A near-offset data reconstruction method based on iterative seismic interferometry

Abstract

The application discloses a near-offset data reconstruction method based on an iterative seismic interference method and relates to the technical field of marine seismic exploration, which comprises the following steps: detecting a seismic wave by using a detector and taking the seismic wave as a real record; calculating a missing signal by using interference interpolation based on the real record, taking the missing signal as a virtual record; and matching the virtual record with the real record by using a least square matching filter. By taking a focusing transform as an equalization operator, the matching degree of the reconstructed virtual shot gather data and the original data is improved by using the globality and the characteristic that the signal and the noise can be separated. Furthermore, in order to make the result more accurate, the interpolation reconstruction process is modified into an iterative method, the reconstructed data after interference is alternately updated with the equalization operator, the final interpolation result is well fitted with the original data, and therefore the quality of subsequent data processing is improved, which is beneficial to improving the robustness.

Description

Technical Field

[0001] This invention relates to the technical field of marine seismic exploration, and in particular to a near-offset data reconstruction method based on iterative seismic interferometry. Background Technology

[0002] A significant problem commonly encountered during seismic acquisition is the potential for data gaps in recorded shot gathers due to survey limitations. This is particularly prevalent in marine seismic surveys, where near-migration gaps are prevalent. This issue negatively impacts various seismic processing steps. Before predicting multiples, these missing near-migration traces must be estimated as part of the surface-correlated multiple suppression process. The presence of missing traces at near-migration locations significantly affects multiple suppression effectiveness, especially in shallow water areas. In summary, near-migration gaps pose a significant challenge in seismic acquisition, affecting not only the processing of primary reflections but also directly impacting the improvement of subsurface imaging. Therefore, an effective and accurate interpolation method is urgently needed to fill in these missing near-migration traces.

[0003] Most interpolation methods focus on reconstructing missing data using information from a single reflection in the data, or using the entire shot gather record. Furthermore, these methods tend to perform well when interpolating randomly missing data, but poorly when extrapolating data (i.e., filling in near-offset missing traces). Moreover, the quality of the extrapolated trace gradually decreases as the distance to the missing trace increases; some relatively simple methods, such as linear interpolation of neighboring traces, may yield results that differ significantly from the true missing trace.

[0004] In recent years, the development of interferometry has provided new approaches to solving the problem of missing near-migration traces. This technique can flexibly generate the properties of virtual shot gather records at arbitrary locations, performing data interpolation by performing seismic interferometry. By cross-correlating the wavefields recorded by the detectors, the seismic impulse response (Green's function) between the two detectors can be extracted. The traces generated by these cross-correlations allow for approximations of the traces recorded if the source were placed at the detector location, as if the virtual source were present at the detector location. A series of studies have proposed and developed interferometric interpolation methods for estimating missing seismic traces using multiple waves.

[0005] These interferometric interpolation methods utilize far-field approximations and limitations in acquisition aperture during the calculation process, resulting in source wavelets in the interpolated results often differing from those in the actual data. To mitigate these interferences, traditional interferometric interpolation methods employ 1D least-squares matched filters to match virtual records with real records. However, due to the lack of original data, the matching process in interpolation algorithms can only select adjacent traces as approximations. When faced with missing near-migrated seismic data, the quality of the virtual shot gather after matched filtering may still be insufficient, adversely affecting the final interpolation results and even reducing the accuracy of subsequent data processing. Summary of the Invention

[0006] The technical problem solved by this invention is that when faced with near-migrated missing seismic data, the quality of the virtual shot gather after matched filtering may still not meet the requirements of existing methods, which will have an adverse effect on the final interpolation results and even reduce the accuracy of subsequent data processing.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: a near-offset data reconstruction method based on iterative seismic interferometry, comprising: detecting seismic waves using a geophone and using the seismic waves as the real record; calculating missing signals based on the real record using interferometric interpolation and using the missing signals as virtual records; matching the virtual records with the real records using a least-squares matched filter; arranging the matched data records into a data volume in chronological order according to shot gather-geophone gather; converting the data volume from the data domain to the focusing domain and removing noise signals; and combining the converted data with the original data to determine whether it meets the iterative requirements.

[0008] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the method utilizes seismic waves from a detector and includes the seismic waves as a true record.

[0009] Install two or more detectors below the water surface:

[0010] The seismic wave signals detected by the detector are used as the actual records.

