A method for removing the influence of surrounding rock heterogeneity based on waveform component clustering

Abstract

The present invention relates to the field of oil and gas exploration technology, and relates to a method for removing the influence of surrounding rock heterogeneity based on waveform component clustering. By combining waveform decomposition and reconstruction with waveform clustering, it is possible to remove strong reflections while clarifying the geological significance represented by the seismic information, and combining multiple factors to verify the rationality of the results, and obtain the prediction results that best conform to the geological viewpoint. The present invention performs seismic phase analysis on waveform component bodies, associates geophysical information with geological information, and solves the problem of unclear geological significance of waveform decomposition data. Waveform clustering is used to characterize the seismic phase of the component body, and the seismic reflection characteristic plane is visualized to achieve data dimensionality reduction. The component body is sorted based on the seismic phase plane distribution map, thereby improving sorting efficiency. The seismic interpretation problem is converted into a qualitative problem of waveform components, and the influence of volcanic surrounding rocks on the seismic reflection characteristics of the target layer is removed to ensure that the proposed interpretation plan is rational. Each stage is subject to quality control using multiple factors.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration, and in particular to a method for removing the influence of surrounding rock heterogeneity based on waveform component clustering. Background Art

[0002] In oil and gas seismic exploration, volcanic rocks have a strong shielding and absorption effect on seismic waves, often resulting in strong and multiple reflections. When the strong impedance difference reflection interface caused by the volcanic rock overlaps with the reflection interface of the target layer, effective reservoir information is submerged in the strong reflection. Or, if the target reservoir is close to the strong reflection layer, the signal is shielded by the strong reflection, resulting in poor seismic data quality in the underlying igneous rock, which brings difficulties to seismic interpretation.

[0003] The methods for separating and identifying weakly reflecting targets in a strong reflection background can be mainly divided into the following two categories: one is to directly identify weakly reflecting targets by improving resolution and enhancing weak signals; the other is to identify weakly reflecting targets by removing strong reflection shielding and reducing its impact on weakly reflecting targets.

[0004] Currently, the primary method for removing strong reflection shielding is based on the decomposition and reconstruction of seismic wavelets. The most common methods include matching pursuit separation and wavelet transform. Numerous researchers have continuously developed improvements based on matching pursuit, addressing its high number of iterations and long computational time. They have also proposed numerous methods to optimize the matching pursuit algorithm. Regarding wavelet transforms, some researchers have used Morlet wavelets to decompose seismic signals into a series of seismic waves of varying frequencies, which, after reconstruction, enable high-precision reservoir identification and prediction. Similarly, many researchers have utilized other methods to remove strong reflection shielding. For example, based on singular value decomposition and EMD maximum energy decomposition, the maximum component is removed and then reconstructed to achieve the purpose of removing strong reflection shielding. Based on the theory of long and short cycles, PCA and Wheeler transform are combined to remove strong reflection shielding.

[0005] Seismic wave decomposition and reconstruction methods for removing strong reflection shielding have drawbacks. Many methods fail to clearly define the geological significance of the removed seismic information and are unable to verify the accuracy of the de-emphasized results. While some methods validate their results based on geological models, including paleogeomorphology, seismic attributes, and well logging information, without clearly defining the geological significance of the removed strong reflections, qualitative characterization of the seismic response is impossible, rendering the results non-unique. Such risky methods are difficult to implement in formal oil and gas exploration. Therefore, identifying the geological significance of waveform component data has become a pressing issue. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention proposes a method for removing the influence of surrounding rock heterogeneity based on waveform component clustering. Through the technical combination of waveform decomposition and reconstruction with waveform clustering, it can remove strong reflections while clarifying the geological significance represented by its seismic information. In addition, it combines multiple factors to verify the rationality of the results and obtain the prediction results that best conform to the geological viewpoint.

[0007] To achieve the above object, the technical solution of the present invention is as follows:

[0008] A method for removing the influence of surrounding rock heterogeneity based on waveform component clustering includes the following steps:

[0009] S1. Collect data from the study area, establish a seismic interpretation work area, and load relevant data; the data includes: regional survey data, drilling and logging data, 3D seismic data, conventional well logging curves, well deviation data, coring data, layer data, single well test production capacity data, etc.

