Phase reconstruction method and system based on multi-wavelet decomposition, and storage medium

Abstract

The invention provides a phase reconstruction method and system based on multi-wavelet decomposition and a storage medium, and belongs to the technical field of geophysical exploration. The method comprises the following steps: firstly, decomposing a seismic signal into wavelets with different shapes according to phases; the phase reconstructed seismic signals are then selectively screened. When the method is used for decomposing the signals, the number of iterations can be effectively reduced, the calculation speed is improved, and the signals are decomposed relatively accurately; when signal reconstruction is carried out, a proper phase range is selected based on phase characteristics in combination with other data information, and signal reconstruction is carried out in a targeted manner.

Description

Technical Field

[0001] The present invention belongs to the technical field of geophysical exploration, and in particular relates to a phase reconstruction method, system and storage medium based on multi-wavelet decomposition. Background Art

[0002] Phase is an important characteristic of a signal. With the development of exploration technology, phase has become increasingly important in the field of seismic exploration. Effectively extracting phase information from seismic data and performing phase analysis and processing have always been hot topics and difficulties in related research work. When extracting and analyzing the phase, the first thing to do is to perform phase decomposition and reconstruction.

[0003] Xie Kai et al. (2015) proposed a reconstruction method for weak reflection seismic signals of lithologic oil and gas reservoirs, which mainly involves performing empirical mode decomposition on the seismic signals after endpoint topology processing, and then reconstructing seismic signals with a higher signal-to-noise ratio through adaptive selection; Yang Peijie et al. (2019) proposed a sparse constrained inversion spectral decomposition method based on Gabor transform, which can obtain more accurate phase-separated seismic data; Wang Jingfei et al. (2020) proposed a phase-separated reconstruction method based on constrained least squares inversion spectral analysis, which can obtain seismic data bodies with different phase components; Luo Hongmei et al. (2022) proposed the concept and process of phase domain reservoir prediction technology.

[0004] The existing phase reconstruction methods generally have a large amount of calculation in signal decomposition; in signal reconstruction, more specific single-phase data is generated, and the depiction of target area details is not flexible enough. Therefore, how to make full use of relevant information, decompose and reconstruct signals more quickly and effectively, and improve application effects is a problem that needs to be solved in this field. Summary of the invention

[0005] In order to overcome the problems and shortcomings in the prior art, the present invention provides a phase reconstruction method, system and storage medium based on multi-wavelet decomposition, which quickly and effectively decompose and reconstruct seismic signals according to phase, thereby obtaining phase gathers and phase-separated seismic data that meet the requirements.

[0006] The present invention is achieved through the following technical solutions:

[0007] The first aspect of the present invention provides a phase reconstruction method based on multi-wavelet decomposition, firstly decomposing the seismic signal into wavelets of different shapes according to the phase; then selectively screening the phase to reconstruct the seismic signal.

[0008] A further improvement of the present invention is:

[0009] The method comprises the following steps:

[0010] Step 1: extracting complex trace attributes of seismic signals;

[0011] Step 2: extracting the initial control parameters of a wavelet;

[0012] Step 3: Search for the locally optimal wavelet control parameters;

[0013] Step 4: Generate a wavelet and extract the residual signal;

[0014] Step 5: Iteratively generate multiple wavelets;

[0015] Step 6: Reconstruct the input seismic signal according to the phase requirement.

[0016] A further improvement of the present invention is:

[0017] The complex trace attributes of the seismic signal are extracted in step 1, specifically:

[0018] The Hilbert transform is used to calculate the complex seismic trace of the input seismic signal x(t), and the instantaneous amplitude, instantaneous phase and instantaneous frequency are calculated respectively.

