A modeling method for identifying natural fractures by using hierarchical classification quantitative discriminators

Abstract

The application discloses a modeling method for identifying natural fractures by using hierarchical classification quantitative discriminators, and comprises the following steps: step S1, constructing a wellbore natural fracture hierarchical classification quantitative discriminator by relying on static and dynamic and macroscopic and microscopic multi-element information fusion; step S2, constructing a spatial natural fracture hierarchical classification quantitative discriminator by relying on multi-level nested time window optimization of seismic attributes; and step S3, using the above identification results, adopting a natural fracture hierarchical classification three-dimensional modeling method to accurately, reliably and quickly construct a natural fracture hierarchical classification three-dimensional network model. The method can not only accurately and finely depict the horizontal and longitudinal distribution characteristics and laws of natural fractures, but also can solve the problems of accurately and reliably identifying induced fractures near faults and natural fractures in stable areas far from fault structures, and provides reliable geological models and technical support for scientifically and efficiently developing conventional fractured reservoirs, dense sandstone, dense carbonate rock and shale oil and gas and other unconventional oil and gas reservoirs.

Description

Technical Field

[0001] This invention relates to the field of geological technology, and in particular to a modeling method for identifying natural fractures using a hierarchical classification and quantitative identifier. Background Technology

[0002] According to the International Energy Agency's (IEA) April Oil Market Report (OMR), oil demand is projected to increase by 2 million barrels per day in 2023, reaching a record 101.9 million barrels per day, driven by the end of the pandemic and the gradual recovery of the global economy. Conventional fractured oil reservoirs account for over 70% of the world's newly added oil reserves, representing about half of the world's total reserves and production. In the Zagros Mountains piedmont region of the Persian Gulf Basin, more than 20 of the 50+ oil and gas fields discovered are fractured reservoirs, including six mega-fields with reserves exceeding 1 billion tons. The Ain Zare oil field in Iraq also exhibits well-developed fractures. In my country, fractured oil and gas reservoirs have been discovered in the Bohai Bay Basin, Songliao Basin, Sichuan Basin, and the Qaidam and Jiuquan Basins in western China, and some have already achieved industrial production capacity. For unconventional oil and gas reservoirs such as tight sandstone, tight carbonate rocks, and shale oil and gas, fracturing is necessary to increase production. However, the implementation and production effect of fracturing are greatly affected by natural fractures. The complex distribution and unclear mechanism of natural fractures may lead to artificial fractures only extending along the direction of natural fractures, making it difficult to form an effective fracture network, and may also lead to engineering phenomena such as channeling and wellbore deformation. Therefore, accurately and reliably characterizing the distribution of natural fractures in three-dimensional space and establishing a three-dimensional quantitative network model is a key step in achieving efficient reservoir development.

[0003] Natural fractures are complex in distribution, controlled by primary geological factors and influenced by multiple phases of tectonic stress, leading to the development of induced fractures near faults and natural fractures far from fault-stable tectonic zones. For different oil and gas reservoirs and research data, the detailed three-dimensional characterization of natural fractures faces the following challenges: ① Limited data constraining fractures in single wells; the limited number of core samples makes it difficult to coring the entire well section and work area; imaging logging is costly, making it virtually impossible to measure every well, and is easily affected by uncontrollable factors such as wellbore wall, mud, electrode contact with the formation, and the experience of interpreters; while artificial intelligence provides a way to identify fracture development in single wells, the reliability of this method depends on the reliability and richness of the sample. ② Complex and variable geological processes and diverse controlling factors of natural fracture development lead to complex distributions, making it difficult to characterize the spatial distribution patterns of natural fractures using a single method. ③ The development degree of fractures varies in different directions, and the underground opening and connectivity of fracture groups also differ, leading to significant differences in permeability across fracture directions. This exacerbates reservoir heterogeneity and easily results in short water breakthrough times, rapid decreases in injection pressure, rapid declines in initial production, and large differences in production between adjacent wells. ④ Some natural fractures are directly induced by faults, and fracture information within a certain distance near the fault can be extracted for modeling. However, in structurally stable areas far from faults, local stress concentration can also lead to fracture development, making this method unsuitable. ⑤ Using seismic prediction to establish a three-dimensional model of natural fractures presents challenges due to varying seismic data resolution, and the reliability of the seismic prediction model has not been verified by core and well logging data. Therefore, considering the actual field conditions, how to fully utilize existing data, ensuring both the richness of geologically constrained data and high-resolution seismic data for inter-well fracture prediction, while also fully integrating existing information for mutual verification, to achieve a detailed characterization of the spatial distribution characteristics of natural fractures in fault-developed and structurally stable areas, and to form a complete and unified technical process, remains a technical challenge to be solved.

[0004] The authorized invention patent "A Method for Modeling Natural Fractures in Tight Sandstone Reservoirs" (application date: August 9, 2019; inventors: Li Hui, Lin Chengyan, Ren Lihua, Li Shitao, Chen Yanyan; patent number: CN201910732445.1) proposes a method for modeling natural fractures in tight sandstone reservoirs; the authorized invention patent "A Method for Modeling Multi-Scale Fractures" (application date: May 18, 2012; inventors: Zhang Jinliang, Tang Mingming, Ren Weiwei; patent number: CN201210154441.8) proposes a method for modeling multi-scale fractures. However, both of these methods require rock mechanics research, mechanical parameters are difficult to obtain, stress field models need repeated adjustments, and high computer performance is required.

[0005] The authorized invention patent "A Discrete Fracture Modeling Method Based on Multi-Point Geostatistics" (application date: March 27, 2019, inventors: Liu Yanfeng, Zhang Wenbiao, Lian Peiqing, Shang Xiaofei, Wang Mingchuan, Zhao Lei, Zhao Huawei, Li Meng; patent number CN201910238530.2) proposes a discrete fracture modeling method based on multi-point geostatistics to characterize the spatial configuration of fractures at different scales. The authorized invention patent "A Three-Dimensional Modeling Method for Reservoir Fractures" (application date: August 4, 2017, inventors: Zhao Xiangyuan, Hu Xiangyang, Zhang Wenbiao; patent number CN201710665503.4) proposes a three-dimensional modeling method for reservoir fractures. These methods, which use multi-point geostatistics to predict fractures between wells, establish a three-dimensional fracture development model of the formation based on the geological constraints of fractures in multiple single wells. However, this method is too random for blocks with a small number of wells, which may lead to significant differences from the actual geological conditions.

[0006] The authorized invention patent "A Discrete Fracture Modeling Method Based on Multi-Scale Factor Constraints" (application date: January 25, 2015; inventors: Feng Jianwei; Gao Changhai; patent number: CN201510036574.9) proposes a discrete fracture modeling method based on multi-scale factor constraints. This method proposes to use the entropy weight method to optimize the main geological factors controlling fracture development and establish discrete fracture models at various scales. However, this method considers geological factor constraints more, and obviously cannot provide a reasonable characterization of fractures in inter-well and un-well blocks.

