Method for wavelet zero phasing

Abstract

The invention discloses a method for wavelet zero phasing, which comprises the following steps: S1, data input: inputting original seismic data, loading an observation system, and forming a seismic record; s2, preprocessing: processing the obtained data by using static correction processing and spherical diffusion compensation processing technologies; according to the method, the seismic data are preprocessed and denoised, so that analysis of wavelet amplitude and phase characteristics is facilitated; a local phase method and shaping regularization are adopted to calculate the phase shift amount of each time on the whole track, so that a time-varying phase angle can be accurately calculated; and the residual phase information can be corrected by adopting the repeated spectrum to extract the wavelets and the least square method. According to the twice step-by-step phase correction method provided by the invention, actual data are processed to prove that phase angle calculation is more accurate, so that the resolution ratio of a profile after phase correction processing is improved, the event continuity becomes better, some weak reflections become clear, and hidden strata can be identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of seismic data processing, in particular to a method for wavelet zero phase. BACKGROUND

[0002] Wavelet is the smallest unit signal of seismic profile, and zero phase wavelet is a non-physical realizable signal symmetrical about zero time. The zero phase wavelet has the highest peak value among all wavelets with the same amplitude spectrum, and the main peak value corresponds to the most accurate stratigraphic interface, so its resolution is also the highest. Moreover, the maximum amplitude position is located at zero time, which is beneficial to seismic interpretation, inversion and reservoir prediction. Therefore, when using seismic result data for subsequent interpretation, reservoir prediction and other researches, it is always desired that the seismic wavelet in the seismic profile is zero phase, but this assumption is often not met. Because the source wavelet excited in seismic data acquisition is generally a physical realizable signal with minimum phase or mixed phase, and the received wavelet after seismic wave propagation and the wavelet after data processing have been changed and are completely different from the source wavelet.

[0003] Moreover, the existing seismic wavelet zero phase method generally assumes that the phase of the seismic wavelet is constant, and uses the variance module method to perform constant phase. If the wavelet is not divergent, i.e. the phase of the wavelet is independent of frequency, the assumption of constant phase correction of the data can be met. Due to the dispersion effect of the earth absorption, the assumption that the phase of the seismic wavelet is constant cannot be met.

[0004] Since the seismic wave will appear dispersion phenomenon when propagating underground, the data is no longer stable phase. For non-stable phase data, the constant phase rotation method is generalized, and a method for estimating the corresponding time-varying wavelet is proposed. The method slides the time window, estimates the constant phase wavelet in each time window through the maximum peak value criterion, and linearly interpolates the points between the positions of the estimated phase values, so that the wavelet phase spectrum and amplitude spectrum at each time can be obtained. Therefore, this method is based on the assumption of piecewise stable phase, and if the selected time window is too small, the estimated phase often changes very sharply, and if the time window is too large, only the relatively stable signal can be estimated, and the time-varying nature of the phase of the non-stable signal is difficult to consider. This makes it difficult for the method to highlight the characteristics of the hidden stratum.

[0005] Zhang Yintang (2017, wavelet zero phase processing method and device), Dai Xiaofeng (2016, method and device for recovering seismic energy of thin layer weak reflection under strong reflection shielding) and others have done a lot of research on wavelet zero phase, mainly using poststack data and logging records to realize wavelet zero phase, and using prestack seismic data containing more information to extract wavelet can more reliably reveal the distribution of underground reservoirs. In addition, a wavelet zero phase processing method and device disclosed in a Chinese patent No. CN106569275A also uses poststack data to process wavelet zero phase. Therefore, an effective method for phase correction of non-stable prestack data needs to be researched. SUMMARY

[0006] The purpose of the present application is to provide a method for wavelet zero phase, aiming to improve the problem that the phase change of existing non-stable data is very severe, which is not conducive to highlighting the characteristics of hidden formations.

[0007] The present application is implemented as follows: a method for wavelet zero phase, comprising

[0008] S1, data input: input original seismic data, load observation system, and form seismic record;

[0009] S2, preprocessing: using static correction processing and spherical diffusion compensation processing technology, the obtained data is processed;

[0010] S3, denoising processing: analyzing the characteristics of various noises, determining the types of noises that need to be suppressed, and determining the parameters required for noise suppression;

[0011] S4, deconvolution processing: in the time domain, combined with the actual situation of geological structure, selecting appropriate deconvolution method, compressing seismic wavelet, and completing the processing of wavelet amplitude spectrum;

[0012] S5, approximate wavelet zero phase: local phase method and shaping regularization are used to calculate the phase shift amount of each time on the whole trace, a series of time-varying phase angles are calculated, the seismic wavelet phase information is obtained, and the seismic record is approximately zero phase processed by using the phase information.

