Seismic high-score attribute shale reservoir small fracture classification prediction method based on structural attitude control

Abstract

The invention discloses a seismic high-score attribute shale reservoir small fracture classification prediction method based on structural attitude control, and relates to the technical field of exploration geophysics and oil and gas development. Aiming at the development requirement of shale reservoir small fracture multi-stage shunting prediction, the invention constructs a small fracture classification prediction method which takes a'high-resolution linear attribute 'method as a core technology and carries out constraint by tectonic occurrence, so that the requirements of different tectonic stage classification description of shale reservoir small fractures are met. According to high-resolution linear attribute (hereinafter referred to as high-score attribute) calculation provided by the invention, fine representation prediction of small fractures of the reservoir is effectively completed, and the fracture occurrence orientation corresponding to each construction period is constrained, so that small fracture classification of different occurrence orientations is realized. According to the prediction method, classification and accurate prediction of small-scale fractures of shale can be completed, an ideal actual drilling data verification coincidence rate is obtained, and field development requirements are met.

Description

Technical Field

[0001] The present invention relates to the field of exploration geophysics and oil and gas development technology, in particular to the field of classification prediction technology of small faults in shale reservoirs, and more specifically to a classification prediction method of small faults in shale reservoirs based on seismic high-resolution attributes controlled by structural occurrence. Background Art

[0002] In the field of oil and gas geological exploration, faults are an important research object. They not only control the formation of oil and gas reservoirs, but also serve as important channels for oil and gas migration, and have an important impact on the oil and gas production capacity of the reservoir. For shale reservoirs, faults also have an important impact on the fracturing operations in later development. Small faults in shale reservoirs often lead to well leakage and casing changes in horizontal well development, which brings great difficulties to actual development.

[0003] Geological studies have shown that small faults in shale reservoirs are controlled by large- and medium-scale faults and are distributed in the form of fault zones around larger-scale faults. Most of their scales range from 5 to 30 m, which is smaller than the resolution of most conventional seismic data, making it difficult to characterize faults based on seismic data. At the same time, due to the different formation times of the faults, they also have different impacts on actual development. Most of the faults formed by early tectonic movements are filled or completely filled; they are not helpful for shale gas development, and the faults formed by later tectonic movements are semi-filled or unfilled, which has a great impact on oil and gas storage and migration. Therefore, only by mastering the development period of the faults and classifying them can we effectively guide the exploration and development of shale gas. In summary, accurately predicting the distribution of small faults in shale reservoirs and classifying the faults formed in different periods are important tasks and difficulties in the current field of oil and gas exploration and development.

[0004] A lot of research has been done on the prediction and classification of small faults in shale reservoirs:

[0005] In 2019, Shi Xuewen and others used principal component analysis (PCA) technology to process ant body attribute slices, and combined the structural background of the work area to characterize the faults and large-scale fractures in stages. They relied on manual interpretation methods to predict the stages of obvious large-scale faults, but could not accurately predict small-scale faults.

[0006] In 2021, Xie Qinghui et al. used the occurrence-controlled ant tracking technology to predict the fracture stages in the YS1 well area of ​​the Sichuan Basin based on the statistical data of large faults. However, the method was mainly aimed at predicting fractures on the plane, and there was a lack of effective verification of the prediction results.

[0007] In 2022, Sun Xiaoqin and others used the Gaussian random method and combined it with the quality control of the ant body to construct a natural fracture prediction model for the Longmaxi Formation in southern Sichuan. She used wells and geology to verify that it basically met the characteristics of fracture development, but there were uncertainties, and the classification of fractures in different periods was not achieved.