[0007] The article "Research on Automatic Identification and Fracture Modeling of Fault System in Puguang Dawan Area" (published June 2013; authors: Wang Lezhi, Liu Honglei, Zhang Jixi, et al.) applied automatic fracture analysis technology and discrete fracture network modeling technology to establish a fracture network model. However, the reliability of this method depends on relatively complete single-well fracture data.

[0008] The authorized invention patent, "Method and Device for Quantitative Prediction of Cracks Based on Post-Stack Seismic Data" (application date: April 17, 2020; inventors: Yu Hao, Fan Xinran, Huang Jiaqiang, Lan Xuemei, Zhang Lianjin, Zhang Xuan, Liu Junying; patent number: CN202010303762.4), proposes a method and device for quantitative prediction of cracks based on post-stack seismic data, realizing quantitative prediction of cracks based on post-stack seismic data. This method improves the accuracy of crack prediction, but suffers from the problem of multiple solutions.

[0009] The authorized invention patent "Prediction Method of Planar Distribution Pattern of Effective Natural Fractures in Oil Reservoirs" (application date: March 21, 2012; inventors: Fan Jianming, Zhao Jiyong, He Yonghong, Li Shuheng, Zeng Lianbo, Wang Jie, Chen Wenlong, Yang Junxia; patent number: CN201210076667.0) provides a method for predicting the planar distribution pattern of effective natural fractures in oil reservoirs. It obtains the planar distribution pattern of effective natural fractures through conventional fracture identification and finite element method, but the method does not form a three-dimensional spatial understanding.

[0010] The authorized invention patent "Reservoir Fracture Modeling Method and System" (application date: May 13, 2016; inventors: Zeng Lianbo; Zhao Xiangyuan; Shi Jinxiong; Wang Jipeng; patent number: CN201610330111.X) proposes a reservoir fracture modeling method and system. However, this method is based on formation fracture characteristics and the distribution characteristics of fracture parameters of each group to obtain the planar density distribution intensity of each group of fractures, establish the fracture models of each group of single sand bodies in a small layer, and finally combine them to form a three-dimensional fracture model of all groups. But this will result in the fracture distribution morphology of each small layer being consistent, and the vertical distribution pattern being unclear.

[0011] The authorized invention patent "Method and Apparatus for Fracture Modeling" (application date: October 20, 2020; inventors: Li Changhai, Zhao Lun, Fan Zifei, Li Jianxin, Wang Shuqin, Zhao Wenqi, Sun Meng, Liu Minghui; patent number: CN202011122331.4) proposes a method and apparatus for fracture modeling. This method effectively characterizes fractures with different dip angles, but it does not consider the significant impact of fracture orientation on reservoir permeability and hydrocarbon accumulation.

[0012] The authorized invention patent "A New Method for Quantitative Identification of Fault-Associated Fractures in Complex Extensional Tectonic Systems" (application date: December 3, 2014; inventors: Ou Chenghua, Chen Wei, Li Chaochun; patent number: CN201410422529.2) provides a new method for quantitative identification of fault-associated fractures in complex extensional tectonic systems. The authorized invention patent "A Multi-Scale Fracture Model and Modeling Method for Tight, Low-Permeability Reservoirs" (application date: October 14, 2016; inventors: Zeng Lianbo, Shi Jinxiong; patent number: CN201610900280.0) proposes a multi-scale fracture model and modeling method for tight, low-permeability reservoirs. These methods have enabled the establishment of induced fracture models near faults, but the reliability of fracture development far from the fault's stable zone needs further verification.

[0013] The authorized invention patent "Method for Identifying Fractures in Sandstone and Conglomerate" (application date: November 27, 2013; inventors: Cao Gang, Wang Wei, Zhuang Xuchao, Zhang Xiaozhen, Lü Shichao, Zhang Huafeng; patent number: CN201310610148.2) proposes a method for identifying fractures in sandstone and conglomerate. The authorized invention patent "A Three-Dimensional Modeling Method for Shale Gas Reservoir Bedding Fractures" (application date: January 15, 2016; inventors: Ou Chenghua, Li Chaochun, Xiong Hongli, Lu Wentao, Zhang Qian, Zhang Mengling, Han Chiyu; patent number: ZL201610028053.3) proposes a three-dimensional modeling method for shale gas reservoir bedding fractures. The authorized invention patent, "A Three-Dimensional Modeling Method for Tectonic Fracturing Based on Geometric Restoration of Tectonic Surfaces" (application date: January 15, 2016; inventors: Ou Chenghua, Li Chaochun, Xiong Hongli, Lu Wentao, Zhang Qian, Zhang Mengling, Han Chiyu; patent number: ZL201610029135.X), proposes a three-dimensional modeling method for tectonic fractures based on geometric restoration of tectonic surfaces. However, all of these methods only address fractures of a single lithology or type, and do not address fractures of other genetic origins.

[0014] Therefore, relying solely on the above-mentioned technical methods is insufficient because it fails to fully explore the static, dynamic, macroscopic, and microscopic multi-dimensional information of natural fractures in various relevant data. This makes it difficult to accurately and reliably identify various natural fractures induced by faults near faults, as well as natural fractures far from the stable zones of fault structures. Furthermore, it is difficult to rely on the established three-dimensional network models of various natural fractures to achieve a fine and accurate depiction of the planar and longitudinal distribution characteristics and patterns of natural fractures. Summary of the Invention

[0015] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier.

[0016] A modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier includes the following steps:

[0017] Step S1: Construct a graded classification and quantitative identification device for natural fractures in wellbore by integrating static, dynamic, macro and micro multi-dimensional information;

[0018] Step S2: Utilize multi-level overlay time-window optimization based on seismic attributes to construct a spatial natural crack classification and quantitative identification device.

[0019] Step S3: Establish a three-dimensional network model of natural cracks using a hierarchical classification modeling method.

[0020] Specifically, step S1 includes the following sub-steps:

[0021] Step S11: Analyze the dynamic evolution relationship between stress field and lithofacies, and establish a classification characteristic model for natural fractures.

[0022] Step S12: Integrate microscopic and macroscopic information from the rock core to construct a quantitative identifier for the in-situ hierarchical characteristics of natural fractures;

[0023] Step S13: Integrate static and dynamic information to establish a quantitative identifier for the classification features of natural fractures in the wellbore;

[0024] The establishment of a classification feature model for natural cracks includes the following sub-steps:

[0025] The longitudinal characteristics of natural cracks were defined as hierarchical cracks, low-angle cracks, and high-angle cracks.

[0026] Determine the classification characteristics of natural cracks that exhibit different orientations.

[0027] Specifically, step S12 further includes the following sub-steps:

[0028] S121, based on the classification feature model of natural fractures, integrates core micro and macro information to establish a feature identification model for fractures along the layer, low-angle fractures, and high-angle fractures;

[0029] S122, Based on the feature identification modes of along-layer fractures, low-angle fractures, and high-angle fractures, establish a quantitative identifier for the features of along-layer fractures, low-angle fractures, and high-angle fractures, namely a quantitative identifier for the graded features of natural fractures;

[0030] S123, relying on core repositioning, constructs a quantitative identifier for the in-situ classification characteristics of natural fractures, and implements in-situ characterization of the classification characteristic patterns of natural fractures.

