A seismic surface wave fitting method and system

Abstract

The present invention provides a seismic surface wave fitting method and system, which belongs to the field of digital signal processing such as seismic exploration data processing. The seismic surface wave fitting method utilizes the similarity of adjacent surface waves in the original seismic signal gather, performs fitting in the wavelet domain to obtain a fitting waveform, and then subtracts the fitting waveform from the original seismic signal gather to obtain the shot gather data after the surface wave is suppressed. The present invention can obtain the shot gather after the surface wave is suppressed, improves the signal-to-noise ratio of the data, and achieves a better processing effect.

Description

Technical Field

[0001] The invention belongs to the field of digital signal processing such as seismic exploration data processing, and in particular relates to a seismic surface wave fitting method and system. Background Art

[0002] In geophysical signal processing, surface waves are regular interference waves with strong energy and dispersion characteristics in seismic records. When processing seismic data, the suppression effect of surface waves will directly affect the subsequent seismic data processing and migration imaging results, and ultimately affect the geological interpretation effect of seismic data.

[0003] The surface wave field is fan-shaped in pre-stack seismic records, and its energy is extremely strong. In conventional earthquakes, it accounts for about 70% of the total seismic wave energy; in a single-point seismic profile, the energy of the surface wave can account for about 80%, which is very serious. Its existence is often intertwined with the phase axis of the reflection wave, reducing the signal-to-noise ratio of the data, and removing the influence of the surface wave is a difficult problem.

[0004] Chinese patent publication CN110261910A discloses a method for removing surface roll from seismic data based on adaptive sparse S transform, which comprises: obtaining a time-frequency diagram from a seismic record, and determining in the time-frequency diagram an area where surface rolls to be suppressed are located; performing spectrum analysis on the selected area where the surface rolls are located, and determining a corresponding frequency range; performing adaptive sparse S transform on the seismic trace along the t-axis, and transforming it to the tfx domain; performing Fourier transform on the transformed data along the x-axis, and transforming it to the tfk domain; filtering the seismic signal in the tfk domain using a mask function M(f,k); performing inverse Fourier transform and inverse S transform on the filtered result, and returning it to the time domain, and obtaining a result after surface roll denoising on this data; and performing corresponding processing on all the data traces to obtain seismic data after surface roll removal.

[0005] At present, bandpass filtering and FK filtering techniques are mainly used in practical applications, which can effectively suppress surface wave interference to a certain extent. The idea adopted by bandpass filtering is to remove the area containing surface waves. This is the most thorough surface wave elimination technology, but it comes at the cost of losing all other information in the area containing surface waves. FK filtering adopts the method of frequency domain wave number removal filtering, which can ensure the removal of the frequency domain - wave number information where the surface wave is located. This method is an improvement on time domain removal. FK apparent velocity filtering to eliminate surface waves has gradually become active in seismic data processing, and is still the main means of eliminating surface waves in various commercial processing systems. However, its main problems are: 1) effective information with the same frequency domain - wave number domain as the surface wave is also removed; 2) side effects of Fourier transform; 3) the apparent velocity of the surface wave is time-space variable. This feature makes it difficult to accurately provide processing parameters, and may cause incomplete denoising or damage the effective signal in the surface wave area while denoising. Summary of the invention

[0006] The purpose of the present invention is to solve the above-mentioned problems existing in the prior art, to provide a seismic surface wave fitting method and system, to improve the signal-to-noise ratio of data, and to achieve better processing effect.

[0007] The present invention is achieved through the following technical solutions:

[0008] The first aspect of the present invention provides a seismic surface wave fitting method, which utilizes the similarity of adjacent surface waves in the original seismic signal gather to perform fitting in the wavelet domain to obtain a fitting waveform, and then subtracts the fitting waveform from the original seismic signal gather to obtain shot gather data after suppressing the surface waves.

[0009] A further improvement of the present invention is that the method comprises:

[0010] (1) Input the original seismic signal gather and perform wavelet transform on it to decompose it into different wavelet scales;

[0011] (2) For each waveform S on each channel under each wavelet scale, i (t i ) is fitted to obtain the waveform S i (t i ) corresponding fitting waveform;

[0012] (3) Use the fitting waveform of all waveforms to obtain the shot gathering data after suppressing the surface waves.

[0013] A further improvement of the present invention is that the operation of step (2) comprises:

[0014] (21) Search for the waveform S i (t i )The waveform with the largest similarity coefficient;

[0015] (22) All the waveforms S i (t i ) The waveforms with the largest similarity coefficient are superimposed and averaged to obtain the fitted waveform.

