A method and device for phase consistency correction of wide-azimuth seismic data in the imaging domain

Abstract

The application relates to a wide-azimuth seismic data imaging domain phase consistency correction method, and relates to the field of seismic exploration. The method solves the problem of phase inconsistency caused by factors such as surface unevenness and different azimuth incidence angles in the existing seismic imaging process. The method comprises the following steps: S1: obtaining wide-azimuth prestack three-dimensional seismic data, and pre-processing the prestack three-dimensional seismic data; S2: performing radial domain multi-azimuth data subset efficient division on the pre-processed prestack three-dimensional seismic data, and obtaining a radial domain data subset; S3: based on the pre-processed prestack three-dimensional seismic data, constructing a reference gather through migration stacking; S4: based on the radial domain data subset, constructing a radial domain imaging gather; S5: based on the constructed reference gather and the radial domain imaging gather, obtaining a phase difference; and performing imaging domain phase consistency correction on the radial domain data subset. The method can effectively improve the phase consistency of seismic data, improve the resolution of seismic data and the imaging quality of underground structures.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of seismic exploration, in particular to a wide-azimuth seismic data imaging domain phase consistency correction method and device. BACKGROUND

[0002] Phase consistency plays a crucial role in shale reservoir description. The phase consistency of seismic data is an indispensable basis for reservoir description and oil and gas exploration and development, and directly affects the accurate understanding and understanding of the structure and properties of underground reservoirs, and plays a crucial role in reservoir description. There are usually the following five aspects of use, specifically including: first, improving the quality of three-dimensional seismic imaging, seismic phase consistency correction can eliminate the phase difference caused by factors such as surface unevenness, underground velocity variation, instrument response difference, etc. Make the phase axis of the seismic reflection wave clearer, thereby improving the resolution and accuracy of seismic imaging. Second, accurately identifying geological interfaces, through phase consistency correction, underground geological interfaces can be better identified, especially those thin layers, complex structure reservoir boundaries, which are directly helpful for reservoir boundary delineation, thickness estimation, etc. Third, enhancing the internal structure characteristics of the reservoir, the cracks, faults and other subtle structure characteristics in the reservoir may be difficult to identify when the seismic phase is inconsistent. Phase consistency correction helps to reveal these features and provides detailed description of the internal crack system distribution, porosity variation, etc. Fourth, improving the accuracy of seismic attribute analysis, the phase consistent corrected seismic data is more suitable for seismic attribute analysis, such as amplitude, frequency, phase, etc. The attributes can more accurately reflect the physical properties of the reservoir, such as fluid saturation, pore pressure, etc. Fifth, improving the reliability of the geological model, more accurate seismic data makes the construction of the geological model more reliable, and the phase consistent seismic data can better constrain the geological modeling parameters, improve the accuracy and reliability of the geological model, which has extremely important guiding significance for reservoir evaluation and oil and gas resource exploration and development.

[0003] In 3D seismic data processing, commonly used phase consistency correction methods mainly include the following: First, minimum phase entropy correction, which calculates and minimizes the phase entropy of seismic traces to achieve phase consistency. This method can correct phase differences in seismic traces to a certain extent, but it is sensitive to noise; excessive noise may affect the correction effect. Second, phase correction based on reflection coefficients, which uses the real and imaginary parts of the reflection coefficient (the real part represents amplitude, and the imaginary part represents phase) for correction. This method requires a reliable reflection coefficient model, and its accuracy may be limited in practical applications. Third, phase axis tracking correction, which achieves phase correction by tracking the phase axis and forcing all seismic traces to be phase-consistent at the same point. However, this method is difficult to guarantee in processing complex geological structures and areas with strong velocity changes because it does not process in the imaging domain. Fourth, deconvolution phase correction, which applies surface-consistent phase deconvolution technology to achieve phase consistency by performing phase deconvolution processing on seismic data. However, the deconvolution process itself is a nonlinear problem, highly dependent on the initial model, and may suffer from error propagation due to model uncertainties. Fifth, machine learning and artificial intelligence-assisted correction. Recent research has attempted to use technologies such as deep learning for phase consistency correction. Although progress has been made in some cases, it still faces problems such as large data sample requirements and relatively poor model interpretability.

