Seam-vuggy inner curtain cluster structure earthquake prediction method and system

Abstract

The invention provides a fracture-cavity inner curtain cluster structure earthquake prediction method and system, and belongs to the field of oil-gas exploration reservoir prediction. The method comprises the steps of firstly determining an inner curtain range of a fracture-cavity body, then predicting boundary characteristics of a cluster structure of an inner curtain of the fracture-cavity body by adopting a small surface element waveform difference analysis method, and finally determining a development position of the cluster structure by adopting a transverse amplitude plane attribute characteristic as a constraint. Aiming at the problem of fracture-vuggy inner curtain structure prediction, a fracture-vuggy inner curtain cluster structure earthquake prediction technology based on combined constraint of longitudinal waveform difference and transverse amplitude plane change is developed, the prediction precision of the inner curtain cluster structure is improved, the prediction reliability and stability are improved, and efficient exploration and development work of oil and gas reservoirs is assisted; based on the idea of weakening trace equalization processing and improving the characterization capability of single-trace seismic data, a small-surface-element attribute calculation method is proposed, and the small-scale information prediction precision is improved.

Description

Technical Field

[0001] The invention belongs to the field of oil and gas exploration reservoir prediction, and in particular relates to a method and system for seismic prediction of clustered structures in fracture-cavity bodies. Background Art

[0002] With the continuous advancement of the exploration and development of the national "Deep Earth No. 1" project-Shunbei Oil and Gas Field, drilling has revealed the differences in the curtain reservoir structure within the fracture-cavity body, showing the characteristics of a clustered structure.

[0003] Chinese patent publication CN109959965A discloses a method and device for obtaining the internal structure of carbonate fracture-cavity bodies, the method comprising: obtaining the external beaded seismic phase of carbonate fracture-cavity bodies through preset features; obtaining the porosity threshold of the target layer inside the carbonate fracture-cavity bodies; processing seismic data through seismic inversion technology to obtain the porosity data volume of the cave-type reservoir inside the carbonate fracture-cavity bodies and the porosity data volume of the pore-type reservoir; obtaining the porosity data volume of the fracture-type reservoir inside the carbonate fracture-cavity bodies through seismic data; determining the spatial structure of the fracture-cavity bodies according to the external beaded seismic phase of carbonate fracture-cavity bodies, the porosity threshold of the target layer inside the carbonate fracture-cavity bodies, the porosity data volume of the cave-type reservoir, the porosity data volume of the pore-type reservoir and the porosity data volume of the fracture-type reservoir. The method and device for obtaining the internal structure of carbonate fracture-cavity bodies provided by the present invention obtain the spatial structure of the fracture-cavity bodies.

[0004] Chinese patent publication CN107272064A discloses a method for characterizing the internal structure of carbonate rock fractures and caves, the method comprising: obtaining seismic data corresponding to the fractures and caves based on existing geological knowledge; and using spectral decomposition of the seismic data to apply high-frequency information to characterize the internal structure of the fractures and caves.

[0005] In June 2011, the Chinese journal Xinjiang Petroleum Geology published the quantitative description technology and application of carbonate fracture-cave bodies in the Yingshan Formation on the northern slope of Tazhong. A large number of forward modeling studies were carried out for various reservoir geological models, and the corresponding geophysical response characteristics were analyzed. The wave impedance inversion prediction method was optimized and converted into a porosity body, which can obtain the effective storage space of the reservoir and finally realize the quantitative description of the effective storage space of karst fracture-cave bodies. The application of actual data has verified that the quantitative description of fracture-cave bodies is one of the effective ways to efficiently explore and develop carbonate oil and gas reservoirs.

[0006] In August 2022, the Chinese journal Petroleum Geophysical Exploration published a paper on the use of gradient structure tensors to characterize the boundaries of carbonate fractures and caves. The paper discussed and analyzed the geometric significance of the gradient structure tensor attributes and the influence of each calculation link in detail. It is believed that the smoothing of tensor elements is the key to characterizing fractures and caves, and its second and third eigenvalues ​​have a certain ability to characterize carbonate fractures and caves. In short, compared with conventional attributes such as amplitude change rate, gradient structure tensor attributes can more continuously characterize carbonate fractures and caves.