[0008] In summary, the current prediction methods for small faults in shale reservoirs have problems such as difficulty in verification, high uncertainty, and low resolution of prediction results. In addition, it is difficult to effectively distinguish faults in different periods in actual production. Summary of the invention

[0009] In order to overcome the defects and deficiencies in the above-mentioned prior art, the present invention provides a classification prediction method for small faults in shale reservoirs based on seismic high-resolution attributes of structural occurrence control. The purpose of the present invention is to solve the difficulties in classification prediction and characterization of small faults in reservoirs in shale exploration and production in the prior art. Aiming at the development needs of multi-stage diversion prediction of small faults in shale reservoirs, the present invention constructs a classification prediction method for small faults with "high-resolution linear attributes" as the core technology and constrained by structural occurrence, so as to achieve the requirements of classification, characterization and description of small faults in shale reservoirs in different structural periods. According to the high-resolution linear attribute (hereinafter referred to as high-resolution attribute) calculation proposed by the present invention, the fine characterization prediction of small faults in the reservoir is effectively completed, and the fracture occurrence azimuth corresponding to each structural period is constrained, so as to realize the classification of small faults with different occurrence azimuths. The prediction method of the present invention can complete the classification and accurate prediction of small-scale faults in shale, and obtain a relatively ideal verification compliance rate of actual drilling data to meet the needs of on-site development.

[0010] In order to solve the above problems existing in the prior art, the present invention is implemented through the following technical solutions.

[0011] The present invention provides a classification prediction method for small faults in shale reservoirs based on seismic high-resolution attributes controlled by structural occurrence, and the classification prediction method comprises the following steps:

[0012] S1. Calculate high-resolution attributes of the three-dimensional seismic data in the study area and obtain preliminary small fault prediction results of high-resolution linear attributes I;

[0013] S2, performing denoising processing based on a directional filter on the preliminary small fracture prediction result of the high-resolution linear attribute I obtained in step S1 to obtain the processing result of the high-resolution linear attribute II;

[0014] S3, the high-resolution linear attribute II processing result obtained in step S2 is verified by actual drilling data to determine whether the compliance rate meets the experimental requirements. If the compliance rate does not meet the requirements, the relevant attribute parameters are optimized and adjusted to enhance the compliance rate of the actual drilling data, and finally a reservoir small fault prediction result with an actual drilling compliance rate exceeding 80% is obtained;

[0015] S4. Using the structural geological background of the research area, combined with drilling and logging data, determine the main tectonic movement periods in the research area, and obtain the multi-directional structural occurrence of multiple structural periods; then decompose the structural occurrence according to the fault orientation corresponding to the single-phase tectonic movement, determine the stress direction of each group of fault systems and the development direction of the corresponding small faults, and finally obtain multiple single-directional structural occurrence periods;

[0016] S5. Based on the prediction results of small reservoir fractures in step S3, the angular range of the occurrence of each single azimuth structural period is used as a constraint. Under the constraint of the occurrence range of each single azimuth structural period, high-resolution attribute azimuth angle control calculation is performed to obtain the prediction results of small fractures in each single azimuth structural period.

[0017] Further preferably, in step S1, high-resolution attribute calculation is performed on the three-dimensional seismic data of the study area, and the calculation formula is:

[0018] ; In the formula, S represents three-dimensional seismic data, i, j, k represent three network point numbers in three-dimensional space, X, Y, Z are spatial high-resolution convolution operator functions, xI, yj, zk are three network point numbers in three-dimensional space, A is the high-resolution linear attribute I to be obtained, and x, y, z are three network point numbers in three-dimensional space.

[0019] Further preferably, in step S2, the processing formula for denoising based on the directional filter is: Where A is the high-resolution linear attribute I, F is the denoising filter function, and B is the high-resolution linear attribute II output after denoising.

[0020] Further preferably, in step S3, the high-resolution linear attribute II processing result obtained in step S2 is verified by actual drilling data, specifically, the leakage positions in the actual drilling data that can indicate the actual development of small faults are verified in the plane and section.

[0021] More preferably, in step S3, if the conformity rate does not meet the requirement, the X, Y, Z spatial high-resolution convolution operator function and the directional filter function F of the high-resolution linear attribute II are adjusted until the conformity rate exceeds 80%, thereby obtaining the high-resolution linear attribute III.

[0022] Further preferably, in step S4, starting from the geological structure of the study area, the main tectonic movement periods of the study area are determined, and the multi-directional structural occurrences of multiple structural periods are obtained. Different occurrences are represented by different angle ranges θ. Assuming there are n occurrences, corresponding to θ1, θ2, ..., θ n Angular range.

