Rank-based NMO-free 3D near-surface correction method and device

Abstract

The application provides a rank-based NMO-free three-dimensional near-surface correction method and device. The rank-based NMO-free three-dimensional near-surface correction scheme provided by the application can perform residual static correction calculation, can obtain more accurate static correction data, and does not need to perform moveout correction, thereby avoiding the low processing efficiency caused by the fact that the existing static correction still needs to perform multiple iteration calculation on NMO velocity values and residual static correction amounts.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of seismic exploration data processing methods, and more particularly, to a rank-based NMO-free 3D near-surface correction method and device. BACKGROUND

[0002] Complex near-surface weathering layers pose challenges to imaging and inversion of land seismic data. Limitations in acquisition and modeling can prevent the creation of an accurate near-surface model that captures its heterogeneous and laterally-varying characteristics. To overcome these limitations, statics is used as an effective alternative solution. However, due to the assumption of surface consistency, statics estimation relies on a simplification of the near-surface model. Therefore, surface-consistent short-wavelength statics estimation and correction are usually followed by a non-surface-consistent step to be able to capture most of the statics.

[0003] Near-surface weathering layers have significant characteristics of complexity and non-uniformity, which bring great challenges to the imaging and inversion of land seismic data. The strata near the surface contain loose minerals and rocks, resulting in severe and complex sound wave attenuation. This complex wave field disturbance is difficult to handle and will seriously affect the analysis results of seismic data. Current seismic exploration techniques are limited by spatial resolution and modeling capabilities, making it difficult to accurately depict the complex and variable details of the near-surface layer geological structure, and unable to effectively parameterize the highly heterogeneous and laterally-varying characteristics of the near-surface layer. This further exacerbates the negative impact of the near-surface layer on seismic data. To try to overcome this difficulty, a statics method is introduced in seismic processing to compensate for surface noise. The basic idea of statics is to eliminate the interference of the near-surface layer on seismic records as much as possible without changing the deep reflection coefficients, so that the reflected wave packet is as close as possible to the source wave packet. A large number of studies have shown that statics can indeed reduce the damage of near-surface noise to seismic data to a great extent, and improve the signal-to-noise ratio, laying the foundation for subsequent geological interpretation and analysis.

[0004] However, the traditional static correction technology itself also has certain problems and limitations, mainly reflected in that the calculation of static correction depends on the assumption of surface consistency, which is often inconsistent with the actual geological conditions. The surface consistency assumption simplifies the near-surface model and uses the approximation of single delay time. However, in fact, due to the complex heterogeneity of the near-surface layer, the input wave field at different positions will indeed experience different attenuation, scattering and deflection, thereby producing different static correction residuals. Therefore, strict surface consistency does not exist. The static correction obtained under the traditional surface consistency method cannot completely eliminate the influence of the near-surface, and the processing effect has limitations. In addition, in the specific processing flow, the short-wave static correction processing of surface consistency is also based on a non-surface consistent processing. Usually, it needs to go through two steps: the first step is to obtain the main static correction by using the non-surface consistent method; the second step is to use the surface consistent method for more detailed residual static correction processing to further improve the accuracy of static correction. This two-step separate strategy has low processing efficiency, and there may be multiple iterations in between, which is a large amount of work. The specific analysis is as follows.

[0005] Firstly, the existing technology generally requires multiple iterations to calculate the NMO velocity value and the residual static correction, because the two are related to each other and restrict each other, and one cannot be accurately calculated without determining the other. However, such repeated iterative calculation requires a large time cost and has a serious low efficiency. In the traditional processing flow, accurate moveout correction must be obtained before static correction. However, NMO correction relies on the time difference mean of the gather, which requires correct estimation of the initial velocity of each channel, so as to implement appropriate NMO correction to achieve the purpose of normal moveout correction. At the same time, accurate estimation of the residual static correction can only be carried out after sufficient NMO correction is performed. Because the estimation of the residual static correction requires analysis and differentiation of the residual noise causes, incorrect time difference correction will cause the mixing of static correction residual and NMO correction residual. Therefore, the two steps are mutually dependent and interdependent. A certain number of circular iterations must be performed to converge to a sufficiently accurate correction result of each. Each iteration requires repeated solution of NMO velocity and static correction, which has high computational complexity and time cost.

[0006] Secondly, the workload of such iterative calculation is huge and it is difficult to automatically terminate and judge. The calculation of NMO velocity can use various algorithms such as Gaussian attenuation algorithm and semblance peak value method, and the static correction can also be searched for the best solution by various methods. These operations themselves have heavy calculation load, and a large amount of manual experience is required for parameter adjustment and effect interpretation, and it is impossible to simply preset the termination condition. The processing workload is large and the efficiency is low.

[0007] Thirdly, the repeated iteration forms a closed loop, and the cumulative error affects the result. Each correction calculation itself has an approximation error of the algorithm itself, and after the loop accumulation, the error is also accumulated and amplified, which leads to the difficulty of convergence of the true solution, forming a vicious cycle of "error-iteration". This is an important reason for low efficiency.

[0008] In summary, the existing three-dimensional seismic data correction scheme is difficult to satisfy people. SUMMARY

[0009] Therefore, the present application aims to provide a three-dimensional near-surface correction scheme without NMO correction link to obtain more accurate static correction data and less calculation.

