Reconstruction method and device for seismic compressed sampling data

Abstract

The present disclosure relates to a reconstruction method and device of seismic compressed sampling data, the method comprising: loading collected seismic compressed sampling data based on an observation system to obtain observation information, the observation information comprising: real sampling positions and sampling data at the real sampling positions; determining target sampling positions corresponding to the real sampling positions on a standard grid of the observation system and a sampling matrix; constructing a non-uniform reconstruction factor representing a non-uniform degree of mapping of the target sampling positions in the standard grid to the real sampling positions; and iteratively solving according to the non-uniform reconstruction factor, the sampling matrix and the sampling data to obtain reconstructed sampling data of the seismic compressed sampling data. The method provided by the present disclosure can improve reconstruction accuracy and computational efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of geophysical exploration technology, and in particular to a method and apparatus for reconstructing seismic compressed sampling data. Background Technology

[0002] In the process of oil and gas exploration, geological structures are usually explored based on seismic waves. The propagation and echo of seismic waves in the geological structure are used to analyze the geological physical structure and characteristics. Corresponding instruments and equipment need to be deployed at the pre-set shot point and receiving position.

[0003] Traditional seismic signal acquisition is based on Shannon's sampling theorem, requiring a sampling rate at least twice the signal bandwidth. This places very high demands on the hardware of exploration sampling equipment, consequently increasing storage and data processing costs. To address this issue, the industry has proposed using compressed sensing seismic exploration technology for sampling and data reconstruction.

[0004] Compressed sensing seismic exploration technology is a data reconstruction technology based on sparse domain constraint inversion. This technology is based on the premise that the data to be processed is sparse or compressible in a certain transform domain. By constructing constraint terms, it uses optimization methods to recover irregular data. Summary of the Invention

[0005] To address, or at least partially address, the following technical problems: given the discrepancy between the actual sampling points and the desired reconstruction grid points in seismic data, current data reconstruction techniques still need improvement in recovering missing or biased data. Furthermore, some reconstruction algorithms are very time-consuming and have low computational efficiency. This disclosure provides a method and apparatus for reconstructing compressed seismic sampling data.

[0006] In a first aspect, embodiments of this disclosure provide a method for reconstructing seismic compressed sampling data. The reconstruction method includes: loading the acquired seismic compressed sampling data based on an observation system to obtain observation information, the observation information including: a true sampling location and sampling data at the true sampling location; determining a target sampling location and a sampling matrix corresponding to the true sampling location on a standard grid of the observation system; constructing a non-uniform reconstruction factor characterizing the degree of non-uniformity in mapping the target sampling location in the standard grid to the true sampling location; and iteratively solving the non-uniform reconstruction factor, the sampling matrix, and the sampling data to obtain reconstructed sampling data regarding the seismic compressed sampling data.

[0007] According to embodiments of this disclosure, determining the target sampling position corresponding to the actual sampling position on the standard grid of the observation system includes: constructing an optimization problem regarding the mapping relationship between the actual sampling position and the standard sampling point positions corresponding to the standard grid of the observation system; solving the optimization problem to obtain the target sampling position corresponding to the actual sampling position on the standard grid.

[0008] According to some embodiments of this disclosure, the above optimization problem satisfies the following expression:

[0009]

[0010] Among them, s j Indicates the actual sampling position p j The positions of n standard sampling points l on the standard grid of the observation system s The position number of the target sampling location corresponding to the minimum distance; This represents the definition symbol; j represents the m actual sampling positions p j The sequence number j, where j ranges from 1 to m, m ≤ n; s represents the position l of the standard sampling point. s The position index of s, where s can range from 1 to n; argmin() represents the value of the variable that minimizes the value of the function inside the parentheses.

[0011] According to embodiments of this disclosure, the above-mentioned non-uniform reconstruction factor satisfies the following expression:

[0012]

[0013]

[0014] in, Let represent the non-uniform reconstruction factor, in m x n matrix form; Let I represent the matrix elements of the non-uniform reconstruction factor, where I denotes the identity matrix. This represents the non-uniform reconstruction vector corresponding to the positions of n standard sampling points at the first true sampling position; This represents the non-uniform reconstruction vector corresponding to the position of n standard sampling points at the m-th real sampling position; k represents the preset value given by the user, and the value of k ranges from 0 to 3; i represents the traversal index of the non-uniform reconstruction vector in the column dimension corresponding to the standard grid, and the value of i ranges from 1 to n; h represents the traversal index in the column dimension that is different from the value of i. Indicates the target sampling location. This represents the sampling location points within h grid intervals of the target sampling location in the standard grid; This represents the sampling location point within the i-th grid interval of the target sampling location in the standard grid.

