Accurate seismic structure information extraction method

Abstract

The invention provides a seismic structure information accurate extraction method, and the method comprises the steps: 1, generating time difference data dT and trace difference data dA according to a post-stack seismic record; step 2, calculating time difference square data dT2, lane difference square data dA2 and time lane difference square data dTA; step 3, smooth time difference data # imgabs0 # smooth trace difference data # imgabs1 # and smooth time trace difference data # imgabs2 # are calculated; step 4, a Hessian matrix is constructed according to the smooth time difference data # imgabs3 # smooth trace difference data # imgabs4 # smooth time trace difference data # imgabs5 #; 5, solving characteristic values T1 and T2 of the Hessian matrix; and step 6, calculating the apparent inclination angle according to the characteristic value. According to the seismic structure information accurate extraction method, influences of seismic noise, faults and seismic amplitude changes can be well avoided, an accurate and stable seismic dip angle data body is obtained, and the seismic dip angle is extracted accurately.

Description

Technical Field

[0001] The invention relates to the technical field of oil reservoir geophysics, and in particular to a method for accurately extracting seismic structure information. Background Art

[0002] Seismic dip reflects the inclination of the seismic reflection surface along a certain direction. The dip in the dip direction is the true dip, and the components of the true dip in other directions are the apparent dips along that direction. In seismic work areas, the apparent dips along the inline and crossline directions are usually considered. However, conventional algorithms are often not stable enough in seismic data due to the influence of seismic noise, faults, and changes in seismic amplitude, and the data often have singular points. It is difficult to remove the influence of the dip data in the later use process.

[0003] In the Chinese patent application with application number: CN201710740537.5, a method and device for determining formation dip is involved, wherein the method includes: calculating the total energy of the seismic data to be measured in each preset direction among multiple preset directions; determining the maximum value among the multiple total energies, and determining the direction corresponding to the maximum value; calculating the formation dip in the direction corresponding to the maximum value, and using the determined formation dip as the formation dip corresponding to the seismic data to be measured. In the embodiment of the invention, the local dip calculation result with high accuracy is guaranteed while ensuring the calculation efficiency, thereby improving the performance of the seismic data processing, interpretation and inversion method characterized by data drive.

[0004] In the Chinese patent application with application number: CN202110074333.9, a method and device for automatically extracting formation dip information from seismic data are involved, wherein the method includes: obtaining seismic data; tracking seismic layers simultaneously on the seismic data to generate a seismic stratigraphic body; performing dip time difference analysis on each seismic trace in the seismic data according to the seismic stratigraphic body to generate a dip time difference sequence corresponding to each seismic trace; filtering the dip time difference sequence corresponding to each seismic trace according to a preset effective range of formation dip time difference to obtain a filtered dip time difference sequence for each seismic trace; using the starting sampling point and the ending sampling point of each seismic trace as the interpolation range, interpolating the filtered dip time difference sequence of each seismic trace to obtain an equally spaced dip time difference sequence for each seismic trace; generating a formation dip body according to the equally spaced dip time difference sequence of each seismic trace. The invention can improve the stability, reliability and controllability of extracting formation dip information from seismic data.

[0005] In the Chinese patent application with application number: CN201310272811.2, a high-precision formation dip estimation method is involved, which first calculates the gradient vector, then flips the vector field, performs vector field flipping, calculates the vector set distance, calculates the vector filter weighting, and finally performs vector weighted filtering processing. This method can maintain the spatial consistency of the formation dip estimation in the area where the phase axis extends in a single direction, and improves the accuracy of the formation dip estimation method in the formation structure of fault, angular unconformity and fold morphology; the algorithm content of this technical solution is easy to implement and has high computational efficiency; at the same time, the analysis window size can be flexibly adjusted to meet the needs of different dip estimation effects. For seismic data with severe noise interference, a large analysis window can be used to improve the spatial consistency of the dip estimation result, and for seismic data with high signal-to-noise ratio, a small analysis window can be used to improve the accuracy of the dip estimation result in the fault and angular unconformity area.

