Fracture identification method based on optimized sub-band RGB fusion

Abstract

The invention belongs to the technical field of oil-gas exploration, and particularly relates to a fracture identification method based on optimized sub-band RGB fusion, and the method comprises the steps: S1, carrying out the preprocessing of seismic data; s2, spectrum analysis and frequency range body attribute analysis; and S3, carrying out sub-band RGB fusion. According to the optimized sub-band RGB fusion method, a data enhancement body is decomposed into a low frequency band body, a middle frequency band body and a high frequency band body, the frequency band bodies have more advantages compared with a single frequency body, sensitive attributes are preferably selected for the three frequency band bodies for attribute fusion, and the characterization capacity of different frequency band bodies for fracture is enhanced. On the basis, layer attribute slices are extracted from sensitive attribute fusion bodies corresponding to different frequency band bodies for RGB fusion and illumination display, so that the generality reflecting fractures in low, medium and high frequency band ranges is further highlighted. According to the method, the generality of the seismic band ratio can be effectively highlighted, and compared with a conventional method, the fracture identification precision can be further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of oil and gas exploration, and particularly relates to a fracture identification method based on optimized frequency-divided band RGB fusion. Background Art

[0002] Due to the factor of tectonic stress, underground rock formations will inevitably crack, resulting in the generation of faults. In the field of oil and gas exploration, faults can not only serve as channels for oil and gas migration but also as boundaries of fault-block oil and gas fields, thus effectively controlling the distribution of oil and gas fields. Therefore, accurately identifying faults is of great significance for the exploration and development of oil and gas fields. Currently, in practical applications, fault detection and identification are mainly based on post-stack attribute methods, such as various post-stack attributes like coherence, curvature, automatic fault extraction (AFE), and likelihood. Through these attribute methods, hidden seismic information can be better extracted from the original seismic data, and the distribution characteristics and boundary contours of faults can be effectively identified. In actual production applications, using a single attribute for fracture detection often fails to achieve the most ideal effect, while fused attributes are more accurate to a certain extent and can more effectively identify abnormal information.

[0003] The frequency-divided RGB fusion technology is widely and maturely applied, and has better visualization effects than single attributes, and can more clearly indicate the target. The main implementation process of the frequency-divided RGB fusion technology is as follows: First, the original data is frequency-divided to extract single-frequency bodies of different frequencies, then the corresponding sensitive attribute bodies are extracted respectively, and then the slices of the attribute bodies are subjected to RGB fusion display, using the fused attribute information to highlight the description of boundary information and abnormal information. The frequency-divided bodies are mainly realized through spectral decomposition technology, which uses mathematical transformation to convert seismic signals from the time domain to the frequency domain to obtain discrete frequency bodies. In actual applications, seismic data will be decomposed into three frequency-divided bodies. To a certain extent, the existing frequency-divided RGB fusion technology can improve the accuracy of fracture identification, but there is often a lack of effective basis for determining the main frequencies of the three frequency-divided bodies, and at the same time, it is unable to better highlight the commonalities of the seismic zone ratios. Summary of the Invention

[0004] The present invention solves the problems that when the prior art performs fracture identification, there is a lack of effective basis for determining the main frequencies of the three frequency-divided bodies, and at the same time, it is unable to better highlight the commonalities of the seismic zone ratios, and provides a fracture identification method based on optimized frequency-divided band RGB fusion, and uses the optimized frequency-divided band RGB fusion method to identify fractures.

[0005] The technical solution claimed by the present invention is as follows:

[0006] A fracture identification method based on optimized frequency-divided band RGB fusion, comprising the following steps:

[0007] S1: Preprocessing of seismic data: Sequentially perform structure-oriented filtering and DSE processing on the original seismic data; the structure-oriented filtering is to optimize the original seismic data to obtain an optimized data volume; the DSE is to calculate the dip and azimuth volume of the optimized data volume, and use the dip and azimuth volume as a constraint to perform DSE calculation on the optimized data volume using a correlation weighted median filtering algorithm to obtain an enhanced data volume; the DSE calculation uses a dip control enhancement technique to calculate the similarity of adjacent traces using the changes in dip and azimuth.

