A high-density azimuth-based prestack attribute seismic sedimentary microfacies characterization method

Abstract

The present invention belongs to the technical field of oil and gas field exploration, and specifically relates to a method for characterizing seismic sedimentary microfacies with high-density azimuth prestack attributes. The method comprises the following steps: obtaining data and information of high-density seismic areas in the study area; analyzing wide-azimuth seismic gathers, and optimizing favorable offset distance data; superimposing azimuth data based on the distribution law of sedimentary facies; optimizing and clustering analysis of sensitive attributes of seismic sedimentary microfacies in different azimuths; actual drilling verification, if the predicted results do not match, performing regional secondary clustering analysis: integrating geological data, accurately characterizing seismic sedimentary microfacies. The method of the present invention significantly improves the accuracy of seismic sedimentary microfacies characterization, can effectively predict the planar distribution law of sedimentary microfacies in the target layer, and assists in high-quality exploration and development of oil and gas fields.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas field exploration, and in particular relates to a high-density azimuth prestack attribute seismic sedimentary microfacies characterization method. Background Art

[0002] As China's oil and gas reservoirs enter the medium-to-high exploration and development stage, medium-to-small complex concealed oil and gas fields dominated by lithology, stratigraphy and various composite oil and gas traps have gradually become an important position for increasing reserves and production in my country. Complex concealed oil and gas reservoirs are characterized by "thin, small, fragmented and deep". The detailed dissection of failed wells shows that the key to high-quality exploration and development of such reservoirs is to accurately characterize the sedimentary microfacies.

[0003] Although some domestic and foreign scholars have done a lot of research on the identification of sedimentary microfacies, they mainly start from lithology and logging data, and use conventional 3D seismic interpretation technology to characterize sedimentary microfacies. For the sedimentary characteristics of concealed oil and gas reservoirs with unclear sand body boundaries, complex stratigraphic structure, and strong heterogeneity, the above technology has low prediction accuracy and cannot achieve the effect of finely characterizing sedimentary microfacies.

[0004] The applicant's prior patent application CN108663714A discloses a sedimentary microfacies characterization method, which integrates the well logging data and geological data of a completed drilling point in the vertical direction with the seismic data in the plane to achieve a fine characterization of sedimentary microfacies by combining well logging and seismic data. The seismic attribute parameters are optimized by univariate regression analysis and cluster analysis to improve the accuracy of sedimentary microfacies characterization; the seismic attribute parameters are integrated by multivariate linear regression to obtain a fitting relationship function model; through the thinning step, the geological parameters of the well points are fully utilized, and the geological parameters converted from the seismic attribute parameters of the virtual well points between the completed drilling points are effectively utilized, so that the sedimentary microfacies finally characterized are more accurate. Moreover, since only some seismic attribute points are extracted, the workload is greatly reduced in actual operation, and the labor cost is reduced. Based on the sedimentary microfacies characterized by this method, the reservoir distribution between wells can be predicted, providing a geological basis for well site deployment and improving the success rate of exploration and development of oil and gas fields.

[0005] With the continuous development of research technology, the accuracy of sedimentary microfacies characterization still needs to be further improved. Summary of the invention

[0006] The main purpose of the present invention is to provide a high-density azimuth prestack attribute seismic sedimentary microfacies characterization method, which fully utilizes the advantages of high-density seismic data azimuth data and significantly improves the accuracy of seismic sedimentary microfacies characterization.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] The present invention provides a high-density azimuth prestack attribute seismic sedimentary microfacies characterization method, which comprises the following steps:

[0009] Step 1, obtain data and information of high-density earthquake zones in the study area;

[0010] Step 2: Analyze wide-azimuth seismic gathers and select favorable offset data;

[0011] Step 3: Based on the distribution law of sedimentary phases, perform azimuth data superposition;

[0012] Step 4: Optimization and cluster analysis of sensitive attributes of seismic sedimentary microfacies in different directions;

[0013] Step 5: Drilling verification. If the prediction results do not match, perform regional secondary cluster analysis:

[0014] Step 6: Integrate geological data to accurately characterize seismic sedimentary microfacies.

[0015] To achieve the above object, the present invention can also adopt the following technical solutions:

[0016] In step 1, a single-point high-density seismic database is established in the study area to obtain all-round seismic data and drilling, logging, and mud recording data of actual wells in the work area.

