Low-sequence fault identification method based on multi-dominant frequency band fault sensitive attribute fusion

Abstract

The invention relates to the technical field of oil exploration, evaluation and development, in particular to a low-order fault identification method based on multi-dominant frequency band fault sensitive attribute fusion. The method comprises the following steps: determining an original three-dimensional seismic data volume, carrying out horizon and fault interpretation on the original three-dimensional seismic data volume, obtaining horizon data H1 of a first round, extracting single-frequency data volumes from the original three-dimensional seismic data volume according to equal intervals, carrying out interpretive processing on each single-frequency data volume, and carrying out data processing on each single-frequency data volume after the interpretive processing, extracting fault sensitive attributes along the obtained H1 layer, selecting a plurality of fault sensitive attributes to establish an attribute library, performing normalization processing on each attribute in the established attribute library, fusing the normalized attribute library into a fault sensitive attribute graph AttF, correcting H1 through the AttF, and obtaining a horizon fault interpretation result H2. According to the method provided by the invention, the low-sequence fault identification capability is effectively improved, and the compilation precision of the structural map is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil exploration, evaluation and development, and particularly to a method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands. Background Art

[0002] With the gradual deepening of oilfield exploration and development, the exploration degree is getting higher and higher, large-scale oil and gas reservoirs have been discovered one after another, and the exploration targets have gradually shifted to small and concealed oil and gas reservoirs. Among them, small oil and gas reservoirs are most typical of small structures and structural-lithologic trap reservoirs formed by small fault blocks and small fault blocks. Therefore, the effective identification of low-order faults is one of the important research contents in the future exploration, evaluation and development fields, and is of great significance for analyzing the hydrocarbon accumulation law, risk assessment and optimizing favorable target areas of oil and gas reservoirs.

[0003] After investigation, the following problems exist in the previous methods for identifying low-order faults: First, forward modeling is used to guide fault interpretation. Small faults can have certain response characteristics, but there are many situations in actual seismic data that can form similar responses. For example, changes in formation dip and reservoir changes can both cause this, resulting in strong multi-solutionity and being difficult to judge. Second, coherent attributes have a clear identification effect on large-offset faults, but the seismic response of small faults is weak or even non-existent. Third, the curvature attribute is more sensitive to small faults than coherent attributes, but it is also more sensitive to folds, which is extremely prone to misjudgment. These problems directly affect the determination of the scope of favorable target areas, and affect the exploration and evaluation of oilfields and the well location deployment. Therefore, in view of the above deficiencies, a method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands is proposed. Summary of the Invention

[0004] (I) Technical Problems to be Solved The present invention provides a method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands to overcome the problems in the prior art that in the process of identifying low-order faults, due to the similar response types, there is strong multi-solutionity, the seismic response of small faults is weak, and the sensitivity of small faults is high, resulting in easy misjudgment and difficulty in judging low-order faults.

[0005] (II) Technical Solutions To solve the above problems, the present invention provides a method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands, including: Step S1: Determine the original three-dimensional seismic data volume, perform horizon and fault interpretation on the original three-dimensional seismic data volume, and obtain the first-round horizon data H1; Step S2: Extract single-frequency data volumes from the original three-dimensional seismic data volume determined in step S1 at equal intervals, and perform interpretive processing on each single-frequency data volume; Step S3: For each single-frequency data volume after the interpretive processing in Step S2, extract fault-sensitive attributes along the H1 layer obtained in Step S1, and select multiple fault-sensitive attributes to establish an attribute library. Step S4: Perform normalization processing on each attribute in the attribute library established in Step S3, and fuse the normalized attribute library into a fault-sensitive attribute map Att F ; Step S5: According to the fault-sensitive attribute map Att F obtained in Step S4, obtain the layer fault interpretation result H2 by correcting the H1 obtained in Step S1.

[0006] Preferably, in Step S2, the spacing of the single-frequency data volumes is equally divided according to the range of the effective frequency bandwidth of the original three-dimensional seismic data volume, and the effective frequency bandwidth of the original three-dimensional seismic data volume is obtained by spectral analysis.

[0007] Preferably, in Step S2, the method for interpretive processing of the single-frequency data volumes includes multi-window dip scanning and structure-oriented filtering techniques.

[0008] Preferably, in Step S3, the fault-sensitive attributes include coherence cube slices, curvature cube slices, ant cube slices, and edge detection attributes.