[0010] S2. Sequence division of single wells, establishment of sequence stratigraphic framework, determination of sedimentary facies types and characteristics of target intervals in the study area;

[0011] S3. Based on actual drilling data, field outcrop data, and core slice data, divide the target reservoir section and establish reservoir well connections;

[0012] S4. Design a forward model of the volcanic rocks in the study area based on the lithologic thickness and geophysical parameters of the actual wells drilled, clarify the seismic reflection characteristics of the volcanic rocks, and verify the rationality of the well-seismic calibration;

[0013] Furthermore, in S4, based on the thickness of underground rock layers in the actual drilling and the rock physical parameters of each rock layer, a seismic forward model is established, and a Ricker wavelet excitation model is used to perform forward simulation.

[0014] S5. Determine the 3D seismic interpretation scheme based on the seismic forward modeling results and complete the detailed interpretation of the target layer;

[0015] Furthermore, in S5, referring to seismic forward modeling, volcanic rocks are identified on the seismic profile according to the seismic reflection characteristics of the volcanic rock development area, and volcanic rocks and non-volcanic rocks are divided. The interpretation phase of the volcanic rock development area is determined according to the volcanic rock development thickness.

[0016] S6. Combine seismic profiles and seismic forward modeling to create a time window to extract the original seismic volume attributes, characterize the volcanic rock boundaries in the study area, and clarify the volcanic rock thickness distribution pattern;

[0017] Furthermore, in S6, the time window range of the seismic slice is established with reference to the forward simulation. The time window size is preferably large enough to just include the period when the seismic reflection characteristics are affected by the volcanic rock. The plane distribution of the volcanic rock is characterized by using the seismic attribute slices. Finally, the plane attributes are combined with the seismic profile to further determine the distribution pattern of the volcanic rock thickness.

[0018] S7, establishing a training sample by opening a time window for the original earthquake volume, performing waveform component decomposition based on PCA, decomposing the training sample into multiple waveform component volumes, and sorting the component volumes carrying the main information;

[0019] Furthermore, in S7, the core idea of ​​principal component analysis (PCA) is to perform dimensionality reduction analysis on the data, which can effectively extract the "main" information in the data, reduce the dimensionality of complex data, thereby removing redundant components and mining the hidden information behind the data. The calculation process is as follows:

[0020] In m n-dimensional data, X n*m =[x1,x2,…x m ], where each x is an n-dimensional column vector,

[0021] (1) Decentralization,

[0022] (2) Calculate the covariance matrix,

[0023] (3) Perform eigenvalue decomposition on the covariance matrix to obtain the characteristic matrix (arrange the columns in descending order of eigenvalues), and take the first k columns to form the matrix P n*k , P is equivalent to a coordinate system, and each column in P is a coordinate axis;

[0024] (4) Projecting the original data into the P coordinate system will yield the reduced-dimensional data.

[0025] This set of dimensionality-reduced data is sorted based on the proportion of waveform component information. The higher the order, the greater the amount of information carried and the higher the proportion. Based on the amount of information carried by the waveform component, the component carrying the main information is selected.

[0026] S8. Open a time window, perform waveform clustering calculation on the selected component volume, extract the component volume seismic phase distribution plan, combine forward simulation and seismic reflection characteristics, and determine the volcanic rock seismic phase;

[0027] Furthermore, in S8, waveform clustering calculation is performed on the component volume based on the time window established in S6 to obtain a seismic phase plane distribution map of the component volume.

[0028] S9. Based on the seismic phase plane distribution map of each component, analyze the coupling relationship between the seismic phase distribution of volcanic rocks and the actual distribution range of volcanic rocks. Combined with the thickness distribution law of volcanic rocks, determine the component volume carrying volcanic rock information and the thickness of volcanic rocks in each component volume;

[0029] Furthermore, in S9, the waveform component decomposition decomposes the original seismic volume into multiple waveform component volumes according to the seismic waveform. Since the seismic waveform is directly affected by the changes in the lithologic sequence of the underground rock formations, each component volume can be considered to carry certain lithologic information.