[0011] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the method includes: calculating missing signals using interferometric interpolation based on the real records, and treating the missing signals as virtual records, including:

[0012] Calculate the cross-correlation between multiple seismic wave signals recorded by one geophone and a single seismic wave signal recorded by another geophone;

[0013] Then, duplicate paths are eliminated to obtain the missing signals at the missing locations;

[0014] The missing signal is used as a virtual record.

[0015] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the mathematical expression of the interferometric interpolation equation is:

[0016]

[0017] in, Indicates the location where the earthquake source was triggered. Indicates the position of the detector. Represents the Green's function in the frequency domain. Represents earthquake data. Indicates the integral symbol, express The survey line in question The symbol is for partial differentials. Represents angular frequency. The surface representing the distribution of earthquake sources. The surface representing the detector distribution, and Located below the water surface, Represents a semicircle at infinity. Indicates complex conjugation. Indicates the direction of the normal.

[0018] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the frequency domain calculation expression is:

[0019]

[0020] in, Let I represent the wave number, L represent the real part, and L represent the plane in which the detector is located.

[0021] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the method includes: matching virtual records with real records using a least-squares matched filter, which comprises:

[0022] By calculating the excitation at position L, L A L B The cross-correlation of the two received records yields L. A Position excitation, L B The virtual record of location reception is expressed mathematically as follows:

[0023]

[0024] in, This represents a true record. This represents a virtual record before matching. This refers to a matched filter, which is defined as follows:

[0025]

[0026] in, The damping factor is represented by the fact that both virtual and real records are transformed into small overlapping blocks between adjacent windows. In each local window, a matched filter estimated through trial and error is used for the virtual record.

[0027] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the matched data records are arranged into a data volume in chronological order according to shot gather-detector gather, and its mathematical expression is:

[0028]

[0029] in, Indicates the position of the detector. The location of the epicenter is indicated by 'u', and the number of shots is indicated by 'u'.

[0030] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the data is transformed from the data domain to the focus domain, and the mathematical expression for the positive focus transformation is:

[0031]

[0032] Where Q represents the focal domain data and F represents the positive focus transform. In the focus transform, the effective signal can be focused to the focal point, thereby creating a difference between the distribution areas of the effective signal and noise in the focal domain.

[0033] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the method involves: converting the data volume from the data domain to the focus domain and removing noise signals, including:

[0034]

[0035]

[0036] in, Indicates the offset distance. Indicates time, Indicates the filtering range; R is an abbreviation for real and has no practical meaning. This indicates the data Perform filtering operation, when The original signal is retained if the signal is active, otherwise it is set to zero. This represents the parameters given by the operator, and the specific value depends on the range of data that needs to be filtered out.

[0037] As a preferred embodiment of the near-offset data reconstruction method based on iterative seismic interferometry described in this invention, the method involves combining the converted data with the original data to determine whether the iterative requirements are met, including:

[0038]

[0039] in, This represents the portion of the original seismic data excluding near-migration missing data. If this condition is not met, data D is returned to the first step, and the entire iterative process is repeated until the result meets the accuracy requirements.

[0040] The beneficial effects of this invention are as follows: By using the focusing transformation as the equalization operator, and leveraging its global nature and ability to separate signal and noise, the matching degree between the reconstructed virtual shot set data and the original data is improved. Furthermore, to achieve higher accuracy, this method modifies the interpolation reconstruction process into an iterative method, using the interferometrically reconstructed data and the equalization operator to update alternately, resulting in a good fit between the final interpolation result and the original data. This improves the quality of subsequent data processing and enhances robustness. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the basic process of a near-offset data reconstruction method based on iterative seismic interferometry, provided as an embodiment of the present invention.

[0042] Figure 2 This is a schematic diagram of the focusing transformation of a near-offset data reconstruction method based on iterative seismic interferometry, provided as an embodiment of the present invention.

[0043] Figure 3 This is a schematic diagram of the basic process of a near-offset data reconstruction method based on iterative seismic interferometry, provided as an embodiment of the present invention.

[0044] Figure 4 Comparison of processing results of a near-offset data reconstruction method based on iterative seismic interferometry provided in an embodiment of the present invention: (a) Original data map; (b) Original data map with missing near-offset data; (c) Reconstruction result map of conventional interferometry; (d) Reconstruction result map of the present invention. Detailed Implementation

[0045] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0046] Example 1

[0047] Reference Figures 1-4 As an embodiment of the present invention, a near-offset data reconstruction method based on iterative seismic interferometry includes:

[0048] Step 1: Detect seismic waves using a geophone and record the seismic waves as actual data, including:

[0049] The seismic waves are detected by a geophone and recorded as real data.