[0010] According to one aspect of the present application, a rank-based NMO-free three-dimensional near-surface correction method is provided, which comprises:

[0011] Step 1, arranging three-dimensional seismic data into a two-dimensional matrix to form a two-dimensional matrix D;

[0012] Step 2, performing Fourier transform on the matrix D to obtain a matrix E;

[0013] Step 3, looping all frequencies and ranks of the matrix E, calculating the singular value decomposition of each frequency slice, and then traversing each rank based on the singular value decomposition result to obtain the low-rank approximation F of each rank of the frequency slice;

[0014] Step 4, collecting the low-rank approximations F of all ranks of all frequency slices, and performing inverse Fourier transform on the collected matrix to obtain a matrix G;

[0015] Step 5, cross-correlating the matrix G and the matrix D, and selecting the maximum delay of the cross-correlation as the static correction amount;

[0016] Step 6, using the static correction amount to perform static correction on the three-dimensional seismic data.

[0017] In some embodiments, the step 1 specifically comprises:

[0018] The x coordinates in the shotpoint coordinates and the receiver coordinates of the three-dimensional seismic data are placed in one dimension of the matrix, and the y coordinates in the shotpoint coordinates and the receiver coordinates are placed in another dimension of the matrix.

[0019] In some embodiments, the singular value decomposition of each frequency slice specifically comprises:

[0020] For each frequency, the frequency slice of the frequency is taken out from the matrix E, and is recorded as a sub-matrix A.

[0021] The singular value decomposition equation of the sub-matrix A is established;

[0022] Based on the singular value decomposition equation, singular values of A and corresponding eigenvectors are obtained by solving an eigenvalue problem of the sub-matrix A.

[0023] In some embodiments, at each frequency slice, ranks are traversed in ascending order.

[0024] In some embodiments, based on the singular value decomposition result, ranks are traversed to obtain a low-rank approximation F of each rank of the frequency slice, including:

[0025] For rank n, based on the first n singular values and corresponding eigenvectors obtained by singular value decomposition, a low-rank approximation F of rank n is obtained.

[0026] According to another aspect of the present application, a rank-based NMO-free three-dimensional near-surface correction device is provided, which comprises:

[0027] A matrixing unit is configured to arrange three-dimensional seismic data into a two-dimensional matrix to form a two-dimensional matrix D.

[0028] A Fourier transform unit is configured to perform Fourier transform on the matrix D to obtain a matrix E.

[0029] A loop calculation unit is configured to loop all frequencies and ranks of the matrix E, calculate singular value decomposition of each frequency slice, and then based on the singular value decomposition result, traverse ranks to obtain a low-rank approximation F of each rank of the frequency slice.

[0030] An inverse Fourier transform unit is configured to collect low-rank approximations F of all ranks of all frequency slices, and perform inverse Fourier transform on the collected matrix to obtain a matrix G.

[0031] A static correction amount calculation unit is configured to perform cross-correlation on the matrix G and the matrix D, and select the maximum delay of the cross-correlation as a static correction amount.

[0032] A static correction unit is configured to perform static correction on the three-dimensional seismic data using the static correction amount.

[0033] In some embodiments, the matrixing unit is specifically configured to:

[0034] The x coordinates in the shotpoint coordinates and the receiver coordinates of the three-dimensional seismic data are placed in one dimension of the matrix, and the y coordinates in the shotpoint coordinates and the receiver coordinates are placed in another dimension of the matrix.

[0035] In some embodiments, the singular value decomposition of each frequency slice specifically includes:

[0036] For each frequency, a frequency slice of the frequency is taken out from the matrix E, denoted as a sub-matrix A.

[0037] establishing a singular value decomposition equation of the sub-matrix A;

[0038] Based on the singular value decomposition equation, singular values of the sub-matrix A and corresponding eigenvectors are obtained by solving an eigenvalue problem of the sub-matrix A.

[0039] In some embodiments, based on the singular value decomposition result, each rank is traversed to obtain a low-rank approximation F of each rank of the frequency slice, which includes:

[0040] For rank n, based on the first n singular values obtained by singular value decomposition and the corresponding eigenvectors, a low-rank approximation F of rank n is obtained.

[0041] In some embodiments, based on the singular value decomposition result, each rank is traversed to obtain a low-rank approximation F of each rank of the frequency slice, which includes:

[0042] For rank n, based on the first n singular values obtained by singular value decomposition and the corresponding eigenvectors, a low-rank approximation F of rank n is obtained.

[0043] According to another aspect of the present application, an electronic device is also provided, which includes:

[0044] a memory storing executable instructions;

[0045] a processor running the executable instructions in the memory to implement the rank-based NMO-free 3D near-surface correction method described above.

[0046] According to another aspect of the present application, a computer readable storage medium is also provided, which stores a computer program, and the computer program is executed by a processor to implement the rank-based NMO-free 3D near-surface correction method described above.

[0047] The present application has at least the following advantages:

[0048] 1. The calculation accuracy of static correction is improved

[0049] The present application discards the surface consistency assumption, directly estimates the static correction amount for the original record, avoids the static correction residual generated by the surface inhomogeneity, and the calculation result is closer to the true optimal solution, thereby improving the effectiveness of the static correction;

[0050] 2. No NMO correction is needed

[0051] NMO velocity error does not affect the estimation of residual static correction, and the multiple iterations of NMO velocity and the estimation of residual static correction are reduced, so that the data after NMO correction also does not need to be cut off in order to avoid the NMO stretching effect;

[0052] 3. The calculation process is simplified and the efficiency is improved

[0053] Because the steps of calculating NMO velocity and static correction amount are skipped for multiple iterations, the redundant operation amount is greatly reduced, and the whole calculation process is more concise and efficient.