[0008] S2: Spectrum analysis and frequency band volume attribute analysis: Perform spectrum scanning and analysis on the enhanced data volume obtained in S1 to determine the main frequency distribution position and effective frequency band distribution range of the enhanced data volume; decompose the enhanced data volume obtained in S1 into low, medium, and high frequency band volumes, perform attribute analysis on the three frequency band volumes respectively, and extract sensitive attributes that can reflect faults; fuse the sensitive attribute volumes corresponding to different frequency band volumes to obtain an attribute fusion volume corresponding to the three frequency band volumes; among them, each frequency band volume corresponds to an attribute fusion volume; when decomposing the enhanced data volume obtained in S1 into low, medium, and high frequency band volumes, the effective frequency band ranges of the three frequency band volumes are kept consistent.

[0009] S3: RGB fusion of frequency bands: Extract the layer-by-layer attribute slices of the attribute fusion volumes corresponding to different frequency band volumes obtained in S2, perform RGB fusion and illumination display on the extracted layer-by-layer attribute slices, and detect and identify faults through the layer-by-layer attribute slices after RGB fusion.

[0010] The original seismic data described in S1 is the original post-stack seismic data.

[0011] Preferably, the step of optimizing the original seismic data to obtain an optimized data volume in S1 includes: calculating a structure-oriented volume from the original seismic data, and using the structure-oriented volume as a constraint to remove the random noise interference of the original seismic data by filtering to obtain an optimized data volume.

[0012] Preferably, the main frequency described in S2 corresponds to the frequency peak.

[0013] Preferably, the effective frequency band distribution range in S2 is the range from the minimum frequency to the maximum frequency when the vertical coordinate amplitude value is 0.707.

[0014] Preferably, the main frequency distribution position and effective frequency band distribution range of the enhanced data volume in S2 can be determined according to actual needs.

[0015] In S2, the data enhancement body obtained in S1 is subjected to frequency band decomposition to obtain low, medium, and high frequency band bodies. Specifically, the data enhancement body is subjected to Fourier transform to obtain full-frequency band frequency domain data, and the low, medium, and high frequency band data are respectively subjected to inverse Fourier transform to obtain the low, medium, and high frequency band bodies.

[0016] The sensitive attribute in S2 refers to the dominant attribute that can characterize fractures, including coherence attribute, curvature attribute, AFE attribute, and likelihood attribute.

[0017] Preferably, the sensitive attributes in S2 are coherence attribute and AFE attribute.

[0018] Preferably, the in-layer attribute slice in S3 is obtained by opening a time window for the obtained attribute body using horizon data and taking the root mean square value within the time window.

[0019] Beneficial effects:

[0020] The present invention provides a fracture recognition method based on optimized frequency-divided band RGB fusion. First, the original seismic data is sequentially subjected to structure-oriented filtering and DSE processing. The DSE calculation uses the dip control enhancement technology to calculate the similarity of adjacent traces using the changes in dip and azimuth, improving the seismic lateral signal-to-noise ratio. The dip control enhancement technology uses the method of calculating dip azimuth, so its result has significantly enhanced the ability to depict faults, which is beneficial for the subsequent research on faults and fractures. Then, spectral scanning and analysis are performed on the enhanced data body obtained in S1 to determine the main frequency distribution position and effective frequency band distribution range of the enhanced data body. Through the scanning and analysis of the enhanced data body, the main frequency distribution of the enhanced data body can be clarified. The enhanced data body is subjected to frequency band decomposition to obtain low, medium, and high frequency band bodies. Combining the main frequency distribution position of the enhanced data body determined above, the main frequencies of the three frequency band bodies can be clarified, solving the problem that there is a lack of effective basis for determining the main frequencies of the three frequency-divided bodies in the prior art for fracture recognition; attribute analysis is respectively performed on the three frequency band bodies, sensitive attributes that can reflect faults are extracted, and the sensitive attribute bodies corresponding to different frequency band bodies are fused to enhance the ability of different frequency band bodies to characterize fractures. On this basis, in-layer slices are extracted from the sensitive attribute fusion body corresponding to different frequency band bodies, and RGB fusion and illumination display are performed on the in-layer slices to enhance the commonality of fractures reflected in different frequency band ranges, and enhance the boundary information and abnormal information in different frequency band ranges, so as to further highlight the commonality of fractures reflected in the low, medium, and high frequency band ranges, solving the problem that the commonality of each seismic zone ratio cannot be well highlighted in the prior art for fracture recognition. Through this method, the commonality of the seismic zone ratio can be effectively highlighted, and at the same time, it involves secondary fusion of attributes, which can further improve the fracture recognition accuracy compared with the conventional method and can be popularized and applied in fracture relatively developed areas.

[0021] The main frequency distribution position and the effective frequency band distribution range of the enhanced data volume can be determined according to actual requirements. The main frequency distribution positions of the three frequency band volumes can be set independently according to user needs, further enhancing the determination basis of the main frequencies of the three frequency-divided volumes. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flowchart of the solution of the present invention.

