Earthquake reflection detection time profile imaging method and system

Abstract

The invention relates to a seismic reflection detection time profile imaging method and system. According to the invention, key processing method combination steps and processes in data processing are provided for deep seismic reflection detection; a combined processing flow of revolution wave tomography static correction, pre-stack denoising, earth surface consistency amplitude compensation, velocity analysis, simulated annealing-Monte Carlo earth surface consistency residual static correction and multi-dimensional data regularization interpolation based on an OMP algorithm is adopted, and finally Kirchhoff pre-stack time and pre-stack depth migration processing is carried out. The influence of irregular spatial sampling of seismic data in a complex terrain area on the imaging effect and the imaging precision is eliminated, the signal-to-noise ratio of the stack time section is improved, and the high-precision pre-stack migration seismic time section is obtained.

Description

Technical Field

[0001] This invention relates to an imaging method and system, belonging to the field of geographic exploration technology, specifically to a seismic reflection detection time profile imaging method and system. Background Technology

[0002] For seismic exploration, data processing is a crucial step. The quality of the processing parameters and workflow directly determines the success or failure of seismic data processing. Selecting appropriate seismic data processing parameters plays a vital role in obtaining high signal-to-noise ratio seismic time profiles for deep seismic reflection exploration and in subsequent seismic geological interpretation. Existing deep seismic reflection exploration data processing workflows are usually based on mature oil seismic exploration data processing workflows. However, deep seismic reflection exploration involves greater depths, and the strata in deep regions consist of metamorphic and igneous rocks, resulting in predominantly weak seismic reflection signals. In addition, the irregular distribution of shot points and significant differences in the excitation energy of individual shots due to complex terrain further complicate the already weak deep seismic reflection data, making it difficult to perform refined data processing. Conventional processing workflows are insufficient for achieving high signal-to-noise ratio and high-precision imaging of the shallow, middle, and deep crustal regions reaching the Moho discontinuity.

[0003] Currently, deep seismic reflection detection has been carried out in multiple orogenic belts, aiming to detect and reveal crustal structure and tectonics, crust-mantle interaction, and the deep-shallow tectonic relationship of major deep faults. This is of great significance for research on crustal deformation evolution, deep seismogenic structures, and the formation environment of mineral resources. Due to its large detection depth, different firing charge modes (large, medium, and small) are often used for seismic data acquisition, resulting in significant differences in the quality of raw single-shot data. Furthermore, due to the complex surface seismic geological conditions in orogenic belts, firing points cannot be arranged according to theoretical designs, resulting in scattered and disordered locations. These factors lead to uneven spatial sampling of seismic data, and the data after multi-channel stacking is irregular. This makes it prone to spatial frequency distortion during migration processing, degrading the imaging quality of seismic reflection waves and causing problems such as imaging footprints, imaging noise, and migration arcs. In addition, the irregular data prevents the full utilization of high-precision migration processing methods, resulting in inaccurate reflection wave imaging amplitude, low imaging accuracy, and seismic time profiles that cannot accurately reflect the morphology and physical property changes of underground geological structures.

[0004] Therefore, when processing deep seismic reflection data, in addition to meticulous static correction and denoising, and velocity model analysis to address the significant velocity differences between shallow, middle, and deep crustal regions, data regularization and interpolation reconstruction are also necessary. This processing method interpolates spatially irregular sample points into regular, uniform sample points, filling in missing irregular seismic traces within a certain range. This ensures that the interpolated seismic data possesses complete spatial and offset information, eliminating the impact of irregular spatial sampling on imaging effects and accuracy. It provides excellent pre-stack data for high-precision pre-stack migration processing. Summary of the Invention

[0005] The following provides a brief overview of one or more aspects to offer a basic understanding of them. This overview is not an exhaustive summary of all conceived aspects, nor is it intended to identify key or decisive elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form to prepare for the more detailed descriptions that follow.

[0006] The main objective of this invention is to address the technical problems existing in the prior art and to provide a seismic reflection detection time-profile imaging method and system. This system and method employ a combined processing flow of rotary wave tomography static correction, pre-stack denoising, surface consistency amplitude compensation, velocity analysis and simulated annealing-Monte Carlo surface consistency residual static correction, and multi-dimensional data regularization interpolation based on the OMP algorithm to pre-process deep seismic reflection detection data. Finally, Kirchhoff pre-stack time and pre-stack depth migration processing is used to achieve accurate imaging with fidelity and amplitude preservation in deep seismic reflection detection data processing.

