A method for quantitatively predicting fractures in a bedded tight reservoir based on a composite fracture criterion

Abstract

This invention discloses a quantitative prediction method for fractures in layered tight reservoirs based on a composite fracture criterion, belonging to the field of oil and gas field development. The method includes: identifying lithology and bedding types based on core, well logging, and seismic data; constructing a three-dimensional mechanical parameter volume by integrating rock mechanics experiments and well-seismic data based on this constraint; establishing a discriminant function according to lithology and selecting different fracture or slip criteria for sandstone, mudstone, and bedding interfaces to form a composite fracture criterion; constructing a fracture density mechanical characterization model based on this criterion and rock mechanics theory; determining the paleostress during the key fracture-forming period through acoustic emission experiments and performing finite element simulation based on the mechanical parameter volume to obtain the paleostress field; finally, embedding the composite criterion and characterization model into a simulation platform, calculating the three-dimensional distribution of fracture density based on the paleostress field, and verifying it with measured data. This invention achieves refined and quantitative prediction of the spatial distribution of fractures in layered tight reservoirs.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas field development. Specifically, it relates to a quantitative prediction method for fractures in layered tight reservoirs based on a composite fracture criterion. It integrates multiple methods such as core observation, well logging interpretation, seismic inversion, rock mechanics experiments, and finite element numerical simulation to construct a three-dimensional finite element geological model of the layered reservoir. Based on the composite rock fracture criterion and fracture parameter characterization model, it simulates the three-dimensional stress field distribution during key tectonic activity periods and accurately predicts the spatial distribution of structural fractures in the reservoir. Background Technology

[0002] Structural fractures in tight reservoirs are key geological factors for improving reservoir permeability and increasing oil and gas production capacity. Accurate prediction of their spatial distribution is of great significance for guiding the efficient exploration and development of oil and gas reservoirs. In recent years, as oil and gas exploration has extended to deeper and unconventional areas, quantitative prediction methods for fractures have received increasing attention.

[0003] In existing technologies, various fracture prediction methods based on stress field simulation and rock fracture criteria have been developed. For example, prior art document CN113534291A discloses a "quantitative prediction method for fractures of different scales in low-permeability reservoirs under rock mechanical layer constraints." This method achieves quantitative prediction of fractures at different scales by dividing the rock mechanical layer structure, establishing multi-level fracture criteria for interbedded soft and hard rocks, and combining it with three-dimensional stress field simulation. This method has achieved certain application results in sandstone-mudstone interbedded low-permeability reservoirs.

[0004] However, my country's terrestrial sedimentary tight sandstone reservoirs generally possess unique geological characteristics, including deep burial, strong diagenesis, low porosity and permeability, frequent thin interbedded sandstone / mudstone layers, and complex bedding structures. Under tectonic activity, fracture development is not only controlled by the regional stress field but also significantly influenced by various bedding planes and lithological interfaces, resulting in a highly heterogeneous spatial distribution of fractures. Existing prediction methods (including the aforementioned comparative documents) largely focus on lithological assemblage and mechanical layering, or employ relatively singular fracture criteria, failing to fully reflect the differentiated control mechanism of bedding structure itself on rock fracture behavior. Specifically, for bedding-type tight reservoirs, a unified composite fracture criterion that comprehensively considers the fracture or slip behavior of sandstone, mudstone matrix, and their interbedded bedding interfaces under different stress states (tensional and compressive) has not yet been established, nor is a rigorously coupled fracture parametric mechanical characterization model. Therefore, the prediction results of existing methods still deviate from the actual fracture development, making it difficult to meet the urgent need for refined exploration and development scheme optimization for bedding-type tight reservoirs. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a quantitative prediction method for fractures in bedding tight reservoirs based on a composite fracture criterion. For bedding tight reservoirs, lithology is identified and determined based on rock mechanics experiments, well logging, and seismic data. A mechanical parameter body is constructed, and a composite fracture criterion for bedding rocks and a mechanical characterization model of fracture parameters are established. The three-dimensional spatial distribution of fracture density parameters is quantitatively predicted through paleotectonic stress field simulation.