[0009] Preferably, in Step S3, the attribute library is a fault-sensitive attribute map corresponding to multiple sets of single-frequency data volumes with dominant frequency bands and clear fault characteristics. The naming order of multiple sets of fault-sensitive attribute maps is Att 1 、Att 2 ...Att n 。

[0010] Preferably, in Step S4, the normalization processing refers to uniformly standardizing the value range spaces of all attributes in the attribute library to a unified value range space by equal proportion.

[0011] Preferably, the value range space is the range of the value ranges of the attribute values of faults and non-faults.

[0012] Preferably, in Step S5, the fusion method of the fault-sensitive attribute map Att F is to vectorially fuse the multiple attribute values corresponding to each point in space according to the fault characteristic attribute extreme value optimization method.

[0013] (III) Beneficial Effects The low-order fault identification method based on the fusion of multi-dominant frequency band fault-sensitive attributes provided by the present invention comprehensively considers the fault response characteristics of multiple attributes under multi-frequency data volumes, not only effectively improves the low-order fault identification ability, but also further improves the accuracy of structure map compilation, providing technical support for oilfield development. Brief Description of the Drawings

[0014] Figure 1 This is a flowchart of the method for identifying low-order faults based on the fusion of multi-dominant frequency band tomographic sensitive attributes in the embodiments of the present invention; Figure 2 This is a planar result display diagram of the first-round interpretation result H1 in the embodiments of the present invention; Figure 3 This is a data diagram of the effective frequency band analysis result of the data volume in the embodiments of the present invention; Figure 4 This is a listing diagram of the equal-frequency interval single-product data volume extracted according to the effective frequency band analysis result in the original data volume in the embodiments of the present invention; Figure 5 This is a display diagram of the fault response characteristics of the original seismic profile in the embodiments of the present invention; Figure 6 This is a display diagram of the fault response characteristics of the interpretive processed seismic profile in the embodiments of the present invention; Figure 7 This is a data diagram of the planar response characteristics of the fault sensitive attributes in different dominant frequency bands in the embodiments of the present invention; Figure 8 This is a display diagram of the range space standardization of the coherence attributes in multiple dominant frequency bands in the embodiments of the present invention; Figure 9 This is a vectorized fusion fault sensitive attribute diagram of the coherence attributes in multiple dominant frequency bands after standardization in the embodiments of the present invention; Figure 10 This is a display diagram of the final horizon and fault interpretation result H2 in the embodiments of the present invention. Detailed Embodiments

[0015] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0016] In addition, those of ordinary skill in the art should understand that the provided drawings are only for explaining the purpose, features, and advantages of the present invention, and the drawings are not actually drawn to scale.

[0017] At the same time, unless the context clearly requires, the words such as "including" and "comprising" in the entire specification and claims should be interpreted as having an inclusive meaning rather than an exclusive or exhaustive meaning; that is, it is the meaning of "including but not limited to".

[0018] Figure 1The flowchart of the low-order fault identification method based on the fusion of multi-dominant frequency band fault-sensitive attributes in the embodiments of the present invention is as follows Figure 1 As shown, the present invention provides a low-order fault identification method based on the fusion of multi-dominant frequency band fault-sensitive attributes, which specifically includes: Step S1: Determine the original 3D seismic data volume, perform horizon and fault interpretation on the original 3D seismic data volume, and obtain the first-round horizon data H1; Step S2: Extract single-frequency data volumes from the original 3D seismic data volume determined in step S1 at equal intervals, and perform interpretive processing on each single-frequency data volume; Step S3: For each single-frequency data volume after the interpretive processing in step S2, extract fault-sensitive attributes along the H1 horizon obtained in step S1, and select multiple fault-sensitive attributes to establish an attribute library; Step S4: Perform normalization processing on each attribute in the attribute library established in step S3, and fuse the normalized attribute library into a fault-sensitive attribute map Att F ; Step S5: According to the fault-sensitive attribute map Att F obtained in step S4, obtain the horizon fault interpretation result H2 by correcting the H1 obtained in step S1.

[0019] In this identification method, in step S2, the extraction method of the single-frequency data volume is the wavelet transform frequency division technology. The wavelet transform is a local analysis of time frequency or spatial frequency. Through stretching and translation operations, the signal or function is gradually refined at multiple scales, achieving time subdivision at high frequencies and frequency subdivision at low frequencies.

[0020] In practical applications, in step S2, the interval of the single-frequency data volume is equally divided according to the effective frequency bandwidth range of the original 3D seismic data volume. The original 3D seismic data volume obtains the effective frequency bandwidth through spectral analysis. The interpretive processing methods of the single-frequency data volume include multi-window dip scanning and structure-oriented filtering technology to improve the fault identification ability of the single-frequency volume.