[0007] To solve the above problems, the solution of the present invention is:

[0008] A seismic reflection detection time-profile imaging method includes:

[0009] Step 1: Decode the raw seismic data and establish the spatial attributes of shot points and detection points;

[0010] Step 2: Calculate the travel time of the first arrival wave to obtain the near-surface velocity structure, and determine the static correction amount that makes the first arrival wave smooth and continuous, and has high effective reflected wave resolution and signal-to-noise ratio on the superimposed profile, so as to eliminate the arrival time error of the seismic reflected wave.

[0011] Step 3: Suppress and remove noise of different types, frequencies and depths in the deep seismic reflection data to improve the signal-to-noise ratio and resolution of the effective reflection signal;

[0012] Step 4: Based on the velocity variation law from shallow to deep in the deep seismic reflection detection area, establish a velocity spectrum with prominent and clear layers of reflected wave energy clusters in each marker layer; then, using the simulated annealing algorithm and the Monte Carlo residual static correction and fine velocity analysis iterative method, obtain the velocity function and static correction amount with fixed position of the reflected wave energy clusters in the marker layer.

[0013] Step 5: The signal-to-noise ratio and resolution of the superimposed time profile are processed using multidimensional data regularization interpolation technology based on the OMP algorithm.

[0014] Step 6: Use Kirchhoff pre-stack time and pre-stack depth migration to obtain deep seismic reflection detection migration seismic time profiles.

[0015] Preferably, in the above-mentioned seismic reflection detection time profile imaging method, step 2 adopts the rotating wave tomography static correction processing method, and calculates the near-surface velocity structure by means of the first arrival wave travel time through the ray tracing algorithm, and obtains a suitable static correction amount.

[0016] Preferably, in the above-mentioned seismic reflection detection time profile imaging method, step 3 employs a "classification, step-by-step, domain-by-domain, zone-by-zone, frequency-by-frequency, and time-by-time" pre-stack denoising processing method to suppress and remove noise of different types, frequencies, and depths in deep seismic reflection data, thereby improving the signal-to-noise ratio and resolution of the effective reflection signal.

[0017] Preferably, in the above-mentioned seismic reflection detection time profile imaging method, step 4 ensures that the residual static correction eventually converges to a sample point range, thereby improving imaging accuracy.

[0018] A seismic reflection detection time-profile imaging system includes:

[0019] The observation system defines a module that decodes raw seismic data and establishes spatial attributes for shot points and detection points.

[0020] The static correction acquisition module calculates the travel time of the first arrival wave to obtain the near-surface velocity structure, and obtains the static correction amount to make the first arrival wave smooth and continuous, with high effective reflected wave resolution and signal-to-noise ratio on the superimposed profile, thus eliminating the arrival time error of the seismic reflected wave.

[0021] The preprocessing module suppresses and removes noise of different types, frequencies, and depths in deep seismic reflection data, thereby improving the signal-to-noise ratio and resolution of the effective reflection signal.

[0022] The iterative processing module establishes a velocity spectrum with prominent and clearly defined energy clusters of reflected waves in each marker layer, based on the velocity variation law from shallow to deep in the deep seismic reflection detection zone. Then, by using the simulated annealing algorithm and the iterative method of Monte Carlo residual static correction and fine velocity analysis, the velocity function and static correction amount with fixed position of the energy clusters of reflected waves in the marker layer are obtained.

[0023] The interpolation processing module uses multidimensional data regularization interpolation processing technology based on the OMP algorithm to process the signal-to-noise ratio and resolution of the superimposed time profile;

[0024] The time profile processing module uses Kirchhoff pre-stack time and pre-stack depth migration to obtain deep seismic reflection detection migration seismic time profiles.

[0025] Preferably, in the above-mentioned seismic reflection detection time profile imaging system, the static correction acquisition module adopts the rotating wave tomography static correction processing method, and calculates the near-surface velocity structure by means of the first arrival wave travel time through the ray tracing algorithm, and obtains the appropriate static correction amount.

[0026] Preferably, in the above-mentioned seismic reflection detection time profile imaging system, the pre-stack processing module adopts a "classification, step-by-step, domain-by-domain, zone-by-zone, frequency-by-frequency, and time-by-time" pre-stack denoising processing method to suppress and remove noise of different types, frequencies, and depths in deep seismic reflection data, thereby improving the signal-to-noise ratio and resolution of the effective reflection signal.