[0006] In the Chinese patent application with application number: CN202010861920.8, a method for seismic identification of low-order faults under high formation dip conditions is involved. It includes the following steps: performing frequency division diffusion filtering on post-stack seismic data; calculating the gradient characteristic attributes of the processed seismic data to obtain the horizontal, vertical and diagonal gradient data of the seismic data; calculating the dip attribute of the seismic data to obtain the dip data of the seismic data; using the dip data to determine whether the formation is tilted and its tilt direction, and then performing gradient replacement according to different situations; dividing the seismic image into stable event axes and unstable event axes, and performing gradient correction on the gradient data after gradient replacement according to different situations to construct a structural tensor of the seismic data; calculating the edge detection factor to perform seismic identification of low-order faults under high formation dip conditions. This inventive method effectively realizes the seismic identification of low-order faults under high formation dip conditions.

[0007] The above existing technologies are greatly different from the present invention and fail to solve the technical problem we want to solve. Therefore, we have invented a new method for accurately extracting seismic structure information. Summary of the invention

[0008] The purpose of the present invention is to provide a method for accurately extracting seismic structural information by removing the influence of these data at the dip calculation stage to improve the stability of the formation dip.

[0009] The purpose of the present invention can be achieved by the following technical measures: a method for accurately extracting seismic structure information, the method for accurately extracting seismic structure information comprising:

[0010] Step 1, generating time difference data dT and trace difference data dA according to post-stack seismic records;

[0011] Step 2, calculating the time difference square data dT2, the track difference square data dA2 and the time track difference square data dTA;

[0012] Step 3: Calculate the smoothed time difference square data Smoothed channel error data Smoothed time channel error square data

[0013] Step 4: Based on the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix;

[0014] Step 5, calculate the eigenvalues ​​T1 and T2 of the Hessian matrix;

[0015] Step 6, calculate the apparent inclination angle according to the eigenvalue.

[0016] The purpose of the present invention can also be achieved by the following technical measures:

[0017] In step 1, the post-stack seismic record refers to an actual observed post-stack seismic record or a synthetic record established based on a geological model.

[0018] In step 1, based on the post-stack seismic records, the difference between the time points of each trace is calculated to generate the time difference dT=F(T i+1 ,A j )-F(T i ,A j ). Where F represents the amplitude value of post-stack seismic data, T i represents the time of the i-th point, T i+1 represents the time of the i+1th point, A j represents the jth seismic data; F(T i ,A j ) represents the amplitude value of the jth seismic data and the i-th point in time, F(T i+1 ,A j ) represents the amplitude value of the j-th seismic data and the i+1-th point in time.

[0019] In step 1, the post-stack seismic records are used to calculate the difference in amplitude between each trace of each line at each time point to generate trace difference data dA=F(T i ,A j+1 )-F(T i ,A j ). Where F represents the amplitude value of post-stack seismic data, T i represents the time of the i-th point, A j represents the jth seismic data, A j+1 represents the j+1th seismic data; F(T i ,A j) represents the amplitude value of the jth seismic data and the i-th point in time, F(T i ,A j+1 ) represents the amplitude value of the j+1th seismic data and the i-th point in time.

[0020] In step 2, the time difference data dT is multiplied to generate time difference square data dT2=dT*dT.

[0021] In step 2, the track error data dA is multiplied to generate track error square data dA2=dA*dA.

[0022] In step 2, the time difference data dT is multiplied by the track difference data dA to generate time and track difference square data dTA=dT*dA.

[0023] In step 3, the time difference square data dT2 is filtered using Gaussian filtering to generate smoothed time difference square data.

[0024] In step 3, the squared error data dA2 is filtered using a Gaussian filter to generate smoothed squared error data.

[0025] In step 3, the time channel error square data is filtered using Gaussian filtering to generate smoothed time channel error square data.

[0026] In step 4, based on the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix:

[0027]

[0028] In step 5, the Jacobi iteration algorithm is used to find the eigenvalues ​​T1 and T2 of the Hessian matrix, and T1 and T2 are sorted so that T1>T2.

[0029] In step 6, the apparent inclination angle D=T2 / T1 is calculated according to the characteristic value.

[0030] The accurate extraction method of seismic structural information in the present invention meets the requirement of stable calculation of formation dip angle, and is mainly used to analyze the formation dip angle of geological strata and understand the inclination of strata. The method mainly calculates the eigenvector value of the Hessian matrix by improving the Hessian matrix construction method, thereby obtaining the formation dip angle volume data of the stratum. The method can better avoid the influence of seismic noise, fault and seismic amplitude change, obtain accurate and stable seismic dip angle data volume, and extract the accuracy of seismic dip angle. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1It is a post-stack seismic record diagram actually observed in a specific embodiment of the present invention;

[0032] Figure 2 A formation dip diagram calculated in a specific embodiment of the present invention;