[0013] S6, residual phase correction: performing complex spectrum processing on seismic data, extracting the amplitude spectrum of wavelet, and then using least square method to obtain the residual phase information of wavelet, and performing residual phase correction on seismic data.

[0014] S7, output: outputting the seismic record after wavelet zero phase, and performing subsequent related processing.

[0015] Preferably, in S2, the static correction processing technology is used to correct the phase axis distortion caused by the surface relief.

[0016] Preferably, in S2, the attenuation of seismic wave energy caused by spherical divergence is effectively compensated by spherical divergence compensation technique.

[0017] Preferably, in S3, the characteristics of abnormal amplitude, shallow multiple refraction wave, external source interference and other noises are analyzed to determine the parameters required for noise suppression.

[0018] Preferably, in S4, the result data of step S3 is processed to expand the frequency width of seismic data by using surface consistency deconvolution processing.

[0019] Preferably, in S4, the result data of step S3 is processed to expand the frequency width of seismic data by using predictive deconvolution processing.

[0020] Preferably, according to the local estimation of signal points near each point, the phase shift amount at each time on the whole trace is calculated by shaping regularization using the calculation form of inversion, a series of time-varying phase angles are obtained, and the approximate zero phase correction is obtained by using the local phase value for phase correction.

[0021] Preferably, the calculation formula of variance modulus and signal and its envelope is written into the calculation form of inversion, the calculation of target function is converted into an inversion problem, and the inversion calculation is performed on the two values respectively, and the phase shift amount is calculated by using shaping regularization.

[0022] Preferably, a sequence of complex spectrum is defined as the logarithm of its Fourier transform, the logarithmic spectrum of the wavelet is distributed near the origin, and the logarithmic spectrum of the reflection coefficient is distributed far from the origin, when the wavelet phase of the input record is calculated, the statistical average method is adopted, the time window selected for calculation is randomly distributed within a certain range, the influence of the reflection coefficient is suppressed, and the wavelet is estimated.

[0023] Preferably, the least square method is used to calculate the filter operator, the zero phase wavelet is taken as the expected output, the seismic wavelet extracted based on the complex spectrum is approximated to the expected wavelet after being convolved by the zero phase operator, that is, the following formula is minimized:

[0024]

[0025] Compared with the prior art, the present application has the beneficial effects that: the present application is beneficial to the analysis of wavelet amplitude and phase characteristics by pre-processing and denoising of seismic data; the local phase method and shaping regularization are adopted to calculate the phase shift amount at each time on the whole trace, which is beneficial to the correct calculation of time-varying phase angle; the complex cepstrum extraction wavelet and the least square method are adopted, so that the residual phase information can be corrected. Through the two-step phase correction method of the present application, the phase angle calculation is more accurate, the resolution of the profile after phase correction processing is improved, the continuity of the phase axis is better, some weak reflections are clear, and the hidden stratum can be identified. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flow chart of the wavelet zero phase method of the embodiment of the present application;

[0027] Figure 2 is a seismic record schematic diagram of a single shot single array seismic record subjected to wavelet amplitude spectrum processing;

[0028] Figure 3 is a seismic record schematic diagram obtained after the seismic record shown in Figure 2 is subjected to phase spectrum zero phase by the present application;

[0029] Figure 4 is a seismic record schematic diagram obtained after the seismic record shown in Figure 2 is subjected to NMO correction and common center point stacking by the present application;

[0030] Figure 5 is a seismic record schematic diagram obtained after the seismic record shown in Figure 3 is subjected to NMO correction and common center point stacking by the present application;

[0031] Figure 6 is a well seismic calibration schematic diagram of the seismic record shown in Figure 5 and well logging data. DETAILED DESCRIPTION

[0032] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connecting", "fixing" and the like should be understood in a broad sense, for example, can be fixed connection, or detachable connection, or integrated; can be mechanical connection, or electrical connection; can be directly connected, or indirectly connected through an intermediate medium, can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0033] The following will be further described in combination with the drawings and specific embodiments:

[0034] Step 101: Preprocess the seismic data;

[0035] Step 102: Suppress various noises in the seismic data, including surface waves, anomalous amplitudes, shallow multiple refraction waves, external interference, 50Hz mains power interference, random environmental noise, etc.