[0031] Specifically, step S13 includes the following sub-steps:

[0032] S131. Based on the in-situ characterization results of the natural fracture classification feature quantitative identifier, the static information response features of the graded fractures are extracted, and a static logging information quantitative identifier for the natural fracture classification features is established.

[0033] The response features include lithology indicator curves, porosity indicator curves, resistivity indicator curves, array acoustic wave amplitude curves, and array acoustic wave anisotropy curves corresponding to along-layer fractures, low-angle fractures, and high-angle fractures.

[0034] S132, integrate the dynamic information response characteristics of wellbore production to verify the development degree of natural fractures at the same wellbore location, and establish a quantitative identifier for dynamic production information of natural fracture classification characteristics.

[0035] S133, integrating static and dynamic information, establishes a quantitative identifier for the classification features of natural fractures in wellbores;

[0036] The establishment of a quantitative identifier for classifying natural fractures in wellbores includes the following sub-steps:

[0037] S1331, based on the quantitative identifier of natural crack classification characteristics, identifies natural cracks at different depths in the longitudinal direction, and uses array acoustic anisotropy interpretation, geostress orientation indication and production dynamic output changes.

[0038] S1332, Analyze the planar orientation variation characteristics of natural cracks in a specific layer at this depth to form a static and dynamic quantitative identification of different groups of natural cracks in the same layer;

[0039] S1333, Establish a quantitative identifier for the classification features of natural fractures in wellbores.

[0040] Specifically, step S2 includes the following sub-steps:

[0041] Step S21: Construct a structural smoothness variable time window optimization quantitative identifier for optimal natural crack suitability;

[0042] Step S22: Construct a variance-variable time-window optimized quantitative identifier to improve the sensitivity to natural cracks;

[0043] Step S23: Construct a grading, classification, and quantitative identification device for ant body dynamic tracking of natural cracks in space.

[0044] Specifically, step S21 includes the following sub-steps:

[0045] Step S211, input the earthquake amplitude data volume;

[0046] Step S212: Compare filtering methods and select the filtering method with the maximum resolution and signal-to-noise ratio suitable for characterizing natural cracks;

[0047] Step S213: Optimize the size of the filtering window in different directions in three-dimensional space, and construct a quantitative identifier for structural smoothness variable time window optimization with the best suitability for natural cracks;

[0048] Step S22 includes the following sub-steps:

[0049] The main input is the processing result of the structural smooth variable time window optimized quantitative identifier with the best natural fracture suitability. Based on the optimization of the filtering bandwidth of the main survey line and the connecting survey line, and the optimization of the vertical smooth filtering variable window length, the difference comparison of the processing results of different directions and different filtering variable time window scales is carried out to weaken the information along the layer.

[0050] By optimizing the proportion of the main survey line, connecting survey line and vertical time window, dip correction of the variance volume in different directions is carried out to form crack structures with different attitudes;

[0051] By relying on tilt plane correction and variable threshold optimization, the variance calculation along the tilt plane is guaranteed to have true information coverage;

[0052] By relying on dip-guided smoothing, highlighting the structural features of faults and fractures, the sensitivity of discontinuous structures is enhanced.

[0053] Specifically, step S23 utilizes a variable time window to optimize the quantitative identifier to improve the sensitivity of natural fractures. The obtained variance volume is established as the main input, and the identification results of the wellbore natural fracture classification quantitative identifier are used as constraints to optimize the ant-body tracking pattern, forming an ant-body tracking pattern for fracture information at different scales. This includes the following sub-steps:

[0054] Optimize the initial tracking range of ants to form an initial tracking range that can capture more tracking traces of ants;

[0055] Adjustments were made to the directional deviation of ant tracking to determine the maximum legal distance that an ant could deviate from when searching in different directions.

[0056] Optimize the stride length of ants to determine the maximum effective search distance for an ant to move forward.

[0057] Optimize the illegal permission step size to determine the maximum search distance for illegal permission when no local anomaly is found in the previous step, thereby enabling the tracking of anomaly information;

[0058] Optimize the legally allowed step size by establishing a legally allowed step size that includes the maximum value based on the implementation of the illegally allowed step size, so as to realize the connection, recording and output of real outliers;

[0059] The ant tracking cessation criteria were optimized by comparing them with the development characteristics and distribution range of natural cracks, defining an illegal allowable step length cessation criterion to terminate the tracking process of illegal ants.

[0060] Specifically, step S3 includes the following sub-steps:

[0061] Step S31: Based on the spatial natural crack classification and quantitative identification device, a spatial prediction model for the distribution and development of natural cracks in the depth domain is established through time-depth information conversion. The model is then normalized, and the grid value of developed cracks is set to 1, while the grid value of non-developed cracks is set to 0.

[0062] Step S32: Quantitatively extract the classification feature parameters of natural cracks.

[0063] The quantitative acquisition of natural crack classification feature parameters includes the following sub-steps:

[0064] S321, the normalized depth domain natural crack distribution and development spatial prediction model is converted into a planar map of the distribution of natural cracks in the sub-layers of the hierarchical grouping system.

[0065] S322, A phase modeling strategy is used to establish a natural crack classification index model;

[0066] S323, using amplitude values ​​as an indicator of natural crack development density, and using the natural crack occurrence index model of different series as constraints, a natural crack classification and development intensity model is constructed.

[0067] Step S33: Establish a three-dimensional network model of natural cracks using a hierarchical classification modeling method;

[0068] The modeling method includes a method that uses a hierarchical classification natural crack occurrence index model as the spatial constraint for natural crack interpolation and a hierarchical classification natural crack intensity model as the main input to form a hierarchical classification three-dimensional discrete network model of natural cracks.

[0069] The beneficial effects of this invention are:

[0070] The present invention not only achieves a fine and accurate depiction of the planar and longitudinal distribution characteristics and patterns of natural fractures, but also solves the problems of accurate and reliable identification of induced fractures near faults and natural fractures far from fault structures in stable zones, as well as accurate, reliable and rapid construction of a three-dimensional network model for the classification of natural fractures. This provides reliable geological models and technical support for the scientific and efficient development of unconventional oil and gas reservoirs such as conventional fractured reservoirs, tight sandstone, tight carbonate rocks and shale oil and gas. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0072] Figure 1 This is a technical flowchart of the present invention;

[0073] Figure 2 This is a schematic diagram of the longitudinal grading characteristics of natural cracks in this invention;

[0074] Figure 3 This is a schematic diagram of the quantitative identifier for the classification characteristics of natural fractures in wellbore based on multi-source information fusion, and an identification case of the present invention.

[0075] Figure 4 This is a schematic diagram of a quantitative identifier for the classification features of natural fractures in a wellbore, established by integrating static and dynamic information, and the identified classification features of natural fractures.