[0033] Figure 3 The flowchart is a specific embodiment of the method for accurately extracting seismic structure information of the present invention. DETAILED DESCRIPTION

[0034] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0035] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0036] like Figure 3 As shown, Figure 3 The flowchart of the method for accurately extracting seismic structure information of the present invention is as follows. The method for accurately extracting seismic structure information comprises:

[0037] Step 101, generating time difference and trace difference data according to post-stack seismic records;

[0038] Step 102, calculating the time difference square data, the channel difference square data and the time channel difference square data;

[0039] Step 103, calculating the smoothed time error square data, the smoothed channel error square data and the smoothed time channel error square data;

[0040] Step 104, constructing a Hessian matrix based on the smoothed time difference square data, the smoothed channel difference square data, and the smoothed time channel difference square data;

[0041] Step 105, finding the eigenvalues ​​T1 and T2 of the Hessian matrix;

[0042] Step 106, calculating the apparent inclination angle according to the eigenvalue.

[0043] The invention is a method for accurately extracting earthquake dip angle by improving the Hessian matrix construction algorithm, which can avoid the influence of earthquake noise, fault and earthquake amplitude change and obtain accurate and stable earthquake dip angle data body.

[0044] The following are several specific embodiments of the present invention.

[0045] Example 1

[0046] In a specific embodiment 1 of the present invention, the method for accurately extracting seismic structure information includes the following steps:

[0047] 1) Calculate the difference between the time points of each trace of the post-stack seismic record to generate the time difference dT = F(T i+1 ,A j )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0048] 2) Calculate the difference in amplitude between each trace of each line at each time point from the post-stack seismic record to generate trace difference data dA = F(T i ,A j+1 )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0049] 3) Multiply the time difference data dT to generate the time difference square data dT2=dT*dT;

[0050] 4) Multiply the track error data dA to generate track error square data dA2=dA*dA;

[0051] 5) Multiply the time difference data dT by the track difference data dA to generate the time track difference square data dTA=dT*dA;

[0052] 6) Use Gaussian filtering (5x5) to filter the time difference square data dT2 to generate smoothed time difference square data

[0053] 7) Use Gaussian filtering (5x5) to filter the error square data dA2 to generate smoothed error square data

[0054] 8) Use Gaussian filtering (5x5) to filter the time channel error square data to generate smoothed time channel error square data

[0055] 9) According to the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix:

[0056]

[0057] 10) Using the Jacobi iteration algorithm, find the eigenvalues ​​T1 and T2 of the Hessian matrix, and sort T1 and T2 so that T1>T2;

[0058] 11) Calculate the apparent dip angle D = T2 / T1 based on the characteristic value. The formation dip angle represented by the seismic phase axis is as follows: Figure 2 shown.

[0059] Example 2

[0060] In the specific embodiment 2 of the present invention, the method for accurately extracting seismic structure information includes the following steps:

[0061] 1) Calculate the difference between the time points of each trace of the post-stack seismic record to generate the time difference dT = F(T i+1 ,A j )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0062] 2) Calculate the difference in amplitude between each trace of each line at each time point from the post-stack seismic record to generate trace difference data dA = F(T i ,A j+1 )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0063] 3) Multiply the time difference data dT to generate the time difference square data dT2=dT*dT;

[0064] 4) Multiply the track error data dA to generate track error square data dA2=dA*dA;

[0065] 5) Multiply the time difference data dT by the track difference data dA to generate the time track difference square data dTA=dT*dA;

[0066] 6) Use Gaussian filtering (7x7) to filter the time difference square data dT2 to generate smoothed time difference square data

[0067] 7) Use Gaussian filtering (7x7) to filter the error square data dA2 to generate smoothed error square data

[0068] 8) Use Gaussian filtering (7x7) to filter the time channel error square data to generate smoothed time channel error square data

[0069] 9) According to the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix:

[0070]

[0071] 10) Using the Jacobi iteration algorithm, find the eigenvalues ​​T1 and T2 of the Hessian matrix, and sort T1 and T2 so that T1>T2;

[0072] 11) Calculate the apparent dip angle D = T2 / T1 based on the eigenvalue. The dip angle of the stratum represented by the seismic event axis.