[0006] To achieve the above objectives, the present invention adopts the following solution:

[0007] A quantitative prediction method for fractures in layered tight reservoirs based on composite fracture criteria includes the following steps:

[0008] (1) Lithology and bedding type identification steps: Based on core, well logging and seismic data, identify and determine the lithology category and bedding type of tight reservoir, and obtain the lithology and bedding type identification results;

[0009] (2) Mechanical parameter body construction steps: Using the identification results obtained in step (1) as constraints, a three-dimensional heterogeneous rock mechanical parameter body is obtained and established through rock mechanics experiments and well-seismic combination methods;

[0010] (3) Steps for establishing the composite fracture criterion: Based on the lithology category determined in step (1), establish a lithology discrimination function, and define corresponding fracture criteria or slip criteria for sandstone, mudstone, and bedding interfaces respectively to form a composite fracture criterion; wherein, the lithology discrimination function is:

[0011]

[0012] in: Indicates sandstone or mudstone, dimensionless; Dimensionless discriminant function representing sand / mudstone classification and bedding type / massiveness;

[0013] (4) Steps for constructing a crack parameter characterization model: Based on the composite fracture criterion and rock mechanics theory in step (3), construct a mechanical model characterizing the relationship between crack density and stress-strain parameters;

[0014] (5) Paleostress field simulation steps: Based on the acoustic emission experiment, the key fracture formation period and paleostress direction are determined. Based on the three-dimensional mechanical parameter body constructed in step (2), a geomechanical model is established, and finite element simulation is performed to obtain the three-dimensional paleostress field distribution of the key period.

[0015] (6) Quantitative prediction and verification steps of cracks: The composite fracture criterion of step (3) and the crack parameter characterization model of step (4) are implanted into the numerical simulation platform. Based on the paleostress field distribution obtained by simulation in step (5), the three-dimensional spatial distribution of crack density is calculated, and the prediction results are verified by actual observation data.

[0016] Furthermore, in the above method, step (1) classifies the bedding types into three categories: horizontal bedding, diagonal bedding, and massive bedding.

[0017] Furthermore, in the above method, in step (3), when At that time, the Mohr-Coulomb criterion or the Griffith criterion was used to determine the fracture of sandstone;

[0018] The Mohr-Coulomb criterion is as follows:

[0019]

[0020] in: This represents the maximum principal stress, expressed in MPa. Minimum principal stress, MPa; This represents the cohesive force of the rock, measured in MPa, and can be determined experimentally. The internal friction angle of the rock is expressed in ° and can be measured experimentally.

[0021] when At that time, the Drucker-Prager criterion or the Hoek-Brown criterion was used to determine the fracture of mudstone;

[0022] when In this case, the slip determination of the bedding interface adopts the Byerlee criterion.

[0023] Furthermore, in the above method, in step (4), the crack density... The calculation formula is: ; Strain energy density, in J / m 3 ; The unit for characterizing the volume of a single unit is m. 3 ; This refers to the surface energy of the crack, expressed in J / m². 2 .

[0024] Furthermore, in the above method, the formula for calculating the strain energy density ω is:

[0025]

[0026] This is the elastic modulus, expressed in GPa. Poisson's ratio is dimensionless. This represents the maximum principal strain, which is dimensionless. Indicates the intermediate principal strain, which is dimensionless; This represents the minimum principal strain, which is dimensionless. , These are the intermediate and minimum principal stresses, respectively, in MPa. , The model parameters are dimensionless and determined based on the peak strain of the rock. It is a natural constant.

[0027] This invention also discloses the application of the above method in oil and gas reservoir exploration and development, using the predicted three-dimensional distribution of fractures for evaluation of favorable zones, fracturing stimulation design, or optimization of development plans.

[0028] The present invention also discloses a storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the present invention.

[0029] The present invention also discloses a crack quantitative prediction device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above method.

[0030] Furthermore, the aforementioned device also includes an interface module for inputting core, logging, seismic, and experimental data, as well as a visualization module for outputting the three-dimensional distribution results of fracture density.

[0031] This invention also discloses a tight reservoir fracture prediction system, comprising:

[0032] The lithology identification module is used to perform the above step (1).

[0033] The mechanical parameter construction module is used to perform the above step (2);

[0034] The rupture criterion module is used to perform the above step (3).

[0035] Crack characterization module, used to perform the above step (4).

[0036] The stress field simulation module is used to perform the above step (5).

[0037] The prediction and verification module is used to perform the above step (6).

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] It solves the problems of accurately identifying the lithology of bedding reservoirs and rationally establishing fracture mechanics models for composite rock masses, and is suitable for quantitative fracture prediction in any bedding tight reservoir. It effectively predicts the spatial distribution characteristics of fracture parameters in bedding tight reservoirs, providing a reliable basis for predicting favorable zones in bedding tight reservoirs, and providing a guarantee for the design and optimization of fracturing stimulation and development schemes for fractured reservoirs, thereby reducing the risks and costs of exploration and development. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the process for quantitative prediction of structural fractures in layered tight reservoirs based on composite fracture criteria.

[0041] Figure 2 It is the type of bedding fracture identified in the core samples from the Qigu Formation core wells in the Yongjin Oilfield;

[0042] Figure 3 It is a three-dimensional lithofacies geological model of the Qigu Formation in Yongjin Oilfield under the constraints of sedimentary facies.