[0015] According to embodiments of this disclosure, reconstructed sampling data about the seismic compression sampling data is obtained by iteratively solving based on the aforementioned non-uniform reconstruction factor, the aforementioned sampling matrix, and the aforementioned sampling data, including:

[0016] Based on the non-uniform reconstruction factor Given a sampling matrix R and sampled data b, determine the objective function and corresponding constraints for the data reconstruction process: The reconstruction problem is expressed as: u represents the reconstructed sampled data; the objective function and constraints of the reconstruction process are: min||x||1s.t. x represents the representation coefficients of the sampled data to be reconstructed in the sparse domain, τ is the tolerable residual size, ||1| represents the L1 norm, which is the sum of the absolute values ​​of each element in the matrix; D represents the matrix of sparse transformation operations; D H This represents the inverse transform corresponding to D; the above sampling matrix is ​​determined based on the above observation information;

[0017] The solution to the above objective function is expressed as:

[0018]

[0019] Among them, u c+1 T represents the reconstructed solution at the (c+1)th iteration; λ This represents the threshold function, where λ represents the set constraint threshold; u c This represents the reconstructed solution at the c-th iteration; express The inverse transform of .

[0020] According to embodiments of this disclosure, when the sampling points in the transverse survey line direction meet the requirements of three-dimensional curvilinear transformation, the above-mentioned sparse transformation operation includes three-dimensional curvilinear transformation; when there are fewer sampling points in the transverse survey line direction and the requirements of three-dimensional curvilinear transformation are not met, the above-mentioned sparse transformation operation includes two-dimensional curvilinear transformation and one-dimensional fast Fourier transform.

[0021] According to embodiments of this disclosure, the objective function is solved iteratively based on the following acceleration algorithm:

[0022]

[0023]

[0024] Among them, t c Let represent the iteration step size at the c-th iteration, where the initial iteration step size t0 is a given value, and I represents the identity matrix.

[0025] According to an embodiment of this disclosure, when loading the collected seismic compressed sampling data, the seismic compressed sampling data is preprocessed to obtain the preprocessed sampling data; the preprocessing includes at least one of the following: denoising, static correction, amplitude compensation, and deconvolution.

[0026] According to embodiments of this disclosure, the aforementioned seismic compressed sampling data is in the case of at least one of the following situations where the sample point distribution is non-uniform: data points whose actual sampling locations deviate from the standard sampling point locations on the standard grid; and data points whose actual sampling locations are missing compared to the standard sampling point locations on the standard grid. Wherein, when there is a non-uniform sample point distribution in the one-dimensional sampling direction, the aforementioned non-uniform reconstruction factor is one-dimensional; and when the distribution of sample points in the two-dimensional sampling direction is non-uniform, the aforementioned non-uniform reconstruction factor is two-dimensional.

[0027] Secondly, embodiments of this disclosure provide a reconstruction apparatus for seismic compressed sampling data. The reconstruction apparatus includes: a loading module, a location mapping relationship calculation module, a non-uniform reconstruction factor construction module, and an iterative solution module. The loading module loads the acquired seismic compressed sampling data based on an observation system to obtain observation information, including: the actual sampling location and sampling data at the actual sampling location. The location mapping relationship calculation module determines the target sampling location and sampling matrix corresponding to the actual sampling location on the standard grid of the observation system. The non-uniform reconstruction factor construction module constructs a non-uniform reconstruction factor characterizing the degree of non-uniformity in mapping the target sampling location in the standard grid to the actual sampling location. The iterative solution module iteratively solves the non-uniform reconstruction factor, the sampling matrix, and the sampling data to obtain reconstructed sampling data related to the seismic compressed sampling data.

[0028] Thirdly, embodiments of this disclosure provide an electronic device. The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, communication interface, and memory communicate with each other via the communication bus; the memory stores computer programs; and the processor, when executing the program stored in the memory, implements the seismic compressed sampling data reconstruction method as described above.

[0029] Fourthly, embodiments of this disclosure provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for reconstructing seismic compressed sampling data as described above.

[0030] Some of the technical solutions provided in the embodiments of this disclosure have at least some or all of the following advantages:

[0031] By determining the target sampling position corresponding to the true sampling position on the standard grid of the observation system; and constructing a non-uniform reconstruction factor that characterizes the degree of non-uniformity in mapping the target sampling position in the standard grid to the true sampling position; and performing reconstruction iteration based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data, the reconstruction accuracy can be effectively improved under non-uniform distribution conditions of shot-receiver points such as deviation sampling positions and missing sampling positions, thus improving the stability of the reconstruction algorithm.

[0032] Some of the technical solutions provided in the embodiments of this disclosure have at least some or all of the following advantages:

[0033] When selecting sparse transformation operations, the choice between parallel three-dimensional (3D) curvelet transform (when the sampling points in the transverse direction meet the requirements of 3D curvelet transform) or parallel two-dimensional (2D) curvelet transform + one-dimensional (1D) fast Fourier transform (when there are fewer sampling points in the transverse direction and the requirements of 3D curvelet transform are not met) can be made based on different applicable conditions. This can meet the challenge of time-consuming computation when processing massive irregular data and improve the computational efficiency of the algorithm.

[0034] Some of the technical solutions provided in the embodiments of this disclosure have at least some or all of the following advantages:

[0035] By using an accelerated algorithm to iteratively calculate and solve the objective function, the number of iterations can be effectively reduced, the convergence speed can be improved, and thus the computational efficiency of the reconstruction process can be increased. Attached Figure Description

[0036] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0037] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0038] Figure 1 A flowchart illustrating a method for reconstructing seismic compressed sampling data according to an embodiment of the present disclosure is shown schematically.