[0023] Further preferably, in step S4, the structural geological background of the study area is combined with drilling and logging data to summarize the impact of the four major tectonic periods, namely, the Khankaian, Caledonian, Yanshanian and Himalayan, on the study area. According to the stress direction and characteristics of each tectonic period, the multi-directional structural occurrence of the four-phase structural superposition is screened out. Then, according to the single-phase tectonic movement, the corresponding fault orientation is decomposed to determine the stress direction of each group of fault systems and the development direction of the corresponding small faults, and finally four single-directional structural occurrence periods are obtained.

[0024] More preferably, in step S5, the angle range of the occurrence of each single-azimuth structural period is used as a constraint, and the high-resolution attribute azimuth angle control calculation is performed under the constraint of the occurrence range of each single-azimuth structural period. The calculation formula is:

[0025] In the formula, x, y, z are the network point numbers of the three-dimensional space with three sets of high-resolution attributes, P is the spatial angle range filter, θ n is the high-scoring attribute corresponding to occurrence n, that is, corresponding to θ n Distribution of small fractures in the range of occurrence angles. C(x, y, z) represents the high-score linear attribute function in step S3 whose compliance rate meets the experimental requirements.

[0026] The decomposition operation of the attitude and orientation is carried out under the constraints of each single-period tectonic movement period and the stress and fault development attitude and orientation. By setting the appropriate azimuth range, the prediction results of the small fault attitude and orientation of each unit tectonic period can be obtained.

[0027] Compared with the prior art, the beneficial technical effects brought by the present invention are as follows:

[0028] 1. Aiming at the development needs of multi-stage diversion prediction of small fractures in shale reservoirs, the present invention constructs a small fracture classification prediction method with the "high-resolution linear attribute" method as the core technology and the structural occurrence as the constraint, so as to meet the requirements of classification, characterization and description of small fractures in shale reservoirs in different structural periods. According to the high-resolution linear attribute (hereinafter referred to as high-resolution attribute) calculation proposed in the present invention, the fine characterization and prediction of small fractures in the reservoir are effectively completed, and the fracture occurrence azimuth corresponding to each structural period is constrained, so as to realize the classification of small fractures with different occurrence azimuths. The prediction method of the present invention can complete the classification and accurate prediction of small-scale fractures in shale, and obtain a relatively ideal actual drilling data verification compliance rate to meet the needs of on-site development.

[0029] 2. When the present invention calculates high-resolution attributes of the three-dimensional seismic data in the research area, by adjusting three high-resolution convolution operators, the obtained high-resolution attributes can effectively highlight the discontinuity of the seismic iso-axis, and linearly enhance the discontinuity characteristics, effectively highlighting the small-scale information implicit in the seismic data. Compared with traditional coherence and ant body attributes, it has less dependence on the quality of seismic data, the prediction results are more stable, and can effectively highlight the development of small-scale faults.

[0030] 3. In the present invention, the preliminary small fracture prediction result of the high-resolution linear attribute I obtained in step S1 is subjected to denoising processing based on a directional filter. Since the high-resolution attribute calculation has a low dependence on data quality, the preliminary fracture identification result may be affected by the noise image of the original seismic data. The preliminary processing result is subjected to denoising processing based on a directional filter to obtain high-resolution linear attribute II. The processing result of high-resolution linear attribute II effectively eliminates noise and enhances the continuity of the prediction result in the direction of fracture development.

