Consistent processing method for continuous data based on multi-channel statistics and signal-to-noise ratio constraint

Abstract

The present invention provides a method for processing the coherence of contiguous data based on multi-trace statistics and signal-to-noise ratio constraint. The method for processing the coherence of contiguous data based on multi-trace statistics and signal-to-noise ratio constraint includes: Step 1: estimating the linear phase shift amount; Step 2: estimating the constant phase shift amount; Step 3: performing phase coherence processing; Step 4: estimating the amplitude spectrum matching operator; Step 5: performing amplitude spectrum coherence processing. The method for processing the coherence of contiguous data based on multi-trace statistics and signal-to-noise ratio constraint can be applied to the coherence processing of two contiguous data with a large difference in signal-to-noise ratio. Through multi-trace statistics and signal-to-noise ratio constraint processing, the stability and accuracy of the matching operator are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic data processing for oil and gas exploration, and particularly to a method for processing the coherence of contiguous data based on multi-trace statistics and signal-to-noise ratio constraint. Background Art

[0002] With the in-depth exploration and development of oilfields, the processing target has shifted from "single 3D block" to "contiguous 3D". The processing of data coherence between adjacent blocks is the core and key to data splicing, comparison and utilization. Currently, the mainstream methods for processing the coherence of contiguous data mainly include Wiener filtering method, wavelet matching method, etc. These methods are simple to calculate and are widely used because the processing results are satisfactory when the quality of contiguous data is good. However, when the data quality is poor or the quality difference of the data to be spliced is large, the existence of factors such as noise often leads to inaccurate data coherence processing operators calculated by the above methods, or large differences in the coherence processing operators calculated by selecting data at different positions. The problems of inaccurate coherence processing operators or poor stability of the above methods caused by the existence of such factors as noise result in poor coherence processing effect of contiguous data and cannot meet the seismic exploration requirements of high-precision and fine processing.

[0003] In the Chinese patent application with the application number: CN201110158050.9, it relates to a method for processing contiguous 2D seismic data in complex surface areas, belonging to the field of data processing in geophysical exploration. The method first establishes a spatial model of the 2D seismic data volume, then performs field first static correction calculation on the spatial model of the 2D seismic data volume, and then performs field first static correction application, pre-stack denoising processing, fidelity and coherence processing on each 2D seismic survey line in the spatial model, and finally improves the signal-to-noise ratio and resolution of the target layer. Using the method of the present invention to process contiguous 2D seismic data collected in different years, the time-depth error of the processed profile is less than the specification index, the structural form is clear, reliable and closed, and a seismic profile with high resolution, high fidelity, high signal-to-noise ratio and capable of clearly reflecting the structural and lithological change characteristics is processed, providing reliable processing results for seismic interpretation.

[0004] In the Chinese patent application with the application number: CN201410374878.1, it involves a method for large-area three-dimensional continuous processing, belonging to the technical field of data processing. The present invention applies the data strip segmentation technology to the large-area three-dimensional continuous processing process. According to the computer system resource configuration, the size of the temporary space, and the size of the data to be processed, the data to be processed is segmented into several strip units, and each strip unit is correspondingly compiled into a job, which is submitted on different PC nodes to implement multi-node job parallel processing. The invention can effectively utilize computer nodes and storage resources, minimize the idle rate of computer nodes, improve data processing efficiency, shorten the processing period, save costs, and has broad prospects for popularization and application.

[0005] In the Chinese patent application with the application number: CN201210536609.1, it involves a method for processing geophysical exploration seismic data, specifically a velocity splicing method in two-dimensional seismic data continuous processing. The velocity file splicing method in this two-dimensional seismic data continuous processing includes seismic data acquisition, pre-stack preprocessing of each block of seismic data, velocity analysis and velocity file splicing processing of the segmented seismic data. The spliced velocity file can be used for dynamic correction processing, stacking processing, and post-stack migration processing of two-dimensional continuous seismic data in different work areas. Using this method, according to parameters such as the splicing spatial position of different seismic data, the CMP distance before and after equalization of different bins, and the CMP number corresponding to the splicing point, the continuous splicing of two-dimensional velocity data can be realized. The spliced velocity file and the aforementioned seismic data are used as inputs to realize the post-stack migration of the spliced seismic data. This technology can greatly shorten the processing cycle of two-dimensional seismic data splicing and provide technical support for the overall understanding of the underground macrostructure and exploration deployment.

