A method of constructing a fracture connectivity prediction model and use thereof

Abstract

The present application relates to the field of oil and gas field exploration and development, especially a structural fracture connectivity prediction model and a method thereof. By calculating the ratio of single well isotope intensity and logging natural gamma, the water injection profile type is divided; by calculating the single well rock mechanics parameters, the size and direction of the paleostress in the study area are determined in combination with the rock internal friction angle and the fracture trend. By establishing a single well geomechanics model, the paleostress size in the fracture formation period is simulated, a structural fracture connectivity database model is established, a structural fracture connectivity prediction model is established through correlation analysis, and finally the structural fracture connectivity is predicted, and the result reliability is verified by using dynamic development data. The present application proposes a structural fracture connectivity prediction model and a method thereof, and the prediction result has certain reference significance for fracture genetic mechanism interpretation, fracture connectivity prediction, geological sweet spot evaluation and other aspects.

Description

Technical Field

[0001] This invention relates to the field of oil and gas field exploration and development, and in particular to a structural fracture connectivity prediction model and its application method. Background Technology

[0002] Low-permeability reservoirs are an important type of oil and gas reservoir in my country's continental sedimentary basins. Influenced by multiple tectonic movements, low-permeability reservoirs develop structural fractures of varying scales. The study of fracture network description and fluid flow in fractured reservoirs is crucial for oil and gas exploration and development. Currently, discrete fracture network, stochastic continuum network, and channel network models have achieved some success in geological modeling of fractured rocks. However, the reservoir fracture characteristic parameters in these models cannot be directly measured; they can only be measured at the intersection of boreholes or outcrops. This leads to high uncertainty in the fracture models established by geologists and reservoir engineers. Several methods exist to evaluate the connectivity of two-dimensional fracture networks, and some methods for simulating stochastic natural fractures have been developed for describing natural fractures. Natural fracture modeling requires studying the impact of fracture density, fracture length, and fracture angle on reservoir fracturing effects. However, multiple isolated fractures exist in the reservoir and have no effect on reservoir production. The permeability of fractured reservoirs is not only controlled by the seepage capacity of individual fractures but also strongly controlled by the connectivity of the fracture network. Fracture density, occurrence dispersion, cross-layer properties, and length all affect fracture connectivity. In this invention patent, fracture connectivity is defined as an index of the ability to connect any two points in space. This parameter determines the fluid flow volume and the reservoir fracturing effect, and is a key parameter bridging the "geological sweet spot" and "engineering sweet spot" of tight sandstone reservoirs, as well as a technical challenge in fracture characterization. Due to the heterogeneity of the reservoir, the peak shape of the water absorption profile exhibits various forms such as box-shaped, bell-shaped, funnel-shaped, bell-funnel-shaped, and funnel-bell-shaped. On the other hand, due to the development of tectonic fractures, especially large-scale fractures, the peak shape of the water absorption profile can be unimodal, bimodal, and multimodal. Core observations show that the variation of the peak shape of the water absorption profile is closely related to the vertical height of the fracture. If the spatial morphology of the fracture is considered as an ellipse or rectangle, the factors affecting fracture connectivity can be analyzed using the initial water absorption profile, and a mathematical model for predicting fracture connectivity can be established. Summary of the Invention

[0003] The present invention aims to solve the above problems by providing a construction crack connectivity prediction model and its usage method, which can accurately predict the connectivity of construction cracks.

[0004] The technical solution of this invention is: a method for constructing a crack connectivity prediction model and its application, the specific steps of which are as follows: The first step is to calculate the ratio of single-well isotopic intensity to the natural gamma ray from the well logging. Using Ba 131Isotope type monitoring is used to monitor changes in water absorption profiles during oil and gas reservoir development. The ratio of two parameters, XXIT / GR, is calculated, where XXIT is the isotope intensity and GR is the natural gamma value from well logging.

[0005] The second step is to classify the water absorption profile type. Calculate the rock logging permeability, and classify the water absorption profile type by combining sedimentary characteristic parameters, core fracture scale, and the mathematical relationship between XXIT / GR and permeability.

[0006] The third step is to calculate the rock mechanics parameters of a single well. Dynamic mechanical parameters of rocks are calculated using well logging data. Combined with rock mechanics experiments, a dynamic-static mechanical parameter conversion model for rocks is established to calculate the rock mechanical parameters of a single well.