[0073] Example 3

[0074] In the specific embodiment 3 of the present invention, the method for accurately extracting seismic structure information includes the following steps:

[0075] 1) Calculate the difference between the time points of each trace of the post-stack seismic record to generate the time difference dT = F(T i+1 ,A j )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0076] 2) Calculate the difference in amplitude between each trace of each line at each time point from the post-stack seismic record to generate trace difference data dA = F(T i ,A j+1 )-F(T i ,A j ); Observed post-stack seismic records refer to: actual observed post-stack seismic records or synthetic records established based on geological models, such as Figure 1 As shown;

[0077] 3) Multiply the time difference data dT to generate the time difference square data dT2=dT*dT;

[0078] 4) Multiply the track error data dA to generate track error square data dA2=dA*dA;

[0079] 5) Multiply the time difference data dT by the track difference data dA to generate the time track difference square data dTA=dT*dA;

[0080] 6) Use Gaussian filtering (3x3) to filter the time difference square data dT2 to generate smoothed time difference square data

[0081] 7) Use Gaussian filtering (3x3) to filter the error square data dA2 to generate smoothed error square data

[0082] 8) Use Gaussian filtering (3x3) to filter the time channel error square data to generate smoothed time channel error square data

[0083] 9) According to the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix:

[0084]

[0085] 10) Using the Jacobi iteration algorithm, find the eigenvalues ​​T1 and T2 of the Hessian matrix, and sort T1 and T2 so that T1>T2;

[0086] 11) Calculate the apparent dip angle D = T2 / T1 based on the eigenvalue. The dip angle of the stratum represented by the seismic event axis.

[0087] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0088] Except for the technical features described in the specification, all other technical features are known technologies to those skilled in the art.

Claims

1. A method for accurately extracting earthquake structural information, characterized in that: The method for accurately extracting seismic structure information includes: Step 1, generating time difference data dT and trace difference data dA according to post-stack seismic records; Step 2, calculating the time difference square data dT2, the track difference square data dA2 and the time track difference square data dTA; Step 3: Calculate the smoothed time difference square data Smoothed channel error data Smoothed time channel error square data Step 4: Based on the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix; Step 5, calculate the eigenvalues ​​T1 and T2 of the Hessian matrix; Step 6, calculate the apparent inclination angle according to the eigenvalue.

2. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 1, the post-stack seismic record refers to an actual observed post-stack seismic record or a synthetic record established based on a geological model.

3. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 1, based on the post-stack seismic records, the difference between the time points of each trace is calculated to generate the time difference dT=F(T i+1 ,A j )-F(T i ,A j ), where F represents the amplitude value of post-stack seismic data, T i represents the time of the i-th point, T i+1 represents the time of the i+1th point, A j represents the jth seismic data; F(T i ,A j ) represents the amplitude value of the jth seismic data and the i-th point in time, F(T i+1 ,A j ) represents the amplitude value of the j-th seismic data and the i+1-th point in time.

4. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 1, the post-stack seismic records are used to calculate the difference in amplitude between each trace of each line at each time point to generate trace difference data dA=F(T i ,A j+1 )-F(T i ,A j ), where F represents the amplitude value of post-stack seismic data, T i represents the time of the i-th point, A j represents the jth seismic data, A j+1 represents the j+1th seismic data; F(T i ,A j ) represents the amplitude value of the jth seismic data and the i-th point in time, F(T i ,A j+1 ) represents the amplitude value of the j+1th seismic data and the i-th point in time.

5. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 2, the time difference data dT is multiplied to generate time difference square data dT2=dT*dT.

6. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 2, the track error data dA is multiplied to generate track error square data dA2=dA*dA.

7. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 2, the time difference data dT is multiplied by the track difference data dA to generate time and track difference square data dTA=dT*dA.

8. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 3, the time difference square data dT2 is filtered using Gaussian filtering to generate smoothed time difference square data.

9. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 3, the squared error data dA2 is filtered using a Gaussian filter to generate smoothed squared error data.

10. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 3, the time channel error square data is filtered using Gaussian filtering to generate smoothed time channel error square data.

11. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 4, based on the smoothed time difference data Smoothed channel error data Smoothed time channel error square data Construct the Hessian matrix:

12. The method for accurately extracting seismic structure information according to claim 1, characterized in that: In step 5, the Jacobi iteration algorithm is used to find the eigenvalues ​​T1 and T2 of the Hessian matrix, and T1 and T2 are sorted so that T1>T2.

13. The method for accurately extracting seismic structure information according to claim 12, characterized in that: In step 6, the apparent inclination angle D=T2 / T1 is calculated according to the characteristic value.

Images