[0006] The above existing technologies are all quite different from the present invention and fail to solve the technical problems we want to solve. Therefore, we have invented a new continuous data consistency processing method based on multi-trace statistics and signal-to-noise ratio constraints. Summary of the Invention

[0007] The object of the present invention is to provide a continuous data consistency processing method based on multi-trace statistics and signal-to-noise ratio constraints, which improves the stability and accuracy of the matching operator through multi-trace statistics and signal-to-noise ratio constraints.

[0008] The object of the present invention can be achieved by the following technical measures: A continuous data consistency processing method based on multi-trace statistics and signal-to-noise ratio constraints, which includes:

[0009] Step 1: Estimate the linear phase shift amount;

[0010] Step 2: Estimate the constant phase shift amount;

[0011] Step 3: Perform phase consistency processing;

[0012] Step 4: Estimate the amplitude spectrum matching operator;

[0013] Step 5: Perform amplitude spectrum consistency processing.

[0014] The object of the present invention can also be achieved by the following technical measures:

[0015] In step 1, for two blocks that need to be processed in a contiguous manner, there is partial overlap in the data of the two blocks; the data of the block with higher data signal-to-noise ratio and resolution is used as the target data, denoted as A, and the data in its overlapping area is denoted as C; the block with relatively lower data signal-to-noise ratio and resolution is used as the data to be processed, denoted as B, and the data in its overlapping area is denoted as D.

[0016] In step 1, estimate the linear phase shift amount in the following manner:

[0017] (1) Select a time window area with a relatively high signal-to-noise ratio in data D, and calculate the phase spectrum θ D (f) of the i-th trace data in the time window;

[0018] (2) Given the scanning range of the linear phase shift amount, scan from small to large within the scanning range according to the linear phase shift amount at equal intervals; let the j-th linear phase shift amount be denoted as Then the linear phase shift amount corresponding to the frequency f k can be calculated by the following formula,

[0019]

[0020] where f Nyquist is the cut-off frequency;

[0021] (3) Calculate the phase spectrum θ D (f) after the linear phase shift to obtain the phase spectrum θ Dl (f), where f is the frequency

[0022]

[0023] and adjust θ Dl (f) by an integer multiple of 2π to make it satisfy -π ≤ θ Dl (f) ≤ π;

[0024] (4) For the same time window area in the target data C, calculate the phase spectrum θ C (f) of the i-th trace data in the time window, and calculate the difference θ C (f) between θ Dl (f) and θ le (f),

[0025] θ le (f) = θ C (f) - θ Dl (f)

[0026] And adjust θ le (f) by an integer multiple of 2π to make it satisfy -π ≤ θ le (f) ≤ π;

[0027] (5) Calculate the amplitude spectrum of the i-th trace in the time window of data D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from small to large, and after removing one-third of the minimum values, the frequencies corresponding to the remaining amplitude spectrum values are denoted as f s1 , f s2 , L, f sn , and use θ le (f s1 ), θ le (f s2 ), L, θ le (f sn ) as the phase difference after removing outliers;

[0028] (6) Calculate the variance of the phase difference after removing outliers corresponding to the i-th trace data

[0029]

[0030] In the formula, sn is the number of frequencies;

[0031] (7) For the i-th trace data in the time window, loop for j, repeat steps (2)-(6) to obtain the variances corresponding to all phase shift amounts of the i-th trace data, and find the phase shift amount corresponding to the minimum variance among them as the optimal linear phase shift amount of the i-th trace data;

[0032] (8) Loop for i, repeat steps (1)-(7) to obtain the optimal linear phase shift amounts corresponding to each trace in the overlapping area, sort the optimal linear phase shift amounts corresponding to each trace from small to large, remove one-fourth of the maximum values and one-fourth of the minimum values, and take the mean of the remaining part as the final optimal linear phase shift amount applicable to all traces

[0033] In step 2, on the basis of completing step 1, estimate the constant phase correction amount for data D according to the following steps.