[0007] The fourth step is to determine the magnitude and direction of paleostress in the study area. By utilizing regional paleostress field information and fracture orientation, combined with the rock internal friction angle and fracture orientation, the direction of paleostress in the study area is determined; by combining the mathematical model between fracture density and paleostress and strain energy, the magnitude of paleostress at the model boundary is obtained through inversion.

[0008] The mathematical model relating fracture density to paleostress and strain energy refers to establishing the relationship between fracture parameters and stress and strain based on rock fracture conditions (Formulas 4-5), and using the optimal match between the calculated fracture density and occurrence data and the observed fracture data as constraints to invert the magnitude of paleostress at the model boundary.

[0009] (4) In formula (4), D vf Let m be the volume density of the crack in the unit cell. 2 / m 3 ; V To characterize the volume of a unit cell, m 3 ; S f m is the area of ​​the newly generated crack. 2 ; J The energy required to generate a crack per unit area was determined by triaxial compression experiments, in J / m². 2 ; The accumulated elastic strain energy density of a single element, J / m 3 ; The strain energy density required to generate a new crack, J / m 3 ; The strain energy density that must be overcome to generate a crack, J / m 3 ; The residual strain energy density, J / m 3The residual strain energy density is calculated by selecting the stress in the steady state during the stress accumulation and release process in different tectonic periods.

[0010] Using the established crack density calculation model, perpendicular to σ In a plane with two directions, when the simulated element is sufficiently small, the variation in crack spacing within the element is ignored, i.e., it is assumed that the cracks in the element are evenly distributed. The total surface area of ​​the cracks is calculated by determining the total length of the cracks within the element. Therefore, the linear density of the cracks is expressed as: (5) In formula (5), D vf Let m be the volume density of the crack within the unit cell. 2 / m 3 ; D lf is the linear density of cracks within the unit cell, in lines / m; L 1 and L 3 are the unit cells in σ 3. σ Length in the 1st direction, in meters; θ Is it the direction of the crack and σ The included angle of 1, °; σ 3. σ 1 represents the minimum principal stress and the maximum principal stress, respectively, in MPa.

[0011] The fifth step is to establish a single-well geomechanical model to simulate the magnitude of paleostress during the fracture formation period. Using the interpretation results of well logging mechanical parameters, a single-well finite element geometric model is established. The grid cell size for stress field simulation is set to match the well logging accuracy. Combined with the distribution of rock mechanical parameters around the well, a single-well finite element geomechanical heterogeneous model is established. Using the paleostress magnitude obtained by inversion, the fine stress field information of the single well is obtained, and the distributions of minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, strain energy, horizontal principal stress difference, and vertical principal stress difference are determined. The horizontal principal stress difference refers to the difference between the horizontal maximum principal stress and the horizontal minimum principal stress, and the vertical principal stress difference refers to the difference between the horizontal minimum principal stress and the horizontal minimum principal stress.

[0012] Step 6: Establish a database model for constructing crack connectivity. By utilizing stress and strain information from a single well, combined with the parameter XXIT / GR and the rock mechanics distribution of the single well, the parameters of the structural fracture connectivity database model are determined. The fluctuation coefficient, average value, minimum value, and maximum value of different parameters are calculated to establish the structural fracture connectivity database model.

[0013] The structural crack connectivity database model parameters include 11 Class I parameters: minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. The average value, minimum value, maximum value, and fluctuation coefficient of the 11 Class I parameters are calculated respectively, and a total of 44 Class II parameters are extracted.

[0014] The fluctuation coefficients of the different parameters are defined as follows: (6) In formula (6), Xcoe The fluctuation coefficient of the parameter. C i For the first parameter i One data value, m This parameter represents the number of data points. C aver This parameter is the average of all data.

[0015] Step 7: Correlation analysis to establish a structural fracture connectivity prediction model. Based on the paleostress field distribution during the crack formation period, and using the fluctuation coefficient of XXIT / GR as a constraint, correlation analysis was conducted to analyze the correlation between the fluctuation coefficient of XXIT / GR and the minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. Parameters closely related to crack connectivity were determined, and a crack connectivity prediction model was established to predict crack connectivity in the study area.

[0016] The eighth step is to predict the connectivity of structural fractures and verify the reliability of the results using dynamic development data. By using parameters closely related to the identified fracture connectivity, the three-dimensional distribution of corresponding factors is predicted, and the fracture connectivity is predicted. The reliability of the results is verified by using fracture-related water channeling and the distribution of flooded zones.