[0016] A further improvement of the present invention is that the operation of step (21) comprises:

[0017] Let S i (t i ) is a waveform on the ith channel under a certain wavelet scale, t i is the time at which the peak of the waveform occurs;

[0018] Using the steepest descent method in S i (t i ) Search for the road on the left and right sides that matches Si (t i )The waveform with the largest similarity coefficient;

[0019] All waveforms with the largest similarity coefficients form waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1.

[0020] A further improvement of the present invention is that the operation of step (22) includes:

[0021] Use the following formula to calculate S I (t I ) and calculate the average value to get the fitting waveform

[0022]

[0023] Among them, t∈(T i1 , T i2 ),T i1 , T i2 The waveform S i (t i ), 2n-1 is the number of fitting channels.

[0024] The operation of step (3) includes:

[0025] The fitted waveforms of all waveforms are subtracted from the original seismic signal gathers to obtain the shot gather data after surface wave suppression.

[0026] A second aspect of the present invention provides a seismic surface wave fitting system, the system comprising:

[0027] Wavelet transform unit, used for performing wavelet transform on the input original seismic signal gather to decompose it into different wavelet scales;

[0028] A fitting unit is connected to the wavelet transform unit and is used to fit each waveform S on each channel under each wavelet scale in turn. i (t i ) is fitted to obtain the waveform S i (t i ) corresponding fitting waveform;

[0029] The output unit is connected to the fitting unit and is used to obtain the shot gathering data after suppressing the surface wave by using the fitting waveforms of all the waveforms.

[0030] A further improvement of the present invention is that the fitting unit performs the following operations:

[0031] Let S i (t i) is a waveform on the ith channel under a certain wavelet scale, t i is the time at which the peak of the waveform occurs;

[0032] Using the steepest descent method in S i (t i ) Search for the road on the left and right sides that matches S i (t i )The waveform with the largest similarity coefficient;

[0033] All waveforms with the largest similarity coefficients form waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1;

[0034] Use the following formula to calculate S I (t I ) and calculate the average value to get the fitting waveform

[0035]

[0036] Among them, t∈(T i1 , T i2 ),T i1 , T i2 The waveform S i (t i ), 2n-1 is the number of fitting channels.

[0037] A further improvement of the present invention is that the output unit performs the following operations:

[0038] The fitted waveforms of all waveforms are subtracted from the original seismic signal gathers to obtain the shot gather data after surface wave suppression.

[0039] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps in the seismic surface wave fitting method.

[0040] Compared with the prior art, the beneficial effects of the present invention are: the present invention can be used to obtain shot gathers after suppressing surface waves, thereby improving the signal-to-noise ratio of data and achieving a better processing effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flowchart of the steps of the method of the present invention;

[0042] Figure 2-1 Simulate earthquake records;

[0043] Figure 2-2The fitted surface wave record;

[0044] Figure 2-3 Reflected wave record after subtracting the surface wave;

[0045] Figure 3-1 The original single shot containing the surface wave;

[0046] Figure 3-2 Commercial software FK visual velocity denoising results;

[0047] Figure 3-3 The method of the present invention removes the surface wave results;

[0048] Figure 3-4 The surface waves and scattered noise fitted by the method of the present invention;

[0049] Figure 4 A schematic diagram of the structure of the system of the present invention. DETAILED DESCRIPTION

[0050] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0051] Reasonable suppression of surface waves can help improve the signal-to-noise ratio of seismic signals, improve the results of migration imaging, and facilitate inversion and interpretation.

[0052] The present invention utilizes the strong similarity of surface rolls of adjacent channels to fit them in the wavelet domain, and finally subtracts the fitted surface rolls from the original seismic signal gathers to obtain shot gather data with a high signal-to-noise ratio after suppressing the surface rolls.

[0053] like Figure 1 As shown, the method of the present invention comprises:

[0054] (1) Input the original seismic signal gather and perform wavelet transform on it to decompose it into different wavelet scales;

[0055] The existing wavelet transform technology is used for wavelet transform, which is briefly introduced as follows:

[0056] Assume x(t) is a square integrable function (x(t)∈L 2 (R)), is a function of the basic wavelet or mother wavelet. Then

[0057]

[0058] It is called the wavelet transform of x(t). Where a>0 is the scale factor, and τ reflects the displacement, which can be positive or negative. Symbol<x,y> stands for inner product, and its meaning is (superscript * stands for conjugate)

[0059] <x(t),y(t)>=∫x(t),y* (t)dt (2)

[0060] It is the displacement and scale expansion of the basic wavelet. In formula (1), not only t is a continuous variable, but also a and τ are continuous variables, so it is called continuous wavelet transform (CWT).