[0076] Figure 5 This is a schematic diagram of the natural crack planar classification features of the present invention;

[0077] Figure 6This is a schematic diagram of the dynamic tracking of large fractures and small cracks according to the present invention;

[0078] Figure 7 This is a schematic diagram illustrating the classification and characterization of natural cracks according to the present invention;

[0079] Figure 8 This is a schematic diagram of the spatial prediction model for the distribution and development of natural cracks in this invention;

[0080] Figure 9 This is a schematic diagram of the natural crack occurrence index model of the hierarchical classification of the present invention;

[0081] Figure 10 This is a schematic diagram of the graded classification model of natural crack development intensity according to the present invention;

[0082] Figure 11 This is a schematic diagram of the hierarchical classification of natural cracks in this invention, representing a three-dimensional discrete network model.

[0083] Figure 12 This is a columnar statistical chart of the distribution of natural crack locations according to the present invention;

[0084] Figure 13 This is a histogram of natural crack length distribution according to the present invention.

[0085] Figure 14 This is a columnar statistical chart of the natural crack dip angle distribution of the present invention. Detailed Implementation

[0086] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0087] To provide a clearer understanding of the technical features, objectives, and beneficial effects of this invention, the following detailed description of the technical solution is provided. Obviously, the described embodiments are only a portion of the embodiments of this invention, not all of them, and should not be construed as limiting the scope of implementation of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the protection scope of this invention.

[0088] This invention addresses the characterization and modeling of various natural fractures in conventional fractured oil and gas reservoirs, as well as unconventional oil and gas reservoirs such as tight sandstone, tight carbonate rocks, and shale oil and gas reservoirs. It proposes a quantitative identifier and three-dimensional modeling method for classifying and classifying natural fractures. First, by analyzing the relationship between stress field and lithofacies dynamic evolution, a characteristic model for classifying and classifying natural fractures is established. Based on this, an in-situ quantitative identifier for classifying and classifying natural fractures is constructed by fusing microscopic and macroscopic information from the core. Further, relying on the fusion of static and dynamic information, a quantitative identifier for classifying and classifying natural fractures in the wellbore is established. Then, using a multi-level overlay variable-time-window optimization technique based on seismic attributes, a structurally smooth variable-time-window optimized quantitative identifier for optimal natural fracture suitability, a variance-volume variable-time-window optimized quantitative identifier to improve the sensitivity of natural fractures, and an ant-body dynamic tracking quantitative identifier for classifying and classifying natural fractures is constructed. By employing the above technical paths, methods, and approaches, not only can the planar and vertical distribution patterns of natural fractures be effectively characterized, but the problem of accurate and reliable identification of induced fractures near faults and natural fractures far from fault-stable structural zones can also be solved. Finally, based on the identification results of the natural crack classification quantitative identifier, we carried out quantitative picking of natural crack classification feature parameters, and further adopted the classification modeling method to complete the accurate, reliable and rapid construction of the natural crack classification three-dimensional network model, so as to realize the fine and accurate characterization of the planar and longitudinal distribution characteristics and laws of natural cracks.

[0089] See the detailed technical solution flowchart. Figure 1 As shown.

[0090] (1) Relying on the fusion of multiple information sources, a quantitative identification device for classifying and classifying natural fractures in wellbores is constructed.

[0091] To address the technical challenge of accurately and clearly characterizing the occurrence and spatial distribution of natural fractures due to their complex formation mechanisms and structural features using single information sources, this invention proposes a quantitative identifier for classifying and classifying natural fractures in wellbores by fusing multi-source information. By analyzing the dynamic evolution relationship between stress field and lithofacies, a classification and characterization model for natural fractures is established. Based on this, microscopic and macroscopic information from the core is integrated to construct an in-situ quantitative identifier for classifying and classifying natural fractures. This identifier is used to calibrate wellbore fracture information. Further integration of static and dynamic multi-source information establishes a quantitative identifier for classifying and classifying natural fractures in wellbores, enabling accurate and reliable identification of the classification and characterization features of natural fractures.

[0092] ①Analyze the dynamic evolution relationship between stress field and lithofacies, and establish a classification model for natural fractures.

[0093] Based on core observations and laboratory experiments, this study analyzes the changing characteristics of natural fracture occurrence, the composition and crystallization of fracture filling materials, and elucidates the controlling role of lithofacies as the material source of fractures and stress as the external factor inducing fracture formation during geological evolution, thus defining the vertical classification of natural fractures into interbedded fractures, low-angle fractures, and high-angle fractures. Furthermore, based on changes in tectonic surface curvature, anisotropy interpretation, paleostress variations, and sedimentary tectonic changes, this study analyzes the scale of tectonic evolution and the changes in the direction and magnitude of stress, elucidating their impact on the planar distribution of natural fractures, and defining the classification characteristics of natural fractures exhibiting different azimuthal variations. Combining these vertical and planar classification characteristics, a hierarchical classification model for natural fractures is constructed.

[0094] Based on core observations and experimental data from six core wells in a certain study area, analysis of rock particle composition, pore structure, and stress conditions revealed that the depositional environment was under still or weakly dynamic water conditions. The mineral particles were small in size and well-sorted, resulting in a dense, layered depositional structure with little cement flow. Poor cementation between layers created weak points (e.g., in shale and coalbed methane). Due to localized stress concentration, inter-layer fractures formed, with the fracture surfaces roughly parallel to the bedding planes. Figure 2 a) Meanwhile, when the sedimentary environment is under dynamic water conditions, the rock mass forms a well-cemented layered or massive structure. The abrupt changes in lithology become weak points, and when stress concentration occurs, low-angle cracks that strike roughly parallel to the rock strata and high-angle cracks that strike roughly perpendicular to the rock strata are easily formed. When subjected to shear force, the cracks are distributed along the maximum shear stress surface at a certain angle to the direction of the maximum principal compressive stress, which can also lead to large-angle shear cracks in homogeneous rock masses. Figure 2 b, c); This resulted in a hierarchical structure of natural fractures, exhibiting longitudinal characteristics of inter-layer fractures, low-angle fractures, and high-angle fractures. The stress direction propagation diagrams for each sub-layer and the anisotropy interpretation diagrams for single wells were obtained using the tectonic curvature method and array acoustic anisotropy interpretation. Figure 5 Due to the influence of tectonic stresses from multiple phases and different directions, the maximum principal stresses in each phase have different directions, intensities, and curvature variations. This results in the distribution of fracture plane orientations varying in northeast, east-west, north-south, and northwest directions, forming a classification characteristic of natural fractures exhibiting different directional variations. Based on the above, a hierarchical classification model for the longitudinal and planar classification of natural fractures in the study area was established. Figure 2 , Figure 5 ).

[0095] ② Integrating microscopic and macroscopic information from rock cores, a quantitative identifier for in-situ hierarchical characteristics of natural fractures was constructed.

[0096] a. Integrate microscopic and macroscopic information from rock cores to establish a quantitative identifier for the classification characteristics of natural fractures.