[0030] The traditional method of identifying the component volumes carrying volcanic rock information involves integrating the volcanic rock forward simulation in S4 and comparing the component volume profile with the original seismic volume profile to determine the component volume carrying volcanic rock information. This method has no process constraints and is guided only by the reservoir prediction goal, ignoring the uncertainty of the component volumes removed in the process.

[0031] Based on waveform clustering, the seismic phase plane distribution of each component is identified, allowing for the rapid and efficient identification of multiple volcanic-related component volumes and their geological significance. Seismic profiles of the component volumes are identified, and the correlation between the volcanic rock boundaries and the seismic phase plane on the component seismic profiles is verified. Profile review verifies that the component waveforms are consistent with the seismic phases (waveform characteristics) on the plane. If the profile reflection characteristics are inconsistent with the seismic phase, consider the horizon and time window issues, and proceed to S8 for adjustment and retesting. For robustness (or in the case of poor data conditions), the correlation threshold can be appropriately lowered to include more associated component volumes. Multiple associated component volumes are then permuted and combined to form multiple removal schemes, and the accuracy of volcanic component removal is further improved based on post-reconstruction geological quality control. This is equivalent to screening associated component volumes from multiple steps and perspectives, ensuring both a planar match for volcanic rock distribution and a result consistent with the inheritance of geomorphology, reservoirs, and sedimentary facies.

[0032] S10, removing the component volume carrying volcanic rock information, and reconstructing the remaining waveform component volumes to obtain a reconstructed seismic volume;

[0033] Furthermore, the waveform reconstruction in S10 is to re-combine and calculate the waveform component bodies after the waveform decomposition. In order to enhance the feasibility of the method, a combination method is specified: the component bodies are divided into two parts, one part is a component body with certain association (a component body that clearly contains volcanic rock information) and the other part is a component body with weak certainty. The weakly associated component bodies are combined with the certain associated component bodies in a permutation method to form multiple schemes. For each scheme, the waveform component bodies that need to be removed are reconstructed.

[0034] S11. Based on well-seismic calibration, the actual well reservoir is projected onto the seismic section, and the seismic reflection characteristics of the reservoir section of the reconstructed seismic volume are combined to select the attributes to predict the reservoir plane distribution;

[0035] Furthermore, in S11, based on well seismic calibration, the seismic reflection characteristics of the actual well reservoir development area on the seismic volume are statistically reconstructed, and the attributes of the reconstructed seismic volume are extracted using root mean square amplitude slices;

[0036] S12. Based on the distribution of reservoirs in actual drilling, paleo-geomorphology, and sedimentary facies inheritance, reservoir prediction geological quality control is performed from multiple angles. If the attribute prediction accuracy exceeds the requirement and conforms to geological laws, the prediction is completed. If the conformity is poor, return to step S10.

[0037] Through seismic forward modeling, we determined the extent to which changes in volcanic rock thickness affect the seismic reflection characteristics of the original formations. By comparing the differences in seismic reflection characteristics between the two without volcanic rock development, we provided a basis for the horizon interpretation scheme. Furthermore, we used reflection coefficients and Ricker wavelet convolution to calculate synthetic records for well seismic calibration, verifying the degree of agreement between the forward modeling and actual drilling seismic profiles and analyzing the development trends of volcanic rocks in the study area.

[0038] Furthermore, in step S12, the consistency between the wellbore and seismic data of the actual wellbore reservoir is first verified. The reservoir distribution predicted by the amplitude attributes must correspond to the wellbore information. On this basis, the paleo-geomorphology and sedimentary facies inheritance are further verified. After clarifying that the reservoir is controlled by facies, the sedimentary phase can be used to evaluate the reservoir. Based on the continuity of sedimentary facies evolution, the distribution characteristics of the upper and lower layers can be used to infer the distribution trend of the target sedimentary facies belt to evaluate the rationality of the reservoir's planar extension direction. In terms of geomorphology, priority is given to reservoirs located in favorable locations such as paleo-geomorphological highlands and slope breaks. If the attribute prediction accuracy exceeds the requirement and matches the above three conditions, the prediction is completed. If the match is poor, return to step S9.