[0023] Figure 2 It is a preprocessing diagram of seismic data in an embodiment of the present invention; wherein: (a) is a cross-sectional view of the original seismic volume (original seismic data), (b) is a cross-sectional view of the optimized data volume, and (c) is a cross-sectional view of the enhanced data volume.

[0024] Figure 3 It is a spectrum analysis and spectrum decomposition diagram in an embodiment of the present invention; wherein: (I) is a spectrum analysis diagram; (II) is a spectrum decomposition diagram. In (II): (a) is a cross-sectional view of the low-frequency band data volume, (b) is a cross-sectional view of the middle-frequency band data volume, and (c) is a cross-sectional view of the high-frequency band data volume.

[0025] Figure 4 It is a comparison of single attribute and RGB fusion illumination display in an embodiment of the present invention; wherein: (I) is a slice along the layer of a single attribute; (II) is an RGB fusion illumination display. In (II): (a) is a slice of the low-frequency band fusion volume, (b) is a slice of the middle-frequency band fusion volume, and (c) is a slice of the high-frequency band fusion volume. (d) is an RGB fusion illumination display. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The present invention will be further described below with reference to the accompanying drawings:

[0027] A fracture identification method based on optimized frequency-divided band RGB fusion, as Figure 1 shown, includes the following steps:

[0028] S1: Preprocessing of seismic data: sequentially perform structure-oriented filtering and DSE processing on the original seismic data; the structure-oriented filtering is to optimize the original seismic data to obtain an optimized data volume; the DSE is to calculate the dip angle and azimuth angle volume of the optimized data volume, and perform DSE calculation on the optimized data volume using a relevant weighted median filtering algorithm with the dip angle and azimuth angle volume as a constraint to obtain an enhanced data volume. The cross-sectional view of the enhanced data volume is as Figure 2 (c) shown; the DSE calculation uses dip angle control enhancement technology to calculate the similarity of adjacent traces using the changes in dip angle and azimuth angle; the original seismic data is original post-stack seismic data, and the cross-sectional view depicted by the original seismic data, that is, the cross-sectional view of the original seismic volume, is as Figure 2(as shown in (a)); optimizing the original seismic data to obtain an optimized data volume, including: calculating a structure-guided volume from the original seismic data, and using the structure-guided volume as a constraint to remove random noise interference from the original seismic data by filtering to obtain an optimized data volume (guided filtering volume), and the cross-sectional view of the optimized data volume is as shown in Figure 2 (b); Figure 2 It shows that the enhanced data volume obtained after processing has a higher signal-to-noise ratio and clearer boundaries compared to the original seismic volume.

[0029] S2: Spectrum analysis and frequency band volume attribute analysis: Perform spectrum scanning and analysis on the enhanced data volume obtained in S1 ( Figure 3 ), determine the main frequency distribution position and effective frequency band distribution range of the enhanced data volume; perform frequency band decomposition on the enhanced data volume obtained in S1 to obtain low, medium, and high frequency band volumes, respectively perform attribute analysis on the three frequency band volumes, and extract sensitive attributes that can reflect faults; fuse the sensitive attribute volumes corresponding to different frequency band volumes to obtain attribute fusion volumes corresponding to the three frequency band volumes; where each frequency band volume corresponds to an attribute fusion volume; when performing frequency band decomposition on the enhanced data volume obtained in S1, the effective frequency band ranges of the low, medium, and high frequency band volumes are kept consistent.

[0030] In a specific embodiment of the present invention, the main frequency corresponds to the frequency peak, and the effective frequency band distribution range is the range from the minimum frequency to the maximum frequency when the ordinate amplitude value is 0.707.

[0031] In a specific embodiment of the present invention, the main frequency distribution position and effective frequency band distribution range of the enhanced data volume can be determined according to actual needs, and the user can determine the main frequency distribution position and effective frequency band distribution range according to the results of spectrum scanning and analysis.

[0032] The sensitive attribute refers to the dominant attribute that can characterize fractures, and the sensitive attributes include coherence attribute, curvature attribute, AFE attribute, likelihood attribute; in a specific embodiment of the present invention, the sensitive attributes are preferably coherence attribute and AFE attribute.

[0033] The performing frequency band decomposition on the enhanced data volume obtained in S1 to obtain low, medium, and high frequency band volumes specifically includes: performing Fourier transform on the enhanced data volume to obtain full-frequency band frequency domain data, and respectively performing inverse Fourier transform on the low, medium, and high frequency band data to obtain low, medium, and high frequency band volumes.

