Marchenko seismic source positioning method based on signal envelope

Through the Marchenko seismic source positioning method based on signal envelope, the virtual gun set is reconstructed using Hilbert transform and convolutional operation, combined with frequency domain filtering technology, the problems of multiple wave interference and noise influence in traditional methods are solved, and the problem of improvement is improved. Accuracy of the focal point positioning.

CN118915156AInactive Publication Date: 2024-11-08JILIN UNIVERSITY
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411404379.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-10
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The traditional source positioning method has low accuracy in the source positioning due to multiple wave interference and noise.

Method used

The Marchenko seismic source positioning method based on signal envelope is adopted to extract the envelope information of seismic data through Hilbert transform, and the virtual gun set is reconstructed by convolutional operation. The initial velocity model and the Marchenko method are used to obtain the underground profile data, and the frequency is Gaussian high-pass filter is used in the domain to enhance the source signal.

Benefits of technology

It effectively avoids multiple wave interference, improves the accuracy of source positioning, reduces noise interference, and enhances the relative strength of source signal.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915156A_ABST
    Figure CN118915156A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of geophysical exploration, and particularly relates to a Marchenko seismic source positioning method based on signal envelope, which comprises the following steps: uniformly arranging detectors on the ground surface, and collecting seismic data sent by an underground seismic source, the seismic data comprising a plurality of seismic trace data; according to the seismic channels, Hilbert transform is adopted to extract envelopes of data of each seismic channel to form envelope information of the seismic data; performing convolution operation on the envelope information of the seismic data to reconstruct a virtual shot gather; calculating travel time information by adopting an initial speed model, carrying out convolution on the travel time information and a seismic source wavelet to estimate a direct wave field, and obtaining underground profile data containing seismic source information by applying a Marchenko method to a virtual shot gather; and converting the underground profile data into a frequency domain, and enhancing seismic source information in the underground profile in the frequency domain by adopting a filtering technology to realize seismic source positioning. The problem of low seismic source positioning precision caused by multiple interference is solved, and the seismic source positioning precision is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of geophysical exploration, and in particular relates to a Marchenko earthquake source positioning method based on signal envelope. Background Art

[0002] In oil and gas production, factors such as oil and gas migration, hydraulic fracturing, and underground mining may induce earthquake events such as fault zone collapse, thus affecting the safety of oil and gas wells. Therefore, earthquake source location is not only the basis for studying natural earthquakes, but also has important significance for promoting oil and gas production efficiently and safely.

[0003] Traditional methods often use travel time-based inversion methods to achieve source location, which rely on accurate first arrival picking. In recent years, a class of waveform-based location methods has emerged, which can detect and locate earthquake events by utilizing the migration and stacking techniques commonly used in seismic exploration. The basic principle of this type of waveform method is to achieve source location by reconstructing the source energy into the underground grid model using specific migration or imaging operators without the need for first arrival picking.

[0004] However, after the underground earthquake source excites seismic waves, the seismic waves form multiple waves due to multiple reflections in the strata, which will interfere with the accurate positioning of the earthquake source. In addition, the seismic waves will be affected by a large amount of random noise during the propagation process, reducing the signal-to-noise ratio. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a Marchenko source locating method based on signal envelope to solve the problem of low source locating accuracy caused by the interference of multiple waves.

[0006] The implementation method of the present invention is: A Marchenko source location method based on signal envelope, the method comprising: S1, evenly arranging detectors on the surface and collecting seismic data emitted by underground seismic sources, wherein the seismic data includes a plurality of seismic trace data; S2, according to the seismic trace, using Hilbert transform to extract the envelope information of the seismic data composed of the envelope of each seismic trace data; S3, reconstructing the envelope information of the seismic data into a virtual shot gather by convolution operation; S4, using the initial velocity model to calculate the travel time information and convolve it with the source wavelet to estimate the direct wave field, and then using the Marchenko method on the virtual shot gather to obtain the underground profile data containing the source information; S5, converting the underground profile data into the frequency domain, and using filtering technology in the frequency domain to enhance the earthquake source information in the underground profile to achieve earthquake source location.