[0039] Figure 2 This diagram schematically illustrates the result of the actual sampling location of the seismic compressed sampling data in the observation information after the observation system is loaded, according to an embodiment of the present disclosure.

[0040] Figure 3 A detailed implementation flowchart of determining the target sampling location according to an embodiment of the present disclosure is illustrated schematically;

[0041] Figure 4 This diagram schematically illustrates the result of reconstructing the target sampling position corresponding to the actual sampling position of the seismic compressed sampling data according to an embodiment of this disclosure.

[0042] Figures 5-8 The illustration schematically shows a comparison of the reconstruction effect of the reconstruction method provided in the embodiments of this disclosure based on three-dimensional simulation data with that of the reconstruction method using the three-dimensional curvelet fast convex set projection (POCS) method in the comparative embodiments;

[0043] Figure 5 A schematic diagram illustrating the effect of three-dimensional simulation data of seismic compressed sampling data according to an embodiment of the present disclosure is shown.

[0044] Figure 6 The illustration schematically shows the effect of the three-dimensional simulation data after non-uniform processing by performing an overall 50% undersampling on the three-dimensional simulation data according to an embodiment of the present disclosure, and then offsetting the spatial position of the original points falling on the regular grid.

[0045] Figure 7 The illustration shows the data effect after reconstructing the non-homogenized three-dimensional simulation data using the three-dimensional curvelet fast POCS method in the comparative embodiment.

[0046] Figure 8 The illustration shows a schematic diagram of the data effect after reconstructing non-homogenized three-dimensional simulation data based on the reconstruction method provided in the embodiments of this disclosure;

[0047] Figure 9 This diagram schematically illustrates the effect of actual land data collected according to an embodiment of the present disclosure after being loaded by an observation system;

[0048] Figure 10 The diagram illustrates the spacing distribution of (a) shot points and (b) receiver points in actual land data acquired according to an embodiment of this disclosure.

[0049] Figure 11 The illustration shows the effect of the observation system loading before (a) and after (b) reconstruction of the actual acquired land data using the three-dimensional curve wave fast POCS method in the comparative embodiment;

[0050] Figure 12 The diagram illustrates the effects of the reconstruction method provided in this disclosure on the actual collected land data before (a) and after (b) reconstruction, after the observation system is loaded.

[0051] Figure 13The diagram illustrates the effect of gather data of actual land data collected according to an embodiment of this disclosure.

[0052] Figure 14 The illustration shows the data effect after reconstructing the actual land data using the three-dimensional curved wave fast POCS method in the comparative embodiment;

[0053] Figure 15 The diagram illustrates the data effect after reconstructing actual collected land data based on the reconstruction method provided in the embodiments of this disclosure;

[0054] Figure 16 A schematic diagram illustrating the structural block diagram of a seismic compressed sampling data reconstruction apparatus according to an embodiment of the present disclosure; and

[0055] Figure 17 A schematic block diagram of an electronic device provided in an embodiment of the present disclosure is shown. Detailed Implementation

[0056] The key to compressed sensing seismic exploration technology lies in the reconstruction and recovery of irregular data; the quality of this reconstruction directly determines the success or failure of the technology. However, during development, it was found that commonly used reconstruction techniques are based on the assumption that the sample data acquired after undersampling is located at the target grid points. In actual acquisition, due to factors such as receiver placement errors and obstacles, the positions of shot points and receivers may deviate from the pre-designed sample point positions, or some data points may be missing. This leads to a decrease in the effectiveness of conventional reconstruction techniques for recovering missing data. Furthermore, some reconstruction algorithms are computationally very time-consuming, resulting in low computational efficiency.

[0057] In view of this, embodiments of the present disclosure provide a method and apparatus for reconstructing seismic compressed sampling data. By constructing a non-uniform reconstruction factor, the reconstructed sampling data calculated based on the non-uniform reconstruction factor has high reconstruction accuracy, improves the stability of the reconstruction algorithm, and effectively solves the problem of poor reconstruction accuracy when shot and receiver points are not uniformly distributed, such as in cases of biased sampling locations or missing sampling locations.

[0058] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0059] The first exemplary embodiment of this disclosure provides a method for reconstructing earthquake compressed sampling data.

[0060] Figure 1 A flowchart illustrating a method for reconstructing seismic compressed sampling data according to an embodiment of the present disclosure is shown.

[0061] Reference Figure 1 As shown, the method for reconstructing seismic compressed sampling data provided in this embodiment includes the following steps: S110, S120, S130 and S140.

[0062] In step S110, based on the observation system, the collected seismic compressed sampling data is loaded to obtain observation information, which includes: the actual sampling location and the sampling data at the actual sampling location.

[0063] Commonly used reconstruction techniques are based on the fact that the sample data collected after undersampling is on the target grid point of the reconstruction. However, in actual acquisition, due to the placement error of the receiver point and the influence of obstacles, the positions of the shot point and receiver point will deviate from the pre-designed sample point positions. This also leads to a worse effect of using conventional reconstruction techniques to recover missing data.

