A method and apparatus for pre-stack time migration denoising

Abstract

This invention belongs to the field of seismic data processing technology for petroleum exploration, specifically relating to a pre-stack time migration denoising method and apparatus. The denoising method includes the following steps: (1) processing seismic data to obtain a pre-stack time migration profile, and selecting a stable wave impedance interface for the work area on the pre-stack time migration profile; (2) performing geological correction on the pre-stack time migration profile based on the stable wave impedance interface for the work area to obtain a geologically corrected profile; (3) performing F-K filtering on the geologically corrected profile to obtain an F-K filtered profile; (4) performing wave impedance interface inverse correction on the F-K filtered profile to obtain the final profile. The pre-stack time migration denoising method of this invention adopts the concept of "geological correction," combined with F-K filtering to effectively suppress migration noise. It features a simple and easy-to-draw time window, a small apparent velocity range that is easy to control, and improved operational efficiency. The denoised profile exhibits good amplitude preservation, greatly improving profile quality.

Description

Technical Field

[0001] This invention belongs to the field of seismic data processing technology for petroleum exploration, and specifically relates to a pre-stack time migration denoising method and apparatus. Background Technology

[0002] In seismic exploration, improving the signal-to-noise ratio (SNR) of data is the primary task. Only with a certain SNR can signal resolution be improved. If the SNR in certain frequency bands is low, it will inevitably limit the bandwidth of the signal. Similarly, only by obtaining a high SNR can the fidelity of signal processing be studied. With low SNR data, the recovery and preservation of signal amplitude will be constrained by noise.

[0003] As seismic exploration becomes more advanced and geological understanding deepens, the requirements for seismic imaging accuracy are increasing. Traditional post-stack time migration methods based on stacking are difficult to image accurately when there are complex underground structures and large velocity variations. Pre-stack migration methods, however, can largely compensate for these shortcomings by directly performing migration and realignment processing on the data (tilt realignment, diffraction wave convergence, amplitude and waveform recovery), which can eliminate structural artifacts and more realistically reflect the morphology of underground geological bodies.

[0004] In traditional seismic data processing, FK filtering is used to divide the data into multiple time windows and apply different apparent velocities for denoising. This approach has several problems: First, for complex stratigraphic systems, seismic migration profiles often exhibit severe arcing. Testing requires setting multiple time windows and utilizing multiple velocity ranges, resulting in numerous and lengthy operations, and easily leaving traces of denoising artifacts. For example, in a work area with 1984 main survey lines, over 300 operations are needed to complete the process, which is cumbersome and time-consuming. Second, variable time windows need to be defined within the seismic data processing module (the start and end times can only be one variable and one fixed; if the start time is fixed, the end time is variable). If the time window is not fixed within the correct range, valid information will be removed, leading to unsatisfactory denoising results. Summary of the Invention

[0005] The purpose of this invention is to provide a pre-stack time offset denoising method to solve the problems of poor performance, low efficiency and inability to meet current needs in the existing technology that uses FK filtering and suppression based on the apparent velocity within a fixed time window.

[0006] The second objective of this invention is to provide a pre-stack time offset denoising device.

[0007] To achieve the above objectives, the technical solution of the pre-stack time offset denoising method of the present invention is as follows:

[0008] A pre-stack time offset denoising method includes the following steps:

[0009] (1) Process the seismic data to obtain the pre-stack time migration profile, and select the stable wave impedance interface of the work area on the pre-stack time migration profile.

[0010] (2) Based on the stable wave impedance interface of the work area selected in step (1), the pre-stack time migration profile is geologically corrected to obtain the geologically corrected profile.

[0011] (3) Perform FK filtering on the geological correction profile obtained in step (2) to obtain the FK filtered profile;

[0012] (4) Perform wave impedance interface inverse correction on the FK filter processing profile obtained in step (3) to obtain the result profile.

[0013] The pre-stack time migration denoising method of this invention adopts the concept of "geological correction" and combines FK filtering to effectively suppress migration noise. The time window is simple and easy to draw, and the apparent velocity range is small and easy to control, which improves the work efficiency. The profile has good amplitude preservation after denoising, which greatly improves the profile quality.

[0014] Preferably, in step (2), the geological correction involves time alignment processing of the stable wave impedance interface of the work area. More preferably, in the time alignment processing, the alignment time is set to be no less than the maximum time of the stable wave impedance interface of the work area. More preferably, the alignment time is 1.1 to 1.4 times the maximum time.

[0015] Preferably, in step (3), the two-dimensional Fourier transform is performed using the following formula in the FK filtering process:

[0016]

[0017] In the above formula, t is the reflection time, x is the channel position, f is the frequency, and k is the space wavenumber.