[0043] Figure 4 It is a three-dimensional mechanical parameter distribution model of the Qigu Formation in Yongjin Oilfield;

[0044] Figure 5 These are the results of acoustic emission experiments on the Qigu Formation rocks in the Yongjin Oilfield;

[0045] Figure 6 It is the key fracture formation movement and paleostress direction of the Qigu Formation in Yongjin Oilfield;

[0046] Figure 7 These are the simulation results of the mechanical loading method and ancient stress field in the finite element model;

[0047] Figure 8 This is the prediction result and verification of different sublayer bedding fractures in the Qigu Formation of Yongjin Oilfield. Detailed Implementation

[0048] A quantitative prediction method for fractures in layered tight reservoirs based on composite fracture criteria includes the following steps:

[0049] Step 1: Identification and determination of tight reservoir lithology and bedding type, the specific methods are as follows:

[0050] (1) The depth of core data comes from the drilling depth. Usually, the depth of core analysis data and logging curves are different and need to be corrected. That is, combined with the regional stratigraphic development characteristics, detailed geological data such as drilling core, cuttings logging and logging data are collected and the depth of the two is corrected to ensure that the logging core information is consistent with the actual underground core.

[0051] (2) Extract the logging response characteristics of known lithological samples, compare and analyze the similarities and differences in logging response characteristics of different lithologies, distinguish lithologies based on these characteristic differences and construct interpretation models, collect field logging and well logging data, and use conventional logging curves such as gamma (GR), spontaneous potential (SP), and resistivity (R) to analyze the characteristics of different lithologies. t The clay content V was calculated using methods such as uranium-free gamma (KTH) and neutron-density (CNL-DEN) cross-plotting. sh Calculation of clay content using uranium-free gamma (KTH):

[0052]

[0053]

[0054] In the formula, GCUR is a regional empirical coefficient, with GCUR=3.7 for new strata (Tertiary strata) and GCUR=2.0 for old strata. The values ​​are for uranium-free gamma-ray logging in pure sandstone sections. These are uranium-free gamma logging values ​​for pure mudstone sections.

[0055] (3) Manual qualitative interpretation of the lithology of the strata in the section without core sampling. If the lithology interpretation result is not ideal, return to the second step of the interpretation process (2) until a reasonable lithology interpretation result for a single well is finally obtained.

[0056] (4) Based on the observation of the drilling core, the phenomena of horizontal bedding, parallel bedding, rhythmic bedding, trough cross-bedding, platy cross-bedding, wedge cross-bedding, wavy cross-bedding, mound cross-bedding, feather cross-bedding, and flushed cross-bedding are identified and classified. In order to reduce the workload of three-dimensional geological modeling, the bedding types of tight sandstone are classified into three categories: horizontal bedding, oblique bedding, and massive bedding.

[0057] (5) Due to the frequent occurrence of thin sand / mudstone interbedded underground and the superposition and development of different types of bedding, in order to reduce the workload of three-dimensional geological modeling and stress field simulation, it is necessary to merge the lithology of the obtained single wells and connected wells, and to convert rocks of the same type with complex mineral composition, similar logging response and inconsistent particle size into the same lithofacies. For example, coarse sandstone, medium sandstone, fine sandstone, siltstone and muddy sandstone are merged into sandstone facies, sandy mudstone and mudstone are merged into mudstone facies, and gravelly sandstone, sandy conglomerate and conglomerate are merged into conglomerate facies. The intrinsic linear relationship between lithofacies type (conglomerate facies, sandstone facies, mudstone facies) and bedding type is established.

[0058] (6) Single well core identification and division are completed by using lithological logging response templates and integrated logging data. Based on clarifying the characteristics of sedimentary facies and sedimentary sand body distribution, lithological-seismic attribute parameters are selected. The lithological-seismic inversion under the joint "well-seismic" model is completed with the consistency between the inverted lithological data and the logging data reaching more than 85%. The lithological model is constructed by using deterministic and stochastic modeling methods.

[0059] Step 2: Constructing the mechanical parameter volume based on rock mechanics experiments and well-seismic integration, the specific method is as follows:

[0060] (1) Screen core sections of the target layer at similar depths in the core wells, including sandstone, mudstone, and layered sandstone and massive sandstone. Make a rough observation of the core sections of all wells, and select cores without pre-existing cracks on the surface for radial orientation consistency calibration to ensure that the radial relative geographical orientation of each core is the same at 0°. Drill a 25mm diameter and 50mm length plunger sample, and cut and grind both ends of the drilled sample.

[0061] (2) Triaxial compression test was performed on the obtained plunger specimen. During the experiment, the confining pressure was kept constant, and the axial load was gradually increased until the rock failed. Rock mechanical parameters (elastic modulus, Poisson's ratio, compressive strength, shear strength, yield strength, internal friction coefficient and internal friction angle, etc.) were obtained.