[0050] Install two or more detectors below the water surface:

[0051] The seismic wave signals detected by the detector are used as the actual records.

[0052] Step 2: Calculate the missing signal using interferometric interpolation based on the actual record, and treat the missing signal as a virtual record, including:

[0053] Calculate the cross-correlation between multiple seismic wave signals recorded by one geophone and a single seismic wave signal recorded by another geophone;

[0054] Then, duplicate paths are eliminated to obtain the missing signals at the missing locations;

[0055] The missing signal is used as a virtual record.

[0056] The mathematical expression for the interference interpolation equation is:

[0057]

[0058] in, Indicates the location where the earthquake source was triggered. Indicates the position of the detector. Represents the Green's function in the frequency domain. Represents earthquake data. Indicates the integral symbol, express The survey line in question The symbol is for partial differentials. Represents angular frequency. The surface representing the distribution of earthquake sources. The surface representing the detector distribution, and Located below the water surface, Represents a semicircle at infinity. Indicates complex conjugation. Indicates the direction of the normal.

[0059] The calculation expression for the frequency domain is:

[0060]

[0061] in, Let I represent the wave number, L represent the real part, and L represent the plane in which the detector is located.

[0062] Step 3: Match the virtual records with the real records using a least-squares matched filter, including:

[0063] By calculating the excitation at position L, L A L B The cross-correlation of the two received records yields L. A Position excitation, L B The virtual record of location reception is expressed mathematically as follows:

[0064]

[0065] in, This represents a true record. This represents a virtual record before matching. This refers to a matched filter, which is defined as follows:

[0066]

[0067] in, The damping factor is represented by the fact that both virtual and real records are transformed into small overlapping blocks between adjacent windows. In each local window, a matched filter estimated through trial and error is used for the virtual record.

[0068] Step 4: Arrange the matched data records into a data volume in chronological order according to the shot gather-detector gather. The mathematical expression for this volume is:

[0069]

[0070] in, Indicates the position of the detector. The location of the epicenter is indicated by 'u', and the number of shots is indicated by 'u'.

[0071] The mathematical expression for a positive focusing transformation, which converts data from the data domain to the focusing domain, is:

[0072]

[0073] Where Q represents the focal domain data and F represents the positive focus transform. In the focus transform, the effective signal can be focused to the focal point, thereby creating a difference between the distribution areas of the effective signal and noise in the focal domain.

[0074] In the focus transformation, the effective signal can be focused to the focal point, thereby creating a difference between the distribution areas of the effective signal and noise in the focus domain. The focus transformation can be considered as weighted autocorrelation, which can eliminate the phase of the in-phase axis in the focus domain and thus reduce the order of the reflected in-phase axis: the primary wave is represented as the focal point of the focus domain, the first-order multiple wave is reduced to the primary wave, and the second-order multiple wave is converted into the first-order multiple wave.

[0075] Step 5: Convert the data volume from the data domain to the focus domain, and remove noise signals, including:

[0076]

[0077]

[0078] in, Indicates the offset distance. Indicates time, Indicates the filtering range; R is an abbreviation for real and has no practical meaning. This indicates the data Perform filtering operation, when The original signal is retained if the signal is active, otherwise it is set to zero. This represents the parameters given by the operator, and the specific value depends on the range of data that needs to be filtered out.

[0079] Step 6: Combine the transformed data with the original data to determine whether the iteration requirements are met, including:

[0080]

[0081] in, This represents the portion of the original seismic data excluding near-migration missing data. If the conditions are not met, data D is returned to the first step, and the entire iterative process is repeated until the result meets the accuracy requirements. The conditions for meeting the requirements depend on the accuracy requirements of the actual exploration; typically, continuous phase axes and amplitude errors within a certain range are sufficient. The overall process for reconstructing near-migration data involves first calculating virtual records at missing locations through cross-correlation of existing records, then applying matched filtering and focusing transformation for equalization, and finally re-interferometrically calculating and iterating the equalization results until the accuracy requirements are met.

[0082] After transforming the data to the focal domain, the effective signal and the noise signal will be located at different positions. Therefore, the noise position is removed and the effective signal is retained.