[0034] (21) Perform linear phase shift correction processing on the phase spectrum of the i-th trace data in the time window of data D to obtain θ' Dl (f),

[0035]

[0036] where θ(f) is the phase spectrum;

[0037] (22) Calculate the phase spectrum θ of the data in the i-th trace in the same time window region of the target data C C (f) and the phase spectrum θ' Dl (f) of the phase difference θ ce (f),

[0038] θ ce (f) = θ C (f) - θ' Dl (f)

[0039] And make an integer multiple adjustment of 2π to θ ce (f) to make it satisfy -π ≤ θ ce (f) ≤ π;

[0040] (23) Calculate the amplitude spectrum of the i-th trace in the time window of data D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from large to small, and after removing one-third of the minimum values, the frequencies corresponding to the remaining amplitude spectrum values are denoted as f s1 , f s2 , L, f sn , and use θ ce (f s1 ), θ ce (f s2 ), L, θ ce (f sn ) as the phase difference after removing outliers;

[0041] [[ID=:49]](24) Calculate the mean value of the phase difference after removing outliers as the best constant phase shift amount for the i-th trace data

[0042] r

[0043] (25) Loop for i, repeat steps (21)-(24) to obtain the best constant phase shift amounts for each trace in the overlapping region, sort the best constant phase shift amounts for each trace from small to large, remove one-fourth of the maximum values and one-fourth of the minimum values, and take the mean value of the remaining part as the final best constant phase shift amount applicable to all traces

[0044] In step 3, on the basis of completing step 2, perform phase consistency processing on the data B to be processed according to the following steps;

[0045] (31) Calculate the amplitude spectrum S B (f) and the phase spectrum θ B (f);

[0046] (32) For the phase spectrum θB (f) Perform constant phase shift and linear phase shift processing to obtain the processed phase spectrum θ Bcl (f),

[0047]

[0048] (33) Calculate the output O' after phase congruency processing B (t),

[0049] O B (t) = Re(IFFT(S B (f) · (cos(θ Bcl (f)) + 1i · sin(θ Bcl (f)))))

[0050] where Re() represents taking the real part, 1i represents the imaginary unit, IFFT() represents the inverse Fourier transform, and O B (t) represents the result after phase congruency processing of data B, and the data after phase congruency processing of the original data D in the overlapping area can be represented as O D (t).

[0051] In step 4, on the basis of completing step 3, calculate the amplitude spectra S D (f, i) within the time window of data C and the data O C (t) after phase congruency processing, and Sort the magnitudes of the amplitude spectra of all traces at a certain frequency from small to large, remove one - quarter of the maximum values and one - quarter of the minimum values, calculate the mean of the remaining part, and perform this processing for all frequencies to obtain the statistical amplitude spectra of data C and data O D (t) and The amplitude spectrum matching operator P(f) is

[0052]

[0053] In step 5, on the basis of completing step 4, perform amplitude spectrum congruency processing on the result O B (t) after phase congruency processing using the amplitude spectrum matching operator P(f),

[0054]

[0055] where represents the amplitude spectrum of data O B (t), represents the phase spectrum of data O B (t), and Q B (t) represents the final result after amplitude spectrum congruency processing.

[0056] In step 5, according to Q B (t), the output of the coherent data consistency processing result with multi-trace statistics and signal-to-noise ratio constraints is obtained.

[0057] The coherent data consistency processing method based on multi-trace statistics and signal-to-noise ratio constraints in the present invention can be applied to the coherent consistency processing of two-phase data with large differences in signal-to-noise ratio. Through multi-trace statistics and signal-to-noise ratio constraint processing, the stability and accuracy of the matching operator are improved. The consistency processing method includes steps such as estimating the linear phase shift amount, estimating the constant phase shift amount, phase consistency processing, amplitude spectrum matching operator estimation, and amplitude spectrum consistency processing. When estimating the consistency processing operator, the method optimizes the data information, eliminates the participation of low signal-to-noise ratio information, and considers the influence of abnormal trace information. Through multi-trace statistics and data information quality screening and elimination, the stability and accuracy of the estimation of the coherent data consistency processing operator for low-quality data are improved, and it also has good adaptability to the coherent data consistency processing with large quality differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a flowchart of a specific embodiment of the coherent data consistency processing method based on multi-trace statistics and signal-to-noise ratio constraints of the present invention;

[0059] Figure 2 is a schematic diagram of two data volumes to be coherently stitched and processed in a specific embodiment of the present invention;

[0060] Figure 3 is a schematic diagram of the stitching display before the coherent processing of two data volumes in a specific embodiment of the present invention;

[0061] Figure 4 is a comparison diagram of the amplitude spectra of the overlapping area before the coherent processing of two data volumes in a specific embodiment of the present invention;