[0031] 4. The present invention can separate small reservoir fractures of different stages and orientations, and can specifically verify the development of fractures in each stage, thus realizing the accurate characterization and prediction of the distribution of small fractures in shale reservoirs and dividing fractures of different structural stages. In actual work, the fracture stages that are favorable to development can be selected in combination with the geological characteristics of the work area, which can effectively improve the success rate of exploration and provide guidance and assistance for actual development. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of the classification prediction method of small faults in shale reservoirs based on structural occurrence control of the present invention;

[0033] Figure 2 The preliminary reservoir small fracture prediction result map is calculated for high-resolution linear attribute I;

[0034] Figure 3 This is the result diagram of high-resolution linear attribute II denoising and optimization of reservoir small fracture prediction;

[0035] Figure 4 This is the prediction result of reservoir small fractures of high-resolution linear attribute III (after optimization and adjustment of relevant parameters) and the verification diagram of actual drilling data;

[0036] Figure 5 This is a schematic diagram of the fault system of a typical tectonic period and the corresponding structural orientation;

[0037] Figure 6 This is the prediction result diagram of small faults in the multi-stage occurrence and orientation of the reservoir in the research area. DETAILED DESCRIPTION

[0038] The following is an exemplary embodiment of the present invention that helps to fully understand the claims and their equivalents in conjunction with the accompanying drawings, wherein the specific details will be considered as exemplary only, and do not limit the scope of the present invention. Therefore, those of ordinary skill in the art can make various changes and modifications to the embodiments without departing from the scope and spirit of the present invention.

[0039] Example 1

[0040] As a preferred embodiment of the present invention, refer to the attached specification Figure 1 As shown, this embodiment discloses a classification prediction method for small faults in shale reservoirs based on seismic high-resolution attributes controlled by structural occurrence, and the classification prediction method includes the following steps:

[0041] S1. Calculate high-resolution attributes of the three-dimensional seismic data in the study area and obtain preliminary small fault prediction results of high-resolution linear attributes I;

[0042] S2, performing denoising processing based on a directional filter on the preliminary small fracture prediction result of the high-resolution linear attribute I obtained in step S1 to obtain the processing result of the high-resolution linear attribute II;

[0043] S3, the high-resolution linear attribute II processing result obtained in step S2 is verified by actual drilling data to determine whether the compliance rate meets the experimental requirements. If the compliance rate does not meet the requirements, the relevant attribute parameters are optimized and adjusted to enhance the compliance rate of the actual drilling data, and finally a reservoir small fault prediction result with an actual drilling compliance rate exceeding 80% is obtained;

[0044] S4. Using the structural geological background of the research area, combined with drilling and logging data, determine the main tectonic movement periods in the research area, and obtain the multi-directional structural occurrence of multiple structural periods; then decompose the structural occurrence according to the fault orientation corresponding to the single-phase tectonic movement, determine the stress direction of each group of fault systems and the development direction of the corresponding small faults, and finally obtain multiple single-directional structural occurrence periods;

[0045] S5. Based on the prediction results of small reservoir fractures in step S3, the angular range of the occurrence of each single azimuth structural period is used as a constraint. Under the constraint of the occurrence range of each single azimuth structural period, high-resolution attribute azimuth angle control calculation is performed to obtain the prediction results of small fractures in each single azimuth structural period.

[0046] Example 2

[0047] As another preferred embodiment of the present invention, this embodiment is a further detailed description and supplement to the technical solution of the present invention based on the above-mentioned embodiment 1. In this embodiment, in step S1, high-resolution attribute calculation is performed on the three-dimensional seismic data of the research area, and the calculation formula is:

[0048]

[0049] ; In the formula, S represents three-dimensional seismic data, i, j, k represent three network point numbers in three-dimensional space, X, Y, Z are spatial high-resolution convolution operator functions, xI, yj, zk are three network point numbers in three-dimensional space, A is the high-resolution linear attribute I to be obtained, and x, y, z are three network point numbers in three-dimensional space.

[0050] The preliminary small fracture prediction results of high-resolution linear attribute I are shown in the appendix of the manual. Figure 2 As shown in (corresponding to the high-resolution linear attribute 1 in the figure), by adjusting the three high-resolution convolution operators, the obtained high-resolution attributes can effectively highlight the discontinuity of the earthquake iso-axis and linearly enhance the discontinuity characteristics, effectively highlighting the small-scale information implicit in the seismic data. Compared with traditional coherence and ant volume attributes, it has less dependence on the quality of seismic data, the prediction results are more stable, and can effectively highlight the development of small-scale faults.

