A method, device and equipment for suppressing ghost waves in three-dimensional seismic data

Abstract

The present invention provides a method, device, and apparatus for suppressing ghost waves in three-dimensional seismic data. The method comprises the following steps: obtaining a sparse three-dimensional seismic wave field recorded during offshore seismic acquisition; densely extrapolating the sparse three-dimensional seismic wave field backward by a first distance to obtain a first dense three-dimensional seismic wave field; densely extrapolating the sparse three-dimensional seismic wave field forward by a second distance and performing the reverse operation to obtain a second dense three-dimensional seismic wave field; determining a dense ghost wave field at the water surface based on the first dense three-dimensional seismic wave field and the second dense three-dimensional seismic wave field; extrapolating the dense ghost wave field forward by a second distance and performing sparse sampling to obtain a sparse ghost wave field at a mirror detection point; and subtracting the sparse ghost wave field from the sparse three-dimensional seismic wave field. The method addresses both the undersampling problem of three-dimensional seismic data and the ghost wave suppression problem, respects the propagation law of seismic waves in a three-dimensional sense, and achieves the goals of suppressing ghost waves, improving seismic data characteristics, and increasing seismic data resolution.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of seismic data processing, and in particular to a method, device and equipment for suppressing ghost waves in three-dimensional seismic data. Background Art

[0002] The air-water interface is a strong reflector. The downward reflection of seismic waves from the water surface creates the well-known ghost wave effect in marine seismic exploration. Ghost waves cause notches in the spectrum and can cause phase and amplitude distortion, making them a long-standing challenge in marine seismic exploration. With advances in wide-azimuth marine seismic data acquisition and processing, three-dimensional (3D) seismic exploration has become a mainstream technology in marine seismic exploration. Suppressing ghost wave effects in 3D seismic data is crucial to preserving its bandwidth and resolution.

[0003] Existing ghost suppression methods for 3D seismic data are almost all based on the assumption of dense wavefield sampling. However, in actual streamer 3D seismic acquisition, the lateral spacing between receiver lines is typically much larger than the longitudinal spacing between receivers, with, for example, a trace-to-line spacing ratio of 1:8. In ocean bottom node (OBN) 3D seismic acquisition, the lateral and longitudinal spacing between receivers is also relatively large. These undersampled 3D seismic acquisition methods do not meet the dense wavefield sampling assumption. Therefore, existing methods struggle to achieve effective ghost suppression when applied to undersampled 3D seismic data. Summary of the Invention

[0004] The embodiments of the present invention provide a method, device and apparatus for suppressing ghost waves in three-dimensional seismic data, so as to solve the problem that the existing methods have poor ghost wave suppression effects.

[0005] In a first aspect, an embodiment of the present invention provides a method for suppressing ghost waves in three-dimensional seismic data, comprising:

[0006] Acquire sparse 3D seismic wavefields recorded in offshore seismic acquisition;

[0007] The sparse three-dimensional seismic wave field is densely extrapolated to a first distance to obtain a first dense three-dimensional seismic wave field, where the first distance is the distance between the detection point position and the water surface position;

[0008] Densely extrapolating the sparse three-dimensional seismic wave field to the forward wave field by a second distance and then performing a reverse extrapolation to obtain a second dense three-dimensional seismic wave field, where the second distance is the distance between the mirror detection point position and the water surface position;

[0009] determining a dense ghost wave field at a water surface location based on the first dense three-dimensional seismic wave field and the second dense three-dimensional seismic wave field;

[0010] The dense ghost wave field is extrapolated to the forward wave field by a second distance and sparsely sampled to obtain a sparse ghost wave field at the mirror detection point.

[0011] Subtracting sparse ghost wavefields from sparse 3D seismic wavefields.

[0012] In one embodiment, the sparse 3D seismic wavefield is determined according to the following expression:

[0013] ;

[0014] in, Indicates the water surface position, Indicates the detection point position, Indicates the mirror detection point position, represents the sparse three-dimensional seismic wave field recorded in seismic acquisition, Represents a wave field, which is composed of a water surface wave field Extrapolating backward, we get Represents the ghost wave field, which is composed of the water surface wave field Extrapolating forward, we find that the polarity of the ghost wave field is opposite to that of the primary wave field.

[0015] In one embodiment, the first dense 3D seismic wavefield is determined according to the following expression:

[0016] ;

[0017] The second dense 3D seismic wave field is determined according to the following expression:

[0018] ;

[0019] in, represents the first dense three-dimensional seismic wave field, represents the backward extrapolated wavefield propagation matrix, represents the forward extrapolated wavefield propagation matrix, represents a densely sampled matrix.

[0020] In one embodiment, the sampling density of the dense sampling matrix is ​​determined according to the following expression:

[0021] ;

[0022] in, Spatial sampling interval, represents the speed of longitudinal waves in water, Indicates the maximum frequency of the expected output.

[0023] In one embodiment, determining a dense ghost wave field at a water surface location based on the first dense 3D seismic wave field and the second dense 3D seismic wave field includes:

[0024] summing the first dense three-dimensional seismic wavefield and the second dense three-dimensional seismic wavefield;

[0025] The summation result is threshold-truncated to retain the wave field with the strongest energy, and the dense ghost wave field at the water surface is obtained.

[0026] In one embodiment, the sparse ghost wave field at the mirror detection point location is determined according to the following expression:

[0027] ;

[0028] in, represents the sparse ghost wave field at the mirror detection point location, represents the dense ghost wave field at the water surface, Represents the acquisition observation matrix.

[0029] In one embodiment, during the solution process, lasso regression is used to minimize the following regularized loss function:

[0030] ;

[0031] in, represents the regularization parameter, The time difference of the initial constraint is , is the radius of the sparsely sampled bin.

[0032] In one embodiment, before subtracting the sparse ghost wavefield from the sparse 3D seismic wavefield, the method further includes:

[0033] Least squares constraints are used to determine the matching operators of sparse 3D seismic wavefields and sparse ghost wavefields.

[0034] Applying the matching operator to the sparse ghost wave field to obtain the matched sparse ghost wave field;

[0035] Subtracting the sparse ghost wave field from the sparse three-dimensional seismic wave field includes: subtracting the matched sparse ghost wave field from the sparse three-dimensional seismic wave field.

[0036] In one embodiment, the matching operator is determined according to the following expression:

[0037] ;

[0038] in, represents the sparse three-dimensional seismic wave field, represents a sparse ghost wave field, Represents a matching operator.