[0034] S3: Frequency band RGB fusion: Extract the layer-by-layer attribute slices of the attribute fusion volumes corresponding to different frequency band volumes obtained in S2, and perform RGB fusion and illumination display on the extracted layer-by-layer attribute slices ( Figure 4), enhance the commonalities of fractures reflected in different frequency bands, enhance the boundary information and abnormal information in different frequency bands, and detect and identify fractures through the slice of the layer - along attribute after RGB fusion. As Figure 4 shown, after RGB fusion, the signal - to - noise ratio is higher, the boundary is clearer, and at the same time, the fault is also clearer.

[0035] In a specific embodiment of the present invention, the slice of the layer - along attribute is obtained by opening a time window for the obtained attribute volume using horizon data and taking the root - mean - square value within the time window.

[0036] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be defined by the claims.

Claims

1. A fracture recognition method based on optimized frequency-band RGB fusion, characterized in that It includes the following steps: S1: Preprocessing of seismic data: Sequentially perform structure-oriented filtering and DSE processing on the original seismic data; the structure-oriented filtering is to optimize the original seismic data to obtain an optimized data volume; the DSE is to calculate the dip and azimuth volume of the optimized data volume, and use the dip and azimuth volume as a constraint to perform DSE calculation on the optimized data volume using a correlation weighted median filtering algorithm to obtain an enhanced data volume; the DSE calculation uses a dip control enhancement technique to calculate the similarity of adjacent traces using the changes in dip and azimuth. S2: Spectrum analysis and frequency band volume attribute analysis: Perform spectrum scanning and analysis on the enhanced data volume obtained in S1 to determine the main frequency distribution position and effective frequency band distribution range of the enhanced data volume; decompose the enhanced data volume obtained in S1 into low, medium, and high frequency band volumes, respectively perform attribute analysis on the three frequency band volumes, and extract sensitive attributes that can reflect faults; fuse the sensitive attribute volumes corresponding to different frequency band volumes to obtain attribute fusion volumes corresponding to the three frequency band volumes; among them, each frequency band volume corresponds to an attribute fusion volume; when decomposing the enhanced data volume obtained in S1 into low, medium, and high frequency band volumes, the effective frequency band ranges of the three frequency band volumes are kept consistent. S3: RGB fusion in different frequency bands: Extract the layer-by-layer attribute slices of the attribute fusion volumes corresponding to different frequency band volumes obtained in S2, perform RGB fusion and illumination display on the extracted layer-by-layer attribute slices, and detect and identify faults through the layer-by-layer attribute slices after RGB fusion.

2. The fracture recognition method based on optimized frequency-divided band RGB fusion according to claim 1, wherein The original seismic data in S1 is original post-stack seismic data.

3. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, characterized in that The step of optimizing the original seismic data to obtain an optimized data volume in S1 includes: calculating a structure-oriented volume from the original seismic data, and using the structure-oriented volume as a constraint to remove the random noise interference of the original seismic data by filtering to obtain an optimized data volume.

4. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, wherein The main frequency in S2 corresponds to the frequency peak.

5. The fracture recognition method based on optimized frequency-divided band RGB fusion according to claim 1, wherein The effective frequency band distribution range in S2 is the range from the minimum frequency to the maximum frequency when the vertical coordinate amplitude value is 0.

707.

6. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, characterized in that The main frequency distribution position and effective frequency band distribution range of the enhanced data volume in S2 can be determined according to actual requirements.

7. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, characterized in that, Decomposing the data enhancement volume obtained in S1 into low, medium, and high frequency band volumes in S2 specifically includes: performing Fourier transform on the data enhancement volume to obtain full-band frequency domain data, and separately performing inverse Fourier transform on the low, medium, and high frequency band data to obtain low, medium, and high frequency band volumes.

8. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, characterized in that The sensitive attributes in S2 refer to the dominant attributes that can characterize faults, including coherence attributes, curvature attributes, AFE attributes, and likelihood attributes.

9. The fracture recognition method based on optimized frequency-divided band RGB fusion according to claim 8, characterized in that, The sensitive attributes in S2 are coherence attributes and AFE attributes.

10. The fracture recognition method based on optimized frequency division band RGB fusion according to claim 1, characterized in that, The layer-by-layer attribute slices in S3 are obtained by opening a time window for the obtained attribute volume using horizon data and taking the root mean square value within the time window.

Images