[0097] Based on the aforementioned hierarchical classification feature model of natural fractures, we extracted characteristic parameters such as dip angle, length, and aperture of natural fractures at the nano- and micro-scale using cast thin sections and scanning electron microscopy. This analyzed the relationship between the direction of natural fractures and the rock strata, establishing a feature identification model for fractures along strata and forming a quantitative identifier for fractures along strata. Furthermore, we collected macro-scale data on the direction of natural fractures and their angle with the strata using core observations, analyzing the mechanical characteristics of weak surfaces such as compression, tension, and shear, and establishing feature identification models for low-angle and high-angle fractures, forming quantitative identifiers for low-angle and high-angle fractures respectively. This comprehensive approach resulted in a hierarchical quantitative identifier for fractures along strata, low-angle fractures, and high-angle fractures.

[0098] Table 1 presents a quantitative identifier for the classification of natural fractures in an oilfield in central and eastern my country. Through analysis of cast thin sections and scanning electron microscopy, the microscopic fracture orientation along the strata was determined, forming an inter-stratum fracture characteristic identifier. By observing and statistically analyzing the macroscopic angle data between the natural fracture orientation and the formation, fractures were found to be distributed at low and high angles along vulnerable surfaces, forming low-angle and high-angle fracture characteristic identifiers. Combining these two methods, a quantitative identifier for the classification of fractures, including inter-stratum fractures, low-angle fractures, and high-angle fractures, was constructed.

[0099] Table 1. Establishment of a quantitative identifier for the classification characteristics of natural fractures by integrating microscopic and macroscopic information from core samples.

[0100]

[0101]

[0102] b. Based on core repositioning, construct a quantitative identifier for the in-situ hierarchical characteristics of natural fractures, and implement in-situ characterization of the hierarchical characteristic patterns of natural fractures.

[0103] A quantitative identifier for the classification characteristics of natural fractures, including those along the layer, low-angle fractures, and high-angle fractures, is used for classification and sorting. Based on core positioning, natural fractures are matched one-to-one with the trajectory of a single well, which describes the specific spatial location and classification characteristics of natural fractures in the wellbore, and realizes in-situ characterization of the classification characteristic pattern of natural fractures.

[0104] Table 2 shows the in-situ characterization results of the natural fracture classification pattern of the core section of Well No. 1 in an oilfield in central and eastern my country. Using the method described above, the specific spatial location and classification characteristics of each fracture on the wellbore were obtained, achieving in-situ characterization of the classification pattern of fractures along the formation, low-angle fractures, and high-angle fractures in the core section of Well No. 1 in this oilfield.

[0105] Table 2. Partial statistical results of in-situ characterization of natural fracture classification model in the core section of Well No. 1 in a certain oilfield.

[0106]

[0107]

[0108] ③ Integrate static and dynamic information to establish a quantitative identifier for the classification and grading characteristics of natural fractures in wellbores.

[0109] a. Integrate static and dynamic information to establish a quantitative identifier for the classification characteristics of natural fractures in wellbores.

[0110] Based on the in-situ characterization results of a natural fracture grading feature quantitative identifier, the depth and grading development characteristics of a specific fracture in the wellbore were obtained. These characteristics were further mapped onto logging curves of the same wellbore, and the response characteristics of lithology indicator curves, porosity indicator curves, resistivity indicator curves, array acoustic wave amplitude curves, and array acoustic wave anisotropy curves to grading fractures such as inter-layer fractures, low-angle fractures, and high-angle fractures were extracted successively. A static logging information quantitative identifier for natural fracture grading features was established, forming the ability to quantitatively identify natural fractures based on logging information. On this basis, the development degree of natural fractures at this location in the same wellbore was further verified using the response characteristics of dynamic information from wellbore production. A dynamic production information quantitative identifier for natural fracture grading features was established, forming a quantitative identification of the dynamic characteristics of inter-layer fractures, low-angle fractures, and high-angle fractures at the same depth in the same wellbore. By fusing static and dynamic information, a quantitative identifier for natural fracture grading features in the wellbore was established, realizing the quantitative identification of the grading features of natural fractures in the wellbore using static logging information combined with dynamic production information.

[0111] The core repositioning results from six wells in the study area were correlated one-to-one with lithological indicator curves (caliber CAL, natural gamma ray GR, spontaneous potential SP), porosity indicator curves (density DEN, neutron CNL, sonic transit time AC), and resistivity indicator curves (lateral resistivity RLL, formation true resistivity RT, and flushed zone resistivity Rxo). When fractures are present, SP shows a low negative anomaly at the fracture location. GR often shows a low value at fracture locations, and a low-peaked pattern when fractures are present (except when uranium is present in the groundwater, increasing GR). RLL (RT and Rxo) are sensitive to fractures. Low-angle fractures cause a significant decrease in lateral resistivity due to drilling fluid intrusion; the decrease is related to the opening angle, often showing a sharp or toothed decrease with no amplitude difference to a slight negative amplitude difference. However, high-angle fractures, due to shallow drilling fluid intrusion, result in medium to high lateral resistivity values, showing a positive difference, and exhibiting many smaller peaks. When low-angle fractures are present, the acoustic transit time (AC) increases, the curve shows small sawtooth patterns or cycle jumps, and there is no response to high-angle fractures. The expansion or contraction of the wellbore diameter (CAL) is closely related to the formation lithology. If it is a mudstone section and the fractures are high-angle and densely developed, the diameter expansion may occur. If it is a sandstone permeable formation, the well logging curve will show a contraction due to the adhesion of mud cake, so it should be analyzed in conjunction with the SP curve. When there are fractures, the density of densities (DEN) decreases compared to the surrounding rock, and the curve shows a sharp peak; the density of low-angle fractures decreases and the neutron porosity logging value increases, while the density of high-angle fractures decreases, and there is an amplitude difference between the density and neutron porosity; however, the density curve can only reflect the fractures in contact with the electrode; CNL reflects the water-bearing porosity of the mud intrusion. The porosity of the dense rock matrix is ​​very low, so neutron logging can directly reflect the degree of fracture development. When fractures develop, they carry bound water, or formation water, oil, etc., which will increase the porosity measured by neutron logging. In imaging logging, the increased P-wave and S-wave time differences and decreased amplitude in the array acoustic wave amplitude interpretation results indicate the development of open fractures. In the array acoustic anisotropy interpretation results, the presence of fractures causes stress dissipation, resulting in weak anisotropic image brightness, increased reflection coefficient, and separation of fast and slow S-waves. The azimuths of the fast and slow S-waves are orthogonal, with the fast S-wave azimuth parallel to the fracture direction. The dynamic change of high initial daily production followed by rapid decline further corroborates the existence of natural fractures at this location. Based on the fusion of the above static and dynamic information, a quantitative identifier for the classification characteristics of natural fractures in the wellbore was established. Figure 3 This study completed the in-situ characterization of natural fractures in wellbore based on well logging and production dynamics, and established a hierarchical characteristic model for natural fractures in wellbore.