[0007] In January 2023, the Chinese journal "Oil and Gas Geology and Recovery" published a fracture-cavity reservoir prediction method based on well control multi-attribute machine learning. Based on the actual drilling well seismic calibration, the different sensitive seismic attribute values ​​at the leakage point location are extracted as the data input array, and the reservoir type defined according to the leakage point characteristics is used as the output array to form the training set data; then the support vector machine method is used to train the training set data model to obtain a prediction model consistent with the well seismic information in accordance with the prior information; finally, a fracture-cavity reservoir prediction method based on well control multi-attribute machine learning is proposed. In actual data application, this method can well reflect the characteristics of the real reservoir type and has a high degree of consistency with the drilling characteristics.

[0008] The current fracture-cavity characterization technology is based on the idea of ​​large-scale characterization. For example, energy attributes can only characterize the macroscopic fracture-cavity outline, but cannot effectively describe the internal structural characteristics of the fracture-cavity body. This brings greater difficulties to the subsequent quantitative description of the fracture-cavity body and the well trajectory design, and the effective prediction of the internal cluster structure becomes more important. Summary of the invention

[0009] The purpose of the present invention is to solve the difficulties existing in the above-mentioned prior art and provide a method and system for seismic prediction of curtain cluster structures in fracture-cavern bodies. The original seismic data is used to qualitatively characterize the overall contour of the fracture-cavern body, and then based on the idea of ​​small facet attribute calculation, a method of joint constraint of longitudinal waveform difference analysis and transverse amplitude plane attribute characteristics is adopted to finally determine the development position of the cluster structure, thereby realizing a fine prediction of the curtain cluster structure in the fracture-cavern body, with a high well-seismic coincidence rate, and achieving good application effect in actual data application.

[0010] The present invention is achieved through the following technical solutions:

[0011] The first aspect of the present invention provides a method for seismic prediction of clustered structures of the interior of a fracture-cavity body. The method first determines the interior range of the fracture-cavity body, then uses a small-surface waveform difference analysis method to predict the boundary characteristics of the clustered structures of the interior of the fracture-cavity body, and then uses the lateral amplitude plane attribute characteristics as constraints to finally determine the development position of the clustered structures.

[0012] A further improvement of the present invention is:

[0013] The method comprises:

[0014] The first step is to obtain the original seismic data volume;

[0015] The second step is to use the original seismic data to calculate energy attributes and predict the contour of the fracture-cavity body. Based on the joint analysis of well and seismic data, the threshold of energy attributes is analyzed to determine the inner range of the fracture-cavity body.

[0016] The third step is to carry out the small-surface longitudinal waveform difference analysis within the fracture-cavity body to determine the boundary information of the inner cluster structure;

[0017] The fourth step is to use the lateral amplitude plane attribute feature constraints to comprehensively determine the developmental position of the inner cluster structure.

[0018] A further improvement of the present invention is:

[0019] The second step is to use the original seismic data to calculate the energy attributes and predict the outline of the fracture-cavity body. According to the joint analysis of well and seismic data, the threshold of the energy attribute is analyzed to determine the inner range of the fracture-cavity body. Specifically:

[0020] Firstly, the sensitive attributes of coherent energy gradient, instantaneous energy and tensor characterizing the contour of fracture-cavity body are extracted from the original seismic data volume to realize the contour prediction of fracture-cavity body. Then, the energy attribute values ​​at the locations of emptying and leakage points of multiple wells, logging type I and II reservoirs, etc. are statistically analyzed, and the average value is taken as the threshold to determine the inner range of fracture-cavity body.

[0021] A further improvement of the present invention is:

[0022] Formula (1) is used to extract tensor attributes:

[0023]

[0024] In the formula, λ i is a tensor attribute, is the gradient of the seismic data s, and eig(*) is the eigenvalue of the matrix *.

[0025] A further improvement of the present invention is:

[0026] Formula (2) is used to extract the instantaneous energy properties:

[0027] E ins (t) = s(t) ∧ 2+H[s(t)] ∧ twenty two)

[0028] In the formula, E ins (t) is the instantaneous energy attribute of the seismic data s(t), and H[] is the Hilbert transform.