[0004] The aforementioned methods are subject to uncertainty when applied to actual seismic data for phase consistency processing due to their sensitivity to noise, dependence on models, difficulty in handling complex geological structures, high computational complexity, and the influence of subjective factors such as human intervention and experience-based judgment. Summary of the Invention

[0005] This invention addresses the phase inconsistency problem in existing seismic imaging methods caused by surface inhomogeneities and different incident angles in the background art, by providing a wide-azimuth seismic data imaging domain phase consistency correction method. This method effectively improves the phase consistency of seismic data, enhances its resolution, and improves the imaging quality of subsurface structures, making it particularly suitable for detailed analysis of shale reservoirs and other complex geological structures. This invention also provides a wide-azimuth seismic data imaging domain phase consistency correction device.

[0006] The present invention solves its problem through the following technical solution: the wide-azimuth seismic data imaging domain phase consistency correction method includes the following steps:

[0007] S1: Acquire wide-azimuth pre-stack 3D seismic data and preprocess the pre-stack 3D seismic data;

[0008] S2: Efficiently partition the radial domain multi-azimuth data subsets of the preprocessed pre-stack 3D seismic data to obtain radial domain data subsets;

[0009] S3: Based on preprocessed pre-stack 3D seismic data, reference gathers are constructed through migration stacking;

[0010] S4: Based on the radial domain data subset obtained in step S2, construct the radial domain imaging gather;

[0011] S5: Calculate the phase difference based on the constructed reference gather and radial domain imaging gather; perform imaging domain phase consistency correction on the radial domain data subset.

[0012] Furthermore, step S1 preprocesses the pre-stack 3D seismic data, including:

[0013] S11. Complete the static correction process to ensure that the differences in travel time caused by uneven terrain are fully corrected;

[0014] S12, Multichannel gain balancing, is used to adjust the energy differences between channels to make the data energy distribution more balanced;

[0015] S13. Apply noise suppression technology to improve the signal-to-noise ratio.

[0016] S14. Tomographic imaging and velocity analysis: Obtain underground velocity models through tomographic imaging technology and perform fine processing on the velocity field.

[0017] Furthermore, the noise suppression techniques applied include adaptive denoising, Wiener filtering, KL transform, and sparse representation.

[0018] Furthermore, the method for efficiently partitioning the radial domain multi-directional data subset in step S2 includes:

[0019] S21. First, the preprocessed seismic data is standardized and organized to ensure that each seismic data includes the corresponding offset, azimuth, geographic coordinates and time information.

[0020] S22. Create an efficient indexing system that quickly divides the data into multiple azimuth intervals based on the two parameters of azimuth angle and offset distance. Each radial interval is given a fixed number to represent a radial domain subset.

[0021] S23. Extract a subset of radial domain data.

[0022] Furthermore, the method for extracting the radial domain data subset in step S23 is as follows:

[0023] Distributed processing is achieved using parallel computing frameworks such as Apache Spark or Hadoop MapReduce, which distribute the task of building subsets across multiple computing nodes and process different radial domain data subsets simultaneously. Then, efficient query access to the dataset is built by storing the data in a database that supports SQL queries, filtering the data based on the radial domain number, and extracting subsets in one go or in batches.

[0024] Furthermore, for frequently used data subsets, a caching strategy can be adopted to pre-read the subsets into memory or fast storage media, reducing the performance loss caused by repeated hard drive reads;

[0025] Based on actual application needs, prioritize processing data from important areas or well site test areas, and rationally arrange the priority and order of data processing and subset construction.

[0026] Furthermore, step S3, based on preprocessed pre-stack 3D seismic data, constructs a reference gather through migration stacking. The specific method includes:

[0027] Common offset data is extracted from the preprocessed data, and volume offset is performed on the extracted common offset data to obtain the common reflection point imaging gather;

[0028] Then, the common reflection point imaging gather is subjected to noise reduction and optimization processing to improve the signal-to-noise ratio of the common reflection point imaging gather;

[0029] The reference gather can be obtained by superimposing the denoised common reflection point imaging gathers.

[0030] Furthermore, the expression for the three-dimensional case of volume migration after extracting the common offset data is as follows:

[0031]

[0032] In the formula: f i ′(τ s +τ g ) is the first derivative of the pre-stack data of the i-th trace, n is the total number of seismic traces in different azimuth subsets, and τ s and τ g These are the travel times from the shot point and receiver point to the imaging point (x, y, T), respectively. s / τ g ) 2 These are the weighting coefficients for compensating for the spherical spread of seismic waves, (x m ,y m ) and h are the center point and half offset of the seismic trace, respectively; T0(x,y,x) m ,y mI(x,y,T,h) represents the starting imaging time for different imaging points in the seismic trace at that azimuth, and it represents the offset aperture. At each horizontal position (x,y) in the imaging profile, I(x,y,T,h) represents the common reflection point gather at that azimuth as the offset distance changes, where T is the reflection time and T0 is the start time of the time window.