[0017] In step 2, before performing wide azimuth seismic gather analysis, an anisotropy analysis of the azimuth gather is performed to control the anisotropy parameter error of the azimuth gather of the target layer within 0.2.

[0018] In step 3, the specific steps of performing azimuth data superposition include:

[0019] 1) Determine the sedimentary facies type in the study area and further subdivide the sedimentary facies into different sedimentary microfacies;

[0020] 2) Combine well and seismic data to determine the differential seismic reflection characteristics of different sedimentary microfacies and preliminarily characterize the distribution morphology of seismic sedimentary microfacies;

[0021] 3) Obtain seismic data in different azimuths of 0-180° in the study area and perform azimuth-based seismic data superposition.

[0022] In step 4, seismic attributes in different directions are extracted, and single attribute weight calculation and evaluation are performed. The calculation formula is as follows:

[0023]

[0024] Among them, Q is the azimuth attribute weight, n is the number of seismic attribute sample points, A ij is the specific score coefficient of a certain attribute parameter, A kj is the comprehensive score coefficient of the total attribute parameters;

[0025] According to the single attribute weight value, the azimuthal seismic attribute with the greatest correlation is determined as the preferred seismic sedimentary microfacies sensitive attribute.

[0026] Furthermore, the seismic attributes in different directions include amplitude type, energy type, spectrum statistics type, sequence statistics type, and correlation type.

[0027] In step 4, a fuzzy self-organizing neural network is used for training and learning to achieve unsupervised clustering analysis. The specific process is as follows:

[0028] S1. Determine the orientation attribute training sample X;

[0029] S2. Normalize each sample data:

[0030]

[0031] Among them, x i is the original data of the i-th orientation attribute sample; is x i Normalized data; max(x i ) is the maximum value of the original data; min(x i ) is the minimum value of the original data; The value range is (0-1) which is reasonable;

[0032] S3. Generate seismic phase map by azimuthal attribute cluster analysis.

[0033] In step 5, the lithology and rock physics characteristics of the target layer are determined based on the drilling data and logging data of the actual well, and the single well sedimentary microfacies division is completed; the division scheme is calibrated and verified with the preliminary azimuthal attribute cluster analysis results, and a secondary cluster analysis is performed in areas that do not match the predicted results.

[0034] Furthermore, a secondary cluster analysis was performed in the areas where the prediction accuracy was lower than 80%.

[0035] Furthermore, the random forest algorithm is used for secondary clustering analysis. The specific steps are as follows:

[0036] a. Construct the original sample set S based on seismic and logging data, S = {(x1, y1), (x2, y2), ..., (x n ,y n )};(x n ,y n ) are the coordinates of the azimuth sample points;

[0037] b. Generate a random forest set of k decision trees K = {k1, k2, ..., k j};

[0038] c. Calculate the importance function f(k i , K m , S) values ​​and sort them:

[0039] f(k i , K m ,S)=r·(w(x,y,K)-w(x,y,{K\k i}))

[0040] Among them, r represents the parameter mean, w(x, y, K) represents the classification interval of the full data sample (x, y); w(x, y, {K\k i} means to filter out k i The classification interval after

[0041] d. Eliminate the decision tree k with the lowest importance min , update the random forest sample set k m-1 ={k m \k min};

[0042] e. In sample set k m-1 Sampling data and constructing feature subspace;

[0043] f. Construct a decision tree, recursively analyze the multiple attributes of well bypass seismic, establish a random forest model, and use the majority voting strategy to obtain the final seismic sedimentary microfacies prediction results.

[0044] The high-density azimuth prestack attribute seismic sedimentary microfacies characterization method of the present invention is based on obtaining various data and information of the high-density seismic zone in the study area; performing wide-azimuth seismic gather analysis to optimize favorable offset distance data; based on the distribution law of different sedimentary phases, superposition processing of azimuth seismic data is carried out; sensitive attributes of seismic sedimentary microfacies in different azimuths are optimized to perform preliminary azimuth attribute clustering analysis; actual drilling verification is performed, and in areas that do not match the predicted results, a secondary clustering analysis is performed on the research results using a random forest algorithm; finally, the geological data are integrated and the azimuth seismic attribute clustering analysis results are integrated to achieve the effect of finely characterizing the seismic sedimentary microfacies.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The method of the present invention makes full use of the advantages of high-density seismic data in terms of azimuthal data, combines drilling, logging, mud logging and other geological data, conforms to the principles of sedimentology, has strong operability, and constrains seismic analysis with geological data, which significantly improves the accuracy of seismic sedimentary microfacies characterization, can effectively predict the planar distribution law of sedimentary microfacies in the target layer, and assists in high-quality exploration and development of oil and gas fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0048] Figure 1 This is a flow chart of the high-density azimuth prestack attribute seismic sedimentary microfacies characterization method described in Example 1 of the present invention;