[0039] In a second aspect, an embodiment of the present invention provides a ghost wave suppression device for three-dimensional seismic data, comprising:

[0040] An acquisition module for acquiring sparse three-dimensional seismic wave fields recorded in offshore seismic acquisition;

[0041] A first dense field reconstruction module is used to densely extrapolate a first distance of the sparse three-dimensional seismic wave field to the backward wave field to obtain a first dense three-dimensional seismic wave field, where the first distance is the distance between the detection point position and the water surface position;

[0042] A second dense field reconstruction module is used to densely extrapolate a second distance from the sparse three-dimensional seismic wave field to the forward wave field and perform reverse extrapolation to obtain a second dense three-dimensional seismic wave field, where the second distance is the distance between the mirror detection point position and the water surface position;

[0043] a first processing module, configured to determine a dense ghost wave field at a water surface position based on the first dense three-dimensional seismic wave field and the second dense three-dimensional seismic wave field;

[0044] The second processing module is used to extrapolate the dense ghost wave field to the forward wave field by a second distance and perform sparse sampling to obtain a sparse ghost wave field at the mirror detection point position;

[0045] Suppression module for subtracting sparse ghost wavefields from sparse 3D seismic wavefields.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0047] at least one processor and memory;

[0048] Memory stores computer-executable instructions;

[0049] At least one processor executes computer-executable instructions stored in the memory, so that at least one processor executes the ghost wave suppression method for three-dimensional seismic data as described in any one of the first aspects.

[0050] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the ghost wave suppression method for three-dimensional seismic data as described in any one of the first aspects.

[0051] The ghost wave suppression method, device, and apparatus for 3D seismic data provided by the embodiments of the present invention obtain a sparse 3D seismic wave field recorded during offshore seismic acquisition; densely extrapolate the sparse 3D seismic wave field to the backward wave field by a first distance to obtain a first dense 3D seismic wave field; densely extrapolate the sparse 3D seismic wave field to the forward wave field by a second distance and perform the reverse operation to obtain a second dense 3D seismic wave field; determine a dense ghost wave field at the water surface based on the first dense 3D seismic wave field and the second dense 3D seismic wave field; extrapolate the dense ghost wave field to the forward wave field by a second distance and perform sparse sampling to obtain a sparse ghost wave field at the mirror detection point; and subtract the sparse ghost wave field from the sparse 3D seismic wave field to achieve ghost wave suppression. This method takes into account both the undersampling problem and the ghost wave suppression problem of 3D seismic data, respects the propagation law of seismic waves in a three-dimensional sense, and achieves the purpose of suppressing ghost waves, improving seismic data characteristics, and increasing seismic data resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0053] Figure 1 Schematic diagram of the propagation path of marine seismic wave fields;

[0054] Figure 2 Schematic diagram of cross-ghosting;

[0055] Figure 3 A flow chart of a method for suppressing ghost waves in three-dimensional seismic data according to an embodiment of the present invention;

[0056] Figure 4 A schematic diagram of wave field extrapolation provided by an embodiment of the present invention;

[0057] Figure 5 A schematic diagram of dense reconstruction of a sparse wave field according to an embodiment of the present invention;

[0058] Figure 6 A schematic diagram illustrating the effect of ghost wave suppression provided by an embodiment of the present invention;

[0059] Figure 7A and Figure 7B Schematic diagrams of ghost wave suppression for fixed depth and variable depth recording respectively;

[0060] Figure 8 A schematic diagram of the FK spectrum of theoretical data of a variable-depth streamer before and after suppressing ghost waves provided by one embodiment of the present invention;

[0061] Figure 9A and Figure 9B Schematic diagrams of ghost wave suppression of near receiving lines and far receiving lines recorded in three-dimensional theory respectively;

[0062] Figure 10 A schematic diagram of the processing effect of a long-offset streamer provided by an embodiment of the present invention;

[0063] Figure 11 A schematic diagram of ghost wave suppression for close-offset gathers provided in one embodiment of the present invention;

[0064] Figure 12 for Figure 11 Schematic diagram of the FK spectrum of the two records before and after suppressing the ghost wave;

[0065] Figure 13 A schematic diagram of superimposed cross sections before and after suppressing ghost waves provided by an embodiment of the present invention;

[0066] Figure 14 for Figure 13 Schematic diagram of a partial magnified display of the superimposed sections before and after the medium suppression of ghost waves;

[0067] Figure 15 A schematic structural diagram of a ghost wave suppression device for three-dimensional seismic data provided by one embodiment of the present invention;

[0068] Figure 16 A schematic structural diagram of an electronic device provided by an embodiment of the present invention.

[0069] The above drawings illustrate specific embodiments of the present invention, which will be described in more detail below. These drawings and the accompanying description are not intended to limit the scope of the present invention in any way, but rather to illustrate the concept of the present invention to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0070] The present invention will be further described in detail below by means of specific embodiments in conjunction with the accompanying drawings. Similar elements in different embodiments are numbered with associated similar elements. In the following embodiments, many detailed descriptions are provided to enable the present application to be better understood. However, those skilled in the art will readily appreciate that some of the features may be omitted in different circumstances, or may be replaced by other elements, materials, or methods. In some cases, some operations related to the present application are not shown or described in the specification. This is to avoid the core portion of the present application being overwhelmed by excessive descriptions, and for those skilled in the art, it is not necessary to describe these related operations in detail. They will fully understand the related operations based on the description in the specification and the general technical knowledge in the art.

[0071] In addition, the features, operations, or characteristics described in the specification may be combined in any appropriate manner to form various embodiments. Furthermore, the steps or actions in the method description may be reordered or adjusted in a manner readily apparent to those skilled in the art. Therefore, the various sequences in the specification and drawings are provided solely for the purpose of clearly describing a particular embodiment and are not intended to be mandatory, unless otherwise specified.

[0072] Component numbers used herein, such as "first" and "second," are used solely to distinguish the components being described and do not convey any sequential or technical meaning. References to "connection" and "coupling" herein, unless otherwise specified, include both direct and indirect connections (couplings).

[0073] The air-water interface is a strong reflecting surface, so the wave field of seismic waves reflected downward at the water surface forms the well-known ghost wave effect in marine seismic exploration. Figure 1 The figure is a schematic diagram of the propagation path of the marine seismic wave field, which takes into account the side wave field of the excitation point (source) and the side wave field of the receiving point (detector). Figure 1 As shown in the figure, due to the presence of the sea surface, the signal received by the receiving point includes the upward wave field of the excitation energy directly transmitted downward, the excitation energy reflected back by the sea surface, and the downward wave field reflected by the underground interface, as well as the downward wave field that reaches the receiving point after being reflected by the sea surface. In other words, the signal recorded by the receiving point includes not only the primary reflection wave (primary wave), but also the ghost wave on the excitation point side, the ghost wave on the receiving point side, and the excitation-reception ghost wave. Ghost waves not only cause gaps in the spectrum, but also may cause phase and amplitude distortion. For example, the presence of ghost waves will cause notches in the seismic data spectrum. :

[0074]

[0075] in, is the ghost wave propagation distance, is the distance a wave propagates, is the speed at which seismic waves propagate in seawater.