[0064] In step S110, the observation information of the earthquake compressed sampling data can be loaded based on the observation system. The actual sampling location in the above observation information includes: line number, point number, location and other information related to the shot point and receiver point, as well as the relationship information between the shot point and receiver point.

[0065] According to embodiments of this disclosure, the aforementioned seismic compressed sampling data is in the case of at least one of the following situations where the sample point distribution is non-uniform: data points whose actual sampling locations deviate from the standard sampling point locations on the standard grid; and data points whose actual sampling locations are missing compared to the standard sampling point locations on the standard grid.

[0066] Figure 2 This diagram schematically illustrates the result of the actual sampling location of the seismic compressed sampling data in the observation information after the observation system is loaded, according to an embodiment of the present disclosure.

[0067] For example, the standard sampling points on the standard grid of this observation system are evenly spaced, referring to... Figure 2 As shown, the earthquake compressed sampling data contains missing data points at the actual sampling locations in the observation information after the observation system is loaded. Figure 2 The data point range marked by the dashed box is used for illustration. Among the shot point number and receiver point number marked by the dashed box, there are multiple missing data points. For example, there is a missing sample data corresponding to the shot point number (e.g., 1324.00 on the standard grid) between shot point number 1322.00 and 1326.00.

[0068] Although the above embodiments use the example of standard sampling points on a standard grid being evenly distributed, it should be understood that in other embodiments, the standard sampling points may also be non-evenly distributed. As long as there is a deviation or missing points between the seismic compressed sampling data and the standard sampling points on the standard grid, it meets the condition of non-uniform distribution.

[0069] In step S120, the target sampling position and sampling matrix corresponding to the above-mentioned real sampling position on the standard grid of the above-mentioned observation system are determined.

[0070] In one embodiment, the sampling matrix is ​​determined based on the aforementioned observation information. For example, depending on the design of the compressed sensing seismic exploration observation system, the sampling matrix is ​​determined by shot line number and shot point number (shot points are irregular), or receiver line number and receiver point number (receiver points are irregular), or a combination of both (both shot points and receiver points are irregular). This sampling matrix is ​​a 0-1 matrix, where the sampling element corresponding to the point of actual data acquisition is 1, and the sampling element corresponding to the missing point is 0.

[0071] Figure 3 A detailed implementation flowchart of determining the target sampling location according to an embodiment of the present disclosure is illustrated schematically.

[0072] According to embodiments of this disclosure, referring to Figure 3 As shown, in step S120 above, determining the target sampling position corresponding to the actual sampling position on the standard grid of the observation system includes the following steps: S310 and S320.

[0073] In step S310, an optimization problem is constructed regarding the mapping relationship between the above-mentioned real sampling locations and the standard sampling point locations corresponding to the standard grid of the observation system.

[0074] In step S320, the above optimization problem is solved to obtain the target sampling position corresponding to the above true sampling position on the above standard grid.

[0075] The actual sampling locations in the observation system correspond to the observation grid, and the actual sampling locations on the observation grid are represented as p1,K,p m The standard sampling point locations (the sampling locations corresponding to the expected output results) on the standard grid of the observation system are denoted as l1,K,l. n The correspondence between the actual sampling locations on the observation grid and the standard sampling point locations on the standard grid (given and known) is unknown and needs to be solved through a constructed optimization problem.

[0076] According to some embodiments of this disclosure, the above optimization problem satisfies the following expression:

[0077]

[0078] Among them, s j Indicates the actual sampling position p j The positions of n standard sampling points on the standard grid of the observation system are l s The position number of the target sampling location corresponding to the minimum distance; The symbol represents the definition; j represents the m actual sampling positions p j The sequence number j, where j ranges from 1 to m, m ≤ n; s represents the position l of the standard sampling point. s The position index of s, where s can range from 1 to n; argmin() represents the value of the variable that minimizes the value of the function inside the parentheses.

[0079] Based on steps S310 to S320, for the scenario problem where the sampling point does not fall on the reconstructed grid point, the constructed optimization problem treats the real sampling position as a point on the observation grid of the observation system. Based on minimizing the deviation between the point on the observation grid and the point on the standard grid, the target sampling position corresponding to the point on the observation grid is determined from the n standard sampling point positions on the standard grid.

[0080] In other embodiments, the reference point corresponding to the point with the smallest product of the distances between points on multiple nearby observation grids (e.g., 0, 1 to 3) including the reference point and points on the standard grid can be determined as the target sampling location corresponding to the actual sampling location.

[0081] In the embodiments of this disclosure, observation system file information corresponding to the reconstructed sampling data can be constructed based on the observation information of the seismic compressed sampling data (which can be described as the original observation information) (shot point information, receiver point information, and relationship information between shot points and receiver points), and a new observation system file (i.e., the target sampling location determined by step S120, and the reconstructed sampling data corresponding to the target sampling location obtained by steps S130 to S140) can be output; the reconstructed sampling data can be reloaded using the newly generated observation system file and used for subsequent fine processing and migration imaging.

[0082] Figure 4 The diagram illustrates the result of reconstructing the target sampling position corresponding to the actual sampling position of the seismic compressed sampling data according to an embodiment of the present disclosure.