[0019] A further improvement of the present invention is:

[0020] The initial control parameters of a wavelet are extracted in step 2, specifically:

[0021] Extract a wavelet W n The initial control parameters, including the time shift u n , phase φ n , frequency f n and scale σ n , the specific process is:

[0022] Get the time corresponding to the maximum instantaneous amplitude, which is the initial time shift u of the sub-wave n ,u n The corresponding phase is the initial phase φ n ,u n The corresponding frequency is the initial frequency f n ;

[0023] Use u n ,φ n 、f n Three parameters, search for the initial scale σ according to the maximum projection principle n ,as follows:

[0024]

[0025] Where D = {g r (t)} r∈Γ is the atomic dictionary, g γ(t) is an atom, R (n) s is the residual of the seismic signal, 《s,h》is the inner product of s and h;

[0026] The seismic signal is decomposed and reconstructed using Morlet wavelet, and the Morlet wavelet function is as follows:

[0027]

[0028] Among them, let g r (t) = m(t), f m is the center frequency;

[0029] In the parameter u n ,φ n 、f n After determination, according to the given phase offset range Δφ, frequency offset range Δf and scale offset range Δσ, the sub-wave is continuously searched in all offset ranges; when the conditions of formula (1) are met, σ at this time n is the optimal value, thus obtaining the optimized initial parameter φ n 、f n And the initial parameter σ n .

[0030] A further improvement of the present invention is:

[0031] The step 3 searches for the locally optimal wavelet control parameters, specifically:

[0032] Using formula (1), we search for the local optimal time shift u' according to the maximum projection principle. n , phase φ' n , frequency f' n and scale σ' n , that is, the optimal value r n = {u' n ,φ n ',f' n ,σ' n};

[0033] First, according to the u obtained in step 2 n , and a given time shift range Δu, calculate the local optimal time shift u' n , as follows: Use u n and Δu, according to the changing trend of the instantaneous phase θ(t), determine the monotonic range of θ(t); within the monotonic range, the phase φ obtained in step 2 n Search and get φ n Where, or with φ n The position with the closest value; the time corresponding to the position is u' n ;

[0034] Then, use u' n And the φ obtained in step 2 n 、f n , σ n , by using formula (1), search again within the offset range composed of Δφ, Δf, Δσ to obtain the optimal value r of the current wavelet n .

[0035] The present invention further comprises:

[0036] In step 4, a wavelet is generated and a residual signal is extracted. The specific operations include:

[0037] According to the optimal value r n , generating wavelet W n , the wavelet W n The amplitude is:

[0038]

[0039] Record a n , W n , and the corresponding phase value φ' n , as a decomposition result;

[0040] Then, the residual signal y(t) is extracted using formula (3):

[0041]

[0042] A further improvement of the present invention is:

[0043] In step 5, multiple wavelets are iteratively generated, and the specific operations include:

[0044] Let x(t)=y(t), repeat steps 1 to 4, and iteratively generate multiple wavelets;

[0045] When the energy of y(t) is less than the set percentage of the energy of the original seismic signal x(t), the iteration is exited;

[0046] For all recorded φ' n The phase gathers are sorted and finally the decomposition result of a single channel is obtained, that is, a group of phase gathers arranged in descending order according to the phase value.

[0047] A further improvement of the present invention is:

[0048] In step six, the input seismic signal is reconstructed according to the phase requirement, specifically: according to the set phase range, the corresponding wavelet is extracted from the multiple wavelets generated in step five, and the extracted wavelets are superimposed to obtain the reconstructed seismic signal.

[0049] A second aspect of the present invention provides a phase reconstruction system based on multi-wavelet decomposition, comprising:

[0050] An attribute extraction unit, used for extracting complex channel attributes of seismic signals;

[0051] An initial parameter extraction unit, used for extracting an initial control parameter of a wavelet;

[0052] A parameter search unit, used to search for locally optimal wavelet control parameters;

[0053] A wavelet generation unit, used for generating a wavelet and extracting a residual signal;

[0054] A wavelet iteration unit, used for iteratively generating multiple wavelets;

[0055] The reconstruction unit is used to reconstruct the input seismic signal according to the phase requirement.