[0007] Furthermore, S1 specifically includes: a number of geophones are arranged at equal intervals on the survey line on the surface of the earth, and each geophone measures a one-dimensional signal, which is recorded as , For the Seismic channel data, the value is , For time, is the number of seismic traces.

[0008] Furthermore, S2 specifically includes: for each seismic trace data collected in S1, extracting the envelope by using Hilbert transform, the Hilbert transform method is: , in, For earthquake The complex-valued signal after Hilbert transform, To represent pi, use As an integral variable, the envelope information of seismic data is expressed as: , in, It is the envelope information of seismic data; Hilbert transform is used to extract the envelope of each seismic trace to obtain the envelope information of the seismic data.

[0009] Further, S3 specifically includes: performing convolution operation on the envelope information of the seismic data to reconstruct the virtual shot gather at each detector, and when reconstructing the virtual shot gather at the first detector, calculating the envelope information of the seismic trace received by the first detector The convolution result between the envelope information of the seismic traces of all the geophones is used to obtain the virtual shot gather located at the first geophone. When reconstructing the second virtual shot gather, the envelope information of the seismic traces received by the second geophone is used. Perform convolution operation with the envelope information of all geophone seismic traces one by one to obtain the virtual shot gather at the second geophone; The above process is repeated until the virtual shot gathers at all the geophones are calculated.

[0010] Furthermore, S5 specifically includes: The subsurface profile data are converted into subsurface profile data in the frequency domain by using a two-dimensional fast Fourier transform; in the frequency domain, the subsurface profile data are decomposed into different frequency components; The Gaussian high-pass filter is used to filter out the low-frequency components in the underground profile data in the frequency domain to obtain the processed frequency information. The transfer function of the Gaussian high-pass filter is expressed as: , in, is a frequency domain point To frequency center The distance and Represent the frequency components in the horizontal and vertical directions respectively, and is the center frequency coordinate in the frequency domain, Responsible for controlling the cutoff frequency of the filter.

[0011] The processed frequency information is converted back to the spatial domain by using two-dimensional inverse fast Fourier transform to obtain the underground profile data of the enhanced source signal; the source is located by using the underground profile data of the enhanced source signal.

[0012] Furthermore, a cutoff frequency is set for the Gaussian high-pass filter to filter out low-frequency components in the underground profile data.