[0049] Figure 2 Schematic diagram of seismic gather energy reflection comparison under different offset conditions in Example 1 of the present invention: ① medium offset>2700m; ② medium offset less than 2000m;

[0050] Figure 3 It is a schematic diagram of seismic reflections in different azimuth data ranges under the same favorable offset condition described in Example 1 of the present invention;

[0051] Figure 4 This is a schematic diagram of the cluster analysis of the azimuth attributes of the equally divided mode (4 azimuths) under the same offset condition described in Example 1 of the present invention;

[0052] Figure 5 It is a schematic diagram of the cluster analysis of the azimuth attributes of the non-equally divided mode (two azimuths, i.e., along / cutting the waterway) under the same offset condition described in Example 1 of the present invention;

[0053] Figure 6 This is a schematic diagram of the secondary orientation clustering analysis obtained based on the random forest algorithm described in Example 1 of the present invention. DETAILED DESCRIPTION

[0054] 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.

[0055] 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.

[0056] In order to enable those skilled in the art to more clearly understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below in conjunction with specific embodiments.

[0057] Example 1

[0058] like Figure 1 As shown, the high-density azimuth prestack attribute seismic sedimentary microfacies characterization method comprises the following steps:

[0059] Step 101, obtaining data and information of high-density earthquake zones in the study area:

[0060] Based on the single-point high-density seismic data in the study area, all-round seismic gather data and drilling, logging and mud recording data of actual wells in the work area are obtained.

[0061] Step 102, wide azimuth seismic gather analysis, and optimization of favorable offset data:

[0062] Anisotropy analysis was conducted on the azimuth data of the study area. The anisotropy of sandstone and mudstone was lower than 0.14, which is a weak anisotropic medium and meets the requirement of controlling the anisotropy parameter error of the azimuth gather of the target layer within 0.2.

[0063] Through the anisotropy analysis of the gathers, it is found that under the same offset condition, the far offset information is missing in the east-west azimuth, while the offset information in the north-south azimuth is complete; within different specific azimuth ranges, the energy reflection strength on the seismic profile is different, such as Figure 2 As shown in the figure, under the same azimuth conditions, when the offset is greater than 2700m, the gather energy is unbalanced and the seismic reflection is messy; when the offset is less than 2000m, the data in each azimuth are displayed completely, the gather energy is balanced, the seismic reflection characteristics are clearer, and the subtle geological information highlighted is richer, such as Figure 3 As shown; In summary, based on the principle of "complete offset data and balanced gather energy", the favorable offset data range of the study area is determined to be 0-2000m.

[0064] Step 103, based on the distribution law of sedimentary phases, carry out azimuth data stacking:

[0065] Based on the logging and lithological data of the study area, a detailed geological dissection of typical wells in the study area determined that the sedimentary facies type of the target area is a far-shore turbidite fan, which is a combination of supply channels and fan-shaped sediments in plane, with a wide spatial distribution and thin-layered distribution; this type of turbidite can be further subdivided into four types of sedimentary microfacies: braided channel, channel center, channel lateral edge, and channel boundary.

[0066] Through well-seismic calibration, the seismic reflection characteristics of four types of turbidite sedimentary microfacies were summarized: braided channels show medium-high frequency and strong reflection characteristics; the center of the channel shows high frequency and strong reflection characteristics; the side edge of the channel shows low frequency and medium-strong reflection characteristics; the boundary of the channel shows low frequency and strong reflection characteristics.

[0067] On the basis of clarifying the sedimentary laws and seismic reflection characteristics of turbidites and four types, the distribution morphology of turbidite seismic sedimentary microfacies was preliminarily clarified. According to the summarized seismic reflection characteristics of sedimentary facies, seismic data of different azimuths of 0-180° in the study area were obtained, and the initial azimuth seismic data superposition was carried out.