[0062] (3) Before interpreting rock mechanics parameters through logging, abnormal logging data should be processed to ensure that the logging data conforms to the actual ideal state. For complex lithological development characteristics, based on empirical logging formulas and using static mechanical parameter data of sandstone and mudstone obtained from rock mechanics experiments, the empirical formulas are optimized. The optimized formulas for calculating dynamic rock mechanics parameters are as follows:

[0063]

[0064]

[0065] In the formula, E is the elastic modulus, MPa; μ is Poisson's ratio; ρ b Density of rock, kg / m³ 3 ;Δt p and Δt s φ represents the longitudinal wave time difference and the transverse wave time difference, respectively, in μs / ft; φ is the internal friction angle, in °.

[0066] (4) In order to meet the actual underground rock mechanics conditions and obtain high-precision rock static mechanical parameters, the rock mechanics parameters are statistically analyzed and classified according to lithology (sandstone, mudstone), and the differences in test points caused by human factors are processed. The least squares method is applied to fit the optimal dynamic-static mechanical parameter correction regression curves for different lithologies, so that they are more in line with the needs of geological engineering.

[0067] (5) Using the lithological model in (4) as a constraint, the seismic inversion data of static mechanical parameters in (2) are resampled into the reservoir geological model to obtain the three-dimensional heterogeneous rock mechanical parameter volume of the target layer in the study area. All seismic units that intersect with the geological unit will contribute to the average calculation, and this contribution will correct the intersection amount. For discrete volume, the "major" or "median" average is used.

[0068] (6) Extract the center point data of the lithological model and rock mechanical parameters of the unit grid. The center point data of the unit grid includes X coordinate, Y coordinate, Z coordinate, lithological code, elastic modulus, Poisson's ratio, and density data. The X coordinate, Y coordinate, and Z coordinate correspond to the x, y, and z (depth) values ​​of the spatial geodetic coordinate system, respectively. The depth curve data is sampled at average intervals, and the data text is a dataset of rock lithology and mechanical properties (txt).

[0069] Step 3: Establish composite fracture criteria for layered tight reservoirs. The specific method is as follows:

[0070] (1) According to rock mechanics theory, when layered / bedding composite rocks are under triaxial stress, their possible deformation and failure modes are mainly brittle failure, plastic failure, and interface slip. Therefore, based on the lithological identification in step 1-2, a sandstone / mudstone discrimination function is constructed:

[0071]

[0072] in: Indicates sandstone or mudstone, dimensionless; This represents the classification of sandstone / mudstone and the discriminant function for bedding type / massiveness; it is dimensionless.

[0073] (2) The criterion for crack formation in sandstone conforms to the elastic fracture criterion, that is, when... Based on different stress states, an elastic fracture criterion applicable to bedding sandstone is established:

[0074] a. When the sandstone mass is under triaxial stress, then... If the minimum principal stress At that time, the Mohr-Coulomb strength criterion was used to determine rock fracture:

[0075]

[0076] in: This represents the maximum principal stress, expressed in MPa. Minimum principal stress, MPa; This represents the cohesive force of the rock, measured in MPa, and can be determined experimentally. The internal friction angle of the rock is expressed in ° and can be measured experimentally.

[0077] b. When the sandstone mass is under triaxial stress, then... If the minimum principal stress At that time, the Griffith strength criterion was used to determine rock fracture:

[0078] when At that time, the rupture criterion is:

[0079]

[0080]

[0081] when At that time, the rupture criterion is:

[0082]

[0083] in: This represents the maximum principal stress, expressed in MPa. Indicates the intermediate principal stress, in MPa; Minimum principal stress, MPa; This represents the tensile strength of rock, expressed in MPa, and can be measured experimentally. The tensile fracture angle is ____ degrees.

[0084] (3) Whether slip occurs at the sand / mud interface and bedding plane has nothing to do with the rock properties, but only with stress. Therefore, the criterion for whether slip occurs at the interface conforms to the Byerlee slip criterion, that is, when At that time, the interface sliding judgment criterion is:

[0085]

[0086] in: This represents the normal stress at the interface, in MPa. This represents the shear stress at the interface, in MPa.

[0087] (4) The criterion for crack formation in mudstone conforms to the plastic fracture criterion, that is, when At that time, based on different stress states, an elastic fracture criterion applicable to sandstone is established:

[0088] a. When the mudstone mass is under triaxial stress, then... If the minimum principal stress At that time, the Drucker-Prager yield criterion was used to determine rock fracture:

[0089]

[0090]

[0091] in: This represents the maximum principal stress, expressed in MPa. Indicates the intermediate principal stress, in MPa; Minimum principal stress, MPa; This represents the cohesive force of the rock, measured in MPa, and can be determined experimentally. The internal friction angle of the rock is expressed in ° and can be measured experimentally.