[0051] Example 3

[0052] As another preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention based on the above-mentioned embodiment 1 or embodiment 2. In this embodiment, due to the low dependence of the calculation of high-resolution linear attribute I on data quality, the preliminary fault identification result may be affected by the noise of the original seismic data. The preliminary processing result is subjected to denoising processing based on the directional filter to obtain the high-resolution linear attribute II (corresponding to the appendix of the specification). Figure 3 The processing results effectively eliminate the noise and enhance the continuity of the prediction results in the direction of fracture development.

[0053] The processing formula for denoising based on directional filters is: In the formula, A is the high-score linear attribute I, F is the denoising filter function, and B is the high-score linear attribute II output after denoising. Designing such an F denoising function can remove noise in a special direction or specific random noise, and improve the actual expression characteristics of the high-score attribute.

[0054] Example 4

[0055] As another preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention on the basis of the above-mentioned embodiment 1, embodiment 2 or embodiment 3. In this embodiment, the small fracture prediction result of the high-resolution linear attribute II obtained by processing step S2 is verified by actual drilling data, and the verification is performed in plane and section through the well leakage position that can indicate the actual development of small fractures. It is judged whether the compliance rate meets the requirements of the experiment. If the compliance rate does not meet the requirements, the relevant attribute parameters are continuously optimized and adjusted to further enhance the compliance rate of the actual drilling data. Finally, the reservoir small fracture prediction result with an actual drilling compliance rate of more than 80% is obtained (the input of this step is B(x, y, z) corresponding to the formula in Example 3, and the output is expressed by C(x, y, z), corresponding to the appendix of the instruction manual) Figure 4 Specifically, if the matching rate does not meet the requirement, the X, Y, Z spatial high-resolution convolution operator function and the directional filter function F of the high-resolution linear attribute II are adjusted until the matching rate exceeds 80%, thereby obtaining the high-resolution linear attribute III.

[0056] Example 5

[0057] As another preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention on the basis of the above-mentioned embodiment 1, embodiment 2, embodiment 3 or embodiment 4.

[0058] In this embodiment, a classification prediction of the fracture period with occurrence constraints is performed based on the prediction results of small reservoir fractures. Starting from the geological structure, the main tectonic movement periods of the research area are determined, and the multi-directional structural occurrence of multiple structural periods is obtained, and different occurrences are represented by different angle ranges θ. The denoised high-resolution attributes are decomposed into structural occurrences, and the high-resolution attributes of the stress direction of each group of fracture systems are determined to achieve the division of multiple single-directional structural occurrences. In the specific implementation, the θ distribution of the structural occurrence is first determined. Assuming that there are n occurrences, corresponding to θ1, θ2, ..., θ n Angle range, such as the high-score attribute corresponding to occurrence 1 can be calculated by the following formula,

[0059]

[0060] In the formula, x, y, z are the network point numbers of the three-dimensional space of the three groups of high-scoring attributes, P is the spatial angle range filter, θ1 is the high-scoring attribute corresponding to the occurrence 1, that is, the distribution of small fractures corresponding to the occurrence angle range of θ1, and C(x, y, z) represents the high-scoring linear attribute function in step S3 with a compliance rate that meets the experimental requirements; that is, the high-scoring attribute corresponding to the occurrence 1 (corresponding to the specification appendix Figure 5 The “small fault in strike orientation 1” in the figure refers to the distribution of small faults in the corresponding strike angle range.

[0061] Furthermore, the "Occurrence Azimuth 1 Small Fault" is calculated in the above steps. By setting a suitable azimuth range, the occurrence azimuth of each single structural period can be obtained. In the same way, the "Occurrence Azimuth 2 Small Fault" to "Occurrence Azimuth n Small Fault" (as shown in the attached manual) can be calculated. Figure 6 As shown in the figure, the occurrence azimuth decomposition operation is performed under the constraints of each single-period tectonic movement period obtained in the previous step and the stress and fault development occurrence azimuth. By setting a suitable azimuth angle range, the prediction results of the occurrence azimuth of small faults in each single tectonic period can be obtained.