[0029] A further improvement of the present invention is:

[0030] Formula (3) is used to extract the threshold:

[0031]

[0032] In the formula, A i is the energy attribute value at the i-th point, V th To determine the threshold.

[0033] A further improvement of the present invention is:

[0034] The third step is to carry out the small-surface longitudinal waveform difference analysis within the fracture-cavity body to determine the boundary information of the inner cluster structure, specifically:

[0035] First, the relevant time window size W and the bin size are set. Then, at each sample point within the fracture-cavity threshold range, the Pearson correlation coefficient between the central seismic trace and the adjacent trace is calculated in turn using formula (4):

[0036]

[0037] In the formula, X, Y are vectors, and the covariance is:

[0038] cov(X,Y)=E((XE(X))(YE(Y)))=E(XY)-E(X)E(Y),

[0039] The standard deviation is:

[0040] The correlation between the waveforms of adjacent channels is calculated, and the multi-channel average result or the least correlated result of formula (5) is subsequently used as the final correlation coefficient of the sampling point in the bin;

[0041]

[0042] In the formula, ρ j is the Pearson correlation coefficient between the center track and j adjacent tracks, and L is the number of adjacent tracks;

[0043] At the same time, prior information such as logging interpretation conclusions is used to determine the correlation coefficient threshold that reflects the boundary information of the inside cluster structure.

[0044] A further improvement of the present invention is:

[0045] Step 4: Use the lateral amplitude plane attribute feature constraints to comprehensively determine the development location of the inner cluster structure, specifically:

[0046] In view of the lateral characteristics of the internal cluster structure of the fracture-cave body, the forward modeling conclusion of the fracture-cave body model containing the internal cluster reservoir is used, that is, the internal cluster reservoir structure will have a waveform "pull-up" phenomenon, and the gradient data is obtained by using formula (6) for the isochronous slices or isodepth slices of the seismic data;

[0047]

[0048] In the formula, S amp (t,l) is the earthquake amplitude at the lth trace and time t in the time window;

[0049] After that, the isochronous or isobath slices with the most obvious inside features are selected, and the amplitude gradient is extracted at the boundary of the inside cluster structure determined in the third step. Then, the gradient variation characteristics of the inside boundaries are analyzed in pairs. When the gradient at the first boundary is positive and the gradient at the second boundary is negative, the data within the two boundaries are set to 1 using formula (7), representing the inside reservoir development location, and the rest are set to 0, representing the non-reservoir (or surrounding rock) location, so that the inside cluster structure development location can be determined.

[0050]

[0051] Where IS is the prediction result of the insider cluster structure at time t, e1 and e2 are the first and second insider boundaries after sequential combination.

[0052] A second aspect of the present invention provides a system for predicting earthquakes of clustered structures in fracture-cavity bodies, the system comprising:

[0053] Data acquisition unit: used to acquire original seismic data volume;

[0054] Fracture-cavity range determination unit: connected to the data acquisition unit, used to use the original seismic data body to calculate energy attributes, predict the outline of the fracture-cavity body, analyze the threshold of energy attributes based on the joint analysis of well and seismic data, and thus determine the inner range of the fracture-cavity body;

[0055] Boundary information determination unit: connected to the fracture-cavity range determination unit, used to carry out small-surface longitudinal waveform difference analysis within the fracture-cavity range to determine the boundary information of the inner cluster structure;

[0056] Comprehensive determination unit: connected to the boundary information determination unit, used to comprehensively determine the development position of the inner cluster structure by using the lateral amplitude plane attribute feature constraint.

[0057] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps in the method for predicting earthquakes of clustered structures within fracture-cavity bodies.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] In order to solve the problem of predicting the internal curtain structure of fracture-cavern bodies, the present invention has developed a seismic prediction technology for the internal curtain cluster structure of fracture-cavern bodies under the joint constraints of longitudinal waveform difference and lateral amplitude plane change, which can improve the prediction accuracy of the internal curtain cluster structure, improve the reliability and stability of the prediction, and facilitate the efficient exploration and development of oil and gas reservoirs.