[0092] b. When the mudstone mass is under triaxial stress, then... If the minimum principal stress At that time, the Hoek-Brown strength criterion was used to determine rock fracture:

[0093]

[0094] in: This represents the maximum principal stress, expressed in MPa. Indicates the intermediate principal stress, in MPa; Minimum principal stress, MPa; This represents the tensile strength of the rock, expressed in MPa.

[0095] Step 4: Construction of the mechanical characterization model for crack parameters, the specific method is as follows:

[0096] According to the theories of elasticity and fracture mechanics, rocks deform and accumulate strain energy under stress. When the release rate of this strain energy equals the energy required to generate a crack per unit area, the rock fractures. For brittle rocks like dense sandstone, the accumulated energy is converted into elastic energy, which then transforms into surface energy and residual strain energy after fracture. However, for mudstone, the energy accumulated during the entire strain process is converted into elastic energy and plasticity. The work done by plasticity only causes deformation, not fracture. Therefore, when calculating fracture parameters, the strain energy density in mudstone should also be calculated using elastic strain energy, as shown in the following formula:

[0097]

[0098] Based on this, the formula for calculating crack density is:

[0099]

[0100]

[0101] in: This represents the maximum principal strain, which is dimensionless. Indicates the intermediate principal strain, which is dimensionless; The minimum principal strain is dimensionless. Model parameters are identified based on the results of triaxial rock mechanics experiments on intact rock and the peak strength points of the stress-strain curves. It is assumed that the strains at the peak points of the stress-strain curves of the rock under confining pressure are as follows: , This represents the maximum principal strain at the peak value, and is dimensionless. This represents the principal strain at the peak value, and is dimensionless. This represents the minimum principal strain at the peak value, and is dimensionless. The elastic modulus is expressed in GPa. It is Poisson's ratio, dimensionless. To characterize the volume of a unit cell (REV), the unit is m. 3 ; J is the energy required to generate a crack per unit area, i.e., the surface energy of the crack, with units of J / m². 2 .

[0102] Step 5: Determining the critical fracture period and simulating the three-dimensional paleostress field. The specific methods are as follows:

[0103] (1) Sampling of the drilling core in the study area by acoustic emission test. The drilling locations are preferably evenly distributed on the plane of the study area or in a quadrilateral shape. Each group of samples includes at least 4 specimens, one of which is taken from the vertical direction (parallel to the wellbore axis), and the other 3 are taken from 3 directions in the horizontal plane at 45° angles to each other.

[0104] (2) The equipment used for acoustic emission test consists of a servo rock rigidity tester and an acoustic emission test system. The processed sample is subjected to repeated loading test in the room at a loading rate of 0.1 MPa / s. The acoustic emission signal of the rock sample during the loading process is measured as a function of the load. On the acoustic emission load change curve of the second loading, the acoustic emission incomplete recording point is found. The load value of the incomplete recording point is used as a reference. In the acoustic emission load curve of the first loading, the Kessel point is determined. The average value of the load of the Kessel point and the incomplete recording point is taken as the maximum normal stress of the rock core in the ground. The number of Kessel points represents the period of tectonic movement.

[0105] (3) Based on the structural trace mechanics method, the cracks are phased and matched. The crack direction is statistically analyzed using rose diagrams to find the dominant group of conjugate cracks. The bisector of the conjugate angle is used as the direction of the maximum principal stress of the paleostress field during the key crack-forming period.

[0106] (4) Based on the three-dimensional geomechanical model established in Ansys and the paleostress test results during the key fracture period, the mechanical boundary of the model was set, the loading rate was set to 0.1 MPa / s, and the elastoplastic finite element three-dimensional stress field simulation was carried out. The simulation was repeated until the maximum principal stress component of the paleostress field was consistent with the distribution trend of the secondary compression fault, the minimum principal stress component was consistent with the distribution trend of the secondary extension fault, and the shear stress component was consistent with the distribution trend of the strike-slip fault.

[0107] Step 6: Quantitative prediction and verification of fractures in layered tight reservoirs. The specific methods are as follows:

[0108] (1) The above-mentioned fracture criteria and crack density mechanical model are implanted into the Ansys software platform. Based on the ancient stress field simulation, the maximum principal stress, minimum principal stress, intermediate principal stress, shear stress, maximum principal strain, minimum principal strain, intermediate principal strain, and yield strength parameters of each node are extracted. After calculation, the crack density value of each node can be obtained.