[0076] Ghost waves are a long-standing challenge in marine seismic exploration. In recent decades, extensive research has been conducted to address the problem of ghost waves in seismic data. Weglein and Amundsen et al. used Green's theorem as a general theoretical framework for deghosting. Ferber et al. used streamers at different depths for ghost suppression. Soubaras proposed using mirror migration and deconvolution to suppress ghost signals. He also proposed the classic method of double-detection and combined receiver-side ghost suppression based on cross-ghosting. Poole derived the primary and ghost inverse transform operators based on the frequency-domain Radon transform formula, and obtained the ghost-suppressed seismic records through inversion. Beasley et al. proposed a wave equation deghosting method. Amundsen et al. also proposed that receiver-side ghost suppression can be implemented in the frequency domain as a spatially deterministic deconvolution of the ocean pressure wave field. The ghost suppression (deconvolution) operator is interpreted as the inverse Fourier transform of the ghost suppression function of the wave equation in the wavenumber domain. Berkhout et al. proposed a ghost suppression technique based on reverberation separation. The authors point out the similarities between suppressing ghost waves and removing mixed sampling interference, and propose a similar technique for dealing with ghost waves. This method treats data containing ghost waves as a mixture of real and mirror sources, and then uses mixed data signal separation to separate ghost waves from other signals.

[0077] Soubaras proposed a classic double-check merging algorithm in 1996:

[0078] (1) The hydrophone and geophone reflection data (excluding the first arrival wave) are superimposed at the detection points to obtain the following: Figure 2 H and G shown in , where H is the hydrophone data and G is the geophone data;

[0079] (2) Cross-ghosting of the superimposed data of the hydrophone and geophone reflection data detection points, that is, summing the superimposed data before and after the hydrophone correction to obtain Figure 2 middle

[0080] ;

[0081] The difference between the stacked data before and after the geophone correction is obtained Figure 2 in

[0082] ;

[0083] in, represents the correction factor, represents the superposition data after hydrophone correction, represents the stacked data after correction of the geophone;

[0084] (3) Obtain the matching operator through least squares:

[0085]

[0086] in, is the matching operator.

[0087] (4) Use the following formula to perform wave field separation:

[0088]

[0089] in, It is the upgoing wave field;

[0090]

[0091] in, It is a downward wave field.

[0092] This method uses stacked data after interactive ghosting to determine the matching operator, thus avoiding interference from downlink waves of opposite polarity in the dual-detection matching operator. This method has become a classic because it provides a stable method for determining the matching operator and is currently an excellent option for suppressing ghost signals at the receiving end of dual-detection seismic data. This method, which uses the summation of the vertical components of the hydrophone and geophone to suppress ghost signals, has the following limitations: it can only be used for multi-component seismic data and can only suppress ghost signals at the receiving end.

[0093] In 2014, Berkhout and Blacquiere proposed using reverberation separation to suppress ghost waves. They pointed out the similarities between suppressing ghost waves and removing intermixing noise, and suggested using a similar technique to treat ghost waves. This technique treats data containing ghost waves as a mixture of real and mirror sources, and then uses mixed data signal separation to separate ghost waves from other signals. Specifically:

[0094] (1) Perform two wavefield extrapolations on the original data containing ghost waves: the first extrapolation is to The wave field generated by the earthquake source at the second extrapolation is to extrapolate the wave field at the second extrapolation to the water surface. The wave field (ghost wave field) generated by the virtual source at is extrapolated to the water surface.

[0095] (2) By summing the extrapolated results and performing threshold truncation processing, the primary wave field located at the sea surface can be obtained.

[0096] (3) Extrapolate the wave field generated by the earthquake source located at the sea surface obtained by truncating the threshold, that is, extrapolate the earthquake source located at the sea level to and At this place, you will get a new primary wave field and ghost wave field.

[0097] (4) The primary wave and ghost wave obtained after wave field extension are adaptively subtracted from the input data to obtain ghost wave record and primary wave record respectively.

[0098] The greatest contribution of this method is that it proposes to treat the ghost wave problem as mixed data signal separation, use the wave field extrapolation method to obtain data with only one wave or only ghost waves, and adaptively subtract them from the original data to achieve the purpose of suppressing ghost waves.

[0099] With advances in wide-azimuth marine seismic data acquisition and processing technology, the application of appropriate 3D ghost suppression techniques to better preserve the bandwidth and resolution of recorded data is more essential than ever. Nearly all reported methods for 3D seismic data ghost suppression require dense wavefield sampling. However, in actual streamer 3D seismic acquisition, the lateral receiver line spacing is typically much larger than the longitudinal receiver spacing (for example, a trace-to-line spacing ratio of 1:8). In OBN 3D acquisition, the lateral and vertical spacing of receiver points is relatively large. These acquisition methods do not meet the dense sampling assumption. Therefore, ghost suppression in 3D seismic data faces two challenges: wavefield undersampling and ghost signal removal.

[0100] The reverberation separation method for ghost suppression cannot address the undersampling problem commonly found in real-world offshore 3D seismic data. To address this issue, Berkhout et al. proposed interpolating the data before using reverberation separation for ghost suppression. However, this pre-interpolation method does not effectively suppress ghost waves.

[0101] In order to take into account the undersampling problem of three-dimensional seismic data and the problem of ghost wave suppression, and thus respect the propagation law of seismic waves in a three-dimensional sense, this application reconstructs the dense seismic wave field at a specified location by extrapolating the original sparsely collected seismic wave field, and uses the dense wave field and echo separation method at different locations to solve the dense ghost wave field on the water surface, and then solves the sparse ghost wave field at the original location to solve the ghost wave suppression problem, with the goal of minimizing the difference between the actual data and the simulated wave field, so as to ultimately suppress ghost waves, improve the wave field characteristics of seismic data, and improve the resolution of seismic data. The method proposed in this application will be described in detail below through specific embodiments.

[0102] Example 1

[0103] Please refer to Figure 3 A method for suppressing ghost waves in three-dimensional seismic data provided by an embodiment of the present invention may include:

[0104] S101. Acquire a sparse three-dimensional seismic wave field recorded in offshore seismic acquisition.