[0039] Compared with the existing technology, the advantages of the present invention are:

[0040] (1) Compared with previous methods for removing strong reflection shielding, this method performs seismic phase analysis on waveform component volumes for the first time, associates geophysical information with geological information, and solves the problem of unclear geological significance of waveform decomposition data.

[0041] (2) The process of sorting component volumes does not completely rely on the profile comparison method. Waveform clustering is used to characterize the seismic phase of the component volume, and the seismic reflection characteristic plane is visualized to achieve data dimensionality reduction. The component volume is sorted based on the seismic phase plane distribution map, which effectively improves the sorting efficiency.

[0042] (3) This method transforms the complex multi-solution seismic interpretation problem into a qualitative problem of waveform components, removes the influence of volcanic surrounding rocks on the seismic reflection characteristics of the target layer, ensures the rationality of the proposed interpretation scheme, and uses multiple factors for quality control at each stage, which can be iteratively tested to ensure that the final result is consistent with geological understanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to the present invention;

[0044] Figure 2 This is a diagram of the formation well connection grid in an embodiment of the present invention;

[0045] Figure 3 This is a diagram of a reservoir well connection grid in an embodiment of the present invention;

[0046] Figure 4 It is the synthetic record calibration in the embodiment of the present invention;

[0047] Figure 5 This is the forward modeling of volcanic rock earthquakes in the embodiment of the present invention;

[0048] Figure 6 The original seismic volume amplitude attribute map in the embodiment of the present invention;

[0049] Figure 7 This is a plane distribution diagram of component seismic phases in an embodiment of the present invention;

[0050] Figure 8 Reconstructing the seismic volume amplitude attribute map in an embodiment of the present invention;

[0051] Figure 9 This is the sedimentary phase diagram of Maoerdi in the embodiment of the present invention;

[0052] Figure 10 This is the Maodi sedimentary phase diagram in the embodiment of the present invention

[0053] Figure 11 This is the paleo-geomorphological restoration map of the target layer in the embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples.

[0055] like Figure 1 As shown in FIG, a method for removing the influence of surrounding rock heterogeneity based on waveform component clustering includes the following steps:

[0056] S1. Collect and organize basic data, establish seismic interpretation work area, and load relevant data;

[0057] S2. Divide the single well sequence and determine the sedimentary facies type and characteristics of the target interval in the study area based on the single well sedimentary facies analysis and combined with the sedimentary phase comparison of connected wells;

[0058] like Figure 2 As shown in the figure, the sequences are divided based on the GR logging curves and an isochronous stratigraphic framework is established.

[0059] S3. Divide the actual drilling reservoir section based on the GR curve and the three porosity logging curves (AC curve, CNL curve, and DEN curve), and analyze the reservoir development advantage area through well connection profiles;

[0060] like Figure 3 As shown in the figure, two sets of reservoirs are developed at the top of the Mao 2nd Member in the study area. The reservoir thickness is thin. According to the rock physical properties, the reservoir property of the formation at Well Shangfeng 8 in this section is the best.

[0061] S4. Calculate the thickness of various lithologies, acoustic wave velocity, and rock density near the target layer during actual drilling, establish a theoretical formation model, and perform forward simulation using Ricker wavelet excitation;

[0062] Using acoustic and density logging curves to calculate synthetic records, perform well-seismic calibration, confirm that the original seismic profile contains seismic reflection characteristics of volcanic rock formations, and adjust the forward model according to the seismic profile until the forward simulation results match the profile;

[0063] like Figure 4 As shown in the figure, the well-seismic calibration determines the position of the Longtan Formation in the original seismic body in the study area, and analyzes the seismic reflection characteristics of the Longtan Formation and its surrounding rocks.