[0109] (2) Based on the above results of rock fracture state and three-dimensional distribution of fracture prediction, the accuracy is verified by single-well core statistics, imaging logging interpretation and CT scan fracture results. If the simulation results are more than 90% consistent with the actual data, the simulation results are considered reliable. Otherwise, the boundary conditions are re-analyzed and adjusted, and the three-dimensional stress field is re-simulated to finally complete the three-dimensional quantitative prediction of fractures in the layered tight reservoir structure.

[0110] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be further described in detail below with reference to the accompanying drawings.

[0111] Example: See Figure 1 It illustrates the flowchart of a quantitative prediction method for fractures in layered tight reservoirs based on a composite fracture criterion according to the present invention, including the following steps;

[0112] In step 1, the lithology and bedding type of the tight reservoir are identified and determined. In this embodiment, the deep tight sandstone reservoir of the Qigu Formation of the Jurassic system in the Yongjin Oilfield in the central Junggar Basin is selected as the study block. Seismic, logging, drilling, and testing data are collected and processed. A series of basic tasks are carried out for 14 core wells, including depth localization of core data, logging interpretation of clay content, and bedding type identification. Through core observation, fractures are classified into three main types: tectonic fractures, overpressure fractures, and bedding fractures. Figure 2 Among them, bedding fractures are the result of a reasonable configuration of large-scale bedding and favorable stress directions. The criteria for judgment are: located in a bedding development section, with a large scale of bedding, sheared into striations or mirror surfaces along the bedding plane, with a certain degree of opening and good extension, good oil content, no conchoidal fracture, and bedding / bedding fractures are easy to form at the interface of massive mudstone or banded sandstone. Therefore, based on the statistical analysis results, it is believed that the bedding types that are conducive to the formation of bedding fractures in the shallow-water deltaic sedimentary facies of the Qigu Formation in the Yongjin area are parallel bedding, flattened bedding, scour surface bedding, and rhythmic bedding. In order to reduce the workload of subsequent three-dimensional geological modeling, the bedding types of tight sandstone are classified into three categories: horizontal bedding, oblique bedding, and massive bedding. The linear density of structural fractures, overpressure fractures, and bedding fractures in the core sections of 14 wells is statistically analyzed, with the unit being fractures / m (see Table 1).

[0113]

[0114] Furthermore, based on the well logging lithology identification results, a quantitative relationship template between well logging and core samples was established. Lithology / lithofacies identification and classification of the Qigu Formation were carried out in 14 single wells. Simultaneously, lithology-seismic attribute parameters were optimized, and a lithology-seismic inversion under sedimentary facies constraints was completed with a well-seismic joint analysis and sedimentary facies constraint, based on a lithology-seismic attribute agreement of over 85% between the inverted lithology data and well logging data. A three-dimensional geological model of the Qigu Formation lithofacies in the Yongjin Oilfield was constructed using deterministic and stochastic modeling methods. Figure 3 ).

[0115] In step 2, a mechanical parameter body based on rock mechanics experiments and well-seismic integration is constructed. First, triaxial compressive strength tests are conducted on different lithofacies types according to rock mechanics experimental specifications to obtain rock mechanics parameters (elastic modulus, Poisson's ratio, compressive strength, shear strength, yield strength, internal friction coefficient, and internal friction angle, etc.). Second, based on the optimized dynamic rock mechanics parameter calculation formula, well logging curves from 14 wells are calculated to obtain single-well dynamic mechanical parameters, including lithological density, Poisson's ratio, elastic modulus, internal friction angle, and cohesion. Third, rock mechanics parameters are statistically analyzed and classified according to lithology (sandstone, mudstone), and differences in test points caused by human factors are addressed. The least squares method is applied to fit and obtain the optimal dynamic-static mechanical parameter correction regression curves for different lithologies, resulting in continuous static rock mechanics parameter curves for 14 wells. Finally, using the lithological model from step 1 as a constraint, the seismic inversion data of the static mechanical parameters is resampled into the reservoir geological model to obtain a three-dimensional heterogeneous rock mechanics parameter body for the target layer in the study area. Figure 4 ).