[0033] Furthermore, the radial domain imaging gather described in step S4 includes the radial domain shot gather and the radial domain receiver gather;

[0034] Offset processing is performed on a subset of radial domain multi-azimuth data to construct a radial domain shot point gather;

[0035] Offset processing is performed on a subset of radial domain multi-azimuth data to construct a radial domain detector point gather.

[0036] Furthermore, based on the radial domain data subset obtained in step S2, a radial domain shot point gather is constructed. The specific methods include:

[0037] Extract the offset trace sets of all shot points in the radial domain subset, and superimpose the offset trace sets of the same shot points in the radial domain subset to obtain a new offset trace set composed of shot points, i.e., the radial domain shot point trace set.

[0038] Furthermore, based on the radial domain data subset obtained in step S2, a radial domain detector point gather is constructed. The specific method includes:

[0039] The offset gathers of all receiver points in the radial domain subset offset results are extracted. The offset gathers of the radial domain data subsets with the same receiver points are superimposed to obtain a new offset gather composed of receiver points, namely the radial domain receiver point gather.

[0040] Furthermore, when performing offset processing on the radial domain multi-directional data subset according to Formula 1, the region with a clear, continuous or partially continuous sediment isochronous interface with the same phase axis is selected as the specified time window.

[0041] Furthermore, the radial domain shot point gather expression is Equation 2:

[0042]

[0043] The radial domain detector point gather expression is Equation 3:

[0044]

[0045] In the formula: (x s ,y s ) and (x g ,y g ) are the coordinates of the shot point and receiver point of the seismic trace, respectively; T0 is the start time of the time window.

[0046] Furthermore, step S5 calculates the phase difference based on the constructed reference gather and radial domain imaging gather; and performs imaging domain phase consistency correction on the radial domain data subset, specifically including:

[0047] 5.1. To obtain phase consistency, a parameter called constant phase rotation is introduced to construct the cross-correlation function, which can be expressed as:

[0048] φ ik (α,τ)=cosα·R(τ)-sinα·r(τ)(4)

[0049] Where, i represents the detector point, k represents the horizontal position of the image (CDP point), and φ ik (α,τ) is the cross-correlation function of the i-th detector in the detector offset channel set corresponding to the k-th CDP point, where α is the constant phase rotation angle;

[0050]

[0051] Where t represents time, [T1,T2] is the time window at that horizontal position, R(τ) is the real part of the cross-correlation function, r(τ) is the imaginary part of the cross-correlation function, τ is the time increment, and g i (t) represents the offset channel selected from the receiver offset channel set at the horizontal position k, and g i (t+τ) represents the offset channel selected from the receiver offset channel set at horizontal position k when the time increment is τ, i.e., at time t+τ; s k (t+τ) is the reference track at horizontal position K, Hg i (t) is g i Hilbert transform of (t);

[0052] The function φ ik The maximum value of (α,τ) is the required phase difference (Δt,θ), that is, the phase difference at Δt is θ; by the principle of extrema, we can obtain:

[0053]

[0054] It is the cross-correlation function φ ik Partial derivatives of (α,τ);

[0055] By solving the above equation, we can obtain:

[0056]

[0057] The phase difference (Δt, θ) can be obtained from equations (7) and (8);

[0058] Similarly, the offset trajectory g of the shot point offset gather is used. j (t) and imaging channel sk (t+τ), similarly, the remaining time difference of the shot point can be calculated; where j represents the shot point;

[0059] 5.2. Using the phase difference obtained in step 5.1, perform phase correction on the radial domain data subsets to obtain the phase-corrected radial domain data subsets.

[0060] The present invention also provides a phase consistency correction device for wide-azimuth seismic data imaging domain, comprising:

[0061] The preprocessing unit is used to acquire wide-azimuth pre-stack 3D seismic data and perform preprocessing on the pre-stack 3D seismic data.

[0062] The acquisition unit is used to efficiently partition the radial domain multi-azimuth data subset of the preprocessed pre-stack 3D seismic data and acquire the radial domain data subset.

[0063] The first building unit is used to construct reference gathers by migration stacking based on preprocessed pre-stack 3D seismic data.

[0064] The second response construction unit is used to construct a radial imaging gather based on the radial domain data subset obtained in step S2.

[0065] The correction unit is used to perform imaging domain phase consistency correction on a subset of radial domain data based on the constructed reference gather and radial imaging gather.