[0027] Preferably, in the above-mentioned seismic reflection detection time profile imaging system, the iterative processing module ensures that the residual static correction eventually converges to a range of sample points, thereby improving imaging accuracy.

[0028] Compared with existing deep seismic reflection detection data processing technologies, this invention brings the following beneficial technical effects:

[0029] The deep seismic reflection data processing flow used in this invention is based on the conventional seismic exploration data processing flow and does not require additional processing flow testing.

[0030] This invention proposes a combination of key processing methods and procedures for deep seismic reflection detection data. It employs a combined processing flow including rotary wave tomography static correction, pre-stack denoising, surface consistency amplitude compensation, velocity analysis and simulated annealing-Monte Carlo surface consistency residual static correction, and multidimensional data regularization interpolation based on the OMP algorithm. Finally, Kirchhoff pre-stack time and pre-stack depth migration processing is performed. This eliminates the impact of irregular spatial sampling of seismic data in complex terrain areas on imaging effects and accuracy, improves the signal-to-noise ratio of the stacked time profile, and obtains a high-precision pre-stack migrated seismic time profile. Attached Figure Description

[0031] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments of the invention and, together with the specification, further serve to explain the principles of the invention and to enable those skilled in the art to make and use this disclosure.

[0032] Figure 1 This is a flowchart illustrating the overall process of processing seismic data from deep earthquake reflection detection.

[0033] Figure 2 A schematic diagram of the static correction method for inversion in rotary wave tomography;

[0034] Figure 3 A high-fidelity noise suppression method and schematic diagram;

[0035] Figure 4 A schematic diagram illustrating the principles of regularization and interpolation of deep seismic reflection data;

[0036] Figure 5(a) shows the overlay results of the cross-sections before and after data regularization interpolation. The figure shows a local part of the overlay cross-section before data regularization.

[0037] Figure 5(b) shows the overlay results of the cross-sections before and after data regularization interpolation. The figure shows a local part of the overlay cross-section after data regularization interpolation.

[0038] Figure 6(a) shows the pre-stack time migration results of the profile after data regularization interpolation. The figure shows a local part of the pre-stack time migration profile before data interpolation regularization.

[0039] Figure 6(b) shows the pre-stack time migration results of the profile after data regularization interpolation. The figure shows a local part of the pre-stack time migration profile after data regularization interpolation.

[0040] Figure 7(a) shows the pre-stack depth migration results of the profile after data regularization interpolation. The figure shows a local part of the pre-stack depth migration profile before data interpolation regularization.

[0041] Figure 7(b) shows the pre-stack depth migration results of the profile after data regularization interpolation. The figure shows a local part of the pre-stack depth migration profile after data regularization interpolation.

[0042] Embodiments of the present invention will be described with reference to the accompanying drawings. Detailed Implementation

[0043] Example

[0044] like Figure 1 As shown, this invention provides key methods and processes for processing seismic data. This embodiment is mainly aimed at the field of deep seismic reflection detection. Without increasing costs, it optimizes the combination of seismic data processing methods and processes, effectively improving the signal-to-noise ratio of deep seismic reflection time profiles and the imaging accuracy of the lower crust-Moho discontinuity regional structure, thus ensuring the reliability of deep seismic reflection geological interpretation.

[0045] This embodiment first provides a seismic reflection detection time profile imaging method, including the following steps:

[0046] Step 1, data preprocessing steps, includes:

[0047] The raw seismic data is decompiled, and the acquired raw seismic data is arranged according to the receiver trace sequence;

[0048] The observation system is defined to address the irregular arrangement of shot and receiver points in the acquisition of deep seismic reflection detection data in complex mountainous terrain. Linear dynamic correction is used to establish accurate spatial attributes of shot and receiver points.

[0049] Step 2: Calculate the travel time of the first arrival wave to obtain the near-surface velocity structure, and determine an appropriate static correction value to eliminate the arrival time error of the seismic reflected wave;

[0050] First arrival picking is performed using a neural network on uncorrected single-shot records. Energy discrimination is performed based on instantaneous amplitude, and pattern recognition is conducted on the shape of the first arrival. Then, polynomial fitting is performed and iterated multiple times to accurately pick the first arrival. An initial near-surface linear model inversion (LMI) layered model can be obtained from the first arrival. Then, a near-surface grid model is obtained using cyclotron tomography. Finally, the surface consistency static correction is calculated based on the final model.