[0021] In this identification method, in step S3, the fault-sensitive attributes include but are not limited to coherence cube slices, curvature cube slices, ant cube slices, and edge detection attributes. The attribute library is the fault-sensitive attribute maps corresponding to multiple sets of single-frequency data volumes with dominant frequency bands and clear fault characteristics. The naming order of multiple sets of fault-sensitive attribute maps is Att 1 、Att 2 ...Att n 。

[0022] It should be noted that spectral analysis can determine the position of the effective wave in the frequency band, and there are many effective waves and strong clarity in the dominant frequency band.

[0023] In practical applications, in step S4, the normalization process refers to uniformly normalizing the value range spaces of all attributes in the attribute library to a unified value range space through equal-proportion standardization. The value range space is the range of the value ranges of the attributes of faults and non-faults. The normalization process makes the value range spaces of the attribute values of each attribute consistent, the value range representing the fault attribute values consistent, and the value range representing the non-fault attribute values consistent.

[0024] In this identification method, the fault-sensitive attribute map Att F is fused by vectorially fusing the multiple attribute values corresponding to each point in the space according to the extreme value optimization method of fault characteristic attributes. The fault-sensitive attribute map Att F can take into account the fault response characteristics of multiple frequency bands, and the fault imaging results are more refined.

[0025] It should be noted that the specific explanation of the extreme value optimization method of fault characteristic attributes is as follows: For example, for the coherence attribute, if the fault characteristic attribute value is the minimum value, then the smallest value among the attributes extracted from multiple frequency bands corresponding to a certain point is selected as the finally selected value. Perform such an operation on all points, and finally form Att F .

[0026] In order to effectively improve the low-order fault identification ability and improve the accuracy of structural map compilation, a low-order fault identification method based on the fusion of multi-dominant frequency band fault-sensitive attributes is developed. In practical applications, combined with Figures 2 to 10 , the operation process of this low-order fault identification method based on the fusion of multi-dominant frequency band fault-sensitive attributes is specifically described as follows: Step 1: Determine the original three-dimensional seismic data volume, perform horizon and fault interpretation on the original three-dimensional seismic data volume, and obtain the first-round horizon data H1.

[0027] In this embodiment, Figure 2 is the plane result display diagram of the first-round interpretation result H1 of this embodiment of the present invention. As Figure 2 shown, determine the original three-dimensional seismic data volume, adopt conventional seismic interpretation means, perform horizon and fault interpretation on the original data volume, and obtain the first-round interpretation result H1.

[0028] Step 2: Extract single-frequency data volumes from the determined original three-dimensional seismic data volume at equal intervals, and perform interpretive processing on each single-frequency data volume.

[0029] In practical applications, Figure 3 is the data diagram of the effective frequency band analysis result of the data volume of this embodiment of the present invention. As Figure 3 shown, through spectral analysis, the effective frequency bandwidth of the original data volume is obtained as 0~100Hz. Figure 4 is the listing diagram of the equal-frequency interval single-product data volumes extracted from the original data volume according to the effective frequency band analysis result in this embodiment of the present invention.Figure 4 As shown, within the range of the acquired basic data volume of the effective bandwidth, equally spaced single-frequency data volumes are extracted. In this embodiment, the interval of the single-frequency data volume is set to 10 Hz. That is to say, the acquired single-frequency data volumes are respectively the 10Hz, 20Hz, 30Hz... 100Hz single-frequency data volumes. Figure 5 This is a display diagram of the fault response characteristics of the original seismic profile in the embodiment of the present invention. Figure 6 This is a display diagram of the fault response characteristics of the interpreted seismic profile in the embodiment of the present invention. As Figure 5 and Figure 6 shown, for all the acquired single-frequency data volumes, multi-window dip scanning and structure-oriented filtering techniques are used to perform interpretive processing on the original single-frequency data volumes, improving the fault recognition ability of the single-frequency volume. The fault response of the processed seismic profile is clearer and the fault surface is more distinct.

[0030] Step 3: For each single-frequency data volume after interpretive processing, extract fault-sensitive attributes along the acquired H1 layer, and select multiple fault-sensitive attributes to establish an attribute library.

[0031] In this embodiment, Figure 7 This is a data diagram of the planar response characteristics of fault-sensitive attributes in different dominant frequency bands in the embodiment of the present invention. As Figure 7 shown, for each set of single-frequency data volumes, extract fault-sensitive attributes along the H1 horizon, and preferably select multiple sets of dominant frequency band fault-sensitive attribute maps with clearer fault characteristics to form a fault-sensitive attribute library. The naming order of the multiple sets of fault-sensitive attribute maps is Att 1 、Att 2 ...Att n .