[0066] Compared with the above-mentioned background technology, the present invention has the following beneficial effects:

[0067] This invention provides a phase consistency correction method for wide-azimuth seismic data imaging domain. Its main purpose is to solve the phase inconsistency problem caused by surface inhomogeneities and different incident angles during seismic imaging, thereby improving the resolution of seismic data and the imaging quality of subsurface structures. It is particularly suitable for the detailed analysis of shale reservoirs and other complex geological structures. After correction using this method, the horizontal phase consistency of the target layer is significantly improved, enhancing the lateral in-phase nature of the seismic signal within the layer. Applying this method, after phase consistency correction in the imaging domain, details of small faults that might have been obscured by noise become clearer, and the focusing effect on the superimposed profile is better. This provides strong support for identifying and characterizing small-scale geological structures such as fractures and small faults. The advantage of this method is that it not only improves the quality of seismic imaging but also helps to deeply analyze the azimuth anisotropy characteristics of geological structures based on wide-azimuth seismic data, which has important reference value for the exploration and development of unconventional oil and gas resources such as shale gas reservoirs. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0069] Figure 2 This is a flowchart of the imaging domain phase difference calculation and correction process of the present invention;

[0070] Figure 3 This is a comparison diagram of the gathers before and after phase difference correction in an embodiment of the present invention (a: gathers before phase correction; b: gathers after phase correction);

[0071] Figure 4 The above is a superimposed comparison of phase difference correction before and after in an embodiment of the present invention (a: cross-section before phase correction; b: cross-section after phase correction). Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0073] This paper provides a detailed explanation of the phase consistency correction method for wide-azimuth seismic data imaging domain, based on actual 3D seismic data from an oilfield.

[0074] Figure 1 This is a flowchart of the overall workflow. Figure 2 The following is a flowchart of the imaging domain phase consistency correction process. Figure 2 Detailed steps:

[0075] 1. The left half of the reference model trace is constructed by first performing common-offset group migration on the preprocessed 3D seismic data. This process is based on seismic wave propagation theory, using mathematical methods to simulate the actual path of seismic waves after reflection from underground to the surface, thus obtaining a more accurate image of the stratigraphic structure. After migration, the data undergoes further stacking and denoising processing. This step typically includes random noise removal and multiple suppression, aiming to enhance the effective signal, reduce interference, and form a high-quality seismic profile. This processed data will serve as the basis for subsequent comparison and correction, i.e., the "model trace".

[0076] 2. The right half of the branch constructs the imaging domain shot gather and receiver gather data. This branch processes imaging domain seismic data that requires phase correction. First, this data is converted into common shot gathers and common receiver gathers in the imaging domain. These two sets are then sorted according to the excitation point (shot point) and receiver point (receiver point), respectively, to facilitate the analysis of seismic wave propagation differences between points.

[0077] 3. During phase correction, the imaging domain gathers formed by the right half-branch are cross-correlated with the high-quality model gathers constructed by the left half-branch. The cross-correlation function accurately reflects the phase difference between the two gathers. The phase difference is obtained by cross-correlating each common shot point or common receiver point with the model gathers. The obtained phase difference is then appropriately averaged to reduce the impact of random errors and ensure the stability and accuracy of phase correction. Finally, the obtained phase difference is applied to the imaging domain data to perform phase consistency correction, ensuring that the seismic waveforms of the entire 3D seismic dataset are consistent in space and time, thus improving imaging quality and the reliability of seismic interpretation.

[0078] This invention discloses a method for phase consistency correction in the imaging domain of wide-azimuth seismic data, comprising the following steps:

[0079] S1: Acquire wide-azimuth pre-stack 3D seismic data and perform preprocessing on the pre-stack 3D seismic data; including the following processing aspects:

[0080] S11. Complete the static correction process to ensure that the differences in travel time caused by uneven terrain are fully corrected;

[0081] S12, Multichannel gain balancing, is used to adjust the energy differences between channels to make the data energy distribution more balanced;

[0082] S13. Apply noise suppression techniques to improve the signal-to-noise ratio; the applied noise suppression techniques include adaptive denoising, Wiener filtering, KL transform, and sparse representation, etc.

[0083] S14. Tomographic imaging and velocity analysis: Obtain underground velocity models through tomographic imaging technology and perform fine processing on the velocity field.

[0084] Through the above series of meticulous processing steps, wide-azimuth seismic data can accurately reflect the characteristics of underground geological structures, laying a high-quality data foundation for subsequent imaging domain phase consistency correction processing.