[0083] Reference Figure 4 As shown in the dashed box, the actual sampling location is processed by the optimization problem, and the target sampling location obtained effectively recovers the missing target sampling location. This only illustrates the shot-receiver point location information in the reconstructed data. The specific results of the reconstructed sampling data will be shown in subsequent embodiments.

[0084] In step S130, a non-uniform reconstruction factor is constructed to characterize the degree of non-uniformity in mapping the target sampling position in the standard grid to the actual sampling position.

[0085] According to embodiments of this disclosure, the above-mentioned non-uniform reconstruction factor satisfies the following expression:

[0086]

[0087]

[0088] in, Let represent the non-uniform reconstruction factor, in m x n matrix form; Let I represent the matrix elements of the non-uniform reconstruction factor, where I denotes the identity matrix. This represents the non-uniform reconstruction vector corresponding to the positions of n standard sampling points at the first true sampling position; This represents the non-uniform reconstruction vector corresponding to the position of n standard sampling points at the m-th real sampling position; k represents the preset value given by the user, and the value of k ranges from 0 to 3; i represents the traversal index of the non-uniform reconstruction vector in the column dimension corresponding to the standard grid, and the value of i ranges from 1 to n; h represents the traversal index in the column dimension that is different from the value of i. Indicates the target sampling location. This represents the sampling location points within h grid intervals of the target sampling location in the standard grid; This represents the sampling location point within the i-th grid interval of the target sampling location in the standard grid.

[0089] In step S140, the non-uniform reconstruction factor, the sampling matrix, and the sampling data are iteratively solved to obtain the reconstructed sampling data of the seismic compression sampling data.

[0090] According to an embodiment of this disclosure, in step S150 above, the reconstructed sampling data about the seismic compression sampling data is obtained by iteratively solving based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data, including:

[0091] Based on the non-uniform reconstruction factor Given the sampling matrix R and the sampled data b, determine the objective function and corresponding constraints for the data reconstruction process.

[0092] The reconstruction problem is represented as:

[0093] Where u represents the reconstructed sampled data to be determined; the sampled data b is known.

[0094] The objective function and constraints of the reconstruction process are: min||x||1 st

[0095] Where x represents the representation coefficient of the sampled data to be reconstructed in the sparse domain, τ is the tolerable residual size, ||||1 represents the L1 norm, and is the sum of the absolute values ​​of each element in the matrix; D represents the matrix of sparse transformation operations; D H This represents the inverse transform corresponding to D; the above sampling matrix is ​​determined based on the above observation information;

[0096] The solution to the above objective function is expressed as:

[0097]

[0098] Among them, u c+1 T represents the reconstructed solution at the (c+1)th iteration; λ Represents the threshold function; u c This represents the reconstructed solution at the c-th iteration; express The inverse transform. The threshold function mentioned above can typically be either a hard threshold function or a soft threshold function, where λ represents the set constraint threshold, defined by the user.

[0099] Based on the above steps S110 to S140, by determining the target sampling position corresponding to the real sampling position on the standard grid of the observation system; and constructing a non-uniform reconstruction factor that characterizes the degree of non-uniformity in mapping the target sampling position in the standard grid to the real sampling position; and performing reconstruction iterative solution based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data, the reconstruction accuracy can be effectively improved under the condition of non-uniform distribution of shot-receiver points such as deviation sampling positions and missing sampling positions, and the stability of the reconstruction algorithm can be improved.

[0100] According to some embodiments of this disclosure, when the sampling points in the transverse survey line direction meet the requirements of the three-dimensional curved wave transform, the sparse transformation operation in formula (5) includes the three-dimensional curved wave transform; when there are fewer sampling points in the transverse survey line direction and the requirements of the three-dimensional curved wave transform are not met, the above sparse transformation operation includes the two-dimensional curved wave transform and the one-dimensional fast Fourier transform.

[0101] When selecting sparse transformation operations, the choice between parallel three-dimensional (3D) curvelet transform (when the sampling points in the transverse direction meet the requirements of 3D curvelet transform) or parallel two-dimensional (2D) curvelet transform + one-dimensional (1D) fast Fourier transform (when there are fewer sampling points in the transverse direction and the requirements of 3D curvelet transform are not met) can be made based on different applicable conditions. This can meet the challenge of time-consuming computation when processing massive irregular data and improve the computational efficiency of the algorithm.

[0102] According to some embodiments of this disclosure, the objective function is solved iteratively based on the following acceleration algorithm:

[0103]

[0104]

[0105] Among them, t c This represents the iteration step size at the c-th iteration, where the initial iteration step size t0 is a given value.

[0106] By using an accelerated algorithm to iteratively calculate and solve the objective function, the number of iterations can be effectively reduced, the convergence speed can be improved, and thus the computational efficiency of the reconstruction process can be increased.

[0107] According to an embodiment of this disclosure, when loading the collected seismic compressed sampling data, the seismic compressed sampling data is preprocessed to obtain the preprocessed sampling data; the preprocessing includes, but is not limited to, at least one of the following operations: denoising, static correction, amplitude compensation, and deconvolution.