[0056] 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 above-mentioned phase reconstruction method based on multi-wavelet decomposition.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] When performing signal decomposition, the present invention can effectively reduce the number of iterations, improve the calculation speed, and decompose the signal relatively accurately; when performing signal reconstruction, based on the phase characteristics and in combination with other data information, an appropriate phase range is selected to perform targeted reconstruction; using the reconstructed data in combination with seismic attributes and time-frequency analysis and other means, geological characteristics and lithology changes can be studied more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of a phase reconstruction method based on multi-wavelet decomposition in an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of single-channel decomposition and reconstruction, where the left side is the original signal; the middle is the reconstruction result of the specified phase range; the right side is the phase gather obtained after decomposing the original signal;

[0061] Figure 3a This is an illustration of the reconstruction of the 1st to 14th phase components;

[0062] Figure 3b This is a schematic diagram of the reconstruction of the 15th to 28th phase components;

[0063] Figure 3cThis is a schematic diagram of the reconstruction of the 29th to 42nd phase components;

[0064] Figure 4a is the original section;

[0065] Figure 4b Phase reconstruction profile for positive angle range;

[0066] Figure 4c Phase reconstruction profile for small angular range;

[0067] Figure 4d Phase reconstructed profile for negative angle range. DETAILED DESCRIPTION

[0068] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0069] The present invention provides a phase reconstruction method based on multi-wavelet decomposition, firstly decomposing the seismic signal into wavelets of different shapes according to the phase; then selectively screening the phase to reconstruct the seismic signal.

[0070] In order to make the purpose and technical solution of the present invention more clearly understood, embodiments are given below and the present invention is further described in detail in conjunction with the accompanying drawings.

[0071] [Example 1]

[0072] The method of the present invention is used to process actual seismic data of a certain work area. The data is post-stack seismic data, with a total of several sections, each section has 1152 traces, a time range of 2700-3500ms, and an interval of 1ms.

[0073] First, a single channel is selected to perform decomposition and reconstruction calculations on the single channel data. In this embodiment, data of 597 lines and 1004 channels are selected.

[0074] Before the single channel calculation begins, the original signal needs to be windowed in the time domain, the result is recorded as x(t), and its energy is recorded. The window type and length value when windowing should be set according to the effect of decomposition and reconstruction. This embodiment uses a Gaussian window with a time length of 40ms.

[0075] The phase reconstruction method based on multi-wavelet decomposition in the embodiment of the present invention specifically includes the following steps:

[0076] Step 1: Extract the complex trace attributes of the seismic signal:

[0077] The Hilbert transform is used to calculate the complex seismic trace of the input seismic signal x(t), and three properties, namely, instantaneous amplitude, instantaneous phase and instantaneous frequency, are calculated.

[0078] Hilbert transform is a common algorithm in the industry and is used to extract the imaginary part x of the real-domain signal. h (t), convert the real domain signal to the complex domain.

[0079] The instantaneous amplitude, instantaneous phase and instantaneous frequency parameters are extracted using the real-domain signal and the corresponding imaginary part. The specific extraction formula is as follows:

[0080] Instantaneous amplitude:

[0081] Instantaneous phase:

[0082] Instantaneous frequency:

[0083] Step 2: Extract the initial control parameters of a wavelet:

[0084] Extract a wavelet W n The initial control parameters, including the time shift u n , phase φ n , frequency f n and scale σ n , the specific process is:

[0085] Get the time corresponding to the maximum instantaneous amplitude, which is the initial time shift u of the sub-wave n ,u n The corresponding phase is the initial phase φ n ,u n The corresponding frequency is the initial frequency f n ;

[0086] Use u n ,φ n 、f n Three parameters, search for the initial scale σ according to the maximum projection principle n ,as follows:

[0087]

[0088] Where D = {g r (t)} r∈Γ is the atomic dictionary, g γ (t) is an atom, R (n) s is the residual of the seismic signal, 《s,h》is the inner product of s and h;

[0089] The seismic signal is decomposed and reconstructed using Morlet wavelet, and the Morlet wavelet function is as follows:

[0090]

[0091] Among them, let g r (t) = m(t), f m is the center frequency. In actual calculations, directly substitute the specific f n ;

[0092] Specifically, in the parameter u n ,φ n 、f n After determination, according to the given phase offset range Δφ, frequency offset range Δf and scale offset range Δσ, the sub-wave is continuously searched in all offset ranges; when the conditions of formula (1) are met, σ at this time n is the optimal value, thus obtaining the optimized initial parameter φ n 、f n And the initial parameter σ n In this embodiment, Δφ is 50, Δf is 5 Hz, and Δσ is 0.1.

[0093] Step 3: Search for the locally optimal wavelet control parameters:

[0094] Using formula (1), we search for the local optimal time shift u' according to the maximum projection principle. n , phase φ' n , frequency f' n and scale σ' n , that is, the optimal value r n = {u' n ,φ' n ,f' n ,σ' n}.

[0095] First, according to the u obtained in step 2 n , and given time offset range Δu and other parameters, calculate the local optimal time shift u' n , as follows: Use u n and Δu, according to the changing trend of the instantaneous phase θ(t), determine the monotonic range of θ(t); within this range, for the phase φ obtained in step 2 n Search and get φ n Where, or with φ n The position with the closest value; the time corresponding to this position is u' n .

[0096] Then, use u' n And the φ obtained in step 2 n 、f n , σ n , by using formula (1), search again within the offset range composed of Δφ, Δf, Δσ to obtain the optimal value r of the current wavelet n .

[0097] Among them, Δφ, Δf, and Δσ are consistent with those in step 2, and Δu is 8 ms in this embodiment.

[0098] Step 4: Generate a wavelet and extract the residual signal:

[0099] According to the optimal value r n , generating wavelet W n , where the wavelet W n The amplitude is:

[0100]

[0101] Record a n , W n , and the corresponding phase value φ' n , as a decomposition result;

[0102] Then, the residual signal y(t) is extracted using formula (3):

[0103]

[0104] Then consider performing local smoothing on y(t). This embodiment performs local smoothing.

[0105] Step 5: Iteratively generate multiple wavelets:

[0106] Let x(t)=y(t), repeat steps 1 to 4, and iteratively generate multiple wavelets;

[0107] When the energy of y(t) is less than a set percentage of the energy of the original seismic signal x(t), the iteration is exited. In this embodiment, the percentage is set to 10%.

[0108] For all recorded φ' n The phase gathers are sorted and finally the decomposition result of a single channel is obtained, that is, a group of phase gathers arranged in descending order according to the phase value.

[0109] like Figure 2 , the left side of the figure is the original single channel, where the vertical axis is time and the horizontal axis is amplitude; the rightmost side is the phase gather obtained by this decomposition, where the vertical axis is time and the horizontal axis represents the seismic component corresponding to the phase from 180 degrees to -180 degrees. The first row above is marked with the serial number of the component, and the second row is marked with the phase value corresponding to the component. In this embodiment, the original signal is decomposed into a total of 42 different Morlet wavelets.

[0110] Step 6: Reconstruct the input seismic signal according to the phase requirements:

[0111] Specifically, according to the set phase range, the corresponding wavelet is extracted from the multiple wavelets generated in step 5, and the extracted wavelets are superimposed (linearly summed) to obtain the reconstructed seismic signal.

[0112] For example, by setting the phase range to be extracted within the range of 180 degrees to -180 degrees, and then performing linear summation on the wavelets within the range, the seismic trace after phase reconstruction can be obtained; the seismic trace only contains information on the set phase, thereby realizing seismic reconstruction of the specified phase component.