[0085] S2: Efficiently partition the radial domain multi-azimuth data subsets of the preprocessed pre-stack 3D seismic data to obtain radial domain data subsets;

[0086] Efficient Radial Domain Multi-Azimuth Data Subset Partition Method: Addressing the characteristics of large volume, wide azimuth, and long offset in wide-azimuth seismic acquisition data, this invention employs a radial domain data sorting strategy to achieve phase consistency processing for each sub-azimuth.

[0087] First, the preprocessed seismic data from the aforementioned steps are standardized and organized to ensure that each seismic data set includes the corresponding offset, azimuth, and other necessary geographic coordinates and time information.

[0088] Create an efficient indexing system that quickly divides data into multiple azimuth intervals based on two parameters: azimuth angle and offset distance. Each radial interval is assigned a fixed number to represent a subset of the radial domain.

[0089] Extracting radial domain data subsets: To improve efficiency, a parallel processing strategy is adopted, using a parallel computing framework (Apache Spark or Hadoop MapReduce) for distributed processing, distributing the task of building subsets across multiple computing nodes, and processing different radial domain data subsets simultaneously; then, an efficient query access mechanism for the dataset is built, storing the data in a database that supports SQL queries, filtering the data based on the radial domain number condition, and extracting subsets in one go or in batches.

[0090] For frequently used data subsets, a caching strategy can be employed, pre-loading the subsets into memory or high-speed storage media to reduce the performance overhead caused by repeated disk reads. Based on actual application requirements, data from important areas or well-site test areas should be processed first, and the priority and order of data processing and subset construction should be rationally arranged. Using these strategies, radial domain data subsets can be efficiently constructed from massive seismic data, improving data processing speed and resource utilization.

[0091] S3: Based on preprocessed pre-stack 3D seismic data, reference gathers are constructed through migration stacking;

[0092] Extract common offset data from the preprocessed data, and perform volume offset on the extracted common offset data (Equation 1) to obtain the common reflection point imaging gather;

[0093] Then, the common reflection point imaging gather is subjected to noise reduction and optimization processing to improve the signal-to-noise ratio of the common reflection point imaging gather;

[0094] The reference gather can be obtained by superimposing the denoised common reflection point imaging gathers.

[0095] The expression for volume migration in the three-dimensional case of extracting common offset data is:

[0096]

[0097] In the formula: f i ′(τ s +τ g ) is the first derivative of the pre-stack data of the i-th trace, n is the total number of seismic traces in different azimuth subsets, and τ s and τ g These are the travel times from the shot point and receiver point to the imaging point (x, y, T), respectively. s / τ g ) 2 These are the weighting coefficients for compensating for the spherical spread of seismic waves, (x m ,ym ) and h are the center point and half offset of the seismic trace, respectively; T0(x,y,x) m ,y m I(x,y,T,h) represents the starting imaging time for different imaging points in the seismic trace at that azimuth, and it represents the offset aperture. At each horizontal position (x,y) in the imaging profile, I(x,y,T,h) represents the common reflection point gather that varies with the offset distance at that azimuth, where T is the reflection time and T0 is the start time of the time window. S4: Based on the radial domain data subset obtained in step S2, construct the radial domain imaging gather.

[0098] The radial domain imaging gather includes the radial domain shot gather and the radial domain receiver gather.

[0099] In the common reflection point set of Equation (1), the offset traces with different shot points and receiver points have been superimposed. Therefore, the azimuth anisotropic time difference of different shot points and receiver points has been mixed in this set and is difficult to distinguish. In order to calculate the azimuth anisotropic time difference of a single shot point and receiver point, a new offset trace set composed of shot points and receiver points must be formed.

[0100] Misapping a subset of radial domain multi-azimuth data to construct a radial domain shot point gather; misapping a subset of radial domain multi-azimuth data to construct a radial domain receiver point gather.

[0101] 4.1 Based on the radial domain data subset obtained in step S2, construct the radial domain shot point gather. The specific method includes:

[0102] Extract the offset trace set of all shot points in the radial domain subset, and superimpose the offset trace sets of the radial domain data subset with the same shot points to obtain a new offset trace set composed of shot points, namely the radial domain shot point trace set, which is expressed as Formula 2.

[0103]

[0104] In the formula: (x s ,y s ) represents the shot point coordinates of the seismic trace; T0 is the start time of the time window;

[0105] 4.2 Based on the radial domain data subset obtained in step S2, construct the radial domain detector point gather. The specific method includes:

[0106] Extract the offset gathers of all receiver points in the radial domain subset offset results, and superimpose the offset gathers of the radial domain data subsets with the same receiver points to obtain a new offset gather composed of receiver points, namely the radial domain receiver point gather, which is expressed as Formula 3.