[0054] 4. Reducing the influence of artificial experience

[0055] In the traditional process, many parameters need to be repeatedly adjusted and interpreted manually. The present method realizes end-to-end automatic calculation, avoiding accidental errors caused by artificial experience.

[0056] 5. Improving the processing effect of low signal-to-noise ratio data

[0057] In low signal-to-noise ratio records, repeated iterations often fail to converge. The present method is one-step, and has better adaptability to low signal-to-noise ratio data.

[0058] 6. Reducing the difficulty of algorithm and implementation

[0059] The calculation process is simple and straightforward, the method is easy to understand and implement, and the algorithm difficulty is greatly reduced.

[0060] 7. Expanding the application range and processing more flexibly

[0061] The present method is not limited by frequency band, and can theoretically process any bandwidth signal, expanding the application range and being flexibly applied to two-dimensional or three-dimensional data.

[0062] 8. Laying a better foundation for subsequent menu analysis

[0063] Through more accurate and effective static correction, the subsequent imaging quality can be effectively improved, laying a better foundation for feature extraction, reservoir identification and other fine description and analysis.

[0064] In summary, the rank-based NMO-free three-dimensional near-surface correction scheme proposed in the present application can calculate residual static correction, obtain more accurate static correction data, and does not require moveout correction, avoiding the low processing efficiency caused by the need for multiple iterations to calculate NMO velocity and residual static correction amount in the existing static correction. The present application has significant improvement, good application prospect, important scientific research value and application value.

[0065] The method and device of the present application have other characteristics and advantages, which will be apparent or will be described in detail in the accompanying drawings and subsequent specific embodiments incorporated herein, which together serve to explain the specific principles of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0066] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the figures, and in which:

[0067] Figure 1 A flow chart of a rank-based NMO-free 3D near-surface correction method according to an embodiment of the present application is shown.

[0068] Figure 2 A 30Hz frequency slice of a certain exploration area without statics correction problem is shown.

[0069] Figure 3 A 30Hz frequency slice of a certain exploration area with statics correction problem is shown.

[0070] Figure 4 A 30Hz frequency slice of a certain exploration area after processing according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0071] Preferred embodiments of the present application will be described herein below with reference to the accompanying drawings. While the preferred embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and fully convey the scope of the application to those skilled in the art.

[0072] The principle of the present application is briefly introduced here. Frequency slices affected by weathering layers often exhibit slowly decaying singular values, while frequency slices without statics correction problem have low-rank property. Therefore, the computation of residual statics can use the low-rank approximation of frequency slices affected by weathering layers to enhance the low-rank structure, obtain frequency slices with less near-surface acquisition footprints, which can also contain erroneous amplitudes. To preserve the amplitude versus offset (AVO) response, the low-rank approximated frequency slices are cross-correlated with the original frequency slices to estimate the residual statics. The procedure is iteratively applied on different rank scales and different frequency bands, i.e. to obtain low-rank approximation of different ranks, and to improve the low-rank approximation of high frequencies.

[0073] The basic steps of 2D residual statics estimation and correction are to convert data from shot-receiver domain to midpoint-hydrophone domain to satisfy the requirement of singular value decay, so as to achieve accurate low-rank approximation. Therefore, in order to extend the method to 3D data, it is necessary to find a suitable organization to obtain 2D matrix, so as to calculate singular value decomposition (SVD). In addition, another important requirement is that these matrices need to satisfy the requirement of singular value decay.

[0074] Further embodiments of the present application are described below.

[0075] Example 1

[0076] Figure 1 A flow chart of a rank-based NMO-free 3D near-surface correction method according to an embodiment of the present application is shown. As shown, the example includes steps 1-6.

[0077] Step 1, arrange 3D seismic data into a 2D matrix to form a 2D matrix D.

[0078] In some embodiments, the x coordinates of the shot point coordinates and receiver point coordinates of the 3D seismic data are placed in one dimension of the matrix, and the y coordinates of the shot point coordinates and receiver point coordinates are placed in another dimension of the matrix.

[0079] The basic steps of 2D residual statics estimation and correction are to convert data from shot-receiver domain to midpoint-offset domain to meet the requirement of singular value decay, so as to achieve accurate low-rank approximation. Therefore, in order to extend this method to 3D data, it is necessary to find a suitable organization to obtain a 2D matrix to calculate singular value decomposition (SVD). In addition, another important requirement is that these matrices need to meet the requirement of singular value decay. According to the present embodiment, 3D data is arranged into a 2D matrix, i.e. matrixed, by organizing data by placing shot point coordinates and receiver point coordinates along one direction (xsrc, xrec) in one dimension of the matrix, and placing shot point coordinates and receiver point coordinates of the other direction (ysrc, yrec) in another dimension. It has the following advantages:

[0080] (1) Ensure that the matrix meets the subsequent calculation requirements of singular value decomposition, and the 3D data is organized into a well-structured 2D matrix, which can be directly decomposed by SVD;

[0081] (2) Use the spatial distribution characteristics of source point and receiver point coordinates for matrix mapping, convert the physical coordinates association to matrix index coordinates, and ensure the corresponding relationship of geographical space information;

[0082] (3) The source coordinates and receiver coordinates of one direction are used as one dimension of the matrix, and the source coordinates and receiver coordinates of the other direction are used as another dimension of the matrix, which maximizes the preservation of the spatial distribution information of 3D data;

[0083] (4) The rows and columns of the matrix have clear physical meaning, which facilitates the interpretation of the results after matrix operation, and maps them back to the actual 3D space;

[0084] (5) The matrix structure is simple, the operation process is intuitive, and it is convenient for matrix operations such as singular value decomposition;

[0085] (6) The matrix organization form is ingenious, which effectively converts three-dimensional sequence data into a layout suitable for two-dimensional matrix operation, creating conditions for subsequent introduction of low-rank methods;

[0086] (7) The matrix conversion process can be automatically implemented, with small repetitive workload and good practicability;

[0087] (8) The matrix result is easy to store and suitable for network transmission, and can be conveniently used as the input of subsequent operations such as low-rank decomposition.