[0108] In some embodiments, in order to highlight the performance improvement of the present disclosure embodiments and the contribution of the non-uniform reconstruction factor to the effect, the following is a comparison of the effects of the reconstructed data obtained by the 3D curve wave fast POCS (convex set projection) method and the reconstructed sampling data obtained by the reconstruction method provided in the present disclosure embodiments for the same seismic compressed sampling data.

[0109] Figures 5-8 The illustration schematically shows a comparison of the reconstruction effect of the reconstruction method provided in the embodiments of this disclosure based on three-dimensional simulation data with that of the reconstruction method using the three-dimensional curvelet fast convex set projection (POCS) method in the comparative embodiments.

[0110] Figure 5 A schematic diagram illustrating the effect of three-dimensional simulation data of seismic compressed sampling data according to an embodiment of the present disclosure is shown.

[0111] In this embodiment, refer to Figure 5 As shown, the size of the three-dimensional simulation data is 256×128×128 (each part is in pixels). The time-distance curves presented by the observation system loading the above three-dimensional simulation data are continuous lines, and the images corresponding to the xy coordinates are very clear.

[0112] Figure 6 The illustration schematically shows the effect of non-uniformized 3D simulation data after performing an overall 50% undersampling on the 3D simulation data according to an embodiment of the present disclosure, and then offsetting the spatial position of the original points falling on the regular grid.

[0113] By undersampling the simulated data by 50% and then performing non-uniformity processing, this embodiment uses the spatial offset of points originally falling on a regular grid as an example of non-uniformity processing. (Refer to...) Figure 6 As shown, the time-distance curve after undersampling and non-uniformization processing presents discontinuous line segments, and the image corresponding to the xy coordinates is very blurry.

[0114] Figure 7 The illustration shows the data effect after reconstructing the non-homogenized three-dimensional simulation data using the three-dimensional curvelet fast POCS method in the comparative embodiment.

[0115] Reference Figure 7 As shown, due to the influence of non-uniform sampling point processing, the 3D curve wave fast POCS method cannot recover the missing data well. The signal-to-noise ratio of the reconstructed signal obtained by this method is calculated to be 4.23dB.

[0116] Figure 8 The diagram illustrates the data effect after reconstructing non-homogenized three-dimensional simulation data using the reconstruction method provided in the embodiments of this disclosure.

[0117] Reference Figure 8 As shown, in the reconstruction method provided by the embodiments of this disclosure, the reconstruction effect of the reconstructed sample data obtained by introducing a non-uniform reconstruction factor is significantly improved. According to calculation, the signal-to-noise ratio of the reconstructed signal obtained by this method is 12.13dB, which can effectively improve the reconstruction accuracy under the condition of non-uniform distribution of shot-receiver points such as deviation sampling position and missing sampling position, and improve the stability of the reconstruction algorithm.

[0118] In addition to using simulated data for verification, actual collected seismic compressed sampling data are used to verify the effectiveness of the reconstruction method provided in the embodiments of this disclosure.

[0119] In this embodiment, the collected seismic compressed sampling data is from a certain land area. The observation system uses an irregular distribution of receiver and shot point spacing, with an average shot point spacing and average receiver spacing of approximately 66m. Shot lines and receiver lines are regularly distributed with a spacing of 200m. To recover the missing shot and receiver data in one go as much as possible, this experiment uses a cross-shaped gather arrangement to test the reconstruction effect.

[0120] Figure 9 This diagram schematically illustrates the effect of actual land data collected according to an embodiment of the present disclosure after being loaded by an observation system; Figure 10 The diagram schematically illustrates the spacing distribution of (a) shot points and (b) receiver points in actual land data acquired according to an embodiment of this disclosure. Figure 13The diagram illustrates the effect of gather data of actual land data collected according to an embodiment of the present disclosure.

[0121] Reference Figure 9 The diagram illustrates a cross-shaped observation system. A sample data space for a particular cross-shaped arrangement is 180×177, where 180 and 177 represent the number of receiver points and shot points, respectively. (Refer to...) Figure 10 As shown in (a) and (b), the actual interval between the shot points and receiver points in the cross-shaped data acquisition fluctuates approximately 50 meters, 100 meters, or 150 meters. In the embodiments of this disclosure, the grid spacing of both shot points and receiver points is set to 50 meters within the given standard grid of the observation system (corresponding to the standard sampling point positions of the new desired output observation system). Based on the standard grid of the observation system and the actual acquired data, the size of the reconstructed data space can be determined to be 240 × 239, meaning the missing rate of the original data compared to the reconstructed data is approximately 44%. (Refer to...) Figure 13 As shown, the original data is incomplete, resulting in a blurry overall image.

[0122] Figure 11 The illustration schematically shows the effect of the observation system loading before (a) and after (b) reconstruction of the actual acquired land data using the three-dimensional curved wave fast POCS method in the comparative embodiment. Figure 14 The diagram illustrates the data effect after reconstructing actual land data using the three-dimensional curved wave fast POCS method in the comparative embodiment. Figure 12 The diagram illustrates the effects of the reconstruction method provided in this disclosure on the actual collected land data before (a) and after (b) reconstruction, after the observation system is loaded. Figure 15 This diagram schematically illustrates the data effect after reconstructing actually collected land data using the reconstruction method provided in this embodiment. Figure 14 and Figure 15 Arrows are used to indicate the phase axes. Figure 11 The specific coordinates are omitted. Figure 11 (a) and Figure 9 It's the same, you can refer to it. Figure 9 The coordinates are shown in the diagram.