[0062] Example 6

[0063] According to the classification prediction method of small faults in shale reservoirs based on structural occurrence control provided by this invention, the method is demonstrated based on an example of a certain work area. The method includes the following detailed steps:

[0064] S1. First, high-resolution attribute calculation is performed on the three-dimensional seismic data in the study area to obtain the preliminary small fault prediction results of high-resolution linear attribute 1. Figure 2 As shown in the figure, high-resolution attributes can effectively obtain small-scale fault information, but due to the poor quality of seismic data acquisition in the northwest part of the work area and the low signal-to-noise ratio, the predicted small fault results are very dense, which does not conform to the actual fault development. In the southeast, there are also dense prediction results, accompanied by strong boundary effects. Therefore, it is necessary to further optimize the preliminary prediction results.

[0065] S2, further process the high-score attribute prediction results obtained in the previous step, and obtain the high-score linear attribute 2 processing results through denoising by steerable filtering. Figure 3 As shown in the figure, the processing results have greatly eliminated the noise of the initial fault prediction. The fault prediction results in the northwest and southeast of the work area are more consistent with the geological development pattern characteristics of the actual reservoir faults, and a large number of prediction illusions caused by boundary effects have been eliminated.

[0066] S3, verify the small fracture prediction results after denoising with the actual drilling leakage points ( Figure 4 ), the actual drilling results and the small fracture prediction results after denoising have good consistency on the plane. By further adjusting the relevant attribute parameters, the statistical actual drilling consistency rate reached 82.3%, achieving high-precision prediction of small fractures in the reservoir and verifying the effectiveness of the high-resolution attribute prediction method.

[0067] S4. Using the structural geological background of the work area combined with drilling and logging data, the study area was summarized to be affected by the four main tectonic periods of Xingkai, Caledonian, Yanshanian and Himalayan. According to the stress direction and characteristics of each tectonic period, the multi-directional structural occurrence of the four-stage structural superposition was screened out. Then, the structural occurrence was decomposed according to the fault orientation corresponding to the single-stage tectonic movement, and the stress direction of each group of fault systems and the development direction of the corresponding small faults were determined ( Figure 5 ), and finally four unidirectional structural occurrence periods were obtained.

[0068] S5. Based on the high-resolution linear attribute prediction results in step 3, the angle range of the occurrence of each tectonic period is used as a constraint. Under the constraint of the occurrence range of each tectonic period, the high-resolution attribute azimuth angle control calculation is performed to obtain the small faults of each tectonic period. In this study, a total of four small fault results of occurrence azimuth were screened ( Figure 6 ), which are the late Himalayan SN-trending structure, the Himalayan NE broom-shaped structure, the Yanshan NEE-trending structure, and the Xingkai period near EW-trending structural small fault. In geological analysis, the small fault structure formed in the Xingkai period has a long development time, the stress is blocked, and the fault is likely to be filled, which is not helpful for development. Therefore, it is eliminated, and the remaining three periods of small faults in the orientation have a greater impact on development.

[0069] The above processing method can separate the reservoir small faults of different stages and orientations, and can specifically verify the development of each stage of fractures, realizing the accurate characterization and prediction of the distribution of small faults in shale reservoirs and dividing the faults of different structural stages. In actual work, combining the geological characteristics of the work area to select the fracture stages that are favorable for development can effectively improve the success rate of exploration and provide guidance and assistance for actual development.

[0070] While the inventive concept has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the claims.