[0107]

[0108] In the formula: (xg ,y g ) represents the coordinates of the receiver point of the seismic trace; T0 is the start time of the time window;

[0109] In equations (2) and (3), offset gathers with the same shot point or the same receiver point in the radial domain data subset will be superimposed. Theoretically, at each horizontal position (x, y) of the imaging profile, there are offset gathers for all shot points or receiver points in that radial domain subset. In reality, due to the effects of the offset aperture and stretch cut, there are only effective offset gathers for a portion of the shot points or receiver points at each horizontal position. In practical applications, when calculating gathers for picking up phase difference imaging, a smaller offset aperture can be selected (empirical parameter for Songliao Basin shale reservoirs: offset distance / depth < 2).

[0110] The offsets in equations (2) and (3) only require imaging of the target layer time window for phase correction. When offsetting a subset of radial domain multi-azimuth data according to equation 1, the region with a clear, continuous or partially continuous isochronous interface of sedimentation is generally selected as the specified time window.

[0111] Figure 4 A comparison of the results of an imaging gather before and after phase consistency correction is presented; (a) is the profile before phase correction; (b) is the profile after phase correction; it can be clearly seen that the in-phase axis is better satisfied with consistency after phase consistency correction.

[0112] Applying phase consistency processing in the imaging domain can eliminate the influence of phase differences, improve the quality of stacked profiles, and thus enhance the focusing imaging capability of 3D seismic imaging on the target layer. Simultaneously, it can also improve the picking accuracy of reflection phase axis amplitudes, obtaining more accurate amplitudes and serving AVO / AVA inversion.

[0113] S5: Calculate the phase difference based on the constructed reference gather and radial domain imaging gather; perform imaging domain phase consistency correction on the radial domain data subset; specific methods include:

[0114] 5.1. To obtain phase consistency, a parameter called constant phase rotation is introduced to construct the cross-correlation function, which can be expressed as:

[0115] φ ik (α,τ)=cosα·R(τ)-sinα·r(τ)(4)

[0116] In the formula, i represents the detector point, k represents the horizontal position of the image (CDP point), and φ ik (α,τ) is the cross-correlation function of the i-th detector in the detector offset channel set corresponding to the k-th CDP point (imaging horizontal position), where α is the constant phase rotation angle;

[0117] in,

[0118]

[0119] Where t represents time, [T1,T2] is the time window at that horizontal position, R(τ) is the real part of the cross-correlation function, r(τ) is the imaginary part of the cross-correlation function, τ is the time increment, and g i (t) represents the offset channel selected from the receiver offset channel set at the horizontal position k, and g i (t+τ) represents the offset channel selected from the receiver offset channel set at horizontal position k when the time increment is τ, i.e., at time t+τ; s k (t+τ) is the reference track at horizontal position K, Hg i (t) is g i Hilbert transform of (t);

[0120] The function φ ik The maximum value of (α,τ) is the required phase difference (Δt,θ) (the phase difference at Δt is θ); by the principle of extrema, we can obtain:

[0121]

[0122] It is the cross-correlation function φ ik Partial derivatives of (α,τ);

[0123] By solving the above equation, we can obtain:

[0124]

[0125] The phase difference (Δt, θ) can be obtained from equations (7) and (8);

[0126] Similarly, the offset trajectory g of the shot point offset gather is used. j (t) and imaging channel s k (t+τ), similarly, the remaining time difference of the shot point can be calculated, where j represents the shot point.

[0127] 5.2. Using the phase difference obtained in step 5.1, perform phase correction on the radial domain data subsets to obtain the phase-corrected radial domain data subsets.

[0128] Figure 3 The comparison of target layers before and after applying the imaging domain phase consistency correction method is shown in the figure. (a) Trace set before phase correction; (b) Trace set after phase correction. It can be intuitively seen that the consistency of the target layer phase in the horizontal direction is significantly improved after correction, which enhances the lateral in-phase of the seismic signal within the layer. Figure 4By comparing the superimposed profiles before and after correction, it is shown that after phase consistency correction in the imaging domain, small fault details that might have been obscured by noise become clearer, and the focusing effect on the superimposed profiles is better. This provides strong support for identifying and characterizing small-scale geological structures such as fractures and small faults. The advantage of this method is that it not only improves the quality of seismic imaging but also helps to analyze the azimuth anisotropy characteristics of geological structures in depth based on wide-azimuth seismic data. This has important reference value for the exploration and development of unconventional oil and gas resources such as shale gas reservoirs.