[0013] Furthermore, the underground seismic source is a passive seismic source.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention is different from the direct positioning method of seismic data. It extracts low-frequency envelope information to filter out high-frequency noise, and then uses the envelope information for earthquake source positioning. It uses a convolution-based method to reconstruct virtual shot collections to convert passive source data into a form similar to active source data as the data basis of the Marchenko method, and removes low-frequency signals by converting between the space domain and the frequency domain. The present invention effectively avoids the interference of multiple waves (i.e., seismic waves that are reflected multiple times in the underground strata), and further uses frequency domain filtering technology to filter out low-frequency components in the seismic profile, thereby improving the accuracy of earthquake source positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for locating a seismic source provided by an embodiment of the present invention; Figure 2 A schematic diagram of a method for observing a source signal and measured seismic data provided by an embodiment of the present invention, wherein (a) is a schematic diagram of seismic detectors evenly distributed on a survey line, and (b) is a schematic diagram of seismic data measured by the detectors; Figure 3 A schematic diagram of a true velocity model provided by an embodiment of the present invention; Figure 4 A total of 481 seismic data obtained by forward modeling using the finite difference method provided in the embodiment of the present invention; Figure 5 A comparison between the 121st seismic data and its envelope information amplitude provided by an embodiment of the present invention, wherein (a) is the 121st seismic data, and (b) is the envelope information corresponding to the seismic data; Figure 6 Envelope information of seismic data provided by an embodiment of the present invention; Figure 7 Partial results of 481 reconstructed virtual shot gathers provided by the embodiment of the present invention, wherein (a), (b) and (c) are the 61st, 121st and 181st virtual shot gathers reconstructed from seismic data, respectively, and (d), (e) and (f) are the 61st, 121st and 181st virtual shot gathers reconstructed from envelope information, respectively; Figure 8 A schematic diagram of a smoothed velocity model provided by an embodiment of the present invention; Fig. 9 A schematic diagram of a portion of direct wave fields provided in an embodiment of the present invention, wherein (a), (b) and (c) are direct wave fields from points (1800m, 300m), (1800m, 600m) and (1800m, 900m) to the surface, respectively; Fig.10 An underground profile including earthquake source information provided by an embodiment of the present invention; Fig.11 A comparison of the earthquake source location results provided in the embodiments of the present invention, wherein (a) is the earthquake source location result of the virtual shot gather reconstructed from the seismic data, and (b) is the earthquake source location result of the method of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0017] See also Figure 1 As shown, a Marchenko source location method based on signal envelope, the method includes: S1, geophones are evenly arranged on the surface to collect seismic data from underground sources, including multiple seismic trace data; S2, according to the seismic trace, using Hilbert transform to extract the envelope information of the seismic data composed of the envelope of each seismic trace data; S3, reconstructing the envelope information of the seismic data into a virtual shot gather by convolution operation; S4, using the initial velocity model to calculate the travel time information and convolve it with the source wavelet to estimate the direct wave field, and then using the Marchenko method on the virtual shot gather to obtain the underground profile data containing the source information; S5, converting the underground profile data into the frequency domain, and using filtering technology in the frequency domain to enhance the earthquake source information in the underground profile to achieve earthquake source location.

[0018] Among them, the underground source refers to the passive source, through the process of S1, see Figure 2 As shown, according to the requirements of the Marchenko method for data format, Figure 2 The method shown in (a) is to place a number of geophones at equal intervals on the surface survey line, and synchronously collect the seismic wave signals emitted by the underground source. Each geophone will measure a one-dimensional signal. Figure 2 The one-dimensional signal shown in (b) is denoted as , For the Seismic channel data, the value is , For time, is the number of seismic traces, Figure 2 Each seismic trace data in (b) is expressed as: .

[0019] For each seismic trace data collected in S1, S2 uses Hilbert transform to extract the envelope. The Hilbert transform method is: , in, For earthquake The complex-valued signal after Hilbert transform, To express pi, we use As an integral variable, the envelope information of seismic data is expressed as: , in, It is the envelope information of seismic data; Hilbert transform is used to extract the envelope of each seismic trace to obtain the envelope information of the seismic data.

[0020] In this process, the envelope information of seismic data includes the complex value signal after Hilbert transform and the original seismic trace data.

[0021] S3 specifically includes: performing convolution operation on the envelope information of the seismic data to reconstruct the virtual shot gather at each geophone, and when reconstructing the virtual shot gather at the first geophone, calculating the envelope information of the seismic trace received by the first geophone The convolution result between the envelope information of all the geophone seismic traces is used to obtain the virtual shot gather located at the first geophone. When reconstructing the second virtual shot gather, the envelope information of the seismic trace received by the second geophone is used. Perform convolution operation with the envelope information of all geophone seismic traces one by one to obtain the virtual shot gather at the second geophone; The above process is repeated until the virtual shot gathers at all the geophones are calculated.

[0022] In S4, first, the smooth velocity model of the underground is estimated by velocity tomography and the finite difference method is used to perform the direct wave field from the underground point to the surface detector. The direct wave field and the virtual shot gather are used as the input data of the Marchenko method.