[0062] Figure 5 is a planar display diagram of the optimal linear phase shift amount of each trace data in the overlapping area estimated in a specific embodiment of the present invention;

[0063] Figure 6 is a display diagram of the optimal linear phase shift amount of each trace data in the overlapping area estimated in a specific embodiment of the present invention sorted from small to large;

[0064] Figure 7 is a planar display diagram of the optimal constant phase shift amount of each trace data in the overlapping area estimated in a specific embodiment of the present invention;

[0065] Figure 8 is a display diagram of the optimal constant phase shift amount of each trace data in the overlapping area estimated in a specific embodiment of the present invention sorted from small to large;

[0066] Figure 9 This is the amplitude spectrum comparison diagram of the overlapping area after the consistency processing of the two-block data bodies in a specific embodiment of the present invention;

[0067] Figure 10 This is the cross-section display diagram after the splicing of the two-block consistency processing in a specific embodiment of the present invention;

[0068] Figure 11 This is the time slice display diagram (time: 2250 ms) after the splicing of the two-block consistency processing in a specific embodiment of the present invention. Detailed implementation manners

[0069] It should be noted that the following detailed description is exemplary and is intended to provide further illustration of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0070] It should be noted that the terms used herein are only for describing specific implementation manners and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they specify the presence of features, steps, operations, and / or combinations thereof.

[0071] As Figure 1 shown, Figure 1 This is the flow chart of the continuous data consistency processing method based on multi-channel statistics and signal-to-noise ratio constraint of the present invention. The continuous data consistency processing method based on multi-channel statistics and signal-to-noise ratio constraint includes the following steps:

[0072] Step 1: Estimate the linear phase shift amount;

[0073] Step 2: Estimate the constant phase shift amount;

[0074] Step 3: Phase consistency processing;

[0075] Step 4: Estimate the amplitude spectrum matching operator;

[0076] Step 5: Amplitude spectrum consistency processing.

[0077] The following are several specific embodiments applying the present invention.

[0078] Embodiment 1

[0079] In a specific embodiment 1 applying the present invention, the continuous data consistency processing method based on multi-channel statistics and signal-to-noise ratio constraint includes the following steps:

[0080] Step 1: Estimate the linear phase shift amount;

[0081] For two blocks that need to be processed in a continuous manner, there is partial overlap in the data of the two blocks. Take the data of the block with higher data signal-to-noise ratio and resolution as the target data, denoted as A, and set the data in its overlapping area as C. The block with relatively lower data signal-to-noise ratio and resolution is used as the data to be processed, denoted as B, and set the data in its overlapping area as D. Estimate the linear phase shift amount in the following way:

[0082] (1) Select a time window area with a relatively high signal-to-noise ratio in data D, and calculate the phase spectrum θ D (f) of the i-th trace data in the time window;

[0083] (2) Given the scanning range of the linear phase shift amount, scan from small to large within the scanning range according to the linear phase shift amount at equal intervals. Let the j-th linear phase shift amount be denoted as Then the linear phase shift amount corresponding to the frequency f k can be calculated by the following formula,

[0084]

[0085] where f Nyquist is the cut-off frequency.

[0086] (3) Calculate the phase spectrum θ D (f) after the linear phase shift and denote it as θ Dl (f),

[0087]

[0088] and adjust θ Dl (f) by an integer multiple of 2π to make it satisfy -π ≤ θ Dl (f) ≤ π.

[0089] (4) For the same time window area in the target data C, calculate the phase spectrum θ C (f) of the i-th trace data in the time window, and calculate the difference θ C (f) between θ Dl (f) and θ le (f),

[0090] θ le (f) = θ C (f) - θ Dl (f)

[0091] and adjust θ le (f) by an integer multiple of 2π to make it satisfy -π ≤ θ le (f) ≤ π.

[0092] (5) Calculate the amplitude spectrum of the i-th trace in the time window of data D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from small to large, and after excluding one-third of the minimum values, the frequencies corresponding to the remaining amplitude spectrum values are denoted as f s1 , f s2 , L, f sn , and take θ le (f s1 ), θ le (f s2 ), L, θ le (f sn ) as the phase difference after excluding outliers.

[0093] (6) Calculate the variance of the phase difference after excluding outliers corresponding to the i-th trace data

[0094]

[0095] (7) For the i-th trace data in the time window, loop for j, and repeat steps (2)-(6) to obtain the variances corresponding to all phase shifts of the i-th trace data, and find the phase shift corresponding to the minimum variance among them as the optimal linear phase shift of the i-th trace data.