[0129] The present invention also provides a phase consistency correction device for wide-azimuth seismic data imaging domain, the wide-azimuth seismic data imaging domain phase consistency correction device comprising:

[0130] The preprocessing unit is used to acquire wide-azimuth pre-stack 3D seismic data and perform preprocessing on the pre-stack 3D seismic data; the acquisition unit is used to efficiently divide the preprocessed pre-stack 3D seismic data into radial domain multi-azimuth data subsets and acquire radial domain data subsets; the first construction unit is used to construct a reference gather based on the preprocessed pre-stack 3D seismic data through migration stacking; the second pass response construction unit is used to construct a radial imaging gather based on the radial domain data subsets obtained in step S2; the correction unit is used to perform consistency correction on the imaging domain phase based on the constructed reference gather and radial imaging gather.

[0131] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the method described in the above embodiments of the wide-azimuth seismic data imaging domain phase consistency correction method. The specific implementation can be referred to the above description of the wide-azimuth seismic data imaging domain phase consistency correction method, which will not be repeated here for the sake of brevity.

[0132] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the implementation methods of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the present invention.

Claims

1. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data, characterized in that: Includes the following steps: S1: Acquire wide-azimuth pre-stack 3D seismic data and preprocess the pre-stack 3D seismic data; S2: Efficiently partition the radial domain multi-azimuth data subsets of the preprocessed pre-stack 3D seismic data to obtain radial domain data subsets; S3: Based on preprocessed pre-stack 3D seismic data, reference gathers are constructed through migration stacking; S4: Based on the radial domain data subset obtained in step S2, construct the radial domain imaging gather; S5: Calculate the phase difference based on the constructed reference gather and radial domain imaging gather; perform imaging domain phase consistency correction on the radial domain data subset.

2. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 1, characterized in that: Step S1 involves preprocessing the pre-stack 3D seismic data, including: S11. Complete the static correction process to ensure that the differences in travel time caused by uneven terrain are fully corrected; S12, Multichannel gain balancing, is used to adjust the energy differences between channels to make the data energy distribution more balanced; S13. Apply noise suppression technology to improve the signal-to-noise ratio. S14. Tomographic imaging and velocity analysis: Obtain underground velocity models through tomographic imaging technology and perform fine processing on the velocity field.

3. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 2, characterized in that: The noise suppression techniques used include adaptive denoising, Wiener filtering, KL transform, and sparse representation.

4. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 1, characterized in that: The method for efficient partitioning of radial domain multi-directional data subsets in step S2 includes: S21. First, the preprocessed seismic data is standardized and organized to ensure that each seismic data includes the corresponding offset, azimuth, geographic coordinates and time information. S22. Create an efficient indexing system that quickly divides the data into multiple azimuth intervals based on the two parameters of azimuth angle and offset distance. Each radial interval is given a fixed number to represent a radial domain subset. S23. Extract a subset of radial domain data.

5. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 4, characterized in that: The method for extracting the radial domain data subset in step S23 is as follows: Distributed processing is achieved using parallel computing frameworks such as Apache Spark or Hadoop MapReduce, which distribute the task of building subsets across multiple computing nodes and process different radial domain data subsets simultaneously. Then, efficient query access to the dataset is built by storing the data in a database that supports SQL queries, filtering the data based on the radial domain number, and extracting subsets in one go or in batches.

6. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 4, characterized in that: For frequently used subsets of data, a caching strategy can be adopted to pre-read the subsets into memory or fast storage media, thereby reducing the performance loss caused by repeated hard drive reads; Based on actual application needs, prioritize processing data from important areas or well site test areas, and rationally arrange the priority and order of data processing and subset construction.

7. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 4, characterized in that: Step S3, based on preprocessed pre-stack 3D seismic data, constructs a reference gather through migration stacking. The specific method includes: Common offset data is extracted from the preprocessed data, and volume offset is performed on the extracted common offset data to obtain the common reflection point imaging gather; Then, the common reflection point imaging gather is subjected to noise reduction and optimization processing to improve the signal-to-noise ratio of the common reflection point imaging gather; The reference gather can be obtained by superimposing the denoised common reflection point imaging gathers.

8. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 7, characterized in that: In the three-dimensional case of volume migration for extracting common offset data, its expression is: In the formula: f i ′(τ s +τ g ) is the first derivative of the pre-stack data of the i-th trace, n is the total number of seismic traces in different azimuth subsets, and τ s and τ g These are the travel times from the shot point and receiver point to the imaging point (x, y, T), respectively. s / τ g ) 2 These are the weighting coefficients for compensating for the spherical spread of seismic waves, (x m ,y m ) and h are the center point and half offset of the seismic trace, respectively; T0(x,y,x) m ,y m I(x,y,T,h) represents the starting imaging time for different imaging points in the seismic trace at that azimuth, and it represents the offset aperture. At each horizontal position (x,y) in the imaging profile, I(x,y,T,h) represents the common reflection point gather at that azimuth as the offset distance changes, where T is the reflection time and T0 is the start time of the time window.

9. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 7, characterized in that: The radial domain imaging gather described in step S4 includes the radial domain shot gather and the radial domain receiver gather; Offset processing is performed on a subset of radial domain multi-azimuth data to construct a radial domain shot point gather; Offset processing is performed on a subset of radial domain multi-azimuth data to construct a radial domain detector point gather.

10. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 9, characterized in that: Based on the radial domain data subset obtained in step S2, a radial domain shot point gather is constructed. The specific method includes: Extract the offset trace sets of all shot points in the radial domain subset, and superimpose the offset trace sets of the same shot points in the radial domain subset to obtain a new offset trace set composed of shot points, i.e., the radial domain shot point trace set.

11. The method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to claim 9, characterized in that: Based on the radial domain data subset obtained in step S2, a radial domain receiver point gather is constructed. The specific method includes: The offset gathers of all receiver points in the radial domain subset offset results are extracted. The offset gathers of the radial domain data subsets with the same receiver points are superimposed to obtain a new offset gather composed of receiver points, namely the radial domain receiver point gather.

12. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to any one of claims 9-11, characterized in that: When performing offset processing on a multi-directional subset of radial domain data according to Formula 1, the region with a clear, continuous or partially continuous isochronous interface of sedimentation isochronous axes is selected as the specified time window.

13. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to any one of claims 9-11, characterized in that: The radial domain shot point gather expression is Equation 2: The expression for the radial domain detector point gather is Equation 3: In the formula: (x s ,y s ) and (x g ,y g ) are the coordinates of the shot point and receiver point of the seismic trace, respectively; T0 is the start time of the time window.

14. A method for phase consistency correction in the imaging domain of wide-azimuth seismic data according to any one of claims 8-13, characterized in that: S5 calculates the phase difference based on the constructed reference gather and radial domain imaging gather; it then performs imaging domain phase consistency correction on the radial domain data subset, specifically including: 5.

1. To obtain phase consistency, a parameter called constant phase rotation is introduced to construct the cross-correlation function, which can be expressed as: f ik (α,τ)=cosα·R(τ)-sinα·r(τ) (4) Where, i represents the detector point, k represents the horizontal position of the image (CDP point), and φ ik (α,τ) is the cross-correlation function of the i-th detector in the detector offset channel set corresponding to the k-th CDP point, where α is the constant phase rotation angle; Where t represents time, [T1,T2] is the time window at that horizontal position, R(τ) is the real part of the cross-correlation function, r(τ) is the imaginary part of the cross-correlation function, τ is the time increment, and g i (t) represents the offset channel selected from the receiver offset channel set at the horizontal position k, and g i (t+τ) represents the offset channel selected from the receiver offset channel set at horizontal position k when the time increment is τ, i.e., at time t+τ; s k (t+τ) is the reference track at horizontal position K, Hg i (t) is g i Hilbert transform of (t); The function φ ik The maximum value of (α,τ) is the required phase difference (Δt,θ), that is, the phase difference at Δt is θ; by the principle of extrema, we can obtain: It is the cross-correlation function φ ik Partial derivatives of (α,τ); By solving the above equation, we can obtain: The phase difference (Δt, θ) can be obtained from equations (7) and (8); Similarly, the offset trajectory g of the shot point offset gather is used. j (t) and imaging channel s k (t+τ) can be used to calculate the remaining time difference of the shot point; where j represents the shot point. 5.

2. Using the phase difference obtained in step 5.1, perform phase correction on the radial domain data subsets to obtain the phase-corrected radial domain data subsets.

15. A phase consistency correction device for wide-azimuth seismic data imaging domain, characterized in that: include: The preprocessing unit is used to acquire wide-azimuth pre-stack 3D seismic data and perform preprocessing on the pre-stack 3D seismic data. The acquisition unit is used to efficiently partition the radial domain multi-azimuth data subset of the preprocessed pre-stack 3D seismic data and acquire the radial domain data subset. The first building unit is used to construct reference gathers by migration stacking based on preprocessed pre-stack 3D seismic data. The second response construction unit is used to construct a radial imaging gather based on the radial domain data subset obtained in step S2. The correction unit is used to perform imaging domain phase consistency correction on a subset of radial domain data based on the constructed reference gather and radial imaging gather.

Images