[0023] The focusing function in the Marchenko method is divided into two parts: the downlink focusing function and the uplink focusing function. The downlink focusing function can be injected from the surface into the underground and focused on a certain point underground, while the uplink aggregation function can be regarded as the reflected wave field of the downlink focusing function after being acted on by the underground medium. In the aggregation function calculation stage, the Marchenko method iteratively performs convolution-type integral operations on the direct wave field and the virtual gun set to obtain converged uplink aggregation functions and downlink focusing functions; then, the uplink Green's function and the downlink Green's function are calculated through the information of the uplink aggregation function, the downlink aggregation function and the virtual gun set; finally, the uplink Green's function and the downlink Green's function are cross-correlated to obtain underground profile data containing source information. The specific calculation method of this process is an existing method and will not be repeated in this embodiment, but it is noted that the data used here is the envelope information of the seismic data.

[0024] Through the above process, the envelope information is mainly the low-frequency component of the seismic data, which also brings low-frequency noise. The present invention uses two-dimensional fast Fourier transform to convert the underground profile data originally in the spatial domain into underground profile data in the frequency domain. In the underground profile data in the frequency domain, the source signal and noise will correspond to different frequencies.

[0025] In the frequency domain, the source signal appears as a higher frequency component, while other types of noise, such as background noise or cross-correlation noise, are usually concentrated in the lower frequency range. The present invention utilizes this feature to enhance the source signal through frequency domain filtering technology. Specifically, the method uses a Gaussian high-pass filter to filter out low-frequency components. The Gaussian high-pass filter is used to suppress the information of the low-frequency part in the frequency domain, while retaining or enhancing the information of the high-frequency part, thereby effectively enhancing the relative strength of the source signal.

[0026] After the high-pass filtering process of the Gaussian high-pass filter, the underground profile information in the frequency domain has been optimized, and the source signal has become more obvious. Next, in order to convert these processed frequency information back to the spatial domain to obtain the underground profile data of the enhanced source signal, it is necessary to apply a two-dimensional inverse fast Fourier transform. Finally, the processed spatial domain underground profile is obtained, which can intuitively and more accurately reflect the location of the source.

[0027] S5 specifically includes: The underground profile data containing earthquake source information is converted into underground profile data in the frequency domain by using two-dimensional fast Fourier transform; in the frequency domain, the underground profile data is decomposed into different frequency components; The Gaussian high-pass filter is used to filter out the low-frequency components in the underground profile data in the frequency domain to obtain the processed frequency information. The transfer function of the Gaussian high-pass filter is expressed as: , in, is a frequency domain point To frequency center The distance and Represent the frequency components in the horizontal and vertical directions respectively, and is the center frequency coordinate in the frequency domain, Responsible for controlling the cutoff frequency of the filter.

[0028] The processed frequency information is converted back to the spatial domain by using two-dimensional inverse fast Fourier transform to obtain the underground profile data of the enhanced source signal; the source is located by using the underground profile data of the enhanced source signal.

[0029] Application Examples, S1, geophones are evenly arranged on the surface to collect seismic data emitted by underground seismic sources, wherein the seismic data includes a plurality of seismic trace data.

[0030] The real speed model used in this embodiment is as follows: Figure 3 As shown, the true velocity model has 4 layers, and the velocity values ​​from top to bottom are 1500m / s, 3500m / s, 2500m / s and 3500m / s respectively. The number of grid points of the true velocity model is 401×481, the grid spacing is 5m×5m, and the model size is 2000m×2400m.

[0031] In order to verify the effect of the method of the present invention, two seismic sources were arranged underground in this embodiment, and the locations of the seismic sources were (1500m, 600m) and (1500m, 1800m) respectively. The finite difference method was used to solve the acoustic wave equation to obtain seismic data, and 481 geophones were arranged at equal intervals on the surface to collect seismic data emitted by the underground seismic sources. The spacing between each geophone was 5m, the total recording time was 3s, the sampling time interval was 0.002s, and the sampling point was 1501. The obtained data has a total of 481 channels, as shown in Figure 1. Figure 4 shown.

