geol. 2011, 28, 1444–1460. [Google Scholar] [CrossRef]
  34. Ma, D.B.; Wu, G.H.; Scarselli, N.; Luo, X.S.; Han, J.F.; Chen, Z.Y. Seismic damage zone and width–throw scaling along the strike-slip faults in the Ordovician carbonates in the Tarim Basin. Pet. Sci. 2019, 16, 752–757. [Google Scholar] [CrossRef] [Green Version]
Energies 15 06415 g001 550
Figure 1. (a) The strike-slip fault system in the Ordovician carbonates (the figure at the top left corner showing the location of Figure 1a in the Tarim Basin) and (b) the Cambrian–Ordovician stratigraphic column in the Tarim Basin (The strike–slip fault zones are mapped by 3D seismic data, the color base map showing the topography of the top of the Ordovician carbonate; the oil/gas distribution and tectonic sketch appear after the reference [16]).
Figure 1. (a) The strike-slip fault system in the Ordovician carbonates (the figure at the top left corner showing the location of Figure 1a in the Tarim Basin) and (b) the Cambrian–Ordovician stratigraphic column in the Tarim Basin (The strike–slip fault zones are mapped by 3D seismic data, the color base map showing the topography of the top of the Ordovician carbonate; the oil/gas distribution and tectonic sketch appear after the reference [16]).
Energies 15 06415 g001
Energies 15 06415 g002 550
Figure 2. (a) The fault network and structure models at the top Ordovician carbonate and (b) typical seismic sections in well M1 area of the Fuman Oilfield (A-A′ showing fault location in Figure 1a; a-a′, bb′, cc′, dd′ showing the locations of the seismic sections; Tc: base of Carboniferous; To3: base of Upper Ordovician).
Figure 2. (a) The fault network and structure models at the top Ordovician carbonate and (b) typical seismic sections in well M1 area of the Fuman Oilfield (A-A′ showing fault location in Figure 1a; a-a′, bb′, cc′, dd′ showing the locations of the seismic sections; Tc: base of Carboniferous; To3: base of Upper Ordovician).
Energies 15 06415 g002
Energies 15 06415 g003 550
Figure 3. Typical seismic section (a,b) and plane attribute (c) processed by maximum likelihood in the Fuman Oilfield.
Figure 3. Typical seismic section (a,b) and plane attribute (c) processed by maximum likelihood in the Fuman Oilfield.
Energies 15 06415 g003
Energies 15 06415 g004 550
Figure 4. 3D fracture network at top of the Middle Ordovician in the well M1 area (A-A’ showing location in Figure 1).
Figure 4. 3D fracture network at top of the Middle Ordovician in the well M1 area (A-A’ showing location in Figure 1).
Energies 15 06415 g004
Energies 15 06415 g005 550
Figure 5. Fracture-cave reservoir seismic section by tensor-thickness process in the Fuman Oilfield (The orange-red blocks are cave reservoirs, and the green envelope suggests the fracture-cave reservoir along the fault damage zone; TTA: tensor-thickness attribute value).
Figure 5. Fracture-cave reservoir seismic section by tensor-thickness process in the Fuman Oilfield (The orange-red blocks are cave reservoirs, and the green envelope suggests the fracture-cave reservoir along the fault damage zone; TTA: tensor-thickness attribute value).
Energies 15 06415 g005
Energies 15 06415 g006 550
Figure 6. 3D cave reservoirs from the tensor-thickness attribute in the well M1 area.
Figure 6. 3D cave reservoirs from the tensor-thickness attribute in the well M1 area.
Energies 15 06415 g006
Energies 15 06415 g007 550
Figure 7. 3D fracture network overlapped with cave reservoirs of the northern segments (a) and southern segment (b) in well M1 area (the green blocks showing fracture network and the red blocks showing cave reservoirs).
Figure 7. 3D fracture network overlapped with cave reservoirs of the northern segments (a) and southern segment (b) in well M1 area (the green blocks showing fracture network and the red blocks showing cave reservoirs).
Energies 15 06415 g007
Energies 15 06415 g008 550
Figure 8. 3D fracture-cave reservoir model showing the well trajectory in well M1 area (the green blocks showing fracture network, the red blocks showing cave reservoirs).
Figure 8. 3D fracture-cave reservoir model showing the well trajectory in well M1 area (the green blocks showing fracture network, the red blocks showing cave reservoirs).
Energies 15 06415 g008
Energies 15 06415 g009 550
Figure 9. 3D fracture-cave reservoir model showing the reservoir connectivity in well M1 area (the green blocks showing fracture network, the red blocks showing cave reservoirs).
Figure 9. 3D fracture-cave reservoir model showing the reservoir connectivity in well M1 area (the green blocks showing fracture network, the red blocks showing cave reservoirs).
Energies 15 06415 g009
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Wang, R.; Yang, J.; Chang, L.; Zhang, Y.; Sun, C.; Wan, X.; Wu, G.; Bai, B. 3D Modeling of Fracture-Cave Reservoir from a Strike-Slip Fault-Controlled Carbonate Oilfield in Northwestern China. Energies 2022, 15, 6415. https://doi.org/10.3390/en15176415

AMA Style

Wang R, Yang J, Chang L, Zhang Y, Sun C, Wan X, Wu G, Bai B. 3D Modeling of Fracture-Cave Reservoir from a Strike-Slip Fault-Controlled Carbonate Oilfield in Northwestern China. Energies. 2022; 15(17):6415. https://doi.org/10.3390/en15176415

Chicago/Turabian Style

Wang, Rujun, Jianping Yang, Lunjie Chang, Yintao Zhang, Chong Sun, Xiaoguo Wan, Guanghui Wu, and Bingchen Bai. 2022. "3D Modeling of Fracture-Cave Reservoir from a Strike-Slip Fault-Controlled Carbonate Oilfield in Northwestern China" Energies 15, no. 17: 6415. https://doi.org/10.3390/en15176415

APA Style

Wang, R., Yang, J., Chang, L., Zhang, Y., Sun, C., Wan, X., Wu, G., & Bai, B. (2022). 3D Modeling of Fracture-Cave Reservoir from a Strike-Slip Fault-Controlled Carbonate Oilfield in Northwestern China. Energies, 15(17), 6415. https://doi.org/10.3390/en15176415

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop