Diffracted wave separation method for complex tectonic seismic data
By combining frequency domain empirical mode decomposition and singular value decomposition, the problem of cross-aliasing of diffracted and reflected wave signals under complex structural conditions was solved, and the effective separation of diffracted wave signals was achieved, thus improving the accuracy and efficiency of seismic exploration.
Patent Information
- Application Number
- CN202411631530.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-15
- Publication Date
- 2025-02-21
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Under complex geological conditions, seismic diffraction and reflection signals are severely intermingled, making it difficult for existing technologies to effectively separate diffraction signals and affecting subsequent energy and mineral resource exploration.
A method combining frequency domain empirical mode decomposition and singular value decomposition is adopted. Seismic data is processed through pseudostatic correction and linear Radon transform, and diffraction wave signals are separated by adaptive low-rank threshold constraints.
Effective separation of diffraction wave signals improves the separation effect under complex structural conditions, enhancing the accuracy and efficiency of subsequent seismic exploration.
Smart Images
Figure CN119493169A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of seismic exploration data processing, and in particular relates to a diffraction wave separation method for complex structure seismic data. Background Art
[0002] With the continuous development of oil and gas exploration and development, areas with good construction conditions and high data quality have completed relatively sufficient exploration and development, and the oil and gas reservoirs dominated by simple shallow layered media have gradually shifted to deep, complex and small-scale structural oil and gas reservoirs. At this time, the seismic diffraction wave signal is essentially a direct response to underground discontinuities, carrying key geological information of complex structures, especially small-scale geological bodies, and has great potential for high-resolution imaging of small-scale discontinuous geological bodies.
[0003] Since the energy of the reflected wave signal is usually more than ten times or even hundreds of times that of the diffraction wave signal, the diffraction wave is often submerged. Therefore, it is very difficult to directly identify the diffraction wave signal in the seismic record. The seismic data processing work generally separates the diffraction wave signal from the seismic wave field first, and then performs specific data processing and interpretation work on the separated diffraction wave signal.
[0004] It is a common and effective method to extract diffraction wave signals from seismic data using local dip information. However, due to the dramatic changes in terrain undulations, the dip angle of the reflection wave event axis in the post-stack data is large, and in seismic data containing non-uniform geological bodies such as faults, holes, cracks, and formation pinch-outs, the diffraction wave signal and the reflection wave signal are seriously cross-aliased. At this time, the calculated local dip angle cannot simply reflect the diffraction wave information, and the event axis corresponding to the large dip angle is not necessarily a diffraction wave signal. However, it is difficult to identify the diffraction wave information of these intersecting areas one by one and calculate their dip angles, which will inevitably lead to poor separation of diffraction wave signals. Therefore, it is necessary to propose an adaptive diffraction wave signal separation method for exploration seismic data observed in complex structures. Summary of the invention
[0005] The purpose of the present invention is to provide a diffraction wave separation method for complex structural seismic data, combining frequency domain empirical mode decomposition and singular value decomposition to solve the problem that under complex structural conditions, the local inclination of diffraction wave signals in exploration seismic data is large, and the cross-aliasing with the reflected wave signal is serious, resulting in poor separation effect of diffraction wave signals that rely on local inclination information, thereby laying a foundation for the subsequent use of diffraction waves for seismic exploration of energy and mineral resources.
[0006] The objective of the present invention is achieved through the following technical solutions:
[0007] A diffraction wave separation method for complex structural seismic data comprises the following steps:
[0008] A. Preprocess the original seismic data;
[0009] B. Extract gathers from the pre-processed seismic data to convert their arrangement from common shot gathers to common offset gathers to ensure that the reflection event morphology is relatively continuous and smooth;
[0010] D(x s ,x r ,t)=D(x s ,x off ,t),(1)
[0011] Among them, D(x s ,x r ,t) represents the common shot point gather, D(x s ,x off ,t) represents the common offset gather, x s represents the shot point position, x r Represents the location of the wave construction point, x off represents the offset information, and t represents the recording time;
[0012] C. Use pseudo-static correction method to correct the common offset gather D(x s ,x off ,t) Process the obtained data D static (x s ,x off ,t), directly set the appropriate correction value Δt for each channel to eliminate the fluctuation of the reflected wave event axis shape caused by the complex terrain;
[0013] D static (x s ,x off ,t)=D(x s ,x off ,t+Δt), (2)
[0014] The correction amount Δt is s and x off The spatial position variation function is used to flatten the convex terrain event axis;
[0015] D. Seismic data static (x s ,x off ,t) is subjected to linear Radon transformation and arranged according to the common ray parameters to obtain the τ-p domain seismic data D τ-p (x s ,p,τ);
[0016]
[0017] Among them, p represents the ray parameter, and τ represents the transformed time;
[0018] E. Seismic data D τ-p (x s ,p,τ) to perform frequency domain empirical mode decomposition and retain the most violent jitter signal as the partial diffraction wave D dif1 (x s ,p,τ), and perform singular value decomposition on the remaining reconstructed intrinsic modal components except the remainder, and adaptively perform low-rank threshold constraints to obtain the remaining diffraction wave signal D dif2 (x s ,p,τ);
[0019] F. Data fusion, D dif1 (x s ,p,τ) and D dif2 (x s ,p,τ) are added to obtain the τ-p domain diffraction wave D dif (x s ,p,τ);
[0020] G. Reconstruct the diffraction wave signal D dif (x s ,p,τ) to obtain the seismic data D containing pseudo-static correction. dif (x s ,x off ,t+Δt), and subtract the previous correction amount Δt to perform inverse correction processing to obtain seismic data D dif (x s ,x off ,t), the final diffraction wave signal D can be obtained by converting the new extraction gather to the common shot gather. dif (x s ,x r ,t).
[0021] Furthermore, in step A, the preprocessing process includes bad track removal, amplitude equalization and random noise suppression.
[0022] Furthermore, step E specifically includes the following steps:
[0023] E1. Seismic data D τ-p (x s ,p,τ) to perform Fourier transform and obtain frequency domain data D τ-p (x s ,p,f), f and t are a changing pair, representing frequency;
[0024] E2 sets a suitable local window and gradually adjusts the seismic data D τ-p (x s,p,f) perform window processing to obtain local window data D i (x s ,p,f), where i represents the window number; the local window can further enhance the linear characteristics of the seismic data, thereby enhancing the robustness of the low-rank threshold processing after singular value decomposition of the seismic data;
[0025] E2. For local window data D i (x s ,p,f) to perform frequency domain empirical mode decomposition. The decomposition process is as shown in formula (4), and the part of the signal IMF1(x) with the most severe jitter in each local window data is directly retained as the diffraction wave signal, and reconstructed into a partial diffraction wave D through inverse Fourier transform. dif1 (x s ,p,τ);
[0026]
[0027] Among them, the remainder r(x) represents the reflected wave information, IMF1(x) represents the diffraction wave information, and the remaining The term is the aliasing term of the reflected wave and the diffracted wave;
[0028] E3, remove the remainder r(x), this process can remove most of the reflected waves in the data and retain the main diffraction wave IMF1(x);
[0029] E4, aliasing data reconstruction, the aliasing terms of reflected waves and diffracted waves Construct the Hankel matrix and perform singular value decomposition to suppress the large-angle reflection wave; let the aliasing term The constructed Hankel matrix is X, and the singular value matrix ∑ is obtained by singular value decomposition;
[0030] X=U∑V T , (5)
[0031]
[0032] U by XX T The eigenvectors of T The eigenvectors of X are composed of ∑, which is composed of diagonal elements representing the singular values of X and satisfies |∑1|≥|∑2|≥…≥|∑ N |;
[0033] E5, singular value decomposition and reconstruction, adaptively select low-rank parameters according to the principle of maximum singular value gradient, adaptive threshold parameter N ∑ Get the formula (7); the first N ∑ The singular values representing the oblique reflected waves are set to 0. Then, the inverse Fourier transform is performed to obtain the remaining diffraction wave signal Ddif2 (x s ,p,τ);
[0034] N ∑ =argmax|∑|, (7).
[0035] Compared with the prior art, the present invention has the following beneficial effects:
[0036] 1. The present invention can effectively solve the problem of failure of separation of diffraction waves and reflection waves according to the inclination of the event axis under complex seismic structural conditions by combining frequency domain empirical mode decomposition with singular value decomposition;
[0037] 2. The pseudo-static correction and linear Radon transform before decomposition can overcome the problem of large reflection wave inclination caused by geological conditions, further increase the linear difference between reflection wave and diffraction wave, and thus improve the effect of subsequent decomposition;
[0038] 3. By performing singular value decomposition under aliasing after frequency domain empirical mode decomposition, the serious aliasing problem of diffraction waves and reflected waves in the modal components can be effectively solved. In addition, adaptive low-rank constraints are performed to avoid artificial setting of empirical parameters and improve calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0040] Figure 1 It is a technical route flow chart of the diffraction wave separation method for complex structure seismic data of the present invention;
[0041] Figure 2 Display of different gathers for seismic data;
[0042] Figure 3 This is the diffraction wave separation result of seismic data. DETAILED DESCRIPTION
[0043] The present invention will be further described below in conjunction with embodiments:
[0044] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.
[0045] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0046] The present invention is directed to a diffraction wave separation method for complex structural seismic data, such as Figure 1 As shown, the following steps are included:
[0047] A. Preprocess the original seismic data to detect the basic situation of the data. The preprocessing process includes bad track removal, amplitude equalization, random noise suppression, etc.
[0048] B. Extract the pre-processed seismic data and convert its arrangement from common shot gather to common offset gather to ensure that the reflection event morphology is relatively continuous and smooth, such as Figure 2 shown.
[0049] D(x s ,x r ,t)=D(x s ,x off ,t),(1)
[0050] Among them, D(x s ,x r ,t) represents the common shot point gather, D(x s ,x off ,t) represents the common offset gather, x s represents the shot point position, x r Represents the location of the wave construction point, x off represents the offset information, and t represents the recording time.
[0051] C. Use pseudo-static correction method to correct the common offset gather D(x s ,x off ,t) further processing to obtain data D static (x s ,x off ,t), and directly set the appropriate correction value Δt for each channel to eliminate the fluctuation of the reflected wave event axis shape caused by the complex terrain and reduce the influence of the formation dip.
[0052] D static (x s ,x off ,t)=D(x s ,x off ,t+Δt), (2)
[0053] The correction amount Δt iss and x off The spatial position change function of is used to flatten the event axis of the undulating terrain.
[0054] D. Seismic data static (x s ,x off ,t) perform linear Radon transformation and arrange them according to the common ray parameters to obtain the τ-p domain seismic data D τ-p (x s ,p,τ).
[0055]
[0056] Where p represents the ray parameter, τ represents the transformed time, and formula (3) represents the process of synthesizing point source seismic data into plane wave seismic data. In plane wave data, the reflection wave appears as a linear event axis, while the diffraction wave appears as a nonlinear event axis. Therefore, after the linear Radon transform, the coherence of the reflection wave event axis can be further improved.
[0057] E. Seismic data D τ-p (x s ,p,τ) to perform frequency domain empirical mode decomposition and retain the most violent jitter signal as the partial diffraction wave D dif1 (x s ,p,τ). The remaining reconstructed intrinsic modal components except the remainder are subjected to singular value decomposition and adaptive low-rank threshold constraints to obtain the remaining diffraction wave signal D dif2 (x s ,p,τ).
[0058] E1. Seismic data D τ-p (x s ,p,τ) to perform Fourier transform and obtain frequency domain data D τ-p (x s ,p,f), f and t are a changing pair, representing frequency.
[0059] E2 sets a suitable local window and gradually adjusts the seismic data D τ-p (x s ,p,f) perform window processing to obtain local window data D i (x s ,p,f), where i represents the window number. The local window can further enhance the linear characteristics of seismic data, thereby enhancing the robustness of low-rank threshold processing after singular value decomposition of seismic data.
[0060] E2. Local window data D i (x s,p,f) is used to perform frequency domain empirical mode decomposition (fx-EMD). The decomposition process is as shown in formula (4). In addition, the most jittery part of the signal IMF1(x) in each local window data is directly retained as the diffraction wave signal and reconstructed into a partial diffraction wave D through inverse Fourier transform. dif1 (x s ,p,τ).
[0061]
[0062] Among them, the remainder r(x) represents the reflected wave information, IMF1(x) represents the diffraction wave information, and the remaining The term is the aliasing term of the reflected wave and the diffracted wave.
[0063] E3. Eliminate the remainder r(x). This process can remove most of the reflected waves in the data and retain the main diffraction wave IMF1(x).
[0064] E4, aliasing data reconstruction. Reconstruct the aliasing terms of the reflected wave and the diffracted wave Construct the Hankel matrix and perform singular value decomposition to suppress the large-angle reflection wave. Let the aliasing term The constructed Hankel matrix is X, and the singular value matrix ∑ is obtained by singular value decomposition.
[0065] X=U∑V T , (5)
[0066]
[0067] U by XX T The eigenvectors of T The eigenvectors of X are composed of ∑, which is composed of diagonal elements representing the singular values of X and satisfies |∑1|≥|∑2|≥…≥|∑ N |.
[0068] E5, singular value decomposition and reconstruction. Adaptively select low-rank parameters according to the singular value gradient maximum principle, and adaptive threshold parameter N ∑ Get the formula (7). ∑ The singular values representing the oblique reflected waves are set to 0. Then, the inverse Fourier transform is performed to obtain the remaining diffraction wave signal D dif2 (x s ,p,τ).
[0069] N ∑ =argmax|∑|, (7)
[0070] F. Data fusion. dif1 (x s ,p,τ) and D dif2 (xs ,p,τ) are added to obtain the τ-p domain diffraction wave D dif (x s ,p,τ).
[0071] G. Reconstruct the diffraction wave signal D dif (x s ,p,τ) to obtain the seismic data D containing pseudo-static correction. dif (x s ,x off ,t+Δt), and subtract the previous correction amount Δt to perform inverse correction processing to obtain seismic data D dif (x s ,x off ,t), the final diffraction wave signal D can be obtained by converting the new extraction gather to the common shot gather. dif (x s ,x r ,t), such as Figure 3 shown.
[0072] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims
1. A diffraction wave separation method for complex structural seismic data, characterized in that: The following steps are involved: A. Preprocess the original seismic data; B. Extract gathers from the pre-processed seismic data to convert their arrangement from common shot gathers to common offset gathers to ensure that the reflection event morphology is relatively continuous and smooth; D(x s ,x r ,t)=D(x s ,x off ,t),(1) Among them, D(x s ,x r ,t) represents the common shot point gather, D(x s ,x off ,t) represents the common offset gather, x s represents the shot point position, x r Represents the location of the wave construction point, x off represents the offset information, and t represents the recording time; C. Use pseudo-static correction method to correct the common offset gather D(x s ,x off ,t) Process the obtained data D static (x s ,x off ,t), directly set the appropriate correction value Δt for each channel to eliminate the fluctuation of the reflected wave event axis shape caused by the complex terrain; D static (x s ,x off ,t)=D(x s ,x off ,t+Δt),(2) The correction value Δt is s and x off The spatial position variation function is used to flatten the convex terrain event axis; D. Seismic data static (x s ,x off ,t) is subjected to linear Radon transformation and arranged according to the common ray parameters to obtain the τ-p domain seismic data D τ-p (x s ,p,τ); Among them, p represents the ray parameter, and τ represents the transformed time; E. Seismic data D τ-p (x s ,p,τ) to perform frequency domain empirical mode decomposition and retain the most violent jitter signal as the partial diffraction wave D dif1 (x s ,p,τ), and perform singular value decomposition on the remaining reconstructed intrinsic modal components except the remainder, and adaptively perform low-rank threshold constraints to obtain the remaining diffraction wave signal D dif2 (x s ,p,τ); F. Data fusion, D dif1 (x s ,p,τ) and D dif2 (x s ,p,τ) are added to obtain the τ-p domain diffraction wave D dif (x s ,p,τ); G. Reconstruct the diffraction wave signal D dif (x s ,p,τ) to obtain the seismic data D containing pseudo-static correction. dif (x s ,x off ,t+Δt), and subtract the previous correction amount Δt to perform inverse correction processing to obtain seismic data D dif (x s ,x off ,t), the final diffraction wave signal D can be obtained by converting the new extraction gather to the common shot gather. dif (x s ,x r ,t).
2. The diffraction wave separation method for complex structure seismic data according to claim 1, characterized in that: Step A: The preprocessing process includes bad track removal, amplitude equalization and random noise suppression.
3. A diffraction wave separation method for complex structure seismic data according to claim 1, characterized in that: Step E specifically comprises the following steps: E1. Seismic data D τ-p (x s ,p,τ) to perform Fourier transform and obtain frequency domain data D τ-p (x s ,p,f), f and t are a changing pair, representing frequency; E2 sets a suitable local window and gradually adjusts the seismic data D τ-p (x s ,p,f) perform window processing to obtain local window data D i (x s ,p,f), where i represents the window number; the local window can further enhance the linear characteristics of the seismic data, thereby enhancing the robustness of the low-rank threshold processing after singular value decomposition of the seismic data; E2. Local window data D i (x s ,p,f) to perform frequency domain empirical mode decomposition. The decomposition process is as shown in formula (4), and the part of the signal IMF1(x) with the most severe jitter in each local window data is directly retained as the diffraction wave signal, and reconstructed into a partial diffraction wave D through inverse Fourier transform. dif1 (x s ,p,τ); Among them, the remainder r(x) represents the reflected wave information, IMF1(x) represents the diffraction wave information, and the remaining The term is the aliasing term of the reflected wave and the diffracted wave; E3, remove the remainder r(x), this process can remove most of the reflected waves in the data and retain the main diffraction wave IMF1(x); E4, aliasing data reconstruction, the aliasing terms of reflected waves and diffracted waves Construct the Hankel matrix and perform singular value decomposition to suppress the large-angle reflection wave; let the aliasing term The constructed Hankel matrix is X, and the singular value matrix ∑ is obtained by singular value decomposition; X=U∑V T , (5) U by XX T The eigenvectors of T The eigenvectors of X are composed of ∑, which is composed of diagonal elements representing the singular values of X and satisfies |Σ1|≥|Σ2|≥…≥|Σ N |; E5, singular value decomposition and reconstruction, adaptively select low-rank parameters according to the principle of maximum singular value gradient, adaptive threshold parameter N ∑ Get the formula (7); the first N ∑ The singular values representing the oblique reflected waves are set to 0; then, the inverse Fourier transform is performed to obtain the remaining diffraction wave signal D dif2 (x s ,p,τ); N ∑ = arg max|∑|, (7).