[0017] The beneficial effects of this invention are as follows: By calculating the ratio of single-well isotopic intensity to natural gamma ray logging, water absorption profile types are classified; by calculating single-well rock mechanical parameters and combining the rock internal friction angle and fracture orientation, the magnitude and direction of paleostress in the study area are determined. A single-well geomechanical model is established to simulate the paleostress magnitude during fracture formation; a tectonic fracture connectivity database model is established; through correlation analysis, a tectonic fracture connectivity prediction model is established; finally, tectonic fracture connectivity is predicted, and the reliability of the results is verified using dynamic development data. This invention patent proposes a flowchart of a tectonic fracture connectivity prediction model and its usage method, which has high practical value, low prediction cost, and strong operability. It can significantly reduce manpower and financial expenditures, and the prediction results have certain reference significance for explaining fracture formation mechanisms, predicting fracture connectivity, and evaluating geological sweet spots. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the construction of a crack connectivity prediction model and its application.

[0019] Figure 2 This is a water absorption profile of element 284 (isotope type is Ba131, XXIT is isotope intensity).

[0020] Figure 3 The peak shape of the water absorption profile in small-scale cracks and the relationship between the parameter XXIT / GR and the permeability K are shown.

[0021] Figure 4 The peak shape of the water absorption profile in a mesoscale fracture and the relationship between the parameter XXIT / GR and the permeability K are shown.

[0022] Figure 5 The peak shape of the water absorption profile in large-scale fractures and the relationship between the parameter XXIT / GR and permeability K are shown.

[0023] Figure 6 The following are triaxial compressive stress-strain curves for rocks: (A) Sample H1; (B) Sample I3.

[0024] Figure 7 The image shows the sandstone strength envelope of Well Yuan 290 at a depth of 2106.27 m.

[0025] Figure 8 (A) The dynamic and static Young's modulus relationship of rocks in the Chang 6 oil layer group of the Yuan 284 pilot test area; (B) The dynamic and static Poisson's ratio relationship of rocks in the Chang 6 oil layer group of the Yuan 284 pilot test area.

[0026] Figure 9 (A) Rose diagram of natural fracture orientation identified by imaging logging in the Yuan 284 pilot test area; (B) Rose diagram of fracture orientation of the 6 oil layer group in the Shigouyi section.

[0027] Figure 10 A detailed simulation scheme for the stress field of a single well.

[0028] Figure 11 This is a diagram analyzing the factors affecting crack connectivity.

[0029] Figure 12 The distribution of (A) minimum principal stress fluctuation coefficient; (B) maximum value distribution of minimum principal stress in a layer; (C) maximum value distribution of Young's modulus; (D) crack connectivity prediction.

[0030] Figure 13 Diagram for identifying the direction of water seepage through cracks.

[0031] Figure 14 This is a comparison chart of the water absorption profile intensity predicted by numerical simulation and the flooding type. Detailed Implementation

[0032] The specific embodiments of the present invention are described below with reference to the accompanying drawings: This invention patent uses the low-permeability sandstone of Block Yuan 284 in the Huaqing Block of the Ordos Basin as an example to illustrate the specific implementation process of the invention. The Ordos Basin is a large Mesozoic intracontinental basin superimposed on the North China Paleozoic craton platform, and is the earliest formed and longest-evolving sedimentary basin in my country. The Late Mesozoic-Cenozoic is an important stage in the formation and evolution of the Ordos Basin. The transformation of the regional tectonic stress regime has led to the formation of tectonic zones with different combinations around the basin, and the basin's geological structure has a dual nature: stability and activity. Overall, the basin's internal structure is stable, while the margins are relatively active, and the topography shows a tectonic trend of high in the north and low in the south, and steep in the west and gentle in the east. The study area is the Yuan 284 pilot experimental area in the Huaqing area, which is structurally located in the central and southern part of the Ordos Basin. The Huaqing area is located in Huachi County, Gansu Province. It is a localized uplift formed by differential compaction; generally a gently dipping westward monocline with low-amplitude, nose-like uplifts trending east-west against the monocline background; covering an area of ​​approximately 300 square kilometers, it is a loess plateau landform, covered by 100-200 m thick Quaternary loess. The terrain is complex, crisscrossed by gullies and uneven ridges. Exposed rocks are visible in the deeply incised river valleys; the ground elevation is 1350-1660 m, with a relative elevation difference of about 310 m. The oil reservoirs in the study area are mainly lithologic, belonging to the delta front lacustrine slump turbidite fan sedimentary system, with sand bodies generally trending northeast-southwest. A structural fracture connectivity prediction model and its application method are described below: The first step is to calculate the ratio of single-well isotopic intensity to the natural gamma ray from the well logging. Using Ba 131 Isotope type monitoring is used to monitor changes in water absorption profiles during oil and gas reservoir development, and the ratio of two parameters, XXIT / GR, is calculated, where XXIT is the isotope intensity and GR is the natural gamma ray value from well logging. Figure 2 ).

[0033] The second step is to classify the water absorption profile type. Calculate the rock logging permeability, and classify the water absorption profile type by combining sedimentary characteristic parameters, core fracture scale, and the mathematical relationship between XXIT / GR and permeability.

[0034] like Figure 3 As shown, in the water absorption profile of small-scale fractures, the parameter XXIT / GR is less than 5, and the parameter XXIT / GR is related to permeability. K It exhibits a two-stage change (3A) and a weak correlation ( Figure 3 B) Positive correlation ( Figure 3 C) and negative correlation ( Figure 3 D, E), Figure 3 (A, D, and E) may be related to the development of natural fractures. Comparison with core samples shows that the upper limit of the height of natural fractures corresponding to this type of water absorption profile is several tens of centimeters. Figure 2 ).

[0035] like Figure 4 As shown, in the mesoscale fracture water absorption profile, the parameter XXIT / GR is less than 10; in mudstone, there are points with XXIT / GR data greater than 3. When XXIT / GR is less than 3, the parameter XXIT / GR and K The relationship is similar to that of water absorption profiles in small-scale cracks; when the parameter XXIT / GR is greater than 3, the parameter XXIT / GR and K The lack of correlation indicates the development of medium-scale fractures. Core observations show that the height of the fractures in the cores varies from 10 cm to 100 cm.

[0036] like Figure 5 As shown, in the large-scale fracture water absorption profile, there are data points with a parameter XXIT / GR value greater than 10, and similar data points exist in the mudstone with a parameter XXIT / GR value greater than 10. When the parameter XXIT / GR is less than 3, the parameter XXIT / GR and... K The relationship is similar to the water absorption profile characteristics of small-scale cracks; when the parameter XXIT / GR is greater than 3, the parameter XXIT / GR and K The lack of correlation indicates the development of large-scale fractures, with fracture height variations exceeding 100 cm within the core. Therefore, it is believed that the fluctuation of the parameter XXIT / GR can reflect the fracture connectivity information.

[0037] The third step is to calculate the rock mechanics parameters of a single well. Dynamic mechanical parameters of rocks are calculated using well logging data. Combined with rock mechanics experiments, a dynamic-static mechanical parameter conversion model for rocks is established to calculate the rock mechanical parameters of a single well.

[0038] Rock mechanical parameters (mainly including Poisson's ratio, Young's modulus, and internal friction angle) are important basic data for studies such as paleostress field simulation, present-day geostress simulation, prediction of fracture dynamic and static parameters, and reservoir water injection pressure.

[0039] Triaxial rock mechanics experiments directly simulate the real three-dimensional stress environment underground, offering high measurement accuracy. However, due to limitations in the number and scale of sampling points, they struggle to reflect information such as the combination patterns and developmental degree of macroscopic natural fractures. Calculating the dynamic mechanical parameters of rocks using well logging data can fully consider the vertical continuity of rock mechanical parameters, and the calculation results can reflect the response of local natural fractures. The relevant formulas are as follows: (1) (2) (3) In formulas (1)~(3), E d The dynamic Young's modulus of the rock is given in MPa. μ d Let be the dynamic Poisson's ratio of the rock, which is dimensionless; ρ b Rock density interpreted for well logging, g / cm³ 3 ;Δ t p The P-wave transit time of the rock is μs / ft; Δ t s denoted as shear wave transit time of the rock, in μs / ft; φ Let be the internal friction angle of the rock, in °; Φ Porosity interpreted from well logging, %.