[0032] S2, according to the seismic trace, using Hilbert transform to extract the envelope information of the seismic data composed of the envelope of each seismic trace data; In this embodiment, Hilbert transform is used to extract the envelope of the seismic data of each seismic channel. Figure 5 As shown in (a) in the figure, its envelope information is as follows Figure 5 As shown in (b) in the figure, it can be seen that in addition to the strong signal generated by the reflection interface, there is also a lot of noise in the seismic data, while the envelope of the seismic data is smoother and has less noise.

[0033] After using the Hilbert transform, the envelope information of the seismic data can be obtained, such as Figure 6 As shown. Figure 4 It can be seen that Figure 6 Reserved Figure 4 The stronger phase axis is obtained, while a lot of noise is removed.

[0034] Step S3, performing convolution operation on the envelope information of the seismic data to reconstruct a virtual shot gather.

[0035] In order to conduct comparative experiments, this embodiment uses convolution to Figure 4 The earthquake data shown and Figure 6 The envelope information of the seismic data shown in FIG. 1 reconstructs the corresponding virtual shot collections, each of which contains 481 shots. Among them, the virtual shot collections of the 61st shot, the 121st shot, and the 181st shot reconstructed from the seismic data are shown in FIG. Figure 7 (a) Figure 7 (b) and Figure 7 As shown in (c) in the figure, the virtual shot collections of the 61st, 121st, and 181st shots reconstructed from the envelope information of the seismic data are shown in Figure 1. Figure 7 (d) Figure 7 (e) and Figure 7 As shown in (f) in the figure, it can be seen that the virtual shot gather reconstructed from the envelope information of the seismic data also contains the main information in the virtual shot gather reconstructed from the seismic data.

[0036] Step S4, using the initial velocity model to calculate travel time information and convolve it with the source wavelet to estimate the direct wave field, and then using the Marchenko method on the virtual shot gather to obtain underground profile data containing source information; Finite difference forward modeling can be performed on the smooth velocity model to obtain the direct wave field from each grid point in the smooth velocity model to the surface detector. The smooth velocity model used in this embodiment is as follows: Figure 8 As shown in Figure 2, the smoothed velocity model is consistent with the true velocity model Figure 3The forward modeling parameters when calculating the direct wave field from each grid point to the surface detector are the same as those in step S1. The total recording time is 3s, the sampling time interval is 0.002s, the sampling points are 1501, and there are 481 seismic traces. The direct wave fields from points (1800m, 300m), (1800m, 600m) and (1800m, 900m) to the surface are as follows: Fig. 9 (a) Fig. 9 (b) and Fig. 9 As shown in (c) in .

[0037] These direct wave fields and the virtual shot gathers reconstructed in step S3 are used as input data of the Marchenko method, and the following can be calculated: Fig.10 The underground profile is shown. It can be seen that there are some abnormal information at about (1500m, 600m) and (1500m, 1800m) in the underground profile, but there is also some low-frequency noise.

[0038] Step S5, converting the underground profile data containing the earthquake source information into the frequency domain, and using filtering technology in the frequency domain to enhance the earthquake source information in the underground profile to achieve earthquake source positioning.