[0105] In this embodiment, the sparse three-dimensional seismic wavefield recorded during seismic acquisition can be obtained, for example, through towed three-dimensional acquisition, OBN three-dimensional acquisition, and other methods. Since the entire ghost wave generation process occurs in water, we can regard water as an isotropic medium that satisfies the assumption that the integral surface of the Rayleigh integral is a plane, and the wavefield in water can be described using the Rayleigh integral. Taking this into account, the total scattered wavefield (sparse three-dimensional seismic wavefield) recorded during seismic acquisition can be described as the sum of the backward propagating wavefield and the forward propagating wavefield from the water surface to the excitation point or receiving point. The ghost wave can be assumed to be the signal received by a mirror detection point that is symmetrical to the actual detection point on the water surface. That is, the sparse three-dimensional seismic wavefield recorded during seismic acquisition can be expressed as the sum of the primary wavefield and the ghost wavefield, where the primary wavefield can be obtained by backward extrapolation of the water surface wavefield, and the ghost wavefield can be obtained by forward extrapolation of the water surface wavefield, and the two have opposite polarities.

[0106] S102 : densely extrapolate the sparse three-dimensional seismic wave field backward by a first distance to obtain a first dense three-dimensional seismic wave field, where the first distance is the distance between the detection point position and the water surface position.

[0107] S103 , densely extrapolate the sparse three-dimensional seismic wavefield toward the forward wavefield by a second distance, and then perform reverse extrapolation to obtain a second dense three-dimensional seismic wavefield, where the second distance is the distance between the mirror detection point position and the water surface position.

[0108] To address aliasing caused by insufficient spatial sampling of seismic data and better respect the laws of seismic wave propagation, we must first address the challenges posed by wavefield undersampling. In this embodiment, dense wavefield reconstruction is performed through steps S102 and S103. It should be noted that this embodiment does not restrict the order in which steps S102 and S103 are executed. The dense field reconstruction algorithm can handle streamers of any shape, as the wavefield at any receiver location can be derived by providing the corresponding wave propagation operator.

[0109] In this embodiment, backward can be understood as a direction toward the inside of the water surface, and forward can be understood as a direction away from the water surface toward the outside of the water surface. Figure 4 A schematic diagram of wavefield extrapolation provided by an embodiment of the present invention. Figure 4 The leftmost side is a schematic diagram of the sparse three-dimensional seismic wave field, which includes the primary wave field below the water surface and the ghost wave field above the water surface. Figure 4 The middle part shows the Figure 4 The first dense three-dimensional seismic wavefield is obtained by extrapolating the sparse three-dimensional seismic wavefield on the far left and performing dense sampling. Figure 4 The rightmost part of the Figure 4 The sparse 3D seismic wavefield on the far left is the second dense 3D seismic wavefield obtained by extrapolating the forward wavefield, performing dense sampling and reversing.

[0110] S104: Determine a dense ghost wave field at the water surface according to the first dense three-dimensional seismic wave field and the second dense three-dimensional seismic wave field.

[0111] In order to extract the dense ghost wave field at the water surface, the echo separation and threshold truncation method can be used. The first dense three-dimensional seismic wave field and the second dense three-dimensional seismic wave field mentioned above are summed, and the result is threshold-truncated to retain the wave field with the strongest energy. The ghost wave field of dense sampling on the water surface can be obtained.

[0112] S105 , extrapolate the dense ghost wave field to the forward wave field by a second distance, and perform sparse sampling to obtain a sparse ghost wave field at the mirror detection point position.

[0113] To obtain the ghost wavefield at the mirror geophone location, we introduce a practical acquisition observation matrix, which contains sampling matrices in both the longitudinal and transverse directions. The dense ghost wavefield at the water surface is extrapolated to the forward wavefield, and sparse sampling is performed using the acquisition observation matrix, effectively extrapolating the dense ghost wavefield at the water surface. Due to the acquisition observation matrix, the final data is undersampled, resulting in a sparse ghost wavefield at the mirror geophone location.

[0114] S106. Subtract the sparse ghost wave field from the sparse three-dimensional seismic wave field.

[0115] After obtaining the sparse ghost wave field, the sparse ghost wave field can be subtracted from the sparse 3D seismic wave field to achieve ghost wave suppression.

[0116] The ghost wave suppression method for 3D seismic data provided in this embodiment obtains a sparse 3D seismic wave field recorded during offshore seismic acquisition; densely extrapolates the sparse 3D seismic wave field to the backward wave field by a first distance to obtain a first dense 3D seismic wave field; densely extrapolates the sparse 3D seismic wave field to the forward wave field by a second distance and performs the reverse operation to obtain a second dense 3D seismic wave field; determines a dense ghost wave field at the water surface based on the first dense 3D seismic wave field and the second dense 3D seismic wave field; extrapolates the dense ghost wave field to the forward wave field by a second distance and performs sparse sampling to obtain a sparse ghost wave field at the mirror detection point; and subtracts the sparse ghost wave field from the sparse 3D seismic wave field to achieve ghost wave suppression. This method takes into account both the undersampling problem and the ghost wave suppression problem of 3D seismic data, respects the propagation law of seismic waves in a three-dimensional sense, and achieves the purpose of suppressing ghost waves, improving seismic data characteristics, and increasing seismic data resolution.

[0117] Example 2

[0118] In order to solve the aliasing caused by insufficient spatial sampling of seismic data and better respect the laws of seismic wave propagation, it is necessary to first solve the challenges brought by undersampling of the wave field. Since the entire ghost wave generation process occurs in water, we can regard water as an isotropic medium. It satisfies the assumption that the integral surface of the Rayleigh integral is a plane, and the wave field in the water can be described by the Rayleigh integral. Taking this into account, the total scattered wave field (sparse three-dimensional seismic wave field) recorded in the seismic acquisition can be described as the sum of the backward propagation wave field and the forward propagation wave field from the water surface to the excitation point or the receiving point. Take the case where the water surface reflection coefficient is -1 as an example. At this time, the receiving point side wave field, that is, the sparse three-dimensional seismic wave field, can be expressed as:

[0119] ;

[0120] in, Indicates the water surface position, Indicates the actual detection point position, Indicates that the detection point is located at the mirror detection point position above the water surface. represents the total wave field actually recorded, that is, the sparse three-dimensional seismic wave field recorded in seismic acquisition, Represents a wave field, which is composed of a water surface wave field Extrapolating backward, we get Represents the ghost wave field, which is composed of the water surface wave field Extrapolating forward, we find that the polarity of the ghost wave field is opposite to that of the primary wave field. A ghost wave can be understood as the signal received by a mirror detection point that is symmetrical to the actual detection point in the water surface.

[0121] The dense field reconstruction algorithm can process streamers of any shape, because as long as the corresponding wave propagation operator is provided, the wave field at any receiver position can be obtained. In this embodiment, the recorded sparse 3D seismic wave field is first extrapolated to the back wave field. , directly establish the first dense three-dimensional seismic wave field , specifically:

[0122]

[0123] ;

[0124] Similarly, the recorded sparse 3D seismic wavefield is extrapolated to the forward wavefield. , establish a dense wave field, and multiply by -1 to obtain the second dense three-dimensional seismic wave field , specifically:

[0125]

[0126] ;

[0127] in, represents the first dense three-dimensional seismic wave field, represents the backward extrapolated wavefield propagation matrix, represents the forward extrapolated wavefield propagation matrix, Represents a dense sampling matrix. The sampling density of a dense sampling matrix is ​​determined by the following expression:

[0128] ;

[0129] in, Spatial sampling interval, represents the speed of longitudinal waves in water, Indicates the maximum frequency of the expected output.