[0061] The original seismic signal gather, i.e., waveform seismic data, is based on waveforms. In order to transform the original waveform, the surface waves are concentrated on some scales, so as to facilitate tracking and fitting. The original seismic signal gather is x(t) in formula (1). Formula (1) is used to obtain the signal after wavelet transformation. This signal is the signal after multi-scale wavelet decomposition. Obtaining this signal means that the surface waves have been decomposed into different wavelet scales.

[0062] (2) Fit each waveform on each channel under each wavelet scale in turn to obtain the fitting waveform corresponding to each waveform: The process of fitting the surface wave is carried out in units of one waveform. Fit each waveform on each channel under each wavelet scale in turn to obtain the fitting waveform of each waveform, as follows:

[0063] (21) Search and waveform S i (t i )The waveform with the largest similarity coefficient:

[0064] Let S i (t i ) is a waveform on the ith channel at a certain wavelet scale (the following descriptions are all at a certain wavelet scale and are not mentioned separately for brevity), t i is the time when the peak value of the waveform is located. In the process of searching, the steepest descent method is used to search upward or downward on the left and right adjacent tracks to find the adjacent track with the above waveform S. i (t i ) The waveform with the largest similarity coefficient, namely S i-1 (t i-1 ), S i+1 (t i+1 ), where i-1, i+1 are the left and right lanes beside the i-th lane.

[0065] Continue searching to the left and right sides according to the above method to obtain the waveform S i (t i ) is the center of the waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1.

[0066] (22) All the waveforms S i (t i) The waveforms with the largest similarity coefficient are superimposed and averaged to obtain the fitted waveform:

[0067] Since the surface waves of adjacent channels have strong similarity, that is, the shape and energy of the surface waves in each channel of the adjacent channels in the original data are the same, the waveform S I (t I ) are the surface wave components, while the different parts are other components containing effective information. To this end, the following formula is used to calculate S I (t I ) and find their mean:

[0068]

[0069] In the formula, t∈(T i1 , T i2 ),T i1 , T i2 is a waveform S of the ith channel i (t i ), 2n-1 is the number of fitting channels, and the value of the number of fitting channels can be set in advance. The waveform obtained at this time is Waveform S i (t i )’s fitted waveform.

[0070] (3) Using the fitted waveforms of all waveforms to obtain the shot gather data after suppressing the surface waves: Subtracting the fitted waveforms of all waveforms from the original seismic signal gathers to obtain the shot gather data after suppressing the surface waves.

[0071] like Figure 4 As shown, the second aspect of the present invention provides a seismic surface wave fitting system, the system comprising:

[0072] The wavelet transform unit 10 is used to perform wavelet transform on the input original seismic signal gather to decompose it into different wavelet scales;

[0073] The fitting unit 20 is connected to the wavelet transform unit 10 and is used to sequentially fit each waveform S on each channel under each wavelet scale. i (t i ) is fitted to obtain the waveform S i (t i ) corresponding fitting waveform;

[0074] The output unit 30 is connected to the fitting unit 20 and is used to obtain the shot gather data after suppressing the surface waves by using the fitting waveforms of all the waveforms.

[0075] A further improvement of the present invention is that the fitting unit performs the following operations:

[0076] Let Si (t i ) is a waveform on the ith channel under a certain wavelet scale, t i is the time at which the peak of the waveform occurs;

[0077] Using the steepest descent method in S i (t i ) Search for the road on the left and right sides that matches S i (t i )The waveform with the largest similarity coefficient;

[0078] All waveforms with the largest similarity coefficients form waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1;

[0079] Use the following formula to calculate S I (t I ) and calculate the average value to get the fitting waveform

[0080]

[0081] Among them, t∈(T i1 , T i2 ),T i1 , T i2 The waveform S i (t i ), 2n-1 is the number of fitting channels.

[0082] A further improvement of the present invention is that the output unit performs the following operations:

[0083] The fitted waveforms of all waveforms are subtracted from the original seismic signal gathers to obtain the shot gather data after surface wave suppression.

[0084] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps in the seismic surface wave fitting method.

[0085] Embodiments of the present invention are as follows:

[0086] [Example 1]

[0087] Tested by model data:

[0088] Synthesize seismic gathers with surface wave records, such as Figure 2-1 As shown, Figure 2-2is the surface wave fitted by this method. It can be seen that the fitted surface wave shape is almost completely consistent with the surface wave shape in the synthetic gather. Subtracting the fitted surface wave record from the simulated seismic record will obtain the reflection wave signal, such as Figure 2-3 shown.