[0040] Representative rock cores were selected and processed into samples with flat ends, a diameter of 2.5 cm, and a length of 5.0 cm using drilling rigs and slicers. Triaxial compressive strength testing was conducted using the MTS286 rock testing system from the CNPC Exploration and Development Research Institute, according to the "Standard for Engineering Rock Mass Testing Methods (GB / T50266-99)". A mathematical model for converting dynamic and static mechanical parameters of the rock was established, laying the foundation for calculating static rock mechanical parameters from well logging data. During the triaxial compressive strength test, the rock samples were placed in a high-pressure chamber, and different confining pressures (0 MPa, 10 MPa, 20 MPa, 30 MPa) were applied around them, gradually increasing the vertical stress on the rock. The axial and radial strain values ​​of the rock samples were recorded, and the corresponding rock stress-strain curves were obtained. Nine representative rock samples were selected for triaxial compression testing in this study. Some test results are shown in Table 1, and the rock stress-strain relationship is shown in... Figure 6 As shown.

[0041] Table 1. Triaxial mechanical test data of rocks in the Chang 6 oil reservoir group of the Yuan 284 pilot test area.

[0042] like Figure 7 By fitting the rock strength envelope under different normal stress conditions, the cohesion and internal friction angle of the rock can be determined. The fitting curve for the sandstone in well Yuan 290 is as follows: τ >19.297+ σ 1tan40.03°; The fitting curve for the sandstone in Well Yuan 414 is as follows: τ >23.407+ σ 1tan44.53°; the fitting curve for mudstone in well Yuan 284 is as follows: τ >13.523+ σ 1tan39.82°. The cohesion of the rock sample from well Yuan 290 is 19.29 MPa and the internal friction angle is 40.03°; the cohesion of the rock sample from well Yuan 414 is 23.40 MPa and the internal friction angle is 44.53°; the cohesion of the rock sample from well Yuan 284 is 13.52 MPa and the internal friction angle is 39.82°.

[0043] By calibrating the dynamic mechanical parameter results from rock mechanics experiments and well logging interpretations, a rock dynamic-static mechanical parameter conversion model is established to obtain the static rock mechanics parameters. Figure 8 ).

[0044] The fourth step is to determine the magnitude and direction of paleostress in the study area. By utilizing regional paleostress field information and fracture orientation, combined with the rock internal friction angle and fracture orientation, the direction of paleostress in the study area is determined; by combining the mathematical model between fracture density and paleostress and strain energy, the magnitude of paleostress at the model boundary is obtained through inversion.

[0045] The east-west trending fissures in the study area were mainly formed during the Yanshanian period, while the northeast-trending fissures were mainly formed during the Himalayan period. Figure 9 Based on the measurement results of the internal friction angle of the rock, the direction of the maximum horizontal principal stress during the Yanshanian period was determined to be SEE109°, the direction of the minimum horizontal principal stress was NNE19°, and the intermediate principal stress was in the vertical direction; during the Himalayan period, the maximum horizontal principal stress was NNE29°, the direction of the minimum horizontal principal stress was SEE119°, and the intermediate principal stress was in the vertical direction. Figure 9 ).

[0046] The mathematical model relating fracture density to paleostress and strain energy refers to establishing the relationship between fracture parameters and stress and strain based on rock fracture conditions (Formulas 4-5), and using the optimal match between the calculated fracture density and occurrence data and the observed fracture data as constraints to invert the magnitude of paleostress at the model boundary.

[0047] The minimum horizontal principal stress during the Yanshanian period was ultimately determined to be 41.29 MPa, and the maximum horizontal principal stress was determined to be 160.58 MPa.

[0048] (4) In formula (4), D vf Let m be the volume density of the crack in the unit cell. 2 / m 3 ; V To characterize the volume of a unit cell, m 3 ; S f m is the area of ​​the newly generated crack. 2 ; J The energy required to generate a crack per unit area was determined by triaxial compression experiments, in J / m². 2 ; The accumulated elastic strain energy density of a single element, J / m 3 ; The strain energy density required to generate a new crack, J / m 3 ; The strain energy density that must be overcome to generate a crack, J / m 3 ; The residual strain energy density, J / m 3 The residual strain energy density is calculated by selecting the stress in the steady state during the stress accumulation and release process in different tectonic periods.