[0130] Figure 5 A schematic diagram of dense reconstruction of a sparse wave field provided in one embodiment of the present invention shows the result of two extrapolations of the original theoretical record. Figure 5 Figure (a) is the original under-sampled record (which can be understood as the sparse 3D seismic wave field in this embodiment), and Figure (b) is the high-density sampled record obtained by backward extrapolation (which can be understood as the first dense 3D seismic wave field in this embodiment). The ghost wave at the position will be extrapolated to the water surface position , and the primary wave is extrapolated to Position; (c) Figure is the high-density sampling record obtained by forward extrapolation. In the high-density sampling record obtained by forward extrapolation, a wave is extrapolated to the water surface position. , and the ghost wave will be located at The position of Figure (c) is multiplied by -1 to obtain Figure (d) (which can be understood as the first dense 3D seismic wave field in this embodiment). The same "event" can be observed in the dense records (b) and (d), that is, the two extrapolation formulas , which can also be regarded as the ghost wave field of densely sampled water surface.

[0131] In order to extract the dense ghost wave field at the water surface, the echo separation and threshold truncation method is used to sum the first dense 3D seismic wave field and the second dense 3D seismic wave field. Specifically:

[0132] ;

[0133] At this time, the dense ghost wave field at the water surface The energy is located in TRON and located Wave field Twice the energy. Threshold truncation is performed on the summation result, and the wave field with the strongest energy is retained, so that the dense ghost wave field at the water surface can be obtained. .

[0134] In order to obtain the ghost wave field at the mirror detection point position, the actual acquisition observation matrix can be introduced , contains sampling matrices in both longitudinal and transverse directions, and extrapolates the dense wave field on the water surface forward. , specifically:

[0135] ;

[0136] in, represents the sparse ghost wave field at the mirror detection point location, represents the dense ghost wave field at the water surface, is the propagation matrix extrapolated to the forward wavefield, Represents the acquisition observation matrix. The above expression actually realizes the sparse extrapolation of the dense ghost wave field on the water surface. The existence of makes the final data undersampled, that is, Therefore, the matrix Usually it is not reversible mathematically. To overcome this problem and avoid overfitting in the solution process, we use the typical lasso regression in mathematics. For our optimization goal:

[0137]

[0138] L1 regularization loss function,

[0139] ;

[0140] in, represents the regularization parameter, The time difference of the initial constraint is , is the radius of the sparsely sampled bin.

[0141] Finally, the sparse ghost wave field at the mirror detection point position is subtracted from the original data to achieve ghost wave suppression of 3D seismic data.

[0142] It should be noted that in order to obtain satisfactory results, it is necessary to construct the propagation operator correctly. and In Rayleigh theory, the integral aperture should be infinite, but in reality, this requirement is unrealistic. Therefore, and The finite aperture must be taken into account, and different choices of the two propagation matrices will lead to slightly different results. They can be constructed using weighted least squares optimization of the one-way propagation operator and .

[0143] The ghost wave suppression method for 3D seismic data provided in this embodiment is designed for undersampled 3D seismic data at sea. It reconstructs a dense seismic wave field at a specified location by extrapolating the original sparsely collected seismic wave field. It then uses the dense wave field at different locations and the echo separation method to solve the dense ghost wave field on the water surface, and then solves the sparse ghost wave field at the original location to solve the ghost wave suppression problem. This method can address both the undersampling problem of 3D data and the ghost wave suppression problem, respecting the propagation laws of seismic waves in a 3D sense. The goal is to minimize the difference between the actual data and the simulated wave field, resolve the uncertainty of this problem, and ultimately achieve the purpose of suppressing ghost waves, improving the characteristics of seismic data, and increasing the resolution of seismic data.

[0144] Compared with the existing dual-detection merging technology based on cross-ghosting, which is only suitable for suppressing ghost waves on the receiving point side of dual-detection seismic data by merging the hydrophone component and the vertical component of the seismic detector in multi-component seismic acquisition, the method provided in this embodiment can also be used to process single-component data, streamer data and source side ghost wave suppression.

[0145] Compared with the existing technical solutions that interpolate seismic data before suppressing ghost waves, the method provided in this embodiment does not need to interpolate the seismic data in advance when suppressing ghost waves in undersampled three-dimensional seismic data. Instead, it directly solves the dense ghost wave field, solves the sparse ghost waves at the original position under the lasso regression constraint, and subtracts the ghost waves from the original data to achieve the purpose of suppressing ghost waves.

[0146] Example 3

[0147] It should be noted that in actual processing, since the water layer is not a strictly isotropic medium, the actual reflection coefficient of the water surface is not strictly -1, and there is a slight energy transmission loss. Therefore, in order to further improve the ghost wave suppression effect, before subtracting the sparse ghost wave field from the sparse 3D seismic wave field, the method provided in this embodiment may further include:

[0148] A matching operator between a sparse 3D seismic wavefield and a sparse ghost wavefield is determined using a least squares constraint; the matching operator is applied to the sparse ghost wavefield to obtain a matched sparse ghost wavefield;

[0149] Subtracting the sparse ghost wave field from the sparse three-dimensional seismic wave field includes: subtracting the matched sparse ghost wave field from the sparse three-dimensional seismic wave field.

[0150] In an optional implementation, the matching operator may be determined according to the following expression:

[0151] ;

[0152] in, represents the sparse three-dimensional seismic wave field, represents a sparse ghost wave field, Represents a matching operator.

[0153] Figure 6 A schematic diagram of the ghost wave suppression effect provided by an embodiment of the present invention. Figure 6 A comparison of the ghost wave field obtained by rarefaction extrapolation and by subtracting the ghost waves from the original record is shown. Figure 6 The middle (a) figure shows the dense ghost wave field on the water surface obtained by the reverberation separation method; Figure 6 The middle (b) figure is the ghost wave field at the mirror receiver position obtained by sparse extrapolation of the dense ghost wave field on the water surface; Figure 6 Figure (c) in the middle is a record obtained by directly subtracting the ghost waves in Figure (b) from the original record. It can be seen that the ghost waves in the original record have been well suppressed, but residual ghost waves still exist; Figure 6 The middle figure (d) is the record of obtaining the matched filter operator by least squares and subtracting the ghost wave after applying the matched filter operator. It can be seen that the ghost wave is better suppressed after applying the matched filter operator.