[0088] Step 2, Fourier transform is performed on the matrix D to obtain the matrix E.

[0089] By performing Fourier transform on the matrix D, the time-domain seismic data can be converted to the frequency domain.

[0090] In some examples, the process of performing Fourier transform on the matrix D to obtain the matrix E is as follows:

[0091] One-dimensional Fourier transform can be performed on each column of the matrix D, and the specific calculation method can use the FFT (Fast Fourier Transform Algorithm) method;

[0092] Through the FFT algorithm, the frequency domain information corresponding to each column time domain signal in the matrix D can be obtained;

[0093] Then, the frequency domain results of all columns are rearranged and combined in column order to form a new complex matrix E.

[0094] The matrix E is the result matrix of the Fourier transform of the matrix D to the frequency domain, which contains the frequency component information of the three-dimensional seismic data in the matrix D.

[0095] The number of rows and columns of the matrix E can be the same as that of the matrix D, but the elements are changed from real numbers to complex numbers, reflecting the frequency characteristics of the signal.

[0096] Thus, the frequency domain conversion of the time-domain seismic data can be realized, which is convenient for subsequent frequency low-rank analysis and other operations.

[0097] Step 3, loop all frequencies and ranks of the matrix E, calculate the singular value decomposition of each frequency slice, and then traverse each rank based on the singular value decomposition result to obtain the low-rank approximation F of each rank of the frequency slice.

[0098] The matrix E includes the information of each frequency in the expected frequency band of the static correction estimation, and each frequency corresponds to a submatrix block in the matrix E as the frequency slice of the frequency. The rank reflects the number of linearly independent column vectors or row vectors of the matrix.

[0099] The singular value decomposition can be performed on the frequency slices of all frequencies in the expected frequency band of the static correction quantity estimation, and then based on the singular value decomposition result, each rank is traversed to finally obtain the low-rank approximation matrix F corresponding to each rank at each frequency.

[0100] In some embodiments, the singular value decomposition of each frequency slice is calculated, specifically including:

[0101] For each frequency, the frequency slice of the frequency is taken out from the matrix E, denoted as a submatrix A.

[0102] The singular value decomposition equation of the submatrix A is established;

[0103] Based on the singular value decomposition equation, the singular values and the corresponding eigenvectors of A are obtained by solving the eigenvalue problem of the submatrix A.

[0104] In some embodiments, at each frequency slice, the ranks are traversed in ascending order.

[0105] In some embodiments, based on the singular value decomposition result, the ranks are traversed to obtain the low-rank approximation F of each rank of the frequency slice, including:

[0106] For rank n, based on the first n singular values and the corresponding eigenvectors obtained by singular value decomposition, the low-rank approximation F of rank n is obtained.

[0107] The singular values on the diagonal of the singular value matrix obtained in singular value decomposition are sorted in descending order by default, that is, the singular values with earlier indexes on the diagonal are larger, and the singular values with later indexes are smaller and smaller. Correspondingly, the singular vectors in the singular vector matrices U and V are also stored in correspondence with the singular values in this order.

[0108] Therefore, when obtaining the low-rank approximation of rank n, the first n singular values can be taken, that is, the first n largest singular values on the diagonal are taken, and the first n singular vectors in the singular vector matrix are taken, that is, the first n columns in the vector matrix are taken. In this way, the low-rank approximation of the matrix can be reconstructed through the largest singular values and singular vectors. Thus, by retaining the first n singular values and vectors and truncating the other parts, the low-rank approximation matrix of the frequency and rank can be obtained.

[0109] In this way, all ranks and frequencies are calculated in a loop, considering the influence of frequency and the influence of different rank values, and the overall low-rank approximation result can be obtained.

[0110] A simple example is given to illustrate. Assume that the expected frequency band of the static correction quantity estimation is 10-50Hz, and the low-rank value range is 1-5.

[0111] The frequency slice corresponding to 10Hz is taken out first, and SVD decomposition is performed thereon to obtain all singular values and eigenvectors at the frequency. Then, low-rank approximation matrices are sequentially obtained by taking ranks from 1 to 5. When the rank is 1, the first singular value and the corresponding eigenvector in the singular value decomposition result at the frequency of 10Hz are retained, and a low-rank approximation matrix is reconstructed through the product of the singular value and the eigenvector. When the rank is 2, the first two singular values and the corresponding eigenvectors in the singular value decomposition result at the frequency of 10Hz are retained, and a low-rank approximation matrix is reconstructed through the product of the singular values and the eigenvectors. In this way, all five low-rank approximation matrices with ranks from 1 to 5 at the frequency of 10Hz are calculated.

[0112] Then, the frequency slice corresponding to 11Hz is taken out, and SVD decomposition is performed thereon to obtain all singular values and eigenvectors at the frequency. Then, low-rank approximation matrices are sequentially obtained by taking ranks from 1 to 5, and the process is repeated until 50Hz, so that all five low-rank approximation matrices with ranks from 1 to 5 at the frequency of 50Hz are calculated.

[0113] In other words, each “frequency-rank” combination can correspond to a low-rank approximation matrix F.