[0049] Using the established crack density calculation model, perpendicular to σ In a two-axis plane, when the simulated element is sufficiently small, the variation in crack spacing within the element is ignored, i.e., it is assumed that the cracks in the element are evenly distributed. The total surface area of ​​the cracks is calculated by determining the total length of the cracks within the element; therefore, the linear density of the cracks is expressed as: (5) In formula (5), D vf Let m be the volume density of the crack within the unit cell. 2 / m 3 ; D lf is the linear density of cracks within the unit cell, in lines / m; L 1 and L 3 are the unit cells in σ 3. σ Length in the 1st direction, in meters; θ Is it the direction of the crack and σ The included angle of 1, °; σ 3. σ 2 and σ 1 represents the minimum principal stress, intermediate principal stress, and maximum principal stress, respectively, in MPa.

[0050] The fifth step is to establish a single-well geomechanical model to simulate the magnitude of paleostress during the fracture formation period. Using the interpretation results of well logging mechanical parameters, a single-well finite element geometric model is established. The grid cell size for stress field simulation is set to match the well logging accuracy. Combined with the distribution of rock mechanical parameters around the well, a single-well finite element geomechanical heterogeneous model is established. Using the paleostress magnitude obtained through inversion, detailed stress field information for the single well is acquired, determining the distribution of minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, strain energy, horizontal principal stress difference, and vertical principal stress difference. The horizontal principal stress difference refers to the difference between the horizontal maximum principal stress and the horizontal minimum principal stress, and the vertical principal stress difference refers to the difference between the vertical principal stress and the horizontal minimum principal stress.

[0051] A single-well finite element geometric model was established using the interpretation results of well logging mechanical parameters. Figure 10 A) The grid cell size for the stress field simulation was set to 12.5 cm to match the vertical accuracy of the stress field simulation with the logging accuracy, taking into account the distribution of rock mechanical parameters around the well. Figure 10 B), establish a single-well finite element geomechanical heterogeneous model ( Figure 10 C and D), combined with the paleostress magnitude at the model boundary, predict the fine stress field information of a single well ( Figure 10 E).

[0052] Step 6: Establish a database model for constructing crack connectivity. By utilizing stress and strain information from a single well, combined with the parameter XXIT / GR and the rock mechanics distribution of the single well, the parameters of the structural fracture connectivity database model are determined. The fluctuation coefficient, average value, minimum value, and maximum value of different parameters are calculated to establish the structural fracture connectivity database model.

[0053] The structural crack connectivity database model parameters include 11 Class I parameters: minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. The average value, minimum value, maximum value, and fluctuation coefficient of the 11 Class I parameters are calculated respectively, and a total of 44 Class II parameters are extracted.

[0054] The fluctuation coefficients of the different parameters are defined as follows: (6) In formula (6), Xcoe The fluctuation coefficient of the parameter. C i For the first parameteri One data value, m This parameter represents the number of data points. C aver This parameter is the average of all data.

[0055] Step 7: Correlation analysis to establish a structural fracture connectivity prediction model. Based on the paleostress field distribution during the crack formation period, and using the fluctuation coefficient of XXIT / GR as a constraint, correlation analysis was conducted to analyze the correlation between the fluctuation coefficient of XXIT / GR and the minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. Parameters closely related to crack connectivity were determined, and a crack connectivity prediction model was established to predict crack connectivity in the study area.

[0056] Table 2 Correlation Analysis of Variables

[0057] In Table 2, ** indicates that the correlation is significant at a confidence level (two-test) of 0.01; * indicates that the correlation is significant at a confidence level (two-test) of 0.05.

[0058] A stepwise regression method was used to establish a crack connectivity prediction model. In constructing the prediction model for water absorption profile strength, the F-probability method was used for variable selection and removal, with a threshold of 0.05 for variables entering the model and a removal threshold of 0.1. The correlation between the number of variables and the fitting coefficients was considered, such as... Figure 11 As shown, in the water absorption profile strength prediction model, the 3-parameter prediction model (Formula 7) is selected. The correlation between fracture connectivity and the fluctuation coefficient of minimum principal stress is the strongest, which is a linear positive correlation; the correlation with the minimum value of minimum principal stress of a single well is the second strongest, which is a linear negative correlation.