[0018] The technical solution of the pre-stack time offset denoising device of the present invention is as follows:

[0019] A pre-stack time-shift denoising device includes:

[0020] The stable wave impedance interface selection module for the work area selects the stable wave impedance interface for the work area based on the pre-stack time offset profile.

[0021] The geological correction module performs geological correction on the stable wave impedance interface of the work area output by the stable wave impedance interface selection module of the work area to obtain a geological correction profile.

[0022] The FK filtering module performs FK filtering on the geological correction profile output by the geological correction module to obtain the FK filtered profile.

[0023] The wave impedance interface inverse correction module performs wave impedance interface inverse correction on the FK filter processing profile output by the FK filter processing module to obtain the result profile.

[0024] The pre-stack time-shift denoising device of the present invention greatly improves the work efficiency compared with the traditional method of using FK filtering and variable time window to suppress based on the apparent velocity, and is less likely to remove effective information, which is beneficial to improving the pre-stack time-shift denoising effect.

[0025] Preferably, in the geological correction module, the geological correction involves time alignment processing of the stable wave impedance interface of the work area. More preferably, in the time alignment processing, the alignment time is set to be no less than the maximum time of the stable wave impedance interface of the work area. More preferably, the alignment time is 1.1 to 1.4 times the maximum time.

[0026] Preferably, in the FK filtering module, a two-dimensional Fourier transform is performed using the following formula:

[0027]

[0028] In the above formula, t is the reflection time, x is the channel position, f is the frequency, and k is the space wavenumber. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the technical process of an embodiment of the present invention.

[0030] Figure 2 This is a pre-stack time offset profile obtained in an embodiment of the present invention;

[0031] Figure 3 This is a time alignment processing profile obtained in an embodiment of the present invention;

[0032] Figure 4 This is an FK-filtered profile obtained by performing FK filtering on the time alignment processing profile in an embodiment of the present invention.

[0033] Figure 5 This is a schematic diagram of the FK filtering process;

[0034] Figure 6 This is the cross-section of the final result obtained in the embodiment of the present invention. Detailed Implementation

[0035] To address the limitations of existing pre-stack time migration denoising methods in complex structural blocks, the inventors broadened their approach and conducted repeated experiments, introducing geological correction concepts into the denoising process. By employing geological correction, multiple time windows are unified, resulting in a single time window that is easily controlled as the apparent velocity range decreases, thus improving operational efficiency. The denoised profile exhibits good amplitude preservation, significantly enhancing profile quality.

[0036] The following description, using a specific area and block as an example and in conjunction with the accompanying drawings, further illustrates the implementation process of the present invention.

[0037] Example 1

[0038] The workflow of the pre-stack time offset denoising method in this embodiment is shown in the diagram below. Figure 1 As shown, it includes the following steps:

[0039] (1) The seismic data of the study area are processed to obtain the pre-stack time migration profile of the area, such as... Figure 2 As shown.

[0040] (2) Select a set of stable wave impedance interfaces in the work area on the pre-stack time migration profile of the seismic data.

[0041] A stable impedance interface in a work area refers to an impedance interface that is continuous and exhibits an arc, with the noise to be processed located above this interface. In this example, using the above method, the selected stable impedance interface for the work area is as follows: Figure 2 As shown by the colored lines.

[0042] (3) Using the geological correction approach, the wave impedance interface selected in step (2) is time-aligned to obtain the time-aligned profile of the pre-stack time migration profile, as shown in the figure. Figure 3 As shown.

[0043] Specifically, time alignment processing aims to align the selected impedance interfaces to the same time, which can be achieved using the HISTA module in the CGG software. Based on the maximum time of the impedance interface determined in step (2) (e.g., 3000ms), the alignment time is set to be no less than this maximum time (e.g., 3500ms). In this example, using the above method, the impedance interface after time alignment processing is shown by the red dashed line. The area circled in yellow solid circle represents the area containing noise.

[0044] In other implementation scenarios, the alignment time can be set to be equal to the maximum time, for example, 3000ms; or the alignment time can be set to 1.1 to 1.4 times the maximum time, for example, 3300, 3800, or 4000ms; thus, the effect can be basically equivalent to that of Example 1.

[0045] (4) Perform FK filtering on the time alignment profile obtained in step (3) to obtain the FK filtered profile diagram, as shown below. Figure 4 As shown.

[0046] Since seismic data comprises both signal and noise, and noise lacks an absolute definition, it represents the unwanted portion of a seismic profile. The purpose of filtering is to eliminate this noise. The FK filtering method separates the signal and noise within the corresponding domain, and then removes the noise portion to obtain the signal portion.