[0114] Step 4: All low-rank approximations F of all ranks of all frequency slices are summarized, and inverse Fourier transform is performed on the summarized matrix to obtain a matrix G.

[0115] The matrix obtained after summarization comprehensively reflects the low-rank information of all frequencies and all ranks. The low-rank approximation matrices F are distributed at different frequencies and different ranks, and they are integrated together to comprehensively reflect the low-rank characteristics of the data matrix. Then, inverse transform is performed on the matrix, so that the result is aligned with D on the time axis, facilitating subsequent cross-correlation.

[0116] Inverse transform is performed on the matrix obtained after summarization, so that low-frequency and high-frequency signals can participate in subsequent processing at the same time, which is of great significance to subsequent processing.

[0117] Step 5: Cross-correlation is performed on the matrix G and the matrix D, and the maximum delay of the cross-correlation is selected as the static correction amount.

[0118] The matrix G is a time-domain matrix obtained by performing inverse Fourier transform on the summarized low-rank approximation matrix F, and the matrix D is the original seismic data matrix.

[0119] According to the present application, the cross-correlation function between the matrix G and the matrix D can be calculated to reflect the similarity between the two. Through parameterized time shift τ, the relationship between the cross-correlation function value and the time shift can be obtained. In other words, the cross-correlation can reflect the similarity between the matrix G and the matrix D, and the best coincidence point of the two can be found through parameterized time shift. The time shift corresponding to the maximum coincidence degree is the maximum similar time shift between the two, that is, the maximum delay.

[0120] On the cross-correlation function diagram, the time shift tau corresponding to the maximum cross-correlation peak value can be found. The tau is the maximum delay of G relative to D. Because the matrix G reflects the best fitting stratum model after low rank processing, the maximum delay tau can be used as the static correction amount required by static correction.

[0121] Step 6, using the static correction amount to perform static correction on the three-dimensional seismic data.

[0122] In the seismic data, due to the influence of near-surface low-velocity layer and the like, the deep reflection signal often has time shift error. The static correction amount is used to perform static correction on the three-dimensional seismic data, so as to eliminate the influence of the near-surface layer time shift error, restore the true time phase of the deep reflection, make the deep reflection signal return to the originally correct occurrence time, improve the time phase information after the time phase information correction, the stratum interface will be clearer, the system noise is reduced, the signal-to-noise ratio is improved, the overall imaging quality is improved, and a better foundation is provided for subsequent quantitative interpretation and analysis, and the accuracy of data analysis and model establishment is improved.

[0123] Therefore, the static correction of the present application plays an important role in maintaining the efficiency of the seismic data processing chain and improving the quality of the final analysis result.

[0124] The embodiment provides a rank-based NMO-free three-dimensional near-surface correction scheme, which can perform residual static correction calculation, can obtain more accurate static correction data, and does not need to perform moveout correction, thereby avoiding the low processing efficiency caused by the fact that the existing static correction still needs to perform multiple iteration calculation on NMO velocity values and residual static correction amounts.

[0125] Example 2

[0126] The present application also provides a rank-based NMO-free three-dimensional near-surface correction device, which comprises:

[0127] The matrix unit is configured to arrange the three-dimensional seismic data into a two-dimensional matrix to form a two-dimensional matrix D.

[0128] The Fourier transform unit is configured to perform Fourier transform on the matrix D to obtain a matrix E.

[0129] The loop calculation unit is configured to loop all frequencies and ranks of the matrix E, calculate singular value decomposition of each frequency slice, and then traverse each rank based on the singular value decomposition result to obtain a low rank approximation F of each rank of the frequency slice.

[0130] The inverse Fourier transform unit is configured to collect the low rank approximations F of all ranks of all frequency slices, and perform inverse Fourier transform on the collected matrix to obtain a matrix G.

[0131] The static correction amount obtaining unit is configured to correlate the matrix G and the matrix D, and select the maximum delay of the correlation as the static correction amount.

[0132] The static correction unit is configured to perform static correction on the three-dimensional seismic data by using the static correction amount.

[0133] In some embodiments, the matrixing unit is specifically configured to:

[0134] The x coordinates in the shot point coordinates and the receiver point coordinates of the three-dimensional seismic data are placed in one dimension of the matrix, and the y coordinates in the shot point coordinates and the receiver point coordinates are placed in another dimension of the matrix.

[0135] In some embodiments, the singular value decomposition of each frequency slice is calculated, specifically including:

[0136] For each frequency, a frequency slice of the frequency is taken out from the matrix E, and is recorded as a sub-matrix A.

[0137] The singular value decomposition equation of the sub-matrix A is established;

[0138] Based on the singular value decomposition equation, the singular values and the corresponding eigenvectors of A are obtained by solving the eigenvalue problem of the sub-matrix A.

[0139] In some embodiments, based on the singular value decomposition result, each rank is traversed to obtain the low-rank approximation F of each rank of the frequency slice, including:

[0140] For rank n, the low-rank approximation F of rank n is obtained based on the first n singular values and the corresponding eigenvectors obtained by the singular value decomposition.

[0141] In some embodiments, based on the singular value decomposition result, each rank is traversed to obtain the low-rank approximation F of each rank of the frequency slice, including:

[0142] For rank n, the low-rank approximation F of rank n is obtained based on the first n singular values and the corresponding eigenvectors obtained by the singular value decomposition.

[0143] The beneficial effects of the embodiment at least include:

[0144] 1. The calculation precision of the static correction is improved

[0145] The present application discards the surface consistency assumption, directly estimates the static correction amount based on the original record, avoids the static correction residual caused by the surface inhomogeneity, and the calculation result is closer to the true optimal solution, thereby improving the effectiveness of the static correction.