[0039] Using two-dimensional fast Fourier transform Fig.10 The underground profile data containing earthquake source information is converted into the frequency domain, and the low-frequency noise is filtered out by using a Gaussian high-pass filter, so that more precise earthquake source information can be extracted to achieve earthquake source location. In order to verify the effectiveness of the present invention, this embodiment uses a virtual shot collection reconstructed from seismic data to perform an earthquake source location experiment as a control group. The results are shown in Figure 2. Fig.11 As shown in (a), the result of the method of the present invention is as follows Fig.11 As shown in (b) of Fig.11 (a) also has the location of the earthquake source, but also has stronger noise, which may interfere with the judgment of the earthquake source location in complex situations; correspondingly, the results obtained by the present invention Fig.11 (b) has less noise and is more conducive to determining the source location.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A Marchenko source location method based on signal envelope, characterized in that: The method includes: S1, evenly arranging detectors on the surface and collecting seismic data emitted by underground seismic sources, wherein the seismic data includes a plurality of seismic trace data; S2, according to the seismic trace, using Hilbert transform to extract the envelope information of the seismic data composed of the envelope of each seismic trace data; S3, reconstructing the envelope information of the seismic data into a virtual shot gather by convolution operation; S4, using the initial velocity model to calculate the travel time information and convolve it with the source wavelet to estimate the direct wave field, and then using the Marchenko method on the virtual shot gather to obtain the underground profile data containing the source information; S5, converting the underground profile data into the frequency domain, and using filtering technology in the frequency domain to enhance the earthquake source information in the underground profile to achieve earthquake source location.

2. The Marchenko source location method based on signal envelope according to claim 1, characterized in that: S1 specifically includes: a number of geophones are arranged at equal intervals on the survey line on the surface of the earth, and each geophone measures a one-dimensional signal, which is recorded as , For the Seismic channel data, the value is , For time, is the number of seismic traces.

3. The Marchenko source location method based on signal envelope according to claim 1, characterized in that: S2 specifically includes: for each seismic trace data collected in S1, extracting the envelope using Hilbert transform, the Hilbert transform method is: , in, For earthquake The complex-valued signal after Hilbert transform, To express pi, we use As an integral variable, the envelope information of seismic data is expressed as: , in, It is the envelope information of seismic data; Hilbert transform is used to extract the envelope of each seismic trace to obtain the envelope information of the seismic data.

4. The Marchenko source location method based on signal envelope according to claim 3, characterized in that: S3 specifically includes: performing convolution operation on the envelope information of the seismic data to reconstruct the virtual shot gather at each geophone, and when reconstructing the virtual shot gather at the first geophone, calculating the envelope information of the seismic trace received by the first geophone The convolution result between the envelope information of the seismic traces of all the geophones is used to obtain the virtual shot gather located at the first geophone. When reconstructing the second virtual shot gather, the envelope information of the seismic traces received by the second geophone is used. Perform convolution operation with the envelope information of all geophone seismic traces one by one to obtain the virtual shot gather at the second geophone; The above process is repeated until the virtual shot gathers at all the geophones are calculated.

5. The Marchenko source location method based on signal envelope according to claim 1, characterized in that: S5 specifically includes: The subsurface profile data are converted into subsurface profile data in the frequency domain by using a two-dimensional fast Fourier transform; in the frequency domain, the subsurface profile data are decomposed into different frequency components; A Gaussian high-pass filter is used to filter out the low-frequency components in the underground profile data in the frequency domain to obtain the processed frequency information. The transfer function of the Gaussian high-pass filter is expressed as: , in, is a frequency domain point To frequency center The distance and Represent the frequency components in the horizontal and vertical directions respectively, and is the center frequency coordinate in the frequency domain, Responsible for controlling the cutoff frequency of the filter; The processed frequency information is converted back to the spatial domain by using two-dimensional inverse fast Fourier transform to obtain the underground profile data of the enhanced source signal; the source is located by using the underground profile data of the enhanced source signal.

6. The Marchenko source location method based on signal envelope according to claim 1, characterized in that: A cutoff frequency is set for the Gaussian high-pass filter to filter out the low-frequency components in the subsurface profile data.

7. The Marchenko source location method based on signal envelope according to claim 1, characterized in that: The underground seismic source is a passive seismic source.

Citation Information

Patent Citations

  • Kirchhoff integration solution-based seismic channel wave depth migration imaging method

    CN110850469A

  • Multi-scale fusion feature enhancement algorithm based on Gaussian filter feedback and hole convolution

    CN111967524A

  • Marchenko imaging-based micro-seismic source positioning method

    CN118131322A

  • Geophone depth calibration

    US20210199829A1