[0036] Step 103: In the time domain, the result of step 102 is subjected to deconvolution to compress the seismic wavelet and complete the processing of the wavelet amplitude spectrum.

[0037] Step 104: Perform wavelet phase analysis and phase correction on the results of step 103.

[0038] This invention calculates the phase shift at various time points along the entire trace using the local phase method and shaping regularization, thereby determining the time-varying phase angle. It employs complex spectrum extraction of wavelet and least squares methods to correct the remaining phase information. Through processing actual data, the two-step phase correction method of this invention demonstrates more accurate phase angle calculations, resulting in improved profile resolution, better continuity of phase axes, clearer weak reflections, and identification of concealed strata after phase correction.

[0039] In one possible implementation, step 101 may include: loading seismic records and loading the observation system; correcting the in-phase axis distortion caused by surface undulations through static correction processing; and effectively compensating for the attenuation of seismic wave energy caused by spherical diffusion through spherical diffusion compensation technology.

[0040] In one possible implementation, the method may further include, before step 102, analyzing the characteristics of various noises such as abnormal amplitude, shallow multiple refraction waves, external interference, 50Hz mains interference, and random environmental noise, determining the types of noise that need to be suppressed, and determining the parameters required for noise suppression in step 102. This ensures that effective signal is not lost as much as possible during noise suppression in step 102.

[0041] In one possible implementation, in step 103, since different deconvolution methods have different applicability, the deconvolution method needs to be selected based on the actual geological structure, such as surface-consistent deconvolution processing and predictive deconvolution processing. The result data from step 102 is then processed to complete the wavelet amplitude spectrum. This aims to accurately characterize reservoir features and meet the requirements of fidelity processing. It effectively expands the bandwidth of seismic data, improves data resolution while preserving the low-frequency components of the data.

[0042] In one possible implementation, in step 104, addressing the current phase correction method's application of the concept of local attributes to phase when performing phase correction on unsteady-state data, the concept of local phase is proposed. This involves changing the global estimation relative to the sliding data window to a local estimation based on signal points near each point. The calculation formulas for the variance modulus and the signal and its envelope are written in an inversion calculation form, transforming the objective function into an inversion problem. Inversion calculations are performed on these two quantities separately. Phase shifts at various times along the entire trace are calculated using shaping regularization, ultimately calculating a series of time-varying phase angles. These local phase values ​​are then used for phase correction, ultimately achieving near-zero phase correction of the unsteady-state seismic data, i.e., the data obtained in step 103.

[0043] In one possible implementation, in step 104, to eliminate the influence of residual phase, the wavelet is extracted using the complex spectrum proposed by homomorphic theory. The complex spectrum, also known as the logarithmic spectrum, is defined as the logarithm of its Fourier transform. According to homomorphic theory, the logarithmic spectrum of the wavelet is distributed near the origin, while the logarithmic spectrum of the reflection coefficient is distributed further away from the origin. This separates the complex spectrum of the seismic wavelet and the complex spectrum of the reflection coefficient sequence in the complex spectrum domain, allowing for the simultaneous acquisition of both. Finally, it is transformed to the time domain to obtain the mixed-phase seismic wavelet in the time domain. Considering that the reflection coefficient and wavelet in actual seismic records are often not strictly separated, to avoid the influence of the reflection coefficient when calculating the wavelet phase of the input record, we use a multichannel statistical averaging method and randomly distribute the selected time window within a certain range during calculation. This effectively suppresses the influence of the reflection coefficient to a considerable extent, allowing for a better estimation of the wavelet.

[0044] Then, the least squares method is used to calculate the filter operator. The zero-phase wavelet is taken as the desired output. The seismic wavelet extracted based on the complex spectrum is convolved with the zero-phase operator to approximate the desired wavelet, that is, the following equation is minimized:

[0045]

[0046] (1) In the equation: x(t) is the input signal, i.e., the wavelet extracted using homomorphic theory; h(t) is the filtering operator; d(t) is the zero-phase wavelet of the desired output. Taking the partial derivative of the error energy of equation (1) and setting it to zero, we can obtain the Tobleitz equation:

[0047]

[0048] (2) Where: r xx The autocorrelation of the input signal; r dxis the cross-correlation between the input signal and the desired output; m is the length of the filter operator. Solve equation (2) to obtain the inverse filter factor, and then convolve it with the input signal to obtain the zero-phase data.