[0146] 2. No NMO correction is needed

[0147] The NMO velocity error does not affect the residual static correction estimation, reduces the multiple iterations of NMO velocity and the estimation of residual static correction, and therefore, the NMO corrected data does not need to be cut off in order to avoid the NMO stretching effect;

[0148] 3. Simplifies the calculation process and improves efficiency

[0149] Due to the skipping of the steps of multiple iterations of NMO velocity and static correction, the redundant operation amount is greatly reduced, and the whole calculation process is more concise and efficient.

[0150] 4. Reduces the influence of artificial experience

[0151] In the traditional process, many parameters need to be repeatedly adjusted and interpreted by artificial experience. The present method realizes end-to-end automatic calculation and avoids accidental errors caused by artificial experience.

[0152] 5. Improves the processing effect of low signal-to-noise ratio data

[0153] In low signal-to-noise ratio records, repeated iterations often fail to converge. The present method is one-step, and has better adaptability to low signal-to-noise ratio data.

[0154] 6. Reduces the difficulty of algorithm and implementation

[0155] The calculation process is simple and direct, the method is easy to understand and implement, and the algorithm difficulty is greatly reduced.

[0156] 7. Expands the application range and is more flexible

[0157] The present method is not limited by the frequency band, and can theoretically process any bandwidth signal, expanding the application range and being flexibly applied to two-dimensional or three-dimensional data.

[0158] 8. Lays a better foundation for subsequent menu analysis

[0159] Through more accurate and effective static correction, the subsequent imaging quality can be effectively improved, laying a better foundation for feature extraction, reservoir identification and other fine description and analysis.

[0160] In summary, the present embodiment proposes a rank-based NMO-free three-dimensional near-surface correction scheme, which can perform residual static correction calculation, obtain more accurate static correction data, and does not need to perform moveout correction, avoiding the low processing efficiency caused by the multiple iterations of NMO velocity and residual static correction in the existing static correction.

[0161] Other detailed descriptions and advantages of the present embodiment can be referred to the corresponding descriptions in the foregoing embodiments, and will not be repeated here.

[0162] Example 3

[0163] According to another aspect of the present application, an electronic device is also provided. The electronic device comprises:

[0164] a memory storing executable instructions;

[0165] a processor running the executable instructions in the memory to implement the rank-based NMO-free 3D near-surface correction method according to the present application.

[0166] The method comprises the following steps 1-6.

[0167] The method comprises:

[0168] Step 1, arrange the 3D seismic data into a 2D matrix to form a 2D matrix D;

[0169] Step 2, perform Fourier transform on the matrix D to obtain a matrix E;

[0170] Step 3, loop through all frequencies and ranks of the matrix E, calculate singular value decomposition of each frequency slice, and then traverse each rank based on the singular value decomposition result to obtain a low-rank approximation F of each rank of the frequency slice;

[0171] Step 4, aggregate the low-rank approximations F of all ranks of all frequency slices, and perform inverse Fourier transform on the aggregated matrix to obtain a matrix G;

[0172] Step 5, perform cross-correlation on the matrix G and the matrix D, and select the maximum delay of the cross-correlation as the static correction amount;

[0173] Step 6, use the static correction amount to perform static correction on the 3D seismic data.

[0174] In some embodiments, the step 1 specifically comprises:

[0175] placing the x coordinates in the shotpoint coordinates and the receiver coordinates of the 3D seismic data in one dimension of the matrix, and placing the y coordinates in the shotpoint coordinates and the receiver coordinates in another dimension of the matrix.

[0176] In some embodiments, the singular value decomposition of each frequency slice specifically comprises:

[0177] for each frequency, taking out the frequency slice of the frequency from the matrix E, denoted as a sub-matrix A.

[0178] establishing a singular value decomposition equation of the sub-matrix A;

[0179] based on the singular value decomposition equation, obtaining singular values and corresponding eigenvectors of A by solving an eigenvalue problem of the sub-matrix A.

[0180] In some embodiments, at each frequency slice, the ranks are traversed in ascending order.

[0181] In some embodiments, based on the singular value decomposition result, the ranks are traversed to obtain a low-rank approximation F of each rank of the frequency slice, comprising:

[0182] For rank n, based on the first n singular values and corresponding eigenvectors obtained by singular value decomposition, a low-rank approximation F of rank n is obtained.

[0183] Specifically, the memory can include one or more computer program products, which can include various forms of computer readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM), cache memory, and / or the like. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, and / or the like.

[0184] The processor can be a central processing unit (CPU) or other forms of processing units with data processing and / or instruction execution capabilities, and can control other components in the electronic device to perform desired functions. In one embodiment of the present application, the processor is used to run the computer readable instructions stored in the memory.

[0185] The beneficial effects of the present embodiment include at least:

[0186] 1. Improved calculation accuracy of static correction

[0187] The present application discards the surface consistency assumption, directly estimates the static correction amount based on the original record, avoids the static correction residual caused by the surface inhomogeneity, and the calculation result is closer to the true optimal solution, thereby improving the effectiveness of the static correction;

[0188] 2. No NMO correction is needed

[0189] NMO velocity error does not affect the estimation of residual static correction, reducing the multiple iterations of NMO velocity and the estimation of residual static correction, therefore, in order to avoid the NMO stretching effect, the data after NMO correction also does not need to be cut off;

[0190] 3. Simplify the calculation process and improve efficiency

[0191] Since the steps of calculating NMO velocity and static correction amount by multiple iterations are skipped, the redundant operation amount is greatly reduced, and the entire calculation process is more concise and efficient.