[0096] (8) Loop for i, repeat steps (1)-(7) to obtain the optimal linear phase shifts corresponding to each trace in the overlapping region, sort the optimal linear phase shifts corresponding to each trace from small to large, exclude one-fourth of the maximum values and one-fourth of the minimum values, and take the mean of the remaining part as the final optimal linear phase shift applicable to all traces

[0097] Step 2: Estimate the constant phase shift;

[0098] On the basis of completing Step 1, estimate the constant phase correction amount for data D according to the following steps.

[0099] (1) Perform linear phase shift correction processing on the phase spectrum of the i-th trace data in the time window of data D to obtain θ' Dl (f),

[0100]

[0101] (2) Calculate the phase difference θ C (f) between the phase spectrum θ Dl (f) of the i-th trace data in the same time window region of the target data C and the phase spectrum θ' ce (f) after linear phase shift correction,

[0102] θ ce (f) = θ C (f) - θ' Dl(f)

[0103] And for θ ce (f) Adjust the integer multiple of 2π to satisfy -π≤θ ce (f)≤π.

[0104] (3) Calculate the amplitude spectrum of the i-th channel in the data time window D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from large to small, and after removing one-third of the minimum values, the frequency corresponding to the remaining amplitude spectrum values is recorded as f s1 ,f s2 ,L,f sn , and θ ce (f s1 ),θ ce (f s2 ),L,θ ce (f sn ) as the phase difference after removing outliers.

[0105] (4) Calculate the mean of the phase difference after removing the outliers as the optimal constant phase shift of the i-th channel data

[0106]

[0107] (5) Repeat steps (1) to (4) for loop i to obtain the optimal constant phase shift for each channel in the overlapping area. Sort the optimal constant phase shift for each channel from small to large, remove one-quarter of the maximum values and one-quarter of the minimum values, and take the average of the remaining values as the final optimal constant phase shift for all channels.

[0108] Step 3: Phase consistency processing;

[0109] After completing step 2, phase consistency processing is performed on the data to be processed B according to the following steps.

[0110] (1) Calculate the amplitude spectrum S of the data to be processed B B (f) and phase spectrum θ B (f);

[0111] (2) Phase spectrum θ B (f) Constant phase shift and linear phase shift processing are performed to obtain the processed phase spectrum θ Bcl (f),

[0112]

[0113] (3) Calculate the output O' after phase consistency processing B (t),

[0114] OB (t) = Re(IFFT(S B (f)·(cos(θ Bcl (f)) + 1i·sin(θ Bcl (f))))),

[0115] where Re() represents taking the real part, 1i represents the imaginary unit, IFFT() represents the inverse Fourier transform, and O B (t) represents the result after performing phase consistency processing on data B. The data obtained by performing phase consistency processing on the original data D in the overlapping area can be represented as O D (t).

[0116] Step 4: Estimation of the amplitude spectrum matching operator;

[0117] On the basis of completing Step 3, calculate the amplitude spectra S D (t) of each trace within the time window of data C and the data O C (f,i) and For all traces at a certain frequency, sort the amplitudes of the amplitude spectra from smallest to largest, eliminate one-fourth of the maximum values and one-fourth of the minimum values, calculate the mean of the remaining part, and perform this processing for all frequencies to obtain the statistical amplitude spectra of data C and data O D (t) and The amplitude spectrum matching operator P(f) is

[0118]

[0119] Step 5: Amplitude spectrum consistency processing;

[0120] On the basis of completing Step 4, perform amplitude spectrum consistency processing on the result O B (t) after phase consistency processing, using the amplitude spectrum matching operator P(f),

[0121]

[0122] where represents the amplitude spectrum of data O B (t), represents the phase spectrum of data O B (t), and Q B (t) represents the final result after amplitude spectrum consistency processing.

[0123] Thus, the output of the coherent data processing result of multi-trace statistics and signal-to-noise ratio constraint can be obtained.