[0064] like Figure 5 As shown in the figure, the forward simulation results confirm that in the volcanic rock distribution area at the bottom of the Longtan Formation in the study area, the wave crest is pulled up, which is about half a phase away from the area without volcanic rocks, and is a parallel strong reflection. As the thickness of the volcanic rock decreases, the wave crest amplitude gradually weakens, the phase gradually moves down, and is in a complex wave shape, while the area without volcanic rocks is a stable parallel strong reflection.

[0065] S5. In this study, volcanic rocks developed in the southwest of the study area, showing complex wave or biaxial weak-moderate reflection characteristics on the seismic profile. Normal shale at the bottom of the Longtan Formation in the east showed parallel strong reflection. Through seismic forward simulation and analysis of the original seismic volume, the interpretation phase of the target layer was determined, and fine tracking was performed to achieve the target layer interpretation.

[0066] S6. Drift up 15 ms and down 10 ms along the target horizon to establish a time window and perform root mean square amplitude (RMS) attribute slicing on the original seismic volume.

[0067] like Figure 6 As shown in the figure, it can be seen from the properties that volcanic rocks are distributed in the western part of the study area, and the volcanic rock overflow channel is obvious. There is no volcanic rock distribution in the east, which realizes the delineation of the volcanic rock boundary.

[0068] S7. Set up a time window along the target layer and select appropriate training samples. At the same time, select the appropriate number of components based on the problem requirements. In this case, the inventor selected the original seismic body within the time window of 100ms up and 200ms down along the layer as the training sample, with the number of components being 30; the waveform component bodies are sorted according to the amount of information they carry from the original seismic body, and the one ranked first is called sub-component body No. 1, and so on. In this study, the component bodies with information accounting for more than 1% are selected, that is, the first eight waveform component bodies;

[0069] S8: Set up a time window along the target horizon with an upward drift of 15ms and a downward drift of 10ms, and perform waveform clustering calculation on the components 1-8;

[0070] like Figure 7 As shown in the figure, the plane distribution of the seismic phases of the components 1-8 is obviously different, and some seismic phases are highly consistent with the distribution of volcanic rocks.

[0071] S9. Based on the planar seismic facies distribution of components 1-8 in S8, combined with factors such as seismic forward modeling and volcanic rock boundaries, assign geological significance to each seismic facies. Based on the consistency between the seismic facies and the volcanic rock distribution, classify the components as correlated and those with weak correlation. During the seismic profile review, check the consistency between the profile reflection characteristics and the seismic phase waveform characteristics. If the consistency is poor, return to S8 and modify the time window range. In this study, components 1 and 3 were confirmed to be correlated, components 4 and 5 were correlated but weaker than the former, and the remaining components were weakly correlated.

[0072] Components S10, 1, and 3 carry information about thick volcanic rocks with distinct seismic facies and a strong correlation with them. Components 4 and 5, on the other hand, carry information about thin volcanic rocks with difficult-to-identify seismic facies and a weak correlation. Therefore, this study employed three reconstruction schemes: eliminating components 1 and 3, eliminating components 1, 3, and 4, and eliminating components 1, 3, 4, and 5, and reconstructing the remaining waveform components.

[0073] S11: The reservoir reflection characteristics in the reconstructed seismic volume are strong amplitude, high frequency "bright spot" reflections. RMS amplitude attribute slicing is performed on the reconstructed seismic volume in the context of the S6 time window to amplify the difference between the reservoir amplitude characteristics and non-reservoir amplitude characteristics.

[0074] like Figure 8 As shown in the figure, after excluding the No. 1, 3, 4 and 5 components, the intensity is significantly reduced, and the reservoir development area is significantly different from the undeveloped area. The Maokou top reservoir is mainly developed in the southwest of the study area, and reservoirs are developed in some areas in the northeast.

[0075] S12. In this study, based on the attribute slices after three different reconstruction schemes, the reservoir development of actual drilling, the inheritance relationship of high-quality facies belts, and the paleo-geomorphology before the deposition of Maoerxia were analyzed as references. Finally, the reconstructed seismic volume without the 1st, 3rd, 4th, and 5th components was selected.