[0113] Select all phase gathers for superposition to generate the reconstructed signal x'(t), such as Figure 2 By comparing the original signal x(t) with the reconstructed signal x'(t), it can be seen that the original signal can be well reconstructed through phase synthesis, thus verifying the effectiveness and accuracy of this method.

[0114] Figure 3a This is an illustration of the reconstruction of the 1st to 14th phase components. Figure 3b This is an illustration of the reconstruction of the 15th to 28th phase components. Figure 3c This is an illustration of the reconstruction of the 29th to 42nd phase components.

[0115] The above is the decomposition and reconstruction of single-channel data.

[0116] Repeat steps 1 to 6 to cycle the seismic data, and the phase reconstruction of the seismic data can be completed.

[0117] Figure 4a is the original section, Figure 4b is the phase reconstruction profile in the positive angle range, Figure 4c For the phase reconstruction profile in a small angle range, Figure 4d This is a phase reconstruction section in the negative angle range. On the above phased seismic gathers, the amplitude difference between the top and bottom interfaces of the reservoir can be clearly observed.

[0118] Based on the differences in the way seismic data behave in different phase components, it is possible to consider obtaining better phase-separated seismic data by performing phase-separated processing on the seismic data; by utilizing the amplification effect of phase-separated seismic data containing specific phases on reservoir characteristics, reservoir prediction can be performed efficiently and accurately.

[0119] The present invention provides a phase reconstruction system based on multi-wavelet decomposition, and the embodiments of the system are as follows:

[0120] [Example 2]

[0121] The system comprises:

[0122] An attribute extraction unit, used for extracting complex channel attributes of seismic signals;

[0123] An initial parameter extraction unit, used for extracting an initial control parameter of a wavelet;

[0124] A parameter search unit, used to search for locally optimal wavelet control parameters;

[0125] A wavelet generation unit, used for generating a wavelet and extracting a residual signal;

[0126] A wavelet iteration unit, used for iteratively generating multiple wavelets;

[0127] The reconstruction unit is used to reconstruct the input seismic signal according to the phase requirement.

[0128] The present invention also provides a computer-readable storage medium, and an embodiment of the computer-readable storage medium is as follows:

[0129] [Example 3]

[0130] 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 is caused to execute the steps in the phase reconstruction method based on multi-wavelet decomposition as described in Example 1.

[0131] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0132] The above technical solution is only one implementation mode of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the principles disclosed in the present invention, and it is not limited to the technical solution described in the above specific embodiments of the present invention. Therefore, the above description is only preferred and does not have a restrictive meaning.

Claims

1. A phase reconstruction method based on multi-wavelet decomposition, characterized in that: First, the seismic signal is decomposed into wavelets of different shapes according to the phase; then the phase is selectively screened to reconstruct the seismic signal.

2. The phase reconstruction method based on multi-wavelet decomposition according to claim 1, characterized in that: The method comprises: The method comprises the following steps: Step 1: extracting complex trace attributes of seismic signals; Step 2: extracting the initial control parameters of a wavelet; Step 3: Search for the locally optimal wavelet control parameters; Step 4: Generate a wavelet and extract the residual signal; Step 5: Iteratively generate multiple sub-waves; Step 6: Reconstruct the input seismic signal according to the phase requirement.

3. The phase reconstruction method based on multi-wavelet decomposition according to claim 2 is characterized in that: The complex trace attributes of the seismic signal are extracted in step 1, specifically: The Hilbert transform is used to calculate the complex seismic trace of the input seismic signal x(t), and the instantaneous amplitude, instantaneous phase and instantaneous frequency are calculated respectively.