[0089] Through the previous simulation of seismic data processing, this method has the following characteristics:

[0090] 1) The local energy of the surface wave is strong, and the energy of adjacent traces is close. The similar waveform is the application basis of this algorithm;

[0091] 2) Two-dimensional wavelet transform creates good conditions for fitting surface waves; the steepest descent method accurately searches for the direction of surface waves;

[0092] 3) The present invention adopts an optimal denoising strategy. The algorithm overcomes the problem of inaccurate surface wave elimination in traditional methods and minimizes the damage to effective waves.

[0093] [Example 2]

[0094] Actual earthquake data test:

[0095] This method has also been tested with actual seismic data, and the surface wave suppression effect is obvious. Figure 3-1 to Figure 3-4 As shown, Figure 3-1 It is a primitive single shot containing surface waves; Figure 3-2 The commercial software FK apparent velocity denoising results show that after removing the surface waves, the residual surface waves are exposed, and the removal is not very clean; Figure 3-3 This is the result of the present invention removing the surface wave, the effective reflected wave is fully revealed, and there is almost no residual noise; Figure 3-4 This is the surface roll fitted by the present invention. By comparing with the original record, the fitted noise is more accurate.

[0096] Finally, it should be explained that the above technical solution is only one implementation method of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the application methods and principles disclosed in the present invention, and it is not limited to the method described in the above specific implementation method of the present invention. Therefore, the method described above is only preferred and does not have a restrictive meaning.

Claims

1. A seismic surface wave fitting method, characterized in that: The method utilizes the similarity of adjacent surface waves in the original seismic signal gathers to perform fitting in the wavelet domain to obtain a fitting waveform, and then subtracts the fitting waveform from the original seismic signal gathers to obtain shot gather data after suppressing the surface waves; The method comprises: (1) Input the original seismic signal gather and perform wavelet transform on it to decompose it into different wavelet scales; (2) For each waveform S on each channel under each wavelet scale, i (t i ) is fitted to obtain the waveform S i (t i ) corresponding fitting waveform; (3) using the fitted waveforms of all waveforms to obtain the shot gather data after suppressing the surface waves; The operation of step (2) includes: (21) Search for the waveform S i (t i )The waveform with the largest similarity coefficient; (22) All the waveform S i (t i ) The waveforms with the largest similarity coefficient are superimposed and averaged to obtain the fitted waveform; The operation of step (21) includes: Let S i (t i ) is a waveform on the ith channel under a certain wavelet scale, t i is the time at which the peak of the waveform occurs; Using the steepest descent method in S i (t i ) Search for the road on the left and right sides that matches S i (t i )The waveform with the largest similarity coefficient; All waveforms with the largest similarity coefficients form waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1.

2. The seismic surface wave fitting method according to claim 1, characterized in that: The operation of step (22) includes: Use the following formula to calculate S I (t I ) and calculate the average value to get the fitting waveform Among them, t∈(T i1 , T i2 ),T i1 , T i2 The waveform S i (t i ), 2n-1 is the number of fitting channels.

3. The seismic surface wave fitting method according to claim 1, characterized in that: The operation of step (3) includes: The fitted waveforms of all waveforms are subtracted from the original seismic signal gathers to obtain the shot gather data after surface wave suppression.

4. A seismic surface wave fitting system, characterized in that: The system comprises: Wavelet transform unit, used for performing wavelet transform on the input original seismic signal gather to decompose it into different wavelet scales; A fitting unit is connected to the wavelet transform unit and is used to fit each waveform S on each channel under each wavelet scale in turn. i (t i ) is fitted to obtain the waveform S i (t i ) corresponding fitting waveform; an output unit connected to the fitting unit, and used to obtain shot gather data after suppressing the surface wave by using the fitting waveforms of all waveforms; The fitting unit performs the following operations: Let S i (t i ) is a waveform on the ith channel under a certain wavelet scale, t i is the time at which the peak of the waveform occurs; Using the steepest descent method in S i (t i ) Search for the road on the left and right sides that matches S i (t i )The waveform with the largest similarity coefficient; All waveforms with the largest similarity coefficients form waveform S I (t I ), I=in, i-n+1,..., i+n, 2n+1; Use the following formula to calculate S I (t I ) and calculate the average value to get the fitting waveform Among them, t∈(T i1 , T i2 ),T i1 , T i2 The waveform S i (t i ), 2n-1 is the number of fitting channels.

5. The seismic surface wave fitting system according to claim 4, characterized in that: The output unit performs the following operations: The fitted waveforms of all waveforms are subtracted from the original seismic signal gathers to obtain the shot gather data after surface wave suppression.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps in the seismic surface wave fitting method as described in any one of claims 1 to 3.

Images