[0060] Based on the idea of ​​weakening trace equalization processing and improving the characterization ability of single-channel seismic data, the present invention proposes to adopt a small-surface attribute calculation method to improve the prediction accuracy of small-scale information. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 The present invention provides a flow chart of a method for predicting earthquakes of clustered structures in fracture-cavity bodies;

[0062] Figure 2 It is a geological model diagram of fracture-cavity body;

[0063] Figure 3 It is the migration imaging section near the fracture-cavity body;

[0064] Figure 4 It is the instantaneous amplitude property diagram of the fracture-cavity body;

[0065] Figure 5 is the correlation coefficient diagram at the fracture-cavity body;

[0066] Figure 6 is the lateral amplitude gradient map at the inner boundary;

[0067] Figure 7 It is a superposition diagram of isoseismic waveform and internal structure prediction results;

[0068] Figure 8 It is an overlay of the earthquake profile and the internal structure prediction results. DETAILED DESCRIPTION

[0069] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0070] The present invention adopts the idea of ​​weakened trace equalization processing to reduce the facets of attribute calculation, and proposes a seismic prediction technology for the internal cluster structure of fracture-cavity bodies that is jointly constrained by longitudinal waveform differences and transverse amplitude plane changes, thereby improving the characterization capability of the internal cluster structure.

[0071] The present invention utilizes the original seismic data and energy-based attributes to characterize the overall contour of the fracture-cavity body, and then sets the energy threshold as the boundary of the fracture-cavity body to determine the interior range of the fracture-cavity body; within this range, the small-surface waveform difference analysis method is used to predict the boundary characteristics of the clustered structure of the interior of the fracture-cavity body, and then the lateral amplitude variation characteristics are used as constraints to finally determine the development position of the clustered structure, thereby achieving a fine prediction of the clustered structure of the interior of the fracture-cavity body.

[0072] The present invention provides a method for predicting earthquakes in clustered structures in fracture-cavity bodies. Figure 1 As shown, the main contents mainly include three parts: energy attribute calculation and threshold determination, longitudinal waveform difference analysis and transverse amplitude plane change feature analysis, and cluster structure prediction. The method embodiment is as follows:

[0073] [Example 1]

[0074] The method specifically comprises the following steps:

[0075] The first step is to obtain the original seismic data volume;

[0076] The second step is to use the original seismic data to calculate energy attributes and predict the contour of the fracture-cavity body. Based on the joint analysis of well and seismic data, the threshold of energy attributes is analyzed to determine the inner range of the fracture-cavity body.

[0077] Firstly, the sensitive attributes of fracture-cavity contour characterization, such as coherent energy gradient, instantaneous energy, tensor, etc., are extracted from the original seismic data volume to realize the contour prediction of fracture-cavity body. Then, the threshold of energy attributes characterizing fracture-cavity contour is calibrated based on the location of emptying and leakage points of actual drilling wells, the division of Class I and Class II reservoirs in well logging, etc. (by statistically analyzing the energy attribute values ​​at the emptying and leakage points of multiple actual drilling wells, Class I and Class II reservoirs in well logging, the average value can be taken as the threshold value, and this process is called calibration process). The threshold value can be the average value of the calibration threshold values ​​of multiple wells according to the actual situation, so as to determine the inner range of fracture-cavity body.

[0078] The tensor attribute formula is (Formula 1):

[0079]

[0080] In the formula, λ i is a tensor attribute, is the gradient of seismic data s, eig(*) is the eigenvalue of matrix *;

[0081] The instantaneous energy property formula is (Formula 2):

[0082] E ins (t) = s(t) ∧ 2+H[s(t)] ∧ twenty two)

[0083] In the formula, E ins (t) is the instantaneous energy attribute of the seismic data s(t), and H[] is the Hilbert transform;

[0084] The threshold determination formula is (Formula 3):

[0085]

[0086] In the formula, A i is the energy attribute value at the i-th point, V th To determine the threshold.

[0087] The third step is to carry out the small-surface longitudinal waveform difference analysis within the fracture-cavity body to determine the boundary information of the inner cluster structure;

[0088] Within the beaded contour, the Pearson correlation coefficient of the small facets at each channel and each sample point is calculated cyclically to highlight the longitudinal waveform difference characteristics, and its threshold is determined to characterize the boundary information of the inner cluster.