[0124] Example 2

[0125] In a specific embodiment 2 of applying the present invention, the method for processing the consistency of contiguous data provided by the present invention is described in detail in combination with typical data. As Figures 2 - 11 shown, where Figure 2 are two data bodies to be processed for consistency splicing, (a) Cross-sectional view of Block A, (b) Cross-sectional view of Block B. Figure 3 is the splicing display before the consistency processing of the two data bodies, (a) Cross-sectional view, (b) Time slice map at 2250 ms. It can be seen from the spliced cross-sectional view that there are obvious discontinuities in the in-phase axis near the adjacent traces of the two-block data, and obvious discontinuities can also be seen in the adjacent areas of the two blocks from the time slice, indicating that there is a certain phase difference between the data of the two blocks. Figure 4 is the comparison diagram of the amplitude spectra in the overlapping area before the consistency processing of the two data bodies. It can be seen from the figure that there are also certain differences in the amplitude spectra of the two data bodies. To perform the consistency processing of contiguous data, it is necessary to eliminate the phase difference between the two blocks and the differences in aspects such as frequency band and energy. The Xline numbers in the overlapping range of Block A and Block B data are: 825 - 935, and the preferred time window range is: 1200 ms - 3000 ms. The scanning range of the linear phase shift amount is set to -10π - 10π, with an interval of The preferred frequency band range is 10 - 50 Hz. Figure 5 is the planar display diagram of the optimal linear phase shift amount of each trace data in the estimated overlapping area. Affected by factors such as the difference in signal-to-noise ratio, abnormal estimated values of local traces can be seen in the planar diagram of the estimated optimal linear phase shift amount, but the estimated values of most traces are concentrated within a certain range. Further examine the focusing situation of the optimal linear phase shift amount values of each trace, Figure 6 is the display diagram of the optimal linear phase shift amount of each trace data in the estimated overlapping area sorted from small to large. It can be seen from the figure that, except for a few maximum and minimum values, most values are concentrated near 0. After removing one-fourth of the maximum value part and one-fourth of the minimum value part, the mean value of the remaining values calculated is -0.049π, which is the optimal linear phase shift amount applicable to the entire block of data. Figure 7 is the planar display diagram of the optimal constant phase shift amount of each trace data in the estimated overlapping area; it can be seen from the figure that, except for a few maximum and minimum values, most values are concentrated within a certain range. Further examine the focusing situation of the optimal constant phase shift amount values of each trace, Figure 8 is the display diagram of the optimal linear phase shift amount of each trace data in the estimated overlapping area sorted from small to large. It can be seen from the figure that, except for a few maximum and minimum values, most values are concentrated in the range of -130° to -140°. After removing one-fourth of the maximum value part and one-fourth of the minimum value part, the mean value of the remaining values calculated is -134.9°, which is the optimal constant phase shift amount applicable to the entire block of data. Figure 9It is the amplitude spectrum comparison chart of the overlapping area after the consistency processing of the two-block data bodies. After the consistency processing, the amplitudes and frequency band widths of the amplitude spectra of the two-block data are basically at the same level. Figure 10 It is the sectional view display chart after the splicing of the two-block consistency processing. Compare Figure 3 a and Figure 10 It can be seen that after the consistency processing, the in-phase axis is very continuous near the adjacent traces of the two-block data, and the consistency of the energy is also good. Figure 11 It is the time slice display after the splicing of the two-block consistency processing. Compare Figure 3 b and Figure 11 It can be seen that after the consistency processing, the disconnection in the splicing area on the time slice is significantly improved, indicating that the consistency processing operator of the continuous data estimated by this method has high precision.

[0126] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0127] Except for the technical features described in the specification, they are all well-known technologies to those skilled in the art.