[0059] The mathematical model for predicting crack connectivity is expressed as follows: (7) The fitting coefficient of formula (7) is R. 2 =0.94, Y 1 represents the fluctuation coefficient of parameter XXIT / GR. This parameter reflects the scale information of the crack and the water absorption profile strength. The larger the value, the greater the crack connectivity, and the smaller the value, the less the crack connectivity. X 1 is the minimum principal stress fluctuation coefficient, which is dimensionless; X 2 represents the maximum value of the minimum principal stress, in MPa; X 3 represents the minimum Young's modulus, in MPa. X 1~ XThe order of 3 is based on the stepwise iteration of the parameters.

[0060] The eighth step is to predict the connectivity of structural fractures and verify the reliability of the results using dynamic development data. By using parameters closely related to the identified fracture connectivity, the three-dimensional distribution of corresponding factors is predicted, and the fracture connectivity is predicted. The reliability of the results is verified by using fracture-related water channeling and the distribution of flooded zones.

[0061] In the simulation, the three sand bodies were treated as relatively independent seepage units. Using a prediction model based on the water absorption profile thickness and fracture connectivity (the fluctuation coefficient of XXIT / GR), the initial water absorption profile results for the three sand bodies were obtained. For example... Figure 12 As shown, in a length of 63 3 The minimum principal stress fluctuation coefficient is sporadically distributed in the following areas: the Yuan 308-41 well area, the Qingping 2 well area, and the Yuan 304-58 well area have high values ​​(greater than 0.014); while the Qingping 50 well area and the Yuan 301-53 well area have low values ​​(less than 0.008). Figure 12 A). The maximum value of the minimum principal stress in single wells in the study area is generally high in the central and western parts and the Yuan 433 well area, with values ​​greater than 31.7 MPa, while it is low in the southern and eastern parts of the study area, with values ​​less than 30.3 MPa. Figure 12 B). The maximum Young's modulus of the rocks in the study area is mainly distributed in the Qingping 15, Qingping 17 and Yuan 435 well areas, with values ​​greater than 23 GPa; in the southern part of the study area, the value is low, less than 21.5 GPa. Figure 12 C). High-value areas of fracture connectivity parameters are generally scattered, with high values ​​(greater than 28) in well areas such as Yuan 297-53, Yuan 308-57, and Yuan 308-41; and low values ​​(less than 10) near the Qingping 15 well area. Figure 12 D).

[0062] In the vertical well area in the northwest of Block Yuan 284, relatively mature development data is available. Dynamic data such as water intake profiles, sand body connectivity between injection and production wells, and pre-injection intensity are used to comprehensively determine the water breakthrough direction of high water-cut well groups. Figure 13 Based on the degree and area of ​​flooding, the flooding types are divided into two categories: Category I flooding covers a large area and often forms cracks with visible water zones; Category II flooding covers a small area and mostly consists of wells with cracks, without forming cracks with visible water zones. Comparison shows that the simulated water absorption profile intensity (crack connectivity) matches the type and area of ​​the flooded area well. Figure 14 When the water absorption profile strength index is greater than 24, it often leads to cracked flooding.