[0123] By comparison Figure 11 (a) and (b) comparison Figure 12 (a) and (b) comparison Figure 11 (b) and Figure 12 (b) and combination Figure 13 , Figure 14 and Figure 15As shown in the data rendering, although the reconstructed data obtained using 3D curve reconstruction can effectively recover missing data, there are still discrepancies between the collected data points and the theoretical grid, for example... Figure 9 and Figure 11 The data points with deviations shown in (a) (the part between the central intersection and the upper right diagonal line, where data points are out of position) remain unchanged after 3D curve reconstruction; their corresponding positions still deviate from the theoretical grid (see reference). Figure 11 As shown in (b), this will cause localized jitter or discontinuity in the in-phase axis, refer to... Figure 14 The in-phase axis indicated by the middle arrow is shown. Figure 14 The circled area exhibits jitter and discontinuity. In contrast, the reconstruction method provided in this embodiment corrects the positions of data points deviating from the standard grid and fills in missing data points after reconstruction. Figure 12 As shown in (b); therefore, the reconstruction method provided by this embodiment can effectively improve the continuity of the in-phase axis and achieve a higher accuracy reconstruction effect, referring to... Figure 15 The in-phase axis indicated by the middle arrow, Figure 15 The data circled in the circle is continuous.

[0124] A second exemplary embodiment of this disclosure provides an apparatus for reconstructing earthquake compressed sampling data.

[0125] Figure 16 A schematic block diagram of a seismic compressed sampling data reconstruction apparatus according to an embodiment of the present disclosure is shown.

[0126] Reference Figure 16 As shown, the seismic compressed sampling data reconstruction device 1600 provided in this embodiment includes: a loading module 1601, a location mapping relationship calculation module 1602, a non-uniform reconstruction factor construction module 1603, and an iterative solution module 1604.

[0127] The loading module 1601 is used to load the collected seismic compressed sampling data based on the observation system to obtain observation information, which includes: the actual sampling location and the sampling data at the actual sampling location.

[0128] The aforementioned location mapping relationship calculation module 1602 is used to determine the target sampling position corresponding to the aforementioned real sampling position on the standard grid of the aforementioned observation system.

[0129] The non-uniform reconstruction factor construction module 1603 is used to construct a non-uniform reconstruction factor that characterizes the degree of non-uniformity in mapping the target sampling position in the standard grid to the actual sampling position.

[0130] The iterative solution module 1604 is used to iteratively solve based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data to obtain reconstructed sampling data about the seismic compression sampling data.

[0131] The various functional modules in the device provided in this embodiment may further include functional modules or sub-modules that can implement the various detailed steps corresponding to the first embodiment. These can be understood with reference to the detailed steps in the first embodiment, and will not be repeated here.

[0132] Any plurality of the functional modules included in the reconfiguration device 1600 may be combined into one module, or any one of the modules may be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules may be combined with at least a portion of the functionality of other modules and implemented in one module. At least one of the functional modules included in the reconfiguration device 1600 may be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any appropriate combination of any of these three implementation methods. Alternatively, at least one of the functional modules included in the reconfiguration device 1600 may be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0133] A third exemplary embodiment of this disclosure provides an electronic device.

[0134] Figure 17 A schematic block diagram of an electronic device provided in an embodiment of the present disclosure is shown.

[0135] Reference Figure 17 As shown, the electronic device 1700 provided in this embodiment includes a processor 1701, a communication interface 1702, a memory 1703, and a communication bus 1704. The processor 1701, the communication interface 1702, and the memory 1703 communicate with each other through the communication bus 1704. The memory 1703 is used to store computer programs. When the processor 1701 executes the program stored in the memory, it implements the seismic compressed sampling data reconstruction method described above.

[0136] A fourth exemplary embodiment of this disclosure also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for reconstructing seismic compressed sampling data as described above.

[0137] The computer-readable storage medium may be included in the device / apparatus described in the above embodiments; or it may exist independently and not assembled into the device / apparatus. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0138] According to embodiments of this disclosure, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0139] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0140] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for reconstructing seismic compressed sampling data, characterized in that, include: Based on the observation system, the collected seismic compressed sampling data is loaded to obtain observation information, which includes: the actual sampling location and the sampling data at the actual sampling location; Determine the target sampling position and sampling matrix corresponding to the actual sampling position on the standard grid of the observation system; Construct a non-uniform reconstruction factor that characterizes the degree of non-uniformity in mapping the target sampling position in the standard grid to the real sampling position; The non-uniform reconstruction factor satisfies the following expression: in, Denotes the non-uniform reconstruction factor, as m OK n Columnar matrix form; …、 … … Let I represent the matrix elements of the non-uniform reconstruction factor, where I denotes the identity matrix. … This indicates the location corresponding to the first actual sampling position. n Non-uniform reconstruction vectors at the locations of standard sampling points; … This represents the non-uniform reconstruction vector corresponding to the position of n standard sampling points at the m-th real sampling position; k represents the preset value given by the user, and the value of k is 0~3; i represents the traversal index of the non-uniform reconstruction vector in the column dimension corresponding to the standard grid, and the value of i is 1~n; h represents the traversal index in the column dimension that is different from the value of i. Indicates the target sampling location. This represents the sampling location points within h grid intervals of the target sampling location in the standard grid; This represents the sampling location point within the i nearest neighbor grid cells of the target sampling location in the standard grid; The reconstructed sampling data of the seismic compressed sampling data is obtained by iteratively solving based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data.