[0051] Step 3: Suppress and remove regular noise such as sound waves, surface waves, linear interference waves, abnormal reflection waves, secondary interference waves and AC interference, irregular noise such as micro-vibrations generated by production activities, low-frequency background noise and instrument noise, as well as noise in different depth areas in deep seismic reflection data, to improve the signal-to-noise ratio and resolution of the effective reflection signal.

[0052] Pre-stack denoising involves manually interactively removing bad channels from individual shots, then using the surface consistency principle to suppress anomalous amplitudes. Linear interference waves are suppressed using an adaptive pre-stack coherent noise suppression method. Adaptive surface wave attenuation techniques are employed to suppress noise by utilizing the differences in frequency and velocity between surface waves and effective waves. Seismic data is decomposed into a predictable portion containing useful signals and an unpredictable portion containing random noise, thus eliminating random noise interference. Coherent filtering is used to suppress multiple waves while maintaining the primary wave without distortion. Frequency-division anomalous amplitude suppression techniques are used to eliminate high-frequency noise. The noise suppression process comprehensively considers the differences between shallow and deep noise, suppressing them separately.

[0053] Step 4: Based on the velocity variation law from shallow to deep in the deep seismic reflection detection zone, establish a velocity spectrum with prominent and clearly defined energy clusters of reflected waves in each marker layer; then, using the simulated annealing algorithm and the Monte Carlo residual static correction and fine velocity analysis iterative method, obtain the velocity function and static correction amount with fixed position of the energy clusters of reflected waves in the marker layer.

[0054] Specifically, this includes:

[0055] The study investigated and analyzed the velocity variation patterns from shallow to deep in the deep seismic reflection detection area, selected the dominant frequency bands of effective reflected waves, and obtained velocity spectra with prominent and clearly defined energy clusters of reflected waves from each marker layer. Techniques such as multi-channel combination to increase the number of stacks were used to improve the quality of gathers and velocity spectra, thereby enhancing the accuracy of velocity interpretation. Furthermore, using iterative methods of residual static correction and fine velocity analysis under Monte Carlo and simulated annealing algorithms, the velocity functions and static correction quantities with fixed positions of energy clusters of reflected waves from the marker layers were obtained. This ensured that the residual static correction ultimately converged to a single sample point range, improving the continuity and focusing of the phase axis of the reflected waves.

[0056] Step 5: The seismic data after the aforementioned processing measures are processed using a multidimensional data regularization technique based on the Orthogonal Matching Pursuit (OMP) algorithm. This technique mainly involves data regularization and interpolation on three parameters: CMP (Common Centroid Gather), Offset, and Time. This eliminates the spatial sampling inhomogeneity caused by complex terrain in deep seismic reflection detection and fills in missing irregular seismic traces within a certain range, so that the regularized and interpolated seismic data has complete spatial and offset information.

[0057] Step 6: Use Kirchhoff pre-stack time and pre-stack depth migration to obtain deep seismic reflection detection migration seismic time profiles.

[0058] As described above, the monitoring processing device logic of this embodiment first extracts only the boundary line of the target area for each frame of image through image masking. If there is an intrusion anomaly on the boundary line, the mask is switched and gradually extended to the entire internal area of ​​the monitored target. If the intrusion anomaly still exists, the alarm action is completed.

[0059] In this embodiment, although the above methods are illustrated and described as a series of actions for the sake of simplicity, it should be understood and appreciated that these methods are not limited by the order of the actions, because according to one or more embodiments, some actions may occur in a different order and / or concurrently with other actions from those illustrated and described herein or not illustrated and described herein but which may be understood by those skilled in the art.

[0060] It should be noted that references to "an embodiment," "an embodiment," "an example embodiment," "some embodiments," etc., in the specification indicate that the described embodiments may include specific features, structures, or characteristics, but each embodiment may not necessarily include said specific features, structures, or characteristics. Furthermore, such phrases do not necessarily refer to the same embodiment. Additionally, when a specific feature, structure, or characteristic is described in connection with an embodiment, whether explicitly described or not, implementing such a feature, structure, or characteristic in conjunction with other embodiments will be within the knowledge of those skilled in the art.