[0192] 4. Reduce the influence of artificial experience

[0193] There are many parameters in the traditional process that need to be repeatedly adjusted manually, and the method realizes end-to-end automatic calculation, avoiding accidental errors caused by manual experience.

[0194] 5. Improved processing effect of low signal-to-noise ratio data

[0195] In low signal-to-noise ratio records, repeated iteration is often difficult to converge, and the method is one-step, and is more suitable for low signal-to-noise ratio data;

[0196] 6. Reduce the difficulty of algorithm and implementation

[0197] The calculation process is simple and direct, the method is easy to understand and implement, and the algorithm difficulty is greatly reduced.

[0198] 7. Expand the application range and process more flexibly

[0199] The method is not limited by the frequency band, and can theoretically process any bandwidth signal, expanding the application range and being flexibly applied to two-dimensional or three-dimensional data.

[0200] 8. Lay a better foundation for subsequent menu analysis

[0201] Through more accurate and effective static correction, the subsequent imaging quality can be effectively improved, and a better foundation is laid for feature extraction, reservoir identification and other fine description analysis.

[0202] In summary, the embodiment proposes a rank-based NMO-free three-dimensional near-surface correction scheme, which can perform residual static correction calculation, can obtain more accurate static correction data, and does not need to perform dynamic correction, avoiding the low processing efficiency caused by the need to calculate NMO velocity and residual static correction in the prior art.

[0203] The detailed description of the embodiment can refer to the corresponding description in the foregoing embodiments, and will not be repeated here.

[0204] Example 4

[0205] According to another aspect of the application, a computer readable storage medium is also provided, which stores a computer program, and the computer program is executed by a processor to realize the rank-based NMO-free three-dimensional near-surface correction method according to the application.

[0206] The method comprises the following steps 1-6.

[0207] The method comprises the following steps 1-6.

[0208] The method comprises:

[0209] Step 1, arrange the three-dimensional seismic data into a two-dimensional matrix to form a two-dimensional matrix D;

[0210] Step 2, Fourier transform the matrix D to obtain a matrix E;

[0211] Step 3, loop all frequencies and ranks of the matrix E, calculate singular value decomposition of each frequency slice, and then traverse each rank based on the singular value decomposition result to obtain a low rank approximation F of each rank of the frequency slice.

[0212] Step 4, aggregate the low rank approximations F of all ranks of all frequency slices, and inverse Fourier transform the aggregated matrix to obtain a matrix G;

[0213] Step 5, cross-correlate the matrix G and the matrix D, and select the maximum delay of the cross-correlation as the static correction amount.

[0214] Step 6, use the static correction amount to perform static correction on the three-dimensional seismic data.

[0215] In some embodiments, the step 1 specifically comprises:

[0216] Place the x coordinates in the shot point coordinates and the receiver point coordinates of the three-dimensional seismic data in one dimension of the matrix, and place the y coordinates in the shot point coordinates and the receiver point coordinates in another dimension of the matrix.

[0217] In some embodiments, the singular value decomposition of each frequency slice specifically comprises:

[0218] For each frequency, take out the frequency slice of the frequency from the matrix E, denoted as a sub-matrix A.

[0219] Establish a singular value decomposition equation of the sub-matrix A.

[0220] Based on the singular value decomposition equation, the singular values and the corresponding eigenvectors of A are obtained by solving the eigenvalue problem of the sub-matrix A.

[0221] In some embodiments, under each frequency slice, the ranks are traversed in ascending order.

[0222] In some embodiments, the low rank approximation F of each rank of the frequency slice is obtained based on the singular value decomposition result, comprising:

[0223] For rank n, the low rank approximation F of rank n is obtained based on the first n singular values and the corresponding eigenvectors obtained by singular value decomposition.

[0224] The computer readable storage medium according to the embodiments of the present application has non-transitory computer readable instructions stored thereon. When the non-transitory computer readable instructions are run by a processor, all or part of the steps of the method of each embodiment of the present application are executed.

[0225] The computer readable storage medium includes, but is not limited to, optical storage media (for example, CD-ROM and DVD), magneto-optical storage media (for example, MO), magnetic storage media (for example, magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (for example, memory card), and media with built-in ROM (for example, ROM cartridge).

[0226] Those skilled in the art should understand that, in order to solve the technical problem of how to obtain a good user experience effect, the embodiment can also include well-known structures such as a communication bus, an interface, and the like, which should also be included in the protection scope of the application.

[0227] The beneficial effects of the embodiment at least include:

[0228] 1. The calculation accuracy of static correction is improved

[0229] The application discards the surface consistency assumption, directly estimates the static correction amount based on the original record, avoids the static correction residual caused by the surface inhomogeneity, and the calculation result is closer to the true optimal solution, thereby improving the effectiveness of the static correction;

[0230] 2. No NMO correction is needed

[0231] The NMO velocity error does not affect the residual static correction estimation, and the multiple iterations of the NMO velocity and the estimation of the residual static correction amount are reduced, so that the NMO corrected data does not need to be cut off in order to avoid the NMO stretching effect;

[0232] 3. The calculation process is simplified and the efficiency is improved

[0233] Since the steps of calculating the NMO velocity and the static correction amount by multiple iterations are skipped, the redundant operation amount is greatly reduced, and the entire calculation process is more concise and efficient.