[0047] For a seismic data profile, a two-dimensional Fourier transform can be performed to convert the function f(t, x) expressed in terms of reflection time t and trace location x into a function F(f, k) expressed in terms of frequency f and space wavenumber k, as shown in the following equation:

[0048]

[0049] In the above formula, t is the reflection time, x is the channel position, f is the frequency, k is the space wavenumber, i is the imaginary unit, f(t, x) is the time and position function, and F(f, k) is the frequency and wavenumber function.

[0050] Because the effective wave and the interference wave have different frequency components, they can be represented on the fK plane, such as... Figure 5 As shown, effective signals and interference signals can be clearly distinguished in terms of frequency and visual speed. Therefore, interference can be suppressed and interference areas can be distinguished from effective signal areas by using different frequencies and visual speeds.

[0051] When determining the apparent velocity value, the same set of linear interferences have different apparent velocity values ​​on the superimposed profile. The filtering parameters are determined based on the apparent velocity; the larger the slope, the larger the apparent velocity.

[0052] Figure 4 In the diagram, the red dashed line represents the wave impedance interface after time alignment processing, and the area circled in yellow solid circle represents... Figure 3 The effect after noise reduction in the same area.

[0053] (5) Perform inverse correction on the FK filter profile obtained in step (4) to obtain the result profile, as shown below. Figure 6 As shown.

[0054] Specifically, the principle of inverse correction is to inversely correct the same time to the original impedance interface time, which is the reverse action of the time alignment process in step (3), and is implemented using the HISTA module in the CGG software.

[0055] The cross-section obtained using this embodiment is as follows Figure 6 As shown, comparison Figure 6 and Figure 2 It can be seen that combining geological correction during the pre-stack time migration denoising process can effectively suppress migration noise.

[0056] This operation is simple and easy to perform, requiring only one job, which greatly improves operational efficiency. It also provides excellent amplitude preservation in noise reduction, significantly improving profile quality. Traditional methods for the same seismic data require over 300 jobs, ultimately resulting in a profile that disrupts wavegroup characteristics.

[0057] Example 2

[0058] The pre-stack time migration denoising device of this embodiment corresponds to the implementation process of the method in Embodiment 1, including a work area stable wave impedance interface selection module, a geological correction module, an FK filtering module, and a wave impedance interface inverse correction module. The above modules correspond to the relevant steps in the method of Embodiment 1. The denoising processing of the pre-stack time migration profile can be realized by using the above modules.

Claims

1. A method for de-noising of pre-stack time migration, characterized in that, The method comprises the following steps: (1) processing seismic data to obtain a pre-stack time migration profile, and selecting a stable wave impedance interface in the working area on the pre-stack time migration profile; (2) performing geological correction on the pre-stack time migration profile according to the stable wave impedance interface in the working area selected in step (1) to obtain a geological correction profile; the geological correction is time alignment processing on the stable wave impedance interface in the working area; (3) performing F-K filter processing on the geological correction profile obtained in step (2) to obtain an F-K filter processing profile; (4) performing wave impedance interface inverse correction on the F-K filter processing profile obtained in step (3) to obtain a result profile.

2. The pre-stack time migration denoising method of claim 1, wherein, In the time alignment processing, the alignment time is set to be not less than the maximum time of the stable wave impedance interface in the working area.

3. The pre-stack time migration denoising method of claim 2, wherein, The alignment time is 1.1-1.4 times of the maximum time.

4. The pre-stack time migration denoising method according to any one of claims 1-3, wherein, In step (3), the F-K filter processing is performed by using the following formula for two-dimensional Fourier transform: In the formula, t is the reflection time, x is the trace position, f is the frequency, and k is the spatial wave number.

5. A device for de-noising pre-stack time migration, characterized in that, The method comprises: a stable wave impedance interface selection module for selecting a stable wave impedance interface in the working area according to a pre-stack time migration profile; a geological correction module for performing geological correction on the stable wave impedance interface in the working area output by the stable wave impedance interface selection module to obtain a geological correction profile; the geological correction is time alignment processing on the stable wave impedance interface in the working area; an F-K filter processing module for performing F-K filter processing on the geological correction profile output by the geological correction module to obtain an F-K filter processing profile; a wave impedance interface inverse correction module for performing wave impedance interface inverse correction on the F-K filter processing profile output by the F-K filter processing module to obtain a result profile.

6. The pre-stack time migration denoising apparatus of claim 5, wherein, In the time alignment processing, the alignment time is set to be not less than the maximum time of the stable wave impedance interface in the working area.

7. The pre-stack time migration denoising apparatus of claim 6, wherein, The alignment time is 1.1-1.4 times of the maximum time.

8. The pre-stack time migration denoising apparatus of any one of claims 5-7, wherein, In the F-K filter processing module, two-dimensional Fourier transform is performed by using the following formula: In the formula, t is the reflection time, x is the trace position, f is the frequency, and k is the spatial wave number.

Images