[0154] The ghost wave suppression method for three-dimensional seismic data provided in this embodiment is based on any of the above embodiments. Before subtracting the sparse ghost wave field from the sparse three-dimensional seismic wave field, a least squares constraint is used to determine a matching operator between the sparse three-dimensional seismic wave field and the sparse ghost wave field; the matching operator is applied to the sparse ghost wave field to obtain a matched sparse ghost wave field; and finally, the matched sparse ghost wave field is subtracted from the sparse three-dimensional seismic wave field, thereby further improving the ghost wave suppression effect.

[0155] In order to further illustrate the effect of the ghost wave suppression method for three-dimensional seismic data provided by the embodiment of the present invention, simulation data and actual data are used for illustration respectively.

[0156] Figure 7A The following is a schematic diagram of ghost wave suppression for fixed depth recording. First, theoretical streamer data with a fixed depth of 12.5 meters is generated, as shown in the following figure: Figure 7A As shown in Figure (a), the method provided by the present invention is used to suppress the ghost wave at the receiving point, and the following is obtained: Figure 7A As shown in Figure (b); Figure 7A Figure (c) shows the suppressed ghost waves. It can be seen that the method provided by the present invention has a good effect on suppressing ghost waves in the theoretical data of fixed water depth streamers.

[0157] Figure 7B This is a schematic diagram of ghost wave suppression during variable depth recording. Simulating a variable depth streamer, the depth is increased by 2.5 meters for every 1000 meters. The theoretical record is generated and ghost waves are suppressed using the method provided by the present invention. The data before and after ghost wave suppression are corrected to a water depth of 12.5 meters. Figure 7B As shown in the figure, the towing depth of tracks 41-80 is 12.5 meters, and the towing depth from track 81 onwards is 15 meters. Figure 7B Figure (a) in the middle is the original theoretical record; Figure 7B Figure (b) in the middle is the record after ghost waves are suppressed by the method provided by the present invention; Figure 7B Figure (c) shows the suppressed ghost waves. It can be seen that the method provided by the present invention also has a good effect on suppressing ghost waves in the variable depth streamer theoretical data.

[0158] Figure 8 Schematic diagram of the FK (frequency wave number) spectrum of the variable depth streamer theoretical data before and after suppressing ghost waves. Figure 8 (a) is the original variable depth recorded FK spectrum, Figure 8 (b) shows the FK spectrum recorded at varying depths after ghost suppression. Comparison shows that the method provided by the present invention effectively suppresses the ghost waves immediately following the primary wave, compensating for the zero-frequency and non-zero notches caused by ghost waves, and improving the recorded resolution.

[0159] Using a regional velocity model, a three-dimensional simulation record was obtained through wave equation simulation. The receiving line spacing was 100 meters, the track spacing was 12.5 meters, and the receiving point depth was 30 meters. Figure 9A It is a simulation record close to the receiving line, with a vertical offset of 25 meters from the receiving line. Figure 9A The middle (a) figure is the original simulation record. Figure 9A The middle (b) figure is the ghost wave field obtained based on the dense field reconstruction method. Figure 9A Figure (c) shows the simulated recording after ghost subtraction. It can be seen that the input data contains significant ghost waves immediately following the upgoing waves. The method provided by this invention removes the energy immediately following the upgoing waves, significantly improving the data resolution.

[0160] Figure 9B For the simulation record of the far receiving line, the vertical offset of the receiving line is 725 meters. Figure 9B The middle (a) figure is the original simulation record containing ghost waves. Figure 9B The middle (b) figure is the record after suppressing ghost waves based on the dense field reconstruction method. Figure 9B The middle (c) figure is a simulated ghost-free wave record. Figure 9B Taking the middle (c) figure as a reference, comparing the original record containing ghost waves and the results after ghost wave suppression, it can be seen that the ghost wave suppression based on dense field reconstruction has achieved a relatively ideal effect.

[0161] The above examples demonstrate the effects obtained by applying the method provided by the present invention to simulated data. The following will further demonstrate the effects that can be obtained by applying the method provided by the present invention to actual data.

[0162] Actual streamer data from a specific region was used to test the ghost suppression effectiveness of the method provided by the present invention. To verify the effectiveness of the dense field reconstruction-based ghost suppression method, surface-related multiple elimination (SRME) was used to suppress multiples associated with the sea surface before ghost suppression. It should be noted that SRME is introduced here solely to illustrate that conventional multiple attenuation techniques cannot be used to suppress extremely short-period ghosts. In practice, multiple attenuation is recommended after ghost suppression.

[0163] The data cable distance is 100 meters, the receiving point distance is 12.5 meters, the immersion depth is 12 meters, and the excitation depth is 8.5 meters. Figure 10 This is a comparison of the processing effects of a far-offset streamer in 3D data. Figure 10 Figure (a) in the middle is the input data. Figure 10 The middle (b) figure shows the data after SRME processing. Figure 10 Figure (c) shows the ghost suppression results using the method provided by the present invention, along with a zoomed-in view of the data within the box. It can be seen that SRME suppressed most of the long-range multiples associated with the sea surface, but had no significant effect on ghosts. Comparing the records before and after ghost suppression, it is clear that ghosts are effectively suppressed using the method provided by the present invention, with the previously "double-fold" seismic wave events reduced to a single reflection event.

[0164] Figure 11 This is a comparison of the actual data before and after ghost wave suppression in the near-offset gather. Figure 11 The middle (a) figure is the input record. Figure 11 The middle (b) picture is the record after suppressing the ghost wave. Figure 11 Figure (c) shows the difference between the two, representing the suppressed ghost waves. Clearly, there are ghost waves immediately following the upgoing waves in the input data. The method provided by this invention removes the energy immediately following the upgoing waves, significantly improving the resolution of the data.

[0165] Figure 12 yes Figure 11 FK spectra of two records before and after suppressing ghost waves. Figure 12 Figure (a) in the middle shows the FK analysis before ghost wave suppression. Figure 12Figure (b) shows the FK analysis after ghost suppression. In the spectrum before ghost suppression, a distinct notch band is visible, ranging from approximately 60Hz to 90Hz. After ghost removal, the frequency components of the notch band are well compensated.

[0166] Figure 13 A schematic diagram of superimposed cross sections before and after ghost wave suppression provided by one embodiment of the present invention. Figure 13 Figure (a) in the middle is the original stacked section and its spectrum analysis. Figure 13 The middle (b) figure shows the cross section and spectrum analysis after SRME. Figure 13 Figure (c) shows the cross-section and spectrum analysis after ghost suppression. A comparison clearly shows no significant attenuation of ghost waves before and after SRME, nor any significant change in the spectrum. However, after ghost suppression, the frequency band is significantly broadened, the cross-section resolution is improved, and the uniqueness of the event corresponding to the interface reflection is enhanced.