[0116] In step 3, a composite fracture criterion for bedding-type tight reservoirs is established, and in step 4, a mechanical characterization model for fracture density parameters is derived and constructed. Based on steps 3 and 4, the three-dimensional spatial distribution of bedding fractures can be obtained using the simulation results in step 5. In step 5, the key fracture formation period is determined and the three-dimensional paleostress field is simulated. Acoustic emission tests are conducted on core samples from wells in the study area. The well locations are preferably evenly distributed on the plane of the study area or in a quadrilateral shape. Acoustic emission experiments based on the Kaiser effect are carried out. Through data analysis, it is concluded that the Qigu Formation in the Yongjin area has mainly experienced 2-3 key tectonic movements since its deposition. Combined with the stress at the Kaiser point along a specific direction, the range of values ​​for the maximum and minimum horizontal and vertical principal stresses is determined. Figure 5Taking Yongjin 302 well as an example, the sample came from a fine sandstone sample at a depth of 5651.8m. The test results showed the highest value of paleostress experienced, namely, the vertical principal stress was 142.991MPa, the horizontal maximum principal stress was 150.64MPa, and the horizontal minimum principal stress was 126.103MPa. Furthermore, combining the regional tectonic background and tectonic trace method, the direction and phase of the maximum paleostress were determined. The results indicate that the Qigu Formation in the Yongjin area mainly experienced two phases of tectonic activity: the two phases of Yanshanian II activity and the Xishanian I-II activity (the key fracture period). During the Yanshanian II period, the Yongjin area was in a complex regional tectonic background, simultaneously subjected to southward compression from the Siberian Plate, pushing from the Kazakhstani Block, and northward compression from the northern Tianshan Mountains. Under the combined action of these three stresses, the regional compressional-torsional characteristics were particularly significant—not only did the rocks undergo intense folding deformation, but the faults also showed obvious signs of torsion due to the superposition of shearing and compression, and the tectonic pattern was deeply reshaped. By the Xishanian I period, the source of tectonic dynamics tended to be concentrated. The Yongjin area was mainly dominated by the intense collision and compression of the northern Tianshan orogenic belt. The continuous uplift and inward pushing of the northern Tianshan Mountains further enhanced the regional compressive stress, intensified fault activity, and increased the amplitude of stratigraphic folding. Figure 6 ).

[0117] Furthermore, based on the three-dimensional geomechanical model established in Ansys finite element simulation software and the paleostress test results during the key fracture formation period, the mechanical boundary of the model was set, the loading rate was set to 0.1 MPa / s, and an elastoplastic finite element three-dimensional stress field simulation was performed. This was repeated until the maximum principal stress component of the paleostress field matched the distribution trend of the secondary compression fault, the minimum principal stress component matched the distribution trend of the secondary extension fault, and the shear stress component matched the distribution trend of the strike-slip fault. Figure 7 Overall, the horizontal shear stress is strong in the northeastern part of the Yanshan Movement Phase II study area, at the intersection of north-south and east-west faults, and is dominated by compression and torsion. The stress intensity is also relatively high, which is conducive to the formation of bedding fractures. In the southern part of the Himalayan Movement Phase II study area, the compressive stress field is stronger, but it is weaker relative to the maximum principal stress and horizontal shear stress of the Yanshan Movement.

[0118] In step 6, quantitative prediction and verification of fractures in layered tight reservoirs are performed. First, the fracture criteria and fracture density mechanics model derived in steps 3 and 4 are programmed into APDL language and embedded into the Ansys software platform. Based on the paleostress field simulation in step 5, the maximum principal stress, minimum principal stress, intermediate principal stress, shear stress, maximum principal strain, minimum principal strain, intermediate principal strain, and yield strength parameters of each node are extracted. After calculation, the density value of the layered fractures at each node can be obtained. Figure 8 According to the simulation results, the bedding fractures are mainly distributed in the central and western parts of the work area, namely the area where the distributary sandbar microfacies are developed, with vertical J3q 2The most developed bedding fractures are found in the smallest layers. For example, the bedding fractures around Yong 301 well are relatively well-developed, resulting in high daily oil production and classifying it as a high-yield well. In contrast, Yongjin 3-Xie 17 well has well-developed structural fractures, but its bedding fracture density is low, leading to lower production capacity and classifying it as a low-yield well.

[0119] Based on the above predicted distribution of bedding fractures, reliability was verified using statistical results from single-well core samples. If the simulation results agree with the actual data by more than 90%, the simulation results are considered reliable. The bedding fracture density agreement in most wells reached over 90%. Figure 8 This verifies the reliability of the bedding fracture prediction results.

[0120] In summary, this invention details a workflow for quantitative prediction of fractures in layered tight reservoirs that integrates multidisciplinary data. Its core lies in: first, through a combination of well-seismic and core observation methods, precise identification and three-dimensional geological modeling of lithology, lithofacies, and bedding types are achieved. Second, through systematic rock mechanics experiments and well logging interpretation, a high-precision three-dimensional heterogeneous rock mechanics parameter body is established, providing a realistic physical field for numerical simulation. A key innovative step is the establishment of a composite fracture criterion centered on the lithology discriminant function (y-value). The most suitable mechanical criteria (such as the Mohr-Coulomb, Drucker-Prager, and Byerlee criteria) are applied to sandstone (brittle), mudstone (elastoplastic), and bedding interfaces, respectively, achieving a differentiated and precise description of the fracture behavior of layered composite rock masses. Furthermore, a mechanical characterization model of fracture density and stress-strain parameters is derived based on strain energy theory. Finally, acoustic emission experiments were used to determine the paleostress stages and directions. Finite element paleostress field simulation was performed using a three-dimensional mechanical parameter volume as input. The aforementioned composite criteria and fracture model were then incorporated to calculate the three-dimensional distribution of fracture density. High-precision verification (consistency > 90%) was performed using core fracture statistics from numerous actual wells. An example using deep tight sandstone in the Yongjin Oilfield of the Junggar Basin fully demonstrates the application process of this method and confirms that its prediction results are in good agreement with actual production dynamics (high-yield wells and low-yield wells), fully proving the effectiveness and practicality of this method.