[0083] By using the focusing transformation as the equalization operator, leveraging its global nature and ability to separate signal and noise, the matching degree between the reconstructed virtual shot set data and the original data is improved. Furthermore, to enhance the accuracy of the results, this method modifies the interpolation reconstruction process into an iterative method, using the interferometrically reconstructed data and the equalization operator to update alternately, resulting in a good fit between the final interpolation result and the original data. This improves the quality of subsequent data processing and enhances robustness.

[0084] It should be recognized that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable storage medium. The method can be implemented using standard programming techniques—including a non-transitory computer-readable storage medium configured with a computer program, wherein such a storage medium causes the computer to operate in a specific and predefined manner—according to the methods and drawings described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. Furthermore, for this purpose, the program can run on a programmed application-specific integrated circuit (ASIC).

[0085] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A near-offset data reconstruction method based on iterative seismic interferometry, characterized in that, include: Seismic waves are detected using a detector and recorded as actual data. Based on the real records, the missing signals are calculated using interferometric interpolation, and the missing signals are used as virtual records. The virtual records are matched with the real records using a least-squares matched filter; The matched data records are arranged into a data volume in chronological order according to the shot gather-detector gather; The data volume is converted from the data domain to the focus domain to remove noise signals; The transformed data is combined with the original data to determine whether the iteration requirements are met. Based on the actual records, the missing signals are calculated using interferometric interpolation, and the missing signals are used as virtual records, including: Calculate the cross-correlation between multiple seismic wave signals recorded by one geophone and a single seismic wave signal recorded by another geophone; Then, duplicate paths are eliminated to obtain the missing signals at the missing locations; Use the missing signal as a virtual record; The mathematical expression for the interference interpolation equation is: in, Indicates the location where the earthquake source was triggered. Indicates the position of the detector. Represents the Green's function in the frequency domain. Represents earthquake data. Indicates the integral symbol, express The survey line in question The symbol is for partial differentials. Represents angular frequency. The surface representing the distribution of earthquake sources. The surface representing the detector distribution. and Located below the water surface, Represents a semicircle at infinity. Indicates complex conjugation. Indicates the direction of the normal; Matching virtual records with real records using a least-squares matched filter includes: By calculating the excitation at position L, L A L B The cross-correlation of the two received records yields L. A Position excitation, L B The virtual record of location reception is expressed mathematically as follows: in, This represents a true record. This represents a virtual record before matching. This refers to a matched filter, which is defined as follows: in, The damping factor is represented by the fact that both virtual and real records are transformed into overlapping blocks between adjacent windows. In each local window, a matched filter estimated through trial and error is used for the virtual record.

2. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 1, characterized in that: The seismic waves are detected by a geophone and recorded as real data. Install two or more detectors below the water surface: The seismic wave signals detected by the detector are used as the actual records.

3. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 1, characterized in that: The calculation expression for the frequency domain is: in, Let I represent the wave number, L represent the real part, and L represent the plane in which the detector is located.

4. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 3, characterized in that: The matched data records are arranged into a data volume in chronological order according to the shot gather-detector gather, and its mathematical expression is: in, Indicates the position of the detector. The location of the epicenter is indicated by 'u', and the number of shots is indicated by 'u'.

5. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 4, characterized in that: The mathematical expression for a positive focusing transformation, which converts data from the data domain to the focusing domain, is: Where Q represents the focal domain data and F represents the positive focus transform. In the focus transform, the effective signal can be focused to the focal point, thereby creating a difference between the distribution areas of the effective signal and noise in the focal domain.

6. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 5, characterized in that: Converting the data volume from the data domain to the focus domain and removing noise signals includes: in, Indicates the offset distance. Indicates time, Indicates the filtering range; R is an abbreviation for real and has no practical meaning. This indicates the data Perform filtering operation, when The original signal is retained if the signal is active, otherwise it is set to zero. This represents the parameters given by the operator, and the specific value depends on the range of data that needs to be filtered out.

7. The near-offset data reconstruction method based on iterative seismic interferometry as described in claim 6, characterized in that: Combining the transformed data with the original data to determine whether the iteration requirements are met includes: in, This represents the portion of the original seismic data excluding near-migration missing data. If this condition is not met, data D is returned to the first step, and the entire iterative process is repeated until the result meets the accuracy requirements.

Images