Claims

1. A method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraints, characterized in that The method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraints includes: Step 1: Estimate the linear phase shift amount; Step 2: Estimate the constant phase shift amount; Step 3: Perform phase coherence processing; Step 4: Estimate the amplitude spectrum matching operator; Step 5: Perform amplitude spectrum coherence processing; In Step 1, for two blocks that need to be processed continuously, there is partial overlap in the data of the two blocks; the data of the block with higher data signal-to-noise ratio and resolution is used as the target data, denoted as A, and the data in its overlapping area is denoted as C; the block with relatively lower data signal-to-noise ratio and resolution is used as the data to be processed, denoted as B, and the data in its overlapping area is denoted as D; In Step 1, the linear phase shift amount is estimated in the following manner: (1) Select the time window region with a relatively high signal-to-noise ratio in the data D, and calculate the phase spectrum θ D (f) of the i-th trace data in the time window; (2) Given the scanning range of the linear phase shift amount, scan from small to large within the scanning range according to the linear phase shift amount at equal intervals; let the j-th linear phase shift amount be denoted as Then the frequency f k The corresponding linear phase shift amount is calculated by the following formula. where f Nyquist is the cut-off frequency; (3) Calculate the phase spectrum θ D (f) After linear phase shift θ l j The phase spectrum θ Dl (f), where f is the frequency; And perform an integer multiple of 2π adjustment on θ Dl (f) so that it satisfies -π ≤ θ Dl (f) ≤ π; (4) For the same time window region in the target data C, calculate the phase spectrum θ C (f) of the i-th trace data in the time window, and calculate the difference θ C (f) between θ Dl (f) and θ le (f). θ le (f) = θ C (f) - θ Dl (f) And perform an adjustment on θ le (f) by an integer multiple of 2π to make it satisfy -π ≤ θ le (f) ≤ π; (5) Calculate the amplitude spectrum of the i-th trace in the time window of data D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from small to large, and after removing one-third of the minimum values, the frequencies corresponding to the remaining amplitude spectrum values are denoted as f s1 , f s2 , …, f sn , and take θ le (f s1 ), θ le (f s2 ), …, θ le (f sn ) as the phase differences after removing outliers; (6) Calculate the variance of the phase difference after removing outliers corresponding to the i-th data Where sn is the number of frequencies; (7) For the i-th trace data in the time window, loop for j, repeat steps (2)-(6), obtain the variances corresponding to all phase shift amounts of the i-th trace data, and find the phase shift amount corresponding to the minimum variance among them as the optimal linear phase shift amount of the i-th trace data; (8) Repeat steps (1) to (7) for loop i to obtain the optimal linear phase shift corresponding to each channel in the overlapping area. Sort the optimal linear phase shift corresponding to each channel from small to large, remove one-quarter of the maximum values and one-quarter of the minimum values, and take the average of the remaining values as the final optimal linear phase shift applicable to all channels.

2. The method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraint according to claim 1, wherein In Step 2, on the basis of completing Step 1, the constant phase correction amount is estimated for the data D according to the following steps; (21) Perform linear phase shift correction processing on the phase spectrum of the i-th trace data in the time window of data D to obtain θ' Dl (f), Where θ(f) is the phase spectrum; (22) Calculate the phase spectrum θ of the data in the i-th trace in the same time window region of the target data C C (f) and the phase spectrum θ' Dl (f) after linear phase shift correction ce (f), θ ce f(θ) = θ C f(θ) - θ D ' l f(θ) And make an integer multiple adjustment of θ ce (f) by an integer multiple of 2π to make it satisfy -π ≤ θ ce (f) ≤ π; (23) Calculate the amplitude spectrum of the i-th trace in the time window of the calculation data D, determine the range of the effective frequency band based on the amplitude spectrum, sort the amplitude spectrum values within the effective frequency band from large to small, and after removing one-third of the minimum values, the frequencies corresponding to the remaining amplitude spectrum values are denoted as f s1 , f s2 , …, f sn , and take θ ce (f s1 ), θ ce (f s2 ), …, θ ce (f sn ) as the phase differences after removing outliers; (24) Calculate the mean of the phase differences after removing outliers as the best constant phase shift amount for the data of the i-th channel (25) For the loop of i, repeat steps (21)-(24) to obtain the optimal constant phase shift amounts for each trace in the overlapping area. Sort the optimal constant phase shift amounts for each trace from smallest to largest, eliminate one-fourth of the maximum values and one-fourth of the minimum values, and take the mean of the remaining part as the final optimal constant phase shift amount applicable to all traces.

3. The method for processing the coherence of contiguous data based on multi-channel statistics and signal-to-noise ratio constraint according to claim 2, wherein In Step 3, on the basis of completing Step 2, the phase coherence processing is performed on the data B to be processed according to the following steps; (31) Calculate the amplitude spectrum S of the data B to be processed B (f) and the phase spectrum θ B (f); (32)Perform constant phase shift and linear phase shift processing on the phase spectrum θ B (f) to obtain the processed phase spectrum θ Bcl (f). (33) Calculate the output O' after the phase consistency processing B (t), O B (t) = Re(IFFT(S B (f)·(cos(θ Bcl (f)) + 1i·sin(θ Bcl (f))))), Among them, Re() represents taking the real part, 1i represents the imaginary unit, IFFT() represents the inverse Fourier transform, and O B (t) represents the result after performing phase consistency processing on data