[0063] The present invention has been described above by way of example, but the present invention is not limited to the specific embodiments described above. Any modifications or variations made based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A method for constructing a crack connectivity prediction model, comprising the following steps: The first step is to calculate the ratio of single-well isotopic intensity to the natural gamma ray from the well logging. Using Ba 131 Isotope type monitors the change of water injection profile during the development of oil and gas reservoir, calculates the ratio of two parameters XXIT / GR, wherein, XXIT represents isotope intensity, and GR represents the natural gamma ray value from well logging. The second step is to classify the water absorption profile type. Calculate the rock logging permeability, and classify the water absorption profile type by combining sedimentary characteristic parameters, core fracture scale, and the mathematical relationship between XXIT / GR and permeability; The third step is to calculate the rock mechanics parameters of a single well. Dynamic mechanical parameters of rocks are calculated using well logging data. Combined with rock mechanics experiments, a dynamic-static mechanical parameter conversion model for rocks is established to calculate the rock mechanical parameters of a single well. The fourth step is to determine the magnitude and direction of paleostress in the study area. By utilizing regional paleostress field information and fracture orientation, combined with rock internal friction angle and fracture orientation, the direction of paleostress in the study area is determined; by combining the mathematical model between fracture density and paleostress and strain energy, the magnitude of paleostress at the model boundary is obtained through inversion. The fifth step is to establish a single-well geomechanical model to simulate the magnitude of paleostress during the fracture formation period. Using the interpretation results of well logging mechanical parameters, a single-well finite element geometric model is established. The grid cell size of the stress field simulation is set to match the well logging accuracy. Combined with the distribution of rock mechanical parameters around the well, a single-well finite element geomechanical heterogeneous model is established. Using the paleostress magnitude obtained by inversion, the fine stress field information of the single well is obtained, and the distribution of minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, strain energy, horizontal principal stress difference, and vertical principal stress difference is determined. The aforementioned difference in horizontal principal stress refers to the difference between the maximum horizontal principal stress and the minimum horizontal principal stress. The aforementioned difference in horizontal principal stress refers to the difference between the vertical principal stress and the minimum horizontal principal stress. Step 6: Establish a database model for constructing crack connectivity. Using single-well stress and strain information, combined with the parameter XXIT / GR and the rock mechanics distribution of single wells, the parameters of the structural fracture connectivity database model are determined, and the fluctuation coefficient, average value, minimum value and maximum value of different parameters are calculated to establish the structural fracture connectivity database model. Step 7: Correlation analysis to establish a structural fracture connectivity prediction model. Based on the paleostress field distribution during the crack formation period, and using the fluctuation coefficient of XXIT / GR as a constraint, correlation analysis was conducted to analyze the correlation between the fluctuation coefficient of XXIT / GR and the minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. Parameters closely related to crack connectivity were determined, and a crack connectivity prediction model was established to predict crack connectivity in the study area. The eighth step is to predict the connectivity of structural fractures and verify the reliability of the results using dynamic development data. By using parameters closely related to the identified fracture connectivity, the three-dimensional distribution of corresponding factors is predicted, and the fracture connectivity is predicted. The reliability of the results is verified by using fracture-related water channeling and the distribution of flooded zones.

2. The method for using the crack connectivity prediction model according to claim 1, characterized in that: The mathematical model between fracture density and paleostress and strain energy refers to establishing the relationship between fracture parameters and stress and strain based on rock fracture conditions, as shown in Formulas 4 and 5. The optimal match between the calculated fracture density and orientation data and the observed fracture data is used as a constraint to invert the paleostress magnitude at the model boundary. （4） In formula (4), D vf Let m be the volume density of the crack in the unit cell. 2 / m 3 ; V To characterize the volume of a unit cell, m 3 ; S f m is the area of ​​the newly generated crack. 2 ; J The energy required to generate a crack per unit area was determined by triaxial compression experiments, in J / m². 2 ; The accumulated elastic strain energy density of a single element, J / m 3 ; The strain energy density required to generate a new crack, J / m 3 ; The strain energy density that must be overcome to generate a crack, J / m 3 ; The residual strain energy density, J / m 3 The residual strain energy density is calculated by selecting the stress in the steady state during the stress accumulation and release process in different tectonic periods; Using the established crack density calculation model, perpendicular to σ In a plane with two directions, when the simulated element is sufficiently small, the variation in crack spacing within the element is ignored, i.e., it is assumed that the cracks in the element are evenly distributed. The total surface area of ​​the cracks is calculated by determining the total length of the cracks within the element. Therefore, the linear density of the cracks is expressed as: （5） In formula (5), D vf Let m be the volume density of the crack within the unit cell. 2 / m 3 ; D lf is the linear density of cracks within the unit cell, in lines / m; L 1 and L 3 are the unit cells in σ 3. σ Length in the 1st direction, in meters; θ Is it the direction of the crack and σ The included angle of 1, °; σ 3. σ 2 and σ 1 represents the minimum principal stress, intermediate principal stress, and maximum principal stress, respectively, in MPa.

3. The method for using the crack connectivity prediction model according to claim 1, characterized in that: The structural crack connectivity database model parameters include 11 Class I parameters: minimum principal stress, intermediate principal stress, maximum principal stress, minimum principal strain, intermediate principal strain, maximum principal stress, Young's modulus, Poisson's ratio, strain energy, horizontal principal stress difference, and vertical principal stress difference. The average value, minimum value, maximum value, and fluctuation coefficient of the 11 Class I parameters are calculated respectively, and a total of 44 Class II parameters are extracted.

4. The method for using the crack connectivity prediction model according to claim 1, characterized in that: The fluctuation coefficients of the different parameters are defined as follows: （6） In formula (6), Xcoe The fluctuation coefficient of the parameter. C i For the first parameter i One data value, m This parameter represents the number of data points. C aver This parameter is the average of all data.

Images