[0076] like Figure 9 and Figure 10 As shown in the figure, the shoal from Maoyidi to Maoerdi in the study area gradually migrated to the northeast, and the reservoir conditions at the top of Maoerdi met the shoal migration pattern of the Maokou Formation. The facies belt favorable for reservoir development has obvious inheritance.

[0077] like Figure 11 As shown, the study area presents a geomorphic trend of high in the west and low in the east, with the topography extending in the northwest-south direction. The reservoirs characterized by the amplitude attributes are developed in the slope-break zone with higher geomorphology, shallow water and high energy, which is in line with geological laws.

[0078] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the implementation methods of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A method for removing the influence of surrounding rock heterogeneity based on waveform component clustering, characterized in that: The following steps are involved: S1. Collect data in the study area, establish a seismic interpretation area, and load relevant data; Data include: regional survey data, drilling and logging data, 3D seismic data, conventional well logging curves, well deviation data, coring data, layer data, and single well test productivity; S2. Sequence division of single wells, establishment of sequence stratigraphic framework, determination of sedimentary facies types and characteristics of target intervals in the study area; S3. Based on actual drilling data, field outcrop data, and core slice data, divide the target reservoir section and establish reservoir well connections; S4. Design a forward model of the volcanic rocks in the study area based on the lithologic thickness and geophysical parameters of the actual wells drilled, clarify the seismic reflection characteristics of the volcanic rocks, and verify the rationality of the well-seismic calibration; S5. Determine the 3D seismic interpretation scheme based on the seismic forward modeling results and complete the detailed interpretation of the target layer; S6. Combine seismic profiles and seismic forward modeling to create a time window to extract the original seismic volume attributes, characterize the volcanic rock boundaries in the study area, and clarify the volcanic rock thickness distribution pattern; S7, establishing a training sample by opening a time window for the original earthquake volume, performing waveform component decomposition based on PCA, decomposing the training sample into multiple waveform component volumes, and sorting the component volumes carrying the main information; S8. Open a time window, perform waveform clustering calculation on the selected component volume, extract the component volume seismic phase distribution plan, combine forward simulation and seismic reflection characteristics, and determine the volcanic rock seismic phase; S9. Based on the seismic phase plane distribution map of each component, analyze the coupling relationship between the seismic phase distribution of volcanic rocks and the actual distribution range of volcanic rocks. Combined with the thickness distribution law of volcanic rocks, determine the component volume carrying volcanic rock information and the thickness of volcanic rocks in each component volume; S10, removing the component volume carrying volcanic rock information, and reconstructing the remaining waveform component volumes to obtain a reconstructed seismic volume; S11. Based on well-seismic calibration, the actual well reservoir is projected onto the seismic section, and the reservoir plane distribution is predicted by selecting the attributes in combination with the seismic reflection characteristics of the reservoir segment of the reconstructed seismic volume; S12. Based on the actual well reservoir distribution, paleo-geomorphology, and sedimentary facies inheritance, reservoir prediction geological quality control is performed from multiple angles. If the attribute prediction accuracy exceeds the requirement and conforms to the geological laws, the prediction is completed. If the conformity does not meet the requirements, return to step S10.

2. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In said S4, based on the thickness of underground rock layers and rock physical parameters of each rock layer in the actual drilling, a seismic forward model is established, and a Ricker wavelet excitation model is used to perform forward simulation; Through seismic forward modeling, the scope of the impact of changes in volcanic rock thickness on the seismic reflection characteristics of the original strata was determined, and the difference in seismic reflection characteristics between the two when there was no volcanic rock development was compared to provide a basis for the stratigraphic interpretation scheme. At the same time, the reflection coefficient and Ricker wavelet convolution were used to calculate the synthetic record for well seismic calibration, to test the degree of consistency between the forward modeling and the actual drilling seismic profile, and to analyze the development trend of volcanic rocks in the study area.

3. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In S5, referring to seismic forward modeling, volcanic rocks are identified on the seismic profile according to the seismic reflection characteristics of the volcanic rock development area, and volcanic rocks and non-volcanic rocks are divided. The interpretation phase of the volcanic rock development area is determined according to the volcanic rock development thickness.

4. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In S6, the time window range of the seismic slice is established with reference to the forward simulation. The time window size can include the period when the seismic reflection characteristics are affected by the volcanic rock. The plane distribution of the volcanic rock is characterized by using the seismic attribute slices. Finally, the plane attributes are combined with the seismic profile to further determine the thickness distribution pattern of the volcanic rock.

5. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In S7, the core idea of ​​principal component analysis (PCA) is to perform dimensionality reduction analysis on the data and extract the "main" information from the data. The calculation process is as follows: In m n-dimensional data, X n*m =[x1,x2,…x m ], where each x is an n-dimensional column vector; (1) Decentralization, ; (2) Calculate the covariance matrix, ; (3) Perform eigenvalue decomposition on the covariance matrix to obtain the characteristic matrix, arrange the columns in descending order of eigenvalues, and take the first k columns to form the matrix P n*k , P is equivalent to a coordinate system, and each column in P is a coordinate axis; (4) Projecting the original data into the P coordinate system will yield the reduced-dimensional data. ; This set of dimensionality-reduced data is sorted according to the proportion of waveform component information. The higher the ranking, the greater the amount of information it carries and the higher the proportion. Based on the amount of information carried by the waveform component, the component carrying the main information is selected.

6. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In S8, waveform clustering calculation is performed on the component volume based on the time window established in S6 to obtain a seismic phase plane distribution map of the component volume.

7. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: In S9, the waveform component decomposition decomposes the original seismic volume into multiple waveform component volumes according to the seismic waveform; since the seismic waveform is directly affected by the changes in the lithologic sequence of the underground rock formations, each component volume is considered to carry lithologic information; Based on waveform clustering, the seismic phase plane distribution of each component is identified, multiple components related to volcanic rocks are identified, and their geological significance is given; Identify the component volume seismic profile and check the matching correlation between the volcanic rock boundary and the seismic phase plane on the component volume seismic profile. The profile review is to check whether the component volume waveform is consistent with the earthquake on the plane. If the profile reflection characteristics are inconsistent with the seismic phase, consider the layer position and time window problems, and adjust and re-check in S8.

8. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: The waveform reconstruction in S10 is to recombine and calculate the waveform component bodies after the waveform is decomposed, and a combination method is specified: the component body is divided into two parts, one part is definitely associated, and the other part is not very certain. The associated component bodies with less certainty are arranged and combined with the definitely associated component bodies to form multiple schemes. For each scheme, the waveform component bodies that need to be removed are reconstructed.

9. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1 is characterized in that: Said S11 is based on well seismic calibration, statistically reconstructing seismic reflection characteristics of the actual well reservoir development area on the seismic volume, and extracting and reconstructing seismic volume attributes using root mean square amplitude slices.

10. The method for removing the influence of surrounding rock heterogeneity based on waveform component clustering according to claim 1, characterized in that: Said S12, first, checks the consistency of the actual drilling reservoir with the wellbore and seismic data. The reservoir distribution predicted by the amplitude attributes must correspond to the wellbore information. On this basis, further checks the paleo-geomorphology and sedimentary facies inheritance. On the basis of clarifying that the reservoir is controlled by the facies, the sedimentary phase is used to evaluate the reservoir. Based on the continuity of the sedimentary facies evolution, the distribution characteristics of the upper and lower small layers are used to infer the distribution trend of the target layer sedimentary facies belt to evaluate the rationality of the reservoir plane extension direction. In terms of geomorphology, priority is given to the reservoir being located in the paleo-geomorphological highland and the favorable position of the slope break. If the attribute prediction accuracy is greater than the requirement and is consistent with the actual drilling reservoir distribution, paleo-geomorphology, and sedimentary facies inheritance, the prediction is completed. If the consistency is poor, return to step S9.

Images