[0089] Specifically:

[0090] To carry out the small-surface longitudinal waveform difference analysis, first set the relevant time window size W (generally equivalent to the size of the beads) and the bin size (generally 3×3). Then, at each sample point within the threshold range of the fracture-cavity body, calculate the Pearson correlation coefficient between the central seismic trace and the adjacent trace in turn (as shown in Formula 4):

[0091]

[0092] In the formula, X, Y are vectors, and the covariance is:

[0093] cov(X,Y)=E((XE(X))(YE(Y)))=E(XY)-E(X)E(Y),

[0094] The standard deviation is:

[0095] To calculate the correlation between waveforms of adjacent channels, the average result of multiple channels or the least correlated result (such as formula 5) can be used as the final correlation coefficient of the sampling point in the bin;

[0096]

[0097] In the formula, ρ j is the Pearson correlation coefficient between the center track and j adjacent tracks, and L is the number of adjacent tracks;

[0098] At the same time, prior information such as logging interpretation conclusions is used to determine the correlation coefficient threshold that reflects the boundary information of the inside cluster structure (by statistically analyzing the correlation coefficient threshold values ​​at the emptying points of multiple wells, the first and second type reservoirs in logging, etc., the average value can be taken as the threshold).

[0099] Step 4: Using the lateral amplitude plane attribute feature constraints, comprehensively determine the developmental position of the inner cluster structure;

[0100] Specifically:

[0101] In order to analyze the lateral characteristics of the internal curtain cluster structure of the fracture-cave body, a fracture-cave body model containing internal curtain cluster reservoir (similar to Figure 2 ) The forward modeling conclusion is that the inner cluster reservoir structure will have a waveform "pull-up" phenomenon, and the gradient data is obtained from the isochronous slices or isodepth slices of the seismic data (such as formula 6);

[0102]

[0103] In the formula, S amp (t,l) is the earthquake amplitude at the lth trace and time t in the time window;

[0104] Afterwards, the isochronous or isobath slices with the most obvious inside features are selected, and the amplitude gradient is extracted at the boundary of the inside cluster structure determined above. Then, the gradient change characteristics of the inside boundaries are analyzed in pairs. When the gradient at the first boundary is positive and the gradient at the second boundary is negative, the data within the two boundaries are set to 1, representing the inside reservoir development location, and the rest are set to 0 (as shown in Formula 7), representing the non-reservoir (or surrounding rock) location, so that the development location of the inside cluster structure can be determined.

[0105]

[0106] Where IS is the prediction result of the insider cluster structure at time t, e1 and e2 are the first and second insider boundaries after sequential combination.

[0107] The fifth step is to use the results of the above-mentioned vertical and horizontal joint constraints, that is, "determine the inner cluster boundary by the longitudinal waveform difference and determine the inner cluster position by the horizontal amplitude attribute" to obtain the spatial three-dimensional cluster structure development characteristics, providing a reliable basis for subsequent comprehensive evaluation.

[0108] The embodiments of the method of the present invention are as follows:

[0109] [Example 2]

[0110] The present invention carries out research and application work on the prediction of the interior structure of carbonate fracture-cavern in the Tarim Basin.

[0111] First, a fracture-cavity model with internal structures was established based on the actual drilling information of carbonate rocks in the Tarim Basin. Figure 2 As shown. Among them, the surrounding rock velocity is 6000m / s; the thickness of the two clusters of fractures and caves is 50m, the width is 50m and 30m respectively, the velocity is 4000m / s, and the interval between the two clusters is 10m. Afterwards, through the wave equation forward modeling technology, the 25Hz main frequency wavelet is used for forward modeling to obtain the shot gather record, and the CMP gather is extracted from it, and the pre-stack depth migration algorithm is used to accurately image it; from the analysis, it can be seen that the seismic response of the inner cluster reservoir structure will have a waveform "pull-up" phenomenon, and the imaging results of the local amplification near the fracture and cave are shown in the attached figure. Figure 3 shown.

[0112] Then, the instantaneous amplitude attribute is extracted from the migration imaging data to realize the contour characterization of the fracture-cavity body, and the threshold of the instantaneous amplitude attribute is calibrated in combination with the prior information to determine the contour range of the fracture-cavity body, as shown in the attached figure. Figure 4 Then, within the bead outline, the Pearson correlation coefficient of each small facet at each sample point is calculated cyclically to highlight the longitudinal waveform difference characteristics, and its threshold is determined to characterize the boundary information of the inner cluster. The results are shown in the attached figure. Figure 5 As shown in the figure; then, for multiple isobath slices of the fracture-cave seismic data, the amplitude gradient data is obtained at its inner cluster boundary, and then the plane variation characteristics of its lateral amplitude are extracted. The results are shown in the attached figure. Figure 6 shown.