[0049] It can be seen that by applying the wavelet zero-phase method described in this invention, and by utilizing the local phase method, shaping regularization, homomorphic theory, and least squares method, and employing corresponding technical means to progressively correct the wavelet phase, wavelet zero-phase can be effectively achieved. This results in improved resolution of the profile after phase correction, better continuity of the phase axis, clearer weak reflections, and identification of concealed strata. Combined with... Figure 6 The seismic data after wavelet zero-phase processing shown in the figure have good matching with well logging and high fidelity.

[0050] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0051] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for wavelet zero-phase transformation, characterized in that, include S1. Data Input: Input raw seismic data, load the observation system, and generate seismic records; S2. Preprocessing: The obtained data is processed using static correction and spherical diffusion compensation techniques. S3. Noise Reduction Processing: Analyze the characteristics of various noises, determine the types of noise that need to be suppressed, and determine the parameters required for noise suppression. S4. Deconvolution processing: In the time domain, a suitable deconvolution method is selected based on the actual geological structure to compress the seismic wavelet and complete the processing of the wavelet amplitude spectrum. S5. Approximate wavelet zero-phase transformation: The local phase method and shaping regularization are used to calculate the phase shift at various times on the entire trace, calculate a series of time-varying phase angles, obtain the seismic wavelet phase information, and use this phase information to perform approximate zero-phase transformation on the seismic record. S6. Residual Phase Correction: Perform complex spectrum processing on the seismic data, extract the amplitude spectrum of the wavelet, and then use the least squares method to obtain the residual phase information of the wavelet, and perform residual phase correction on the seismic data. S7. Output: Output the seismic record after wavelet zero-phase conversion for subsequent related processing.

2. The method for wavelet zero-phase transformation according to claim 1, characterized in that, In S2, static correction processing technology is used to correct the in-phase axis distortion caused by surface undulations.

3. The method for wavelet zero-phase transformation according to claim 2, characterized in that, In S2, the attenuation of seismic wave energy caused by spherical diffusion is effectively compensated through spherical diffusion compensation technology.

4. The method for wavelet zero-phase transformation according to claim 1, characterized in that, In S3, the characteristics of various noises, such as abnormal amplitude, shallow multiple refraction waves, and external interference, are analyzed to determine the parameters required for noise suppression.

5. The method for wavelet zero-phase transformation according to claim 1, characterized in that, In S4, surface-consistent deconvolution is used to process the wavelet amplitude spectrum of the data obtained in step S3, thereby expanding the bandwidth of the seismic data.

6. The method for wavelet zero-phase transformation according to claim 1, characterized in that, In S4, predictive deconvolution is used to process the wavelet amplitude spectrum of the data obtained in step S3, thereby expanding the bandwidth of the seismic data.

7. The method for wavelet zero-phase transformation according to claim 1, characterized in that, Local estimation is performed based on signal points near each point. The phase shift at each time step on the entire channel is calculated using the inversion calculation method and shaping regularization. A series of time-varying phase angles are obtained. Phase correction is performed using local phase values ​​to obtain an approximate zero phase correction.

8. A method for wavelet zero-phase transformation according to claim 7, characterized in that, The formulas for calculating the variance modulus and the signal and its envelope are written in the form of inversion calculation. The objective function is transformed into an inversion problem. Inversion calculations are performed on these two quantities respectively, and then the phase shift is calculated using shaping regularization.

9. A method for wavelet zero-phase transformation according to claim 1, characterized in that, The complex spectrum of a sequence is defined as the logarithm of its Fourier transform. The logarithmic spectrum of the wavelet is distributed near the origin, while the logarithmic spectrum of the reflection coefficient is distributed far from the origin. When obtaining the wavelet phase of the input record, a statistical averaging method is used. The time window selected during the calculation is randomly distributed within a certain range to suppress the influence of the reflection coefficient and estimate the wavelet.

10. A method for wavelet zero-phase transformation according to claim 9, characterized in that, The filter operator is calculated using the least squares method. The zero-phase wavelet is taken as the desired output. The seismic wavelet extracted based on the complex spectrum is convolved with the zero-phase operator to approximate the desired wavelet, that is, the following equation is minimized:

Images