[0068] Step 104, optimization and cluster analysis of sensitive attributes of seismic sedimentary microfacies in different directions:

[0069] Under the favorable offset condition (<2000m) determined in step 102, extract the seismic attributes of different directions that are sensitive to the boundary and internal reflection of turbidites, including amplitude, energy, spectrum statistics, sequence statistics, and correlation. Perform single attribute weight calculation and evaluation, and the calculation formula is as follows:

[0070]

[0071] Among them, Q is the attribute weight, n is the number of earthquake attribute sample points, A ij is the specific score coefficient of a certain attribute parameter, A kj It is the comprehensive score coefficient of the total attribute parameters.

[0072] According to the single attribute weight value, it is determined that the azimuthal seismic attributes that are sensitive to the boundary and internal reflection of turbidites are amplitude and spectrum statistics attributes.

[0073] Fuzzy self-organizing neural network is used for training and learning to realize unsupervised cluster analysis.

[0074] Determine the orientation attribute training sample X.

[0075] Normalize each sample data:

[0076]

[0077] Among them, x i is the original data of the i-th orientation attribute sample; is x i Normalized data; max(x i ) is the maximum value of the original data; min(x i ) is the minimum value of the original data; The value range is (0-1) which is reasonable.

[0078] Generate azimuth attribute cluster analysis seismic facies map: Compare and study the turbidite sedimentary microfacies sensitivity attribute maps in accordance with the equal division mode (i.e., four azimuth ranges of 0-45°, 45-90°, 90-135°, and 135-180°) and the non-equal division mode (along the waterway and perpendicular to the waterway) respectively (e.g. Figure 4As shown in Figure 2, it was found that under the same offset condition, the orientation of the channel is the most sensitive orientation for the microfacies correlation of turbidite sediments (such as Figure 5 As shown), the multi-azimuth attribute clustering analysis of the sensitive azimuth seismic attributes is preferably performed.

[0079] Step 105: actual drilling verification, if the prediction results do not match, perform regional secondary cluster analysis:

[0080] Extract the logging sensitivity curve and lithology data of the actual drilling target layer, compare them with the preliminary azimuth attribute map of step 104, and use the random forest algorithm in the target layer section of the weak reflection area where the prediction result accuracy is less than 80%. The specific steps are as follows:

[0081] a. Construct S original sample set based on seismic and well logging data,

[0082] S={(x1, y1), (x2, y2),..., (x n ,y n )};(x n ,y n ) are the coordinates of the azimuth sample points;

[0083] b. Generate a random forest set of k decision trees K = {k1, k2, ..., k j};

[0084] c. Calculate the importance function f(k i , K m , S) values ​​and sort them:

[0085] f(k i , K m ,S)=r·(w(x,y,K)-w(x,y,{K\k i}))

[0086] Among them, r represents the parameter mean, w(x, y, K) represents the classification interval of the full data sample (x, y); w(x, y, {K\k i} means to filter out k i The classification interval after

[0087] d. Eliminate the decision tree k with the lowest importance min , update the random forest sample set k m-1 ={k m \k min};

[0088] e. In sample set k m-1 Sampling data and constructing feature subspace;

[0089] f. Construct a decision tree, recursively analyze the multiple attributes of well-side seismic traces, establish a random forest model, and use the majority voting strategy to obtain the final seismic sedimentary microfacies prediction results. Through multiple iterations, the prediction template is constructed, the target area is reclassified, and the seismic trace data of the actual calibration well is used as supervision to constrain the prediction model trace, and a secondary azimuth clustering analysis is performed, such as Figure 6 shown.

[0090] Step 106, integrate geological data to accurately characterize earthquake sedimentary microfacies:

[0091] Based on the geological data of the comprehensive study area, the secondary azimuth seismic attribute clustering analysis results in step 105 were integrated with the single well phase, lithofacies, and logging phase to achieve the unification of the turbidite seismic phase and sedimentary phase, further refine the lithological boundaries of the turbidite, and obtain the planar distribution map of the turbidite seismic sedimentary microfacies that best matches the actual sedimentary characteristics. The recognition accuracy was improved by 5% compared with the initial azimuth attribute map to 87%, which significantly improved the characterization accuracy of the turbidite seismic sedimentary microfacies.