[0113] Finally, the isobath slice with the most obvious internal features, i.e., at a depth of 1460 m, was selected, and the variation characteristics of the lateral amplitude gradient of the slice were combined and analyzed to determine the location of the internal cluster structure. The seismic waveform of the slice and the prediction results of the internal structure are shown in the attached figure. Figure 7 As shown in the figure, the result is projected onto the seismic data, i.e., the dotted rectangular box. By combining the longitudinal waveform difference analysis and the lateral amplitude plane feature constraints, the final internal structure prediction result can be obtained, as shown in the attached figure. Figure 8 As shown, the prediction results are Figure 2 The cluster reservoirs designed in the model have consistent development characteristics, and the prediction results are highly reliable.

[0114] The present invention also provides a fracture-cavity internal curtain cluster structure earthquake prediction system, and the embodiments of the system are as follows:

[0115] [Example 3]

[0116] The system comprises:

[0117] Data acquisition unit: used to acquire original seismic data volume;

[0118] Fracture-cavity range determination unit: connected to the data acquisition unit, used to use the original seismic data body to calculate energy attributes, predict the outline of the fracture-cavity body, analyze the threshold of energy attributes based on the joint analysis of well and seismic data, and thus determine the inner range of the fracture-cavity body;

[0119] Boundary information determination unit: connected to the fracture-cavity range determination unit, used to carry out small-surface longitudinal waveform difference analysis within the fracture-cavity range to determine the boundary information of the inner cluster structure;

[0120] Comprehensive determination unit: connected to the boundary information determination unit, used to comprehensively determine the development position of the inner cluster structure by using the lateral amplitude plane attribute feature constraint.

[0121] The present invention also provides a computer-readable storage medium, and an embodiment of the computer-readable storage medium is as follows:

[0122] [Example 4]

[0123] The computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps in the method for predicting earthquakes of clustered structures in fracture-cavity bodies.

[0124] The above technical solution is only one implementation mode of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the principles disclosed in the present invention, and it is not limited to the technical solution described in the above specific embodiments of the present invention. Therefore, the above description is only preferred and does not have a restrictive meaning.

Claims

1. A method for predicting earthquakes of clustered structures in fracture-cavity bodies, characterized in that: First, the inner range of the fracture-cavity body is determined. Then, the small-surface waveform difference analysis method is used to predict the boundary characteristics of the clustered structure inside the fracture-cavity body. Finally, the lateral amplitude plane attribute characteristics are used as constraints to finally determine the development position of the clustered structure.

2. The earthquake prediction method for clustered structure in fracture-cavity body according to claim 1 is characterized in that: The method comprises: The first step is to obtain the original seismic data volume; The second step is to use the original seismic data to calculate energy attributes and predict the contour of the fracture-cavity body. Based on the joint analysis of well and seismic data, the threshold of energy attributes is analyzed to determine the inner range of the fracture-cavity body. The third step is to carry out the small-surface longitudinal waveform difference analysis within the fracture-cavity body to determine the boundary information of the inner cluster structure; The fourth step is to use the lateral amplitude plane attribute feature constraints to comprehensively determine the developmental position of the inner cluster structure.

3. The method for predicting earthquakes of clustered structures in fracture-cavity bodies according to claim 2, characterized in that: The second step is to use the original seismic data to calculate the energy attributes and predict the outline of the fracture-cavity body. According to the joint analysis of well and seismic data, the threshold of the energy attribute is analyzed to determine the inner range of the fracture-cavity body. Specifically: Firstly, the sensitive attributes of coherent energy gradient, instantaneous energy and tensor characterizing the contour of fracture-cavity body are extracted from the original seismic data volume to realize the contour prediction of fracture-cavity body. Then, the energy attribute values ​​at the emptying points of multiple wells and the locations of Class I and Class II reservoirs in well logging are statistically analyzed, and the average value is taken as the threshold to determine the inner range of fracture-cavity body.