[0032] Step 4: Normalize each attribute in the established attribute library, and fuse the normalized attribute library into a fault-sensitive attribute map Att F .

[0033] In practical applications, Figure 8 This is a display diagram of the normalization of the value range space of coherent attributes in multiple dominant frequency bands in the embodiment of the present invention. As Figure 8 shown, read the value range space of all attribute maps in the attribute library and normalize it to a unified value range space. The normalization process is equal-proportion normalization. Through the normalization process, the value range space of the attribute values of each attribute is consistent, the value range of the fault attribute values is consistent, and the value range of the non-fault attribute values is consistent. Figure 9 This is a vectorized fusion fault-sensitive attribute map of multiple dominant frequency band coherent attributes after normalization in the embodiment of the present invention. As Figure 9 shown, vectorize and fuse all the attributes in the normalized attribute library to form the final fault-sensitive attribute map, the fault-sensitive attribute map Att FIt can take into account the tomographic response characteristics of multiple frequency bands, and the tomographic imaging results are more refined.

[0034] Step Five: According to the obtained tomographic sensitive attribute map Att F , the formation horizon tomographic interpretation result H2 is obtained by correcting the obtained H1.

[0035] In this embodiment, Figure 10 is the display diagram of the final formation horizon and tomographic interpretation result H2 of the embodiment of the present invention. As Figure 10 shown, the formation horizon tomographic interpretation result H2 is obtained by correcting the obtained H1, and the fracture system is interpreted with the aid of the final tomographic attribute map to obtain a more refined structural map.

[0036] The low-order fault identification method based on the fusion of tomographic sensitive attributes of multiple dominant frequency bands provided by the present invention comprehensively considers the tomographic response characteristics of multiple attributes in the multi-frequency data volume, not only effectively improves the low-order fault identification ability, but also further improves the accuracy of the structural map compilation, providing technical support for oilfield development.

[0037] The above embodiments are only used to illustrate the present invention, rather than limiting the present invention. Those of ordinary skill in the relevant technical field can also make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all equivalent technical solutions also belong to the scope of the present invention, and the patent protection scope of the present invention shall be defined by the claims.

Claims

1. A method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands, characterized in that, it includes: Step S1: Determine the original three-dimensional seismic data volume, perform horizon and fault interpretation on the original three-dimensional seismic data volume, and obtain the first-round horizon data H1; Step S2: Extract single-frequency data volumes from the original three-dimensional seismic data volume determined in Step S1 at equal intervals, and perform interpretive processing on each single-frequency data volume; Step S3: For each single-frequency data volume after the interpretive processing in Step S2, extract fault-sensitive attributes along the H1 horizon obtained in Step S1, and select multiple fault-sensitive attributes to establish an attribute library; Step S4: Normalize each attribute in the attribute library established in Step S3, and fuse the normalized attribute library into a fault-sensitive attribute map Att F ; Step S5: According to the fault-sensitive attribute map Att obtained in the said step S4 F , the formation fault interpretation result H2 is obtained by correcting H1 obtained in the said step S1.

2. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 1, characterized in that, in Step S2, the interval of the single-frequency data volume is equally divided according to the range of the effective frequency bandwidth of the original three-dimensional seismic data volume, and the effective frequency bandwidth of the original three-dimensional seismic data volume is obtained by spectral analysis.

3. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 1, characterized in that, in Step S2, the methods for the interpretive processing of the single-frequency data volume include multi-window dip scanning and structure-oriented filtering techniques.

4. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 1, characterized in that, in Step S2, the fault-sensitive attributes include coherence cube slices, curvature cube slices, ant cube slices, and edge detection attributes.

5. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 4, characterized in that, In the step S3, the attribute library is a fault-sensitive attribute map corresponding to multiple sets of single-frequency data volumes with frequency band advantages and clear fault characteristics. The naming order of multiple sets of fault-sensitive attribute maps is Att 1 , Att 2 ... Att n .

6. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 1, characterized in that, in Step S4, the normalization process refers to uniformly standardizing the value range spaces of all attributes in the attribute library to a unified value range space by equal proportion.

7. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 6, characterized in that, the value range space is the range of the value ranges of the attribute values of faults and non-faults.

8. The method for identifying low-order faults based on the fusion of fault-sensitive attributes in multiple dominant frequency bands according to Claim 1, characterized in that, In the step S5, the fusion method of the fault sensitive attribute map Att F is to perform vectorized fusion on multiple attribute values corresponding to each point in space according to the extreme value optimization method of fault characteristic attributes.

Images