Claims

1. A classification prediction method for small faults in shale reservoirs based on seismic high-resolution attributes controlled by structural occurrence, characterized by: The classification prediction method includes the following steps: S1. Calculate high-resolution attributes of the three-dimensional seismic data in the study area and obtain preliminary small fault prediction results of high-resolution linear attributes I; S2, performing denoising processing based on a directional filter on the preliminary small fracture prediction result of the high-resolution linear attribute I obtained in step S1 to obtain the processing result of the high-resolution linear attribute II; S3, the high-resolution linear attribute II processing result obtained in step S2 is verified by actual drilling data to determine whether the compliance rate meets the experimental requirements. If the compliance rate does not meet the requirements, the relevant attribute parameters are optimized and adjusted to enhance the compliance rate of the actual drilling data, and finally a reservoir small fault prediction result with an actual drilling compliance rate exceeding 80% is obtained; S4. Using the structural geological background of the research area, combined with drilling and logging data, determine the main tectonic movement periods in the research area, and obtain the multi-directional structural occurrence of multiple structural periods; then decompose the structural occurrence according to the fault orientation corresponding to the single-phase tectonic movement, determine the stress direction of each group of fault systems and the development direction of the corresponding small faults, and finally obtain multiple single-directional structural occurrence periods; S5. Based on the prediction results of small reservoir fractures in step S3, the angular range of the occurrence of each single azimuth structural period is used as a constraint. Under the constraint of the occurrence range of each single azimuth structural period, high-resolution attribute azimuth angle control calculation is performed to obtain the prediction results of small fractures in each single azimuth structural period.

2. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 1, characterized in that: In step S1, high-resolution attribute calculation is performed on the three-dimensional seismic data in the study area. The calculation formula is: Where S represents three-dimensional seismic data, i, j, k represent three network point numbers in three-dimensional space, X, Y, Z are spatial high-resolution convolution operator functions, xI, yj, zk are three network point numbers in three-dimensional space, A is the high-resolution linear attribute I to be obtained, and x, y, z are three network point numbers in three-dimensional space.

3. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 1 or 2, characterized in that: In step S2, the processing formula for denoising based on the directional filter is: Where A is the high-resolution linear attribute I, F is the denoising filter function, and B is the high-resolution linear attribute II output after denoising.

4. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 3 is characterized by: In step S3, if the compliance rate does not meet the requirements, the X, Y, Z spatial high-resolution convolution operator function and the directional filter function F of the high-resolution linear attribute II are adjusted until the compliance rate exceeds 80%, thereby obtaining the high-resolution linear attribute III.

5. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 1 or 2, characterized in that: In step S3, the high-resolution linear attribute II processing result obtained in step S2 is verified by actual drilling data, specifically, the leakage position that can indicate the actual development of small faults in the actual drilling data is verified in the plane and section.

6. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 1 or 2, characterized in that: In step S4, starting from the geological structure of the study area, the main tectonic movement periods of the study area are determined, and the multi-directional structural occurrences of multiple structural periods are obtained. Different occurrences are represented by different angle ranges θ. Assuming there are n occurrences, corresponding to θ1, θ2, ..., θ n Angular range.

7. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 6 is characterized by: In step S4, the structural geological background of the study area is combined with drilling and logging data to summarize the impact of the four major tectonic periods on the study area, namely the Khankaian, Caledonian, Yanshanian and Himalayan periods. According to the stress direction and characteristics of each tectonic period, the multi-directional structural occurrence of the four-period structural superposition is screened out; then, according to the single-period tectonic movement, the structural occurrence of the corresponding fault orientation is decomposed, and the stress direction of each group of fault systems and the development direction of the corresponding small faults are determined, and finally four single-directional structural occurrence periods are obtained.

8. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 1 or 2, characterized in that: In step S5, the angle range of each unidirectional structural period is used as a constraint, and the high-resolution attribute azimuth angle control calculation is performed under the constraint of the range of each unidirectional structural period. The calculation formula is: In the formula, x, y, z are the network point numbers of the three-dimensional space with three sets of high-resolution attributes, P is the spatial angle range filter, θ n is the high-scoring attribute corresponding to occurrence n, that is, corresponding to θ n Distribution of small fractures in the range of occurrence angles. C(x, y, z) represents the high-score linear attribute function in step S3 whose compliance rate meets the experimental requirements.

9. The method for classifying and predicting small faults in shale reservoirs based on seismic high-resolution attributes and structural occurrence control according to claim 8, characterized in that: The occurrence azimuth decomposition operation is performed under the constraints of each single-period tectonic movement period and stress and fault development. By setting a suitable azimuth range, the prediction results of small faults in the occurrence azimuth of each unit tectonic period can be obtained.

Images