4. The earthquake prediction method for clustered structures in fracture-cavity bodies according to claim 3 is characterized in that: Formula (1) is used to extract tensor attributes: In the formula, λ i is a tensor attribute, ▽s is the gradient of the seismic data s, and eig(*) is used to find the eigenvalue of the matrix *.

5. The method for predicting earthquakes of clustered structures in fracture-cavity bodies according to claim 4, characterized in that: Formula (2) is used to extract the instantaneous energy properties: E ins (t)=s(t) ∧ 2+H[s(t)] ∧ 2 (2) In the formula, E ins (t) is the instantaneous energy attribute of the seismic data s(t), and H[] is the Hilbert transform.

6. The earthquake prediction method for clustered structure of fracture-cavity body according to claim 5, characterized in that: Formula (3) is used to extract the threshold: In the formula, A i is the energy attribute value at the i-th point, V th To determine the threshold.

7. The earthquake prediction method for clustered structure in fracture-cavity body according to claim 6, characterized in that: In the third step, within the fracture-cavity body, small-surface longitudinal waveform difference analysis is carried out to determine the boundary information of the inner cluster structure, specifically: First, the relevant time window size W and the bin size are set. Then, at each sample point within the fracture-cavity threshold range, the Pearson correlation coefficient between the central seismic trace and the adjacent trace is calculated in turn using formula (4): In the formula, X, Y are vectors, and the covariance is: cov(X,Y)=E((XE(X))(YE(Y)))=E(XY)-E(X)E(Y), The standard deviation is: The correlation between the waveforms of adjacent channels is calculated, and the multi-channel average result or the least correlated result of formula (5) is subsequently used as the final correlation coefficient of the sampling point in the bin; In the formula, ρ j is the Pearson correlation coefficient between the center track and j adjacent tracks, and L is the number of adjacent tracks; At the same time, prior information such as logging interpretation conclusions is used to determine the correlation coefficient threshold that reflects the boundary information of the inside cluster structure.

8. The method for predicting earthquakes of clustered structures in fracture-cavity bodies according to claim 7, characterized in that: In the fourth step, the development position of the inner cluster structure is comprehensively determined by using the lateral amplitude plane attribute feature constraints, specifically: In view of the lateral characteristics of the internal cluster structure of the fracture-cave body, the forward modeling conclusion of the fracture-cave body model containing the internal cluster reservoir is used, that is, the internal cluster reservoir structure will have a waveform "pull-up" phenomenon, and the gradient data is obtained by using formula (6) for the isochronous slices or isodepth slices of the seismic data; In the formula, S amp (t,l) is the earthquake amplitude at the lth trace and time t in the time window; After that, the isochronous or isobath slices with the most obvious inside features are selected, and the amplitude gradient is extracted at the boundary of the inside cluster structure determined in the third step. Then, the gradient change characteristics of the inside boundaries are analyzed in pairs. When the gradient at the first boundary is positive and the gradient at the second boundary is negative, formula (7) is used, and the data within the two boundaries are set to 1, representing the inside reservoir development location, and the rest are set to 0, representing the non-reservoir (or surrounding rock) location, so that the inside cluster structure development location can be determined; Where IS is the prediction result of the insider cluster structure at time t, e1 and e2 are the first and second insider boundaries after sequential combination.

9. A fracture-cavity internal curtain cluster structure earthquake prediction system, characterized in that: The system comprises: Data acquisition unit: used to acquire original seismic data volume; Fracture-cavity range determination unit: connected to the data acquisition unit, used to use the original seismic data body to calculate energy attributes, predict the outline of the fracture-cavity body, analyze the threshold of energy attributes based on the joint analysis of well and seismic data, and thus determine the inner range of the fracture-cavity body; Boundary information determination unit: connected to the fracture-cavity range determination unit, used to carry out small-surface longitudinal waveform difference analysis within the fracture-cavity range to determine the boundary information of the inner cluster structure; Comprehensive determination unit: connected to the boundary information determination unit, used to comprehensively determine the development position of the inner cluster structure by using the lateral amplitude plane attribute feature constraint.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps of the method for predicting earthquakes of clustered structures in fracture-cavity bodies as described in any one of claims 1 to 8.

Images