[0167] Figure 14 for Figure 13 Schematic diagram of the partial enlargement of the superimposed sections before and after the medium suppression of ghost waves. Figure 14 The middle (a) figure is the superposition section of the suppressed ghost wavefront. Figure 14 The middle (b) figure is the superposition section after suppressing the ghost wave. Figure 14 Figure (c) shows the difference between the cross-sections before and after ghost wave suppression. This comparison demonstrates that the method provided by this invention effectively suppresses ghost waves, significantly reduces their interference with the desired wave, and significantly improves cross-section resolution.

[0168] The method provided by the present invention reconstructs a dense wavefield through wavefield extrapolation, solves the dense ghost wave field on the water surface, and uses lasso regression to solve the sparse wavefield at the original location, subtracting ghost waves from the original data. This method simultaneously addresses both undersampling and ghost suppression issues in actual three-dimensional data. By providing the corresponding propagation operator, the integral at any receiver location can be accurately evaluated, thus addressing anti-ghosting issues at both the receiving and excitation points at varying depths. Both simulated and actual data experiments demonstrate the method's excellent effectiveness. Ghost suppression based on dense field reconstruction effectively compensates for notches caused by ghost waves, broadens the frequency band of seismic profiles, and improves the characteristics of seismic data.

[0169] Example 4

[0170] The present invention also provides a device for suppressing ghost waves in three-dimensional seismic data. Figure 15 shown. Figure 15 A schematic diagram of the structure of a ghost wave suppression device for three-dimensional seismic data provided by an embodiment of the present invention. Figure 15As shown, the ghost wave suppression device 40 for three-dimensional seismic data provided in this embodiment may include: an acquisition module 401, a first dense field reconstruction module 402, a second dense field reconstruction module 403, a first processing module 404, a second processing module 405 and a suppression module 406.

[0171] An acquisition module 401 is used to acquire a sparse three-dimensional seismic wave field recorded in offshore seismic acquisition;

[0172] A first dense field reconstruction module 402 is configured to densely extrapolate a first distance of the sparse 3D seismic wavefield to the backward wavefield to obtain a first dense 3D seismic wavefield, where the first distance is the distance between the detection point position and the water surface position;

[0173] A second dense field reconstruction module 403 is configured to densely extrapolate a second distance from the sparse 3D seismic wavefield to the forward wavefield and perform the reverse operation to obtain a second dense 3D seismic wavefield, where the second distance is the distance between the mirror detection point and the water surface.

[0174] A first processing module 404 is configured to determine a dense ghost wave field at a water surface location based on the first dense 3D seismic wave field and the second dense 3D seismic wave field;

[0175] The second processing module 405 is configured to extrapolate the dense ghost wave field to the forward wave field by a second distance and perform sparse sampling to obtain a sparse ghost wave field at the mirror detection point.

[0176] The suppression module 406 is configured to subtract the sparse ghost wavefield from the sparse 3D seismic wavefield.

[0177] The device of this embodiment can be used to perform Figure 3 The technical solution of the method embodiment shown has similar implementation principles and technical effects, which will not be repeated here.

[0178] In an optional implementation, the sparse 3D seismic wave field is determined according to the following expression:

[0179] ;

[0180] in, Indicates the water surface position, Indicates the detection point position, Indicates the mirror detection point position, represents the sparse three-dimensional seismic wave field recorded in seismic acquisition, Represents a wave field, which is composed of a water surface wave field Extrapolating backward, we get Represents the ghost wave field, which is composed of the water surface wave field Extrapolating forward, we find that the polarity of the ghost wave field is opposite to that of the primary wave field.

[0181] In an optional implementation, the first dense 3D seismic wavefield is determined according to the following expression:

[0182] ;

[0183] The second dense 3D seismic wave field is determined according to the following expression:

[0184] ;

[0185] in, represents the first dense three-dimensional seismic wave field, represents the backward extrapolated wavefield propagation matrix, represents the forward extrapolated wavefield propagation matrix, represents a densely sampled matrix.

[0186] In an optional implementation, the sampling density of the dense sampling matrix is ​​determined according to the following expression:

[0187] ;

[0188] in, Spatial sampling interval, represents the speed of longitudinal waves in water, Indicates the maximum frequency of the expected output.

[0189] In an optional implementation, the first processing module 404 is specifically configured to:

[0190] summing the first dense three-dimensional seismic wavefield and the second dense three-dimensional seismic wavefield;

[0191] The summation result is threshold-truncated to retain the wave field with the strongest energy, and the dense ghost wave field at the water surface is obtained.

[0192] In an optional implementation, the sparse ghost wave field at the mirror detection point location is determined according to the following expression:

[0193] ;

[0194] in, represents the sparse ghost wave field at the mirror detection point location, represents the dense ghost wave field at the water surface, Represents the acquisition observation matrix.

[0195] In an optional implementation, during the solution process, lasso regression is used to minimize the following regularized loss function:

[0196] ;

[0197] in, represents the regularization parameter, The time difference of the initial constraint is , is the radius of the sparsely sampled bin.

[0198] In an optional embodiment, the ghost wave suppression device 40 for three-dimensional seismic data may further include a matching module (not shown in the figure) for determining a matching operator between the sparse three-dimensional seismic wavefield and the sparse ghost wavefield using a least squares constraint before subtracting the sparse ghost wavefield from the sparse three-dimensional seismic wavefield; applying the matching operator to the sparse ghost wavefield to obtain a matched sparse ghost wavefield; and subtracting the matched sparse ghost wavefield from the sparse three-dimensional seismic wavefield.

[0199] In an optional implementation, the matching operator is determined according to the following expression:

[0200] ;

[0201] in, represents the sparse three-dimensional seismic wave field, represents a sparse ghost wave field, Represents a matching operator.

[0202] In summary, the present invention solves the undersampling problem of offshore three-dimensional seismic data by reconstructing a dense seismic wave field; solves the dense ghost wave field at the water surface through the echo separation method, thereby realizing the wave field separation of the dense field; solves the sparse ghost wave field at the original mirror detection point position through sparse wave field extrapolation under the lasso regression constraint; and realizes ghost wave suppression of the sparse wave field through ghost wave subtraction under the least squares constraint.

[0203] Example 5

[0204] The present invention also provides an electronic device. Figure 16 As shown, the embodiment of the present invention is only Figure 16 This is just an example for explanation, and it does not mean that the present invention is limited to this. Figure 16 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 16 As shown, the electronic device 50 provided in this embodiment includes: a memory 501, a processor 502 and a bus 503. The bus 503 is used to realize the connection between various components.

[0205] The memory 501 stores a computer program, which can implement the technical solution of any of the above method embodiments when executed by the processor 502.