[0112] b. Integrate static and dynamic information to establish a quantitative identifier for the classification features of natural fractures in wellbores.

[0113] A quantitative identifier for the classification characteristics of natural fractures in wellbore was established, forming a quantitative characterization of various types of natural fractures at different depths in the longitudinal direction. For the identified natural fractures at different depths in the longitudinal direction, a quantitative identifier for the classification characteristics of natural fractures in wellbore was further established using array acoustic anisotropy interpretation, geostress orientation indication, and production dynamic output changes, analyzing the planar orientation variation characteristics of natural fractures in a specific layer at that depth. On the one hand, a classification characteristic model of natural fractures based on static information was formed; on the other hand, the development status of fractures at that well location in that sub-layer was further corroborated by production dynamic output changes. Based on this, a static and dynamic quantitative identification of different groups of natural fractures in the same layer was formed, thereby establishing a classification characteristic model of natural fractures in wellbore by relying on the fusion of static and dynamic information.

[0114] A quantitative identification system for classifying natural fractures in wellbores was implemented in a study area. The fast shear wave azimuth plane showed variations in northeast, east-west, north-south, and northwest directions. The tectonic curvature variation characterized the geostress azimuth, and the natural fractures developed perpendicular to the direction of the maximum principal stress, thus indicating a northeast, east-west, north-south, and northwest distribution of natural fractures on a certain plane. Based on this, a classification feature model for natural fractures with static information from different plane layers was established. The dynamic production changes showed high initial production followed by rapid decline, further supporting the existence of the aforementioned natural fractures. Therefore, a quantitative identification system for the planar classification features of natural fractures in wellbores oriented northeast, east-west, north-south, and northwest was established. Figure 4 This study accurately characterized and depicted the four planar classification features of natural fractures in wellbore in different sub-layers of the study area: northeast-oriented, east-west-oriented, north-south-oriented, and northwest-oriented.

[0115] (2) Based on the multi-level overlay time window optimization of seismic attributes, a spatial natural fracture classification and quantitative identification instrument is constructed.

[0116] Based on the quantitative identification of natural fracture classification features in wellbore, a spatial natural fracture classification quantitative identification device is constructed by further relying on multi-level overlay time window optimization of seismic attributes, so as to achieve accurate and reliable identification of the classification features of natural fractures.

[0117] First, using seismic amplitude data as the main input, a structural smoothing variable-time-window optimized quantitative identifier for optimal natural fracture suitability is constructed by optimizing the filtering method and the geometric size of the time window. Second, using the variable-time-window optimized structural smoothing as the main input, a variance-volume variable-time-window optimized quantitative identifier is established to improve the sensitivity of natural fractures by comparing differences in smoothing scales in different directions and directions, performing multi-directional dip correction, setting plane confidence thresholds, and dip-guided smoothing. Finally, using the variable-time-window optimized variance volume as the main input, a dynamic tracking and classification quantitative identifier for natural fractures is constructed by optimizing the tracking method, tracking element arrangement distance, tracking deviation parameters, ant step size, illegal allowable step size, legal allowable step size, and ant tracking stop criteria. Based on the sequential application of the above three natural fracture classification and quantitative identifiers, a multi-level overlay of seismic amplitude data volume, structural smoothing, variance volume, and ant tracking is further implemented to construct a spatial natural fracture classification and quantitative identifier, realizing the classification and quantitative identification of natural fractures based on the multi-level overlay variable-time-window optimization technology of seismic attributes.

[0118] ① Construct a structurally smooth, time-varying, quantitative identifier with optimal suitability for natural cracks.

[0119] First, input the seismic amplitude data volume; then, compare the filtering methods and select the filtering method with the maximum resolution and signal-to-noise ratio suitable for characterizing natural fractures; finally, optimize the filtering window size in different directions in three-dimensional space and construct a structural smooth variable time window optimization quantitative identifier for the best suitability of natural fractures.

[0120] For a certain study area, the selected filtering method is tilt-guided and edge-enhanced. The optimized filtering time window size is 1, thereby constructing a structural smoothness optimization quantitative identifier with the best suitability for natural cracks.

[0121] ② Construct a variance-variable time-window optimized quantitative identifier to improve sensitivity to natural cracks.

[0122] Using the processing results of a structurally smoothed variable-time-window optimized quantitative identifier for optimal natural fracture suitability as the main input, and relying on the optimization of the filtering bandwidth of the main survey line and connecting survey line, and the optimization of the vertical smoothing filter variable-time-window length, the differences in processing results under different directions and different filter variable-time-window scales are compared to weaken the information along the layer. By optimizing the variable-time-window ratio of the main survey line, connecting survey line, and vertical direction, dip correction of the variance volume in different directions is carried out to form fracture structures with different attitudes. Variable-threshold optimization based on dip plane correction ensures that the variance calculation along the dip plane has a high coverage of true information. Dip-guided smoothing highlights structural features such as faults and fractures, enhancing the sensitivity to discontinuous structures. Using the above techniques, a variance volume variable-time-window optimized quantitative identifier for improving the sensitivity to natural fractures is constructed.

[0123] For a certain study area, the final planar filter bandwidth is 3×3, the vertical smoothing filter window length is 25, the ratio of time windows for the main survey line, connecting survey line and vertical direction is 1.5×1.5×1.5, and the dip surface correction threshold is 0.9. A variance-variable time window optimized quantitative identifier was constructed, which improved the sensitivity of natural crack prediction in the study area.

[0124] ③ Construct a hierarchical classification and quantitative identification system for ant body dynamic tracking of natural cracks in space.

[0125] This study utilizes a variable-time-window optimization quantitative identifier to improve the sensitivity of natural fractures. The resulting variance volume is established as the main input, and the identification results of the wellbore natural fracture classification quantitative identifier are used as constraints. The following optimization steps are employed: Ant tracking pattern optimization to form ant tracking patterns for fracture information at different scales; initial ant tracking range optimization to capture more tracking traces; ant tracking orientation deviation adjustment to determine the maximum legal distance the ant deviates from when searching in different directions; ant movement step length optimization to determine the maximum effective search distance for the ant; illegal allowable step length optimization to determine the maximum allowed search distance when no local anomaly is found in the previous step, thus tracking as many anomalies as possible; legal allowable step length optimization to establish a legal allowable step length including the maximum value based on the illegal allowable step length implementation, enabling the connection, recording, and output of real anomalies; and ant stopping tracking criterion optimization to define an illegal allowable step length stopping criterion by comparing it with the development characteristics and distribution range of natural fractures, thereby terminating the tracking process of illegal ants. Based on the implementation of these seven optimization steps, a dynamic ant tracking spatial natural fracture classification quantitative identifier is constructed. By applying this quantitative identifier, a three-dimensional characterization of large fractures is first formed, and then small cracks are characterized simultaneously. At the same time, dynamic tracking of large fractures and small cracks is implemented, and finally a time-domain natural crack classification and prediction model is established.