[0121] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

[0122] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A quantitative prediction method for fractures in layered tight reservoirs based on composite fracture criteria, characterized in that, Includes the following steps: (1) Lithology and bedding type identification steps: Based on core, well logging and seismic data, identify and determine the lithology category and bedding type of tight reservoir, and obtain the lithology and bedding type identification results; (2) Mechanical parameter body construction steps: Using the identification results obtained in step (1) as constraints, a three-dimensional heterogeneous rock mechanical parameter body is obtained and established through rock mechanics experiments and well-seismic combination methods; (3) Steps for establishing the composite fracture criterion: Based on the lithology category determined in step (1), establish a lithology discrimination function, and define corresponding fracture criteria or slip criteria for sandstone, mudstone, and bedding interfaces respectively to form a composite fracture criterion; wherein, the lithology discrimination function is: ； in: Indicates sandstone or mudstone, dimensionless; Dimensionless discriminant function representing sand / mudstone classification and bedding type / massiveness; (4) Steps for constructing a crack parameter characterization model: Based on the composite fracture criterion and rock mechanics theory in step (3), construct a mechanical model characterizing the relationship between crack density and stress-strain parameters; Crack density The calculation formula is: ; Strain energy density, in J / m 3 ; The unit for characterizing the volume of a single unit is m. 3 ; This refers to the surface energy of the crack, expressed in J / m². 2 ; The formula for calculating the strain energy density ω is: ； This is the elastic modulus, expressed in GPa. Poisson's ratio is dimensionless. This represents the maximum principal strain, which is dimensionless. Indicates the intermediate principal strain, which is dimensionless; This represents the minimum principal strain, which is dimensionless. , These are the intermediate and minimum principal stresses, respectively, in MPa. , The model parameters are dimensionless and determined based on the peak strain of the rock. It is a natural constant; (5) Paleostress field simulation steps: Based on the acoustic emission experiment, the key fracture formation period and paleostress direction are determined. Based on the three-dimensional mechanical parameter body constructed in step (2), a geomechanical model is established, and finite element simulation is performed to obtain the three-dimensional paleostress field distribution of the key period. (6) Quantitative prediction and verification steps of cracks: The composite fracture criterion of step (3) and the crack parameter characterization model of step (4) are implanted into the numerical simulation platform. Based on the paleostress field distribution obtained by simulation in step (5), the three-dimensional spatial distribution of crack density is calculated, and the prediction results are verified by actual observation data.

2. The method according to claim 1, characterized in that, In step (1), the bedding types are classified into three categories: horizontal bedding, diagonal bedding, and massive bedding.

3. The method according to claim 1, characterized in that, In step (3), when At that time, the Mohr-Coulomb criterion or the Griffith criterion was used to determine the fracture of sandstone; The Mohr-Coulomb criterion is as follows: ； in: This represents the maximum principal stress, expressed in MPa. Minimum principal stress, MPa; This represents the cohesive force of the rock, measured in MPa, and can be determined experimentally. The internal friction angle of the rock is expressed in ° and can be measured experimentally. when At that time, the Drucker-Prager criterion or the Hoek-Brown criterion was used to determine the fracture of mudstone; when In this case, the slip determination of the bedding interface adopts the Byerlee criterion.

4. The application of the method as described in any one of claims 1-3 in oil and gas reservoir exploration and development, characterized in that, The predicted three-dimensional distribution of fractures is used for the evaluation of favorable zones, fracturing stimulation design, or optimization of development plans.

5. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-3.

6. A crack quantitative prediction device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-3.

7. The apparatus according to claim 6, characterized in that, It also includes an interface module for inputting core, well logging, seismic and experimental data, as well as a visualization module for outputting the three-dimensional distribution results of fracture density.

8. A tight reservoir fracture prediction system, characterized in that, include: A lithology identification module is used to perform step (1) of claim 1. A mechanical parameter construction module is used to perform step (2) of claim 1. The rupture criterion module is used to perform step (3) of claim 1. Crack characterization module, used to perform step (4) of claim 1. Stress field simulation module, used to perform step (5) of claim 1; The prediction verification module is used to perform step (6) of claim 1.

Images