2. The reconstruction method according to claim 1, characterized in that, Determining the target sampling location corresponding to the true sampling location on the standard grid of the observation system includes: An optimization problem is to construct the mapping relationship between the actual sampling locations and the standard sampling point locations corresponding to the standard grid of the observation system; Solving the optimization problem yields the target sampling position corresponding to the actual sampling position on the standard grid.

3. The reconstruction method according to claim 2, characterized in that, The optimization problem satisfies the following expression: ， in, Indicates the actual sampling location On the standard grid of the observation system n Location of each standard sampling point The position number of the target sampling location corresponding to the minimum distance; Indicates the definition symbol; express m Real sampling locations The serial number, j The value can be 1~ m, m ≤ n ; s Indicates the location of the standard sampling point Position number, s The value can be 1~ n ; argmin() represents the value of the variable that minimizes the value of the function inside the parentheses.

4. The reconstruction method according to claim 1, characterized in that, Iteratively solving based on the non-uniform reconstruction factor, the sampling matrix, and the sampling data yields reconstructed sampling data regarding the seismic compression sampling data, including: Based on the non-uniform reconstruction factor Sampling matrix R and sampling data b Determine the objective function and corresponding constraints for the data reconstruction process: The reconstruction problem is expressed as: , u This represents the reconstruction of sampled data; the objective function and constraints of the reconstruction process are: ; x This represents the characterization coefficients of the sampled data to be reconstructed in the sparse domain. Let |||1 represent the L1 norm and |||1 represent the tolerable residual size. Let |||1 represent the sum of the absolute values ​​of each element in the matrix. A matrix representing sparse transformation operations; Indicates and The corresponding inverse transform; the sampling matrix is ​​determined based on the observation information; The solution to the objective function is expressed as: ， in, Indicates the first c The reconstructed solution at +1 iteration; Represents the threshold function. This indicates the set constraint threshold. Indicates the first c The reconstructed solution at the next iteration; express The inverse transform of .

5. The reconstruction method according to claim 4, characterized in that, When the sampling points in the transverse survey line direction meet the requirements of three-dimensional curvilinear transformation, the sparse transformation operation includes three-dimensional curvilinear transformation; When there are too few sampling points in the transverse survey line direction, and the requirements of three-dimensional curvilinear transformation are not met, the sparse transformation operation includes two-dimensional curvilinear transformation and one-dimensional fast Fourier transform.

6. The reconstruction method according to claim 4, characterized in that, The objective function is solved iteratively using the following accelerated algorithm: ， ， in, Indicates the first c The iteration step size at the next iteration, the iteration step size at the initial iteration Given a value, I represents the identity matrix.

7. The reconstruction method according to claim 1, characterized in that, When loading the collected seismic compressed sampling data, the seismic compressed sampling data is preprocessed to obtain the preprocessed sampling data. The preprocessing includes at least one of the following: noise reduction, static correction, amplitude compensation, and deconvolution.

8. The reconstruction method according to any one of claims 1-7, characterized in that, The earthquake compressed sampling data is in the case of at least one of the following situations where the sample point distribution is non-uniform: Includes data points whose actual sampling locations deviate from the standard sampling point locations on the standard grid; Includes data points that are missing from the actual sampling location compared to the standard sampling point location on the standard grid; Specifically, when there is a non-uniform distribution of sample points in the one-dimensional sampling direction, the non-uniform reconstruction factor is one-dimensional; when the distribution of sample points in the two-dimensional sampling direction is non-uniform, the non-uniform reconstruction factor is two-dimensional.

9. A device for reconstructing seismic compressed sampling data, characterized in that, The reconstruction apparatus implements the reconstruction method according to any one of claims 1-8, and the reconstruction apparatus comprises: The loading module is used to load the collected seismic compressed sampling data based on the observation system to obtain observation information, which includes: the actual sampling location and the sampling data at the actual sampling location; The location mapping relationship calculation module is used to determine the target sampling position and sampling matrix corresponding to the real sampling position on the standard grid of the observation system; The non-uniform reconstruction factor construction module is used to construct a non-uniform reconstruction factor that characterizes the degree of non-uniformity in mapping the target sampling position in the standard grid to the real sampling position; The iterative solution module is used to iteratively solve the non-uniform reconstruction factor, the sampling matrix, and the sampling data to obtain reconstructed sampling data about the seismic compression sampling data.

10. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-8.

Images