[0126] The specific parameters of the ant-body dynamic tracking spatial natural fracture classification and quantitative identification device constructed for a certain study area are as follows: the ant-body tracking mode is a customized tracking mode based on active tracking; the initial tracking range of the ant-body is 2; the tracking orientation deviation is 2; the ant movement step length is 3; the illegal allowable step length is 2; the legal allowable step length is 2; and the illegal allowable step length stopping criterion is 10. Based on this, the fault interpretation results are synthesized to characterize large faults and small fractures in sequence, realizing the dynamic tracking of large faults and small fractures, and establishing a time-domain natural fracture classification and prediction model.

[0127] Figure 6The above-mentioned technical methods are illustrated in the following diagram. By utilizing a smooth variable time window optimization technique, maximum resolution and signal-to-noise ratio are achieved, providing suitable seismic amplitude data volumes for characterizing natural fractures. Based on this, a variance volume variable time window optimization technique is used to enhance the sensitivity to discontinuous structures such as faults and fractures. Furthermore, an ant-body dynamic tracking technique is incorporated to form a dynamic tracking process between large fractures and small fractures, achieving synchronous characterization of large fractures and small fractures. This is achieved by extracting slices along the bedding planes and longitudinal profiles of the target segment from the time-domain natural fracture classification and prediction model. Figure 7 The study found that natural cracks exhibit a hierarchical distribution pattern in the longitudinal direction, including cracks along the bedding plane, low-angle cracks, and high-angle cracks, and a classification distribution pattern in the planar direction, including northeast, east-west, north-south, and northwest directions. This study achieved a hierarchical classification and characterization of natural cracks based on dynamic tracking of ant bodies.

[0128] (3) A three-dimensional network model of natural cracks was established using a hierarchical classification modeling method.

[0129] Based on the results of constructing a spatial natural fracture classification and quantitative identification device through multi-level nested variable time window optimization of seismic attributes, a spatial prediction model for the distribution and development of natural fractures in the depth domain is established through time-depth information conversion. Quantitative picking of characteristic parameters for natural fracture classification is carried out. Furthermore, a classification modeling method is adopted to establish a natural fracture classification occurrence index model representing the spatial development range and a natural fracture intensity model representing the spatial development density. Finally, a three-dimensional discrete network model for natural fracture classification is constructed.

[0130] ① Quantitatively collect characteristic parameters for classifying and grading natural cracks

[0131] By combining a wellbore natural fracture classification and quantitative identification tool with a spatial natural fracture classification and quantitative identification tool, we obtained integrated classification characteristics of natural fracture cores and wellbore, as well as optimization results of natural fracture time windows and multi-level overlapping seismic attributes. Through time-depth conversion, a spatial prediction model for the distribution and development of natural fractures in the depth domain was established, and this model was normalized. Under the control constraints of the natural fracture classification characteristics, by setting the grid value of developed fractures to 1 and the grid value of undeveloped fractures to 0, the normalized spatial prediction model for the distribution and development of natural fractures in the depth domain was converted into a planar map of the distribution of small-layer natural fractures in a hierarchical grouping system. A natural fracture classification and occurrence index model was established using a phase modeling strategy. Using the amplitude value as an indicator of natural fracture development density (when the value is 0, there are basically no fractures; as the value increases, the fractures are more developed), and using the natural fracture occurrence index models of different groups as constraints, a natural fracture classification and development intensity model was constructed.

[0132] Figure 8This paper presents a spatial prediction model for the distribution and development of natural fractures in an oilfield. The model first normalizes the spatial prediction model of natural fracture distribution and development in the study area, ensuring that the model values ​​are distributed between 0 and 1. The spatial development of natural fractures is constrained by the hierarchical classification characteristics of the natural fractures. Planar distribution maps of natural fractures in six sub-layers are drawn using hierarchical grouping. Finally, a natural fracture hierarchical classification occurrence index model is established using facies modeling. Figure 9 Subsequently, using a graded and classified occurrence index model as a constraint, a graded and classified natural crack development intensity model was established. Figure 10 ).

[0133] ② A three-dimensional network model of natural cracks was established using a hierarchical classification modeling method.

[0134] Using a hierarchical natural fracture occurrence index model as the spatial constraint for natural fracture interpolation and a hierarchical natural fracture intensity model as the main input, a method for constructing a hierarchical three-dimensional discrete network model of natural fractures is formed. Based on this, a three-dimensional network model of natural fractures is established, which accurately reproduces the system characteristics of natural fractures in three-dimensional space, as well as the distribution location, orientation, dip angle, and morphology of each fracture fragment in each system. This enables the characterization and modeling of induced fractures near faults and natural fractures in tectonically stable zones far from faults, clearly depicting the heterogeneity of the spatial distribution and the complexity of the development of natural fractures.

[0135] Figure 11 This is a three-dimensional network model for classifying natural fractures in a reservoir, established using the above method. Stress is concentrated near faults and in areas where faults turn, leading to more developed natural fractures. The study area exhibits four oriented natural fracture groups: east-west, north-south, northwest, and northeast, with southwest-northeast oriented fractures being the most prevalent. Figure 12 The crack lengths are generally concentrated around 200-300 meters. Figure 13 Vertically, high-angle fractures (60-90°), low-angle fractures (45-60°), and inter-layer fractures (0-45°) are developed, with high-angle fractures being the predominant type. Figure 14 This led to the construction of a three-dimensional network model for the hierarchical classification of natural cracks.

[0136] The proposed solution includes a wellbore natural fracture classification and quantitative identification device, a spatial natural fracture classification and quantitative identification device, and a natural fracture classification and three-dimensional modeling method.

[0137] First, by analyzing the relationship between stress field and lithofacies dynamic evolution, a classification feature model for natural fractures is established. Based on this, relying on the fusion of core micro and macro information, an in-situ quantitative identifier for natural fracture classification features is constructed. Further relying on the fusion of static and dynamic information, a quantitative identifier for wellbore natural fracture classification features is established. Then, using a multi-level overlay variable-time-window optimization technique based on seismic attributes, a structurally smooth variable-time-window optimized quantitative identifier for optimal natural fracture suitability, a variance-volume variable-time-window optimized quantitative identifier to improve natural fracture sensitivity, and an ant-body dynamic tracking quantitative identifier for natural fracture classification are constructed. Finally, based on the identification results of the above-mentioned quantitative identifiers for natural fracture classification, quantitative picking of natural fracture classification feature parameters is carried out. Further employing a hierarchical classification modeling method, an accurate, reliable, and rapid three-dimensional network model for natural fracture classification is constructed, achieving a refined and accurate depiction of the planar and longitudinal distribution characteristics and patterns of natural fractures.

[0138] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

[0139] It should be noted that, for the sake of simplicity, the aforementioned method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and units involved are not necessarily essential to this application.

[0140] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0141] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, ROM, RAM, etc.