[0061] The prior description of this disclosure is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not intended to be limited to the examples and designs described herein, but should be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for time profile imaging of seismic reflection detection, characterized in that: include: Step 1: decode the original seismic data and establish the spatial attributes of the shot points and detection points; Step 2, calculate the travel time of the first arrival wave to obtain the near-surface velocity structure, and obtain the static correction amount that makes the first arrival smooth and continuous, and the effective reflection wave resolution and signal-to-noise ratio on the stacked section high, so as to eliminate the arrival time error of the seismic reflection wave; Step 3, suppress and remove noises of different types, frequencies and depths in deep seismic reflection data to improve the signal-to-noise ratio and resolution of effective reflection signals; Step 4: According to the velocity variation law from shallow to deep in the deep seismic reflection detection area, establish a velocity spectrum with prominent reflection wave energy clusters and clear layers in each marker layer; then use the simulated annealing algorithm and Monte Carlo residual static correction and fine velocity analysis iteration method to obtain the velocity function and static correction value with fixed position of reflection wave energy clusters in the marker layer; Step 5, using the multi-dimensional data regularized interpolation processing technology based on the OMP algorithm to process the signal-to-noise ratio and resolution of the stacked time section; Step 6: Kirchhoff prestack time and prestack depth migration are used to obtain the deep seismic reflection detection migration seismic time section.

2. A seismic reflection detection time profile imaging method according to claim 1, characterized in that: In step 2, the rotating wave tomography static correction processing method is adopted. Through the ray tracing algorithm, the travel time of the first arrival wave is calculated to obtain the near-surface velocity structure and obtain the appropriate static correction amount.

3. The method for time profile imaging of seismic reflection detection according to claim 1, characterized in that: In step 3, a pre-stack denoising processing method of "classification, step, domain, region, frequency and time" is adopted to suppress and remove noise of different types, frequencies and depths in deep seismic reflection data, thereby improving the signal-to-noise ratio and resolution of effective reflection signals.

4. The method for time profile imaging of seismic reflection detection according to claim 1, characterized in that: In step 4, the residual static correction is finally converged to a sample point range, thereby improving the imaging accuracy.

5. A seismic reflection detection time profile imaging system, characterized in that: include: Observation system definition module, raw seismic data decoding, establishment of spatial attributes of shot points and detection points; The static correction acquisition module calculates the travel time of the first arrival wave to obtain the near-surface velocity structure, and obtains the static correction value that makes the first arrival smooth and continuous, and the effective reflection wave resolution and signal-to-noise ratio on the stacked section high, thereby eliminating the arrival time error of the seismic reflection wave; The stacking pre-processing module suppresses and removes noises of different types, frequencies and depths in deep seismic reflection data, thereby improving the signal-to-noise ratio and resolution of effective reflection signals; The iterative processing module establishes a velocity spectrum with prominent reflection wave energy groups and clear layers in each marker layer according to the velocity variation law from shallow to deep in the deep seismic reflection detection area; then, the simulated annealing algorithm, Monte Carlo residual static correction and fine velocity analysis iterative method are used to obtain the velocity function and static correction amount of the fixed position of the reflection wave energy group of the marker layer; The interpolation processing module uses the multi-dimensional data regularization interpolation processing technology based on the OMP algorithm to process the signal-to-noise ratio and resolution of the stacked time section; The time section processing module uses Kirchhoff prestack time and prestack depth migration processing to obtain deep seismic reflection detection migration seismic time section.

6. A seismic reflection detection time profile imaging system according to claim 5, characterized in that: In the static correction acquisition module, the rotating wave tomography static correction processing method is adopted. Through the ray tracing algorithm, the first arrival wave travel time is calculated to obtain the near-surface velocity structure and obtain the appropriate static correction value.

7. The seismic reflection detection time profile imaging system according to claim 5, characterized in that: In the stack pre-processing module, the pre-stack denoising processing method of "classification, step, domain, region, frequency and time" is adopted to suppress and remove noise of different types, frequencies and depths in deep seismic reflection data, so as to improve the signal-to-noise ratio and resolution of effective reflection signals.

8. The seismic reflection detection time profile imaging system according to claim 5, characterized in that: In the iterative processing module, the residual static correction is finally converged to a sample point range, thereby improving the imaging accuracy.

Images