4. The phase reconstruction method based on multi-wavelet decomposition according to claim 3 is characterized in that: The initial control parameters of a wavelet are extracted in step 2, specifically: Extract a wavelet W n The initial control parameters, including the time shift u n , phase φ n , frequency f n and scale σ n , the specific process is: Get the time corresponding to the maximum instantaneous amplitude, which is the initial time shift u of the sub-wave n ,u n The corresponding phase is the initial phase φ n ,u n The corresponding frequency is the initial frequency f n ; Use u n ,φ n 、f n Three parameters, search for the initial scale σ according to the maximum projection principle n ,as follows: Where D = {g r (t)} r∈Γ is the atomic dictionary, g γ (t) is an atom, R (n) s is the residual of the seismic signal, 《s,h》is the inner product of s and h; The seismic signal is decomposed and reconstructed using Morlet wavelet, and the Morlet wavelet function is as follows: Among them, let g r (t) = m(t), f m is the center frequency; In the parameter u n ,φ n 、f n After determination, according to the given phase offset range Δφ, frequency offset range Δf and scale offset range Δσ, the sub-wave is continuously searched in all offset ranges; when the conditions of formula (1) are met, σ at this time n is the optimal value, thus obtaining the optimized initial parameter φ n 、f n And the initial parameter σ n .

5. The phase reconstruction method based on multi-wavelet decomposition according to claim 4 is characterized in that: The step 3 searches for the locally optimal wavelet control parameters, specifically: Using formula (1), we search for the local optimal time shift u' according to the maximum projection principle. n , phase φ n ', frequency f n 'and scale σ' n , that is, the optimal value r n = {u' n ,φ n ',f n ',σ' n }; First, according to the u obtained in step 2 n , and a given time shift range Δu, calculate the local optimal time shift u' n , as follows: Use u n and Δu, according to the changing trend of the instantaneous phase θ(t), determine the monotonic range of θ(t); within the monotonic range, the phase φ obtained in step 2 n Search and get φ n Where, or with φ n The position with the closest value; the time corresponding to the position is u' n ; Then, use u' n And the φ obtained in step 2 n 、f n , σ n , by using formula (1), search again within the offset range composed of Δφ, Δf, Δσ to obtain the optimal value r of the current wavelet n .

6. The phase reconstruction method based on multi-wavelet decomposition according to claim 5, characterized in that: In step 4, a wavelet is generated and a residual signal is extracted. The specific operations include: According to the optimal value r n , generating wavelet W n , the wavelet W n The amplitude is: Record a n , W n , and the corresponding phase value φ n ', as a decomposition result; Then, the residual signal y(t) is extracted using formula (3):

7. The phase reconstruction method based on multi-wavelet decomposition according to claim 6, characterized in that: In step 5, multiple wavelets are iteratively generated, and the specific operations include: Let x(t)=y(t), repeat steps 1 to 4, and iteratively generate multiple wavelets; When the energy of y(t) is less than the set percentage of the energy of the original seismic signal x(t), the iteration is exited; For all recorded φ n 'Sorting, and finally get the decomposition result of a single channel, that is, a set of phase gathers arranged in descending order according to the phase value.

8. The phase reconstruction method based on multi-wavelet decomposition according to claim 7, characterized in that: In step 6, the input seismic signal is reconstructed according to the phase requirement, specifically: According to the set phase range, the corresponding wavelet is extracted from the multiple wavelets generated in step 5, and the extracted wavelets are superimposed to obtain the reconstructed seismic signal.

9. A phase reconstruction system based on multi-wavelet decomposition, characterized in that: include: An attribute extraction unit, used for extracting complex channel attributes of seismic signals; An initial parameter extraction unit, used for extracting an initial control parameter of a wavelet; A parameter search unit, used to search for locally optimal wavelet control parameters; A wavelet generation unit, used for generating a wavelet and extracting a residual signal; A wavelet iteration unit, used for iteratively generating multiple wavelets; The reconstruction unit is used to reconstruct the input seismic signal according to the phase requirement.

10. 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 of the phase reconstruction method based on multi-wavelet decomposition as described in any one of claims 1 to 8.

Images