[0142] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier, characterized in that, Includes the following steps: Step S1: Construct a graded classification and quantitative identification device for natural fractures in wellbore by integrating static, dynamic, macro and micro multi-dimensional information; Step S2 involves constructing a spatial natural fracture classification and quantitative identification tool based on multi-level overlay of seismic attributes and time-window optimization, including the following sub-steps: Step S21: Construct a structural smoothness variable time window optimization quantitative identifier for optimal natural crack suitability; Step S22: Construct a variance-variable time-window optimized quantitative identifier to improve sensitivity to natural cracks; Step S23: Construct a hierarchical classification and quantitative identification device for ant body dynamic tracking of natural cracks in space; Step S3 involves establishing a three-dimensional network model of natural cracks using a hierarchical classification modeling method, including the following sub-steps: Step S31: Based on the spatial natural crack classification and quantitative identification device, a spatial prediction model for the distribution and development of natural cracks in the depth domain is established through time-depth information conversion. The model is then normalized, and the grid value of developed cracks is set to 1, while the grid value of non-developed cracks is set to 0. Step S32: Quantitatively extract the classification feature parameters of natural cracks. The quantitative acquisition of natural crack classification feature parameters includes the following sub-steps: S321, the normalized spatial prediction model of the distribution and development of natural fractures in the depth domain is converted into a planar map of the distribution of natural fractures in the sub-layers of a hierarchical grouping system; S322, A phase modeling strategy is used to establish a natural crack classification index model; S323, using amplitude values ​​as an indicator of natural crack development density, and using the natural crack occurrence index model of different series as constraints, a natural crack classification and development intensity model is constructed. Step S33: Establish a three-dimensional network model of natural cracks using a hierarchical classification modeling method; The modeling method includes a method that uses a hierarchical classification natural crack occurrence index model as the spatial constraint for natural crack interpolation and a hierarchical classification natural crack intensity model as the main input to form a hierarchical classification three-dimensional discrete network model of natural cracks.

2. The modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier according to claim 1, characterized in that, Step S1 includes the following sub-steps: Step S11: Analyze the dynamic evolution relationship between stress field and lithofacies, and establish a classification characteristic model for natural fractures. Step S12: Integrate microscopic and macroscopic information from the rock core to construct a quantitative identifier for the in-situ hierarchical characteristics of natural fractures; Step S13: Integrate static and dynamic information to establish a quantitative identifier for the classification features of natural fractures in the wellbore; The establishment of a classification feature model for natural cracks includes the following sub-steps: The longitudinal characteristics of natural cracks were defined as hierarchical cracks, low-angle cracks, and high-angle cracks. Determine the classification characteristics of natural cracks that exhibit different orientations.

3. The modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier according to claim 2, characterized in that, Step S12 further includes the following sub-steps: S121, based on the classification feature model of natural fractures, integrates core micro and macro information to establish a feature identification model for fractures along the layer, low-angle fractures, and high-angle fractures; S122, Based on the feature identification modes of along-layer fractures, low-angle fractures, and high-angle fractures, establish a quantitative identifier for the features of along-layer fractures, low-angle fractures, and high-angle fractures, namely a quantitative identifier for the graded features of natural fractures; S123, relying on core repositioning, constructs a quantitative identifier for the in-situ classification characteristics of natural fractures, and implements in-situ characterization of the classification characteristic patterns of natural fractures.

4. The modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier according to claim 3, characterized in that, Step S13 includes the following sub-steps: S131. Based on the in-situ characterization results of the natural fracture classification feature quantitative identifier, the static information response features of the graded fractures are extracted, and a static logging information quantitative identifier for the natural fracture classification features is established. The response features include lithology indicator curves, porosity indicator curves, resistivity indicator curves, array acoustic wave amplitude curves, and array acoustic wave anisotropy curves corresponding to along-layer fractures, low-angle fractures, and high-angle fractures. S132, integrate the dynamic information response characteristics of wellbore production to verify the development degree of natural fractures at the same wellbore location, and establish a quantitative identifier for dynamic production information of natural fracture classification characteristics. S133, integrating static and dynamic information, establishes a quantitative identifier for the classification features of natural fractures in wellbores; The establishment of a quantitative identifier for classifying natural fractures in wellbores includes the following sub-steps: S1331, based on the quantitative identifier of natural crack classification characteristics, identifies natural cracks at different depths in the longitudinal direction, and uses array acoustic anisotropy interpretation, geostress orientation indication and production dynamic output changes. S1332, Analyze the planar orientation variation characteristics of natural cracks in a specific layer at this depth to form a static and dynamic quantitative identification of different groups of natural cracks in the same layer; S1333, Establish a quantitative identifier for the classification features of natural fractures in wellbores.

5. The modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier according to claim 1, characterized in that, Step S21 includes the following sub-steps: Step S211, input the earthquake amplitude data volume; Step S212: Compare filtering methods and select the filtering method with the maximum resolution and signal-to-noise ratio suitable for characterizing natural cracks; Step S213: Optimize the size of the filtering window in different directions in three-dimensional space, and construct a quantitative identifier for structural smoothness variable time window optimization with the best suitability for natural cracks; Step S22 includes the following sub-steps: Using the processing results of the structural smooth variable time window optimized quantitative identifier with the best natural fracture suitability as the main input, and relying on the optimization of the filtering bandwidth of the main survey line and the connecting survey line, and the optimization of the vertical smooth filtering variable window length, the differences in the processing results of different directions and different filtering variable time window scales are compared, and the information along the layer is weakened; By optimizing the proportions of the main survey line, connecting survey lines, and vertical time windows, dip corrections were performed on the variance volumes in different directions to form fracture structures with different attitudes; By relying on tilt plane correction and variable threshold optimization, the variance calculation along the tilt plane is guaranteed to have true information coverage; By relying on dip-guided smoothing, highlighting the structural features of faults and fractures, the sensitivity of discontinuous structures is enhanced.

6. The modeling method for identifying natural cracks using a hierarchical classification and quantitative identifier according to claim 5, characterized in that, Step S23 utilizes a variable time window to optimize the quantitative identifier and improve its sensitivity to natural fractures. The obtained variance volume is established as the main input, and the identification results of the wellbore natural fracture classification quantitative identifier are used as constraints. Ant tracking pattern optimization is then performed to form ant tracking patterns for fracture information at different scales, including the following sub-steps: Optimize the initial tracking range of ants to form an initial tracking range that can capture more tracking traces of ants; Adjustments were made to the directional deviation of ant tracking to determine the maximum legal distance that an ant could deviate from when searching in different directions. Optimize the stride length of ants to determine the maximum effective search distance for an ant to move forward. Optimize the illegal permission step size to determine the maximum search distance for illegal permission when no local anomaly is found in the previous step, thereby enabling the tracking of anomaly information; Optimize the legally allowed step size by establishing a legally allowed step size that includes the maximum value based on the implementation of the illegally allowed step size, so as to realize the connection, recording and output of real outliers; The ant tracking cessation criteria were optimized by comparing them with the development characteristics and distribution range of natural cracks, defining an illegal allowable step length cessation criterion to terminate the tracking process of illegal ants.

Images