[0206] Memory 501 and processor 502 are directly or indirectly electrically connected to each other to enable data transmission or interaction. For example, these components may be electrically connected via one or more communication buses or signal lines, such as bus 503. Memory 501 stores a computer program for implementing a method for suppressing ghost waves in 3D seismic data, including at least one software functional module that may be stored in memory 501 in the form of software or firmware. Processor 502 executes various functional applications and data processing by running the software program and modules stored in memory 501.

[0207] The memory 501 may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The memory 501 is used to store programs, and the processor 502 executes the programs after receiving execution instructions. Furthermore, the software programs and modules in the memory 501 may also include an operating system, which may include various software components and / or drivers for managing system tasks (e.g., memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components.

[0208] The processor 502 can be an integrated circuit chip with signal processing capabilities. The processor 502 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. It can implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of the present invention. The general-purpose processor can be a microprocessor or any conventional processor. It is understood that Figure 16 The structure is only for illustration and may also include Figure 16 More or fewer components than shown, or with Figure 16 Different configurations shown. Figure 16 Each component shown in the figure may be implemented using hardware and / or software.

[0209] It should be noted that the electronic devices provided in this embodiment include, but are not limited to, at least one of the following: user-side devices and network-side devices. User-side devices include, but are not limited to, computers, smartphones, tablets, etc. Network-side devices include, but are not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing that consists of a group of loosely coupled computers forming a super virtual computer.

[0210] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the technical solution of any of the above method embodiments.

[0211] The various embodiments in the present disclosure are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0212] The scope of protection of the present disclosure is not limited to the above-described embodiments. Obviously, those skilled in the art may make various modifications and variations to the present disclosure without departing from the scope and spirit of the present disclosure. If such modifications and variations fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is intended to include such modifications and variations.

Claims

1. A method for suppressing ghost waves in three-dimensional seismic data, characterized in that: include: Acquire sparse 3D seismic wavefields recorded in offshore seismic acquisition; Densely extrapolating a first distance from the sparse three-dimensional seismic wave field to the back wave field to obtain a first dense three-dimensional seismic wave field, where the first distance is the distance between the detection point position and the water surface position; Densely extrapolating the sparse three-dimensional seismic wavefield to the forward wavefield by a second distance and performing a reverse extrapolation to obtain a second dense three-dimensional seismic wavefield, wherein the second distance is the distance between the mirror detection point position and the water surface position; determining a dense ghost wave field at a water surface location based on the first dense 3D seismic wave field and the second dense 3D seismic wave field; Extrapolating the dense ghost wave field to the forward wave field by a second distance and performing sparse sampling to obtain a sparse ghost wave field at the mirror detection point position; The sparse ghost wavefield is subtracted from the sparse three-dimensional seismic wavefield.

2. The method according to claim 1, characterized in that The sparse three-dimensional seismic wave field is determined according to the following expression: ； in, Indicates the water surface position, Indicates the detection point position, Indicates the mirror detection point position, represents the sparse three-dimensional seismic wave field recorded in seismic acquisition, represents the primary wave field, which is composed of the water surface wave field Extrapolating backward, we get represents the ghost wave field, which is composed of the water surface wave field Extrapolating forward, it is found that the polarity of the ghost wave field is opposite to the polarity of the primary wave field.

3. The method according to claim 2, characterized in that The first dense three-dimensional seismic wave field is determined according to the following expression: ； The second dense three-dimensional seismic wave field is determined according to the following expression: ； in, represents the first dense three-dimensional seismic wave field, represents the backward extrapolated wavefield propagation matrix, represents the forward extrapolated wavefield propagation matrix, represents a densely sampled matrix.

4. The method according to claim 3, characterized in that The sampling density of the dense sampling matrix is ​​determined according to the following expression: ； in, Spatial sampling interval, represents the speed of longitudinal waves in water, Indicates the maximum frequency of the expected output.

5. The method according to claim 4, characterized in that Determining a dense ghost wave field at a water surface position according to the first dense 3D seismic wave field and the second dense 3D seismic wave field comprises: summing the first dense three-dimensional seismic wavefield and the second dense three-dimensional seismic wavefield; The summation result is threshold-truncated to retain the wave field with the strongest energy, and the dense ghost wave field at the water surface is obtained.

6. The method according to claim 5, characterized in that The sparse ghost wave field at the mirror detection point is determined according to the following expression: ； in, represents the sparse ghost wave field at the mirror detection point location, represents the dense ghost wave field at the water surface, Represents the acquisition observation matrix.

7. The method according to claim 6, characterized in that In the solution process, lasso regression is used to minimize the following regularized loss function: ； in, represents the regularization parameter, The time difference of the initial constraint is , is the radius of the sparsely sampled bin.

8. The method according to any one of claims 1 to 7, characterized in that Before subtracting the sparse ghost wavefield from the sparse 3D seismic wavefield, the method further includes: Determining a matching operator between the sparse three-dimensional seismic wavefield and the sparse ghost wavefield using a least squares constraint; Applying the matching operator to the sparse ghost wavefield to obtain a matched sparse ghost wavefield; Subtracting the sparse ghost wavefield from the sparse three-dimensional seismic wavefield includes: subtracting the matched sparse ghost wavefield from the sparse three-dimensional seismic wavefield.

9. The method according to claim 8, characterized in that The matching operator is determined according to the following expression: ； in, represents the sparse three-dimensional seismic wave field, represents the sparse ghost wave field, Represents the matching operator.

10. A ghost wave suppression device for three-dimensional seismic data, characterized in that: include: An acquisition module for acquiring sparse three-dimensional seismic wave fields recorded in offshore seismic acquisition; a first dense field reconstruction module, configured to densely extrapolate a first distance of the sparse three-dimensional seismic wavefield to a backward wavefield to obtain a first dense three-dimensional seismic wavefield, wherein the first distance is the distance between a detection point position and a water surface position; a second dense field reconstruction module, configured to densely extrapolate a second distance from the sparse three-dimensional seismic wavefield to the forward wavefield and perform reverse extrapolation to obtain a second dense three-dimensional seismic wavefield, wherein the second distance is the distance between the mirror detection point position and the water surface position; a first processing module, configured to determine a dense ghost wave field at a water surface position based on the first dense 3D seismic wave field and the second dense 3D seismic wave field; A second processing module is configured to extrapolate the dense ghost wave field to the forward wave field by a second distance and perform sparse sampling to obtain a sparse ghost wave field at a mirror detection point position; A suppression module is configured to subtract the sparse ghost wavefield from the sparse 3D seismic wavefield.

11. An electronic device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the ghost wave suppression method for three-dimensional seismic data according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the ghost wave suppression method for three-dimensional seismic data according to any one of claims 1 to 9.

Images