[0234] 4. The influence of artificial experience is reduced

[0235] In the traditional process, many parameters need to be repeatedly adjusted and interpreted manually, and the method realizes end-to-end automatic calculation, avoiding accidental errors caused by artificial experience;

[0236] 5. The processing effect of low signal-to-noise ratio data is improved

[0237] In low signal-to-noise ratio records, repeated iterations often fail to converge, and the method is one-step, and is more suitable for low signal-to-noise ratio data;

[0238] 6. The algorithm and implementation difficulty are reduced

[0239] The calculation process is simple and direct, the method is easy to understand and implement, and the algorithm difficulty is greatly reduced.

[0240] 7. Extended application range, more flexible processing

[0241] The method is not limited by frequency band, and theoretically can process signals of any bandwidth, thereby extending the application range and being flexibly applied to two-dimensional or three-dimensional data.

[0242] 8. Better foundation for subsequent menu analysis

[0243] Through more accurate and effective static correction, the subsequent imaging quality can be effectively improved, and a better foundation is laid for feature extraction, reservoir identification and other fine description analysis.

[0244] In summary, the embodiment proposes a rank-based NMO-free three-dimensional near-surface correction scheme, which can perform residual static correction calculation, obtain more accurate static correction data, and does not need to perform moveout correction, thereby avoiding the low processing efficiency caused by the need for multiple iterations of NMO velocity value and residual static correction amount in the existing static correction.

[0245] Detailed descriptions of the embodiment can be referred to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0246] Example 5

[0247] In order to verify the effectiveness of the method in the application, the data of a certain exploration area are tested. Figure 2 is a 30Hz frequency slice without static correction problem; Figure 3 is a 30Hz frequency slice with static correction problem; Figure 4 is a 30Hz frequency slice processed according to the embodiment of the application. It can be seen that the signal-to-noise ratio of the frequency slice processed by the application is significantly improved, which confirms the effectiveness of the application scheme.

[0248] Other detailed descriptions of the embodiment can be referred to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0249] The above has described the embodiments of the application, and the above descriptions are exemplary, not exhaustive, and are not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles, practical applications or technical improvements of the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A rank-based, NMO-free three-dimensional near-surface correction method, characterized in that, The method includes: Step 1: Arrange the 3D seismic data into a 2D matrix to form a 2D matrix D; Step 2: Perform a Fourier transform on matrix D to obtain matrix E; Step 3: Iterate through all frequencies and ranks of matrix E, calculate the singular value decomposition of each frequency slice, and then iterate through each rank based on the singular value decomposition results to obtain the low-rank approximation F of each rank of the frequency slice. Step 4: Summarize the low-rank approximations F of all ranks of all frequency slices, and perform an inverse Fourier transform on the summarized matrix to obtain matrix G; Step 5: Perform cross-correlation on matrices G and D, and select the maximum delay of the cross-correlation as the static correction. Step 6: Perform static correction on the three-dimensional seismic data using the static correction amount.

2. The method according to claim 1, characterized in that, Step 1 specifically includes: The x-coordinates of the shot point and receiver point coordinates from the 3D seismic data are placed on one dimension of the matrix, and the y-coordinates of the shot point and receiver point coordinates are placed on the other dimension of the matrix.

3. The method according to claim 1, characterized in that, Calculating the singular value decomposition for each frequency slice specifically includes: For each frequency, a frequency slice of that frequency is taken from matrix E and denoted as submatrix A; Establish the singular value decomposition equation for submatrix A; Based on the singular value decomposition equation, the singular values ​​and corresponding eigenvectors of A are obtained by solving the eigenvalue problem of submatrix A.

4. The method according to claim 1, characterized in that, Within each frequency slice, the ranks are traversed in ascending order.

5. The method according to claim 1, characterized in that, Based on the singular value decomposition results, by traversing each rank, the low-rank approximations F of each rank of this frequency slice are obtained, including: For rank n, based on the first n singular values ​​obtained from singular value decomposition and the corresponding eigenvectors, we obtain the low-rank approximation F of rank n.

6. A rank-based, NMO-free three-dimensional near-surface correction device, characterized in that, The device includes: Matrixing units are used to arrange three-dimensional seismic data into a two-dimensional matrix, forming a two-dimensional matrix D; The Fourier transform unit is used to perform a Fourier transform on matrix D to obtain matrix E. The loop calculation unit is used to loop through all frequencies and ranks of matrix E, calculate the singular value decomposition of each frequency slice, and then traverse each rank based on the singular value decomposition result to obtain the low-rank approximation F of each rank of the frequency slice. The inverse Fourier transform unit is used to summarize the low-rank approximations F of all ranks of all frequency slices, and to perform an inverse Fourier transform on the summarized matrix to obtain matrix G. The static correction unit is used to perform cross-correlation on matrices G and D, and select the maximum delay of the cross-correlation as the static correction. A static correction unit is used to perform static correction on the three-dimensional seismic data using the static correction amount.

7. The apparatus according to claim 6, characterized in that, The matrix unit is specifically used for: The x-coordinates of the shot point and receiver point coordinates from the 3D seismic data are placed on one dimension of the matrix, and the y-coordinates of the shot point and receiver point coordinates are placed on the other dimension of the matrix.

8. The apparatus according to claim 6, characterized in that, Based on the singular value decomposition results, traversing each rank, the low-rank approximations F of each rank of this frequency slice are obtained, including: For rank n, based on the first n singular values ​​obtained from singular value decomposition and the corresponding eigenvectors, we obtain the low-rank approximation F of rank n.

9. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement the method of any one of claims 1-5.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of any one of claims 1-5.

Images