[0092] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A high-density azimuth prestack attribute seismic sedimentary microfacies characterization method, characterized by: The following steps are involved: Step 1, obtain data and information of high-density earthquake zones in the study area; Step 2: Analyze wide-azimuth seismic gathers and select favorable offset data; Step 3: Based on the distribution law of sedimentary phases, perform azimuth data superposition; Step 4: Optimization and cluster analysis of sensitive attributes of seismic sedimentary microfacies in different directions; Step 5: Drilling verification. If the prediction results do not match, perform regional secondary cluster analysis: Step 6: Integrate geological data to accurately characterize earthquake sedimentary microfacies; In step 3, the specific steps of performing azimuth data superposition include: 1) Determine the sedimentary facies type in the study area and further subdivide the sedimentary facies into different sedimentary microfacies; 2) Combine well and seismic data to determine the differential seismic reflection characteristics of different sedimentary microfacies and preliminarily characterize the distribution morphology of seismic sedimentary microfacies; 3) Obtain seismic data in different azimuths of 0-180° in the study area and perform azimuth seismic data superposition; In step 4, seismic attributes in different directions are extracted, and single attribute weight calculation and evaluation are performed. The calculation formula is as follows: Among them, Q is the azimuth attribute weight, n is the number of seismic attribute sample points, A ij is the specific score coefficient of a certain attribute parameter, A kj is the comprehensive score coefficient of the total attribute parameters; According to the single attribute weight value, the azimuth seismic attribute with the greatest correlation is determined as the preferred seismic sedimentary microfacies sensitive attribute; In step 4, a fuzzy self-organizing neural network is used for training and learning to achieve unsupervised clustering analysis. The specific process is as follows: S1. Determine the orientation attribute training sample X; S2. Normalize each sample data: Among them, x i is the original data of the i-th orientation attribute sample; is x i Normalized data; max(x i ) is the maximum value of the original data; min(x i ) is the minimum value of the original data; The value range is (0-1) which is reasonable; S 3. Generate azimuth attribute cluster analysis seismic phase map; In step 5, the lithology and rock physics characteristics of the target layer are determined based on the drilling data and logging data of the actual well, and the single well sedimentary microfacies division is completed; the division scheme is calibrated and verified with the preliminary azimuthal attribute cluster analysis results, and a secondary cluster analysis is performed in areas that do not match the predicted results.

2. The method according to claim 1, characterized in that: In step 1, a single-point high-density seismic database is established in the study area to obtain all-round seismic data and drilling, logging, and mud recording data of actual wells in the work area.

3. The method according to claim 1, characterized in that: In step 2, before performing wide azimuth seismic gather analysis, an anisotropy analysis of the azimuth gather is performed to control the anisotropy parameter error of the azimuth gather of the target layer within 0.

2.

4. The method according to claim 1, characterized in that: The seismic attributes in different directions include amplitude type, energy type, spectrum statistics type, sequence statistics type and correlation type.

5. The method according to claim 4, characterized in that: A secondary cluster analysis was performed in areas where the prediction accuracy was less than 80%.

6. The method according to claim 1, characterized in that: Use the random forest algorithm to perform secondary clustering analysis. The specific steps are as follows: a. Construct the original sample set S based on seismic and logging data, S = {(x1, y1), (x2, y2), ..., (x n ,y n )};(x n ,y n ) are the coordinates of the azimuth sample points; b. Generate a random forest set of k decision trees K = {k1, k2, ..., k j }; c. Calculate the importance function f(k i , K m , S) values ​​and sort them: f(k i ,K m ,S)=r·(w(x,y,K)-w(x,y,{K\k i })) Among them, r · represents the parameter mean, w(x, y, K) represents the classification interval of the full data sample (x, y); w(x, y, {K\k i }) means to filter out k i The classification interval after d. Eliminate the decision tree k with the lowest importance min , update the random forest sample set k m-1 ={k m \k min }; e. In sample set k m-1 Sampling data and constructing feature subspace; f. Construct a decision tree, recursively analyze the multiple attributes of well bypass seismic, establish a random forest model, and use the majority voting strategy to obtain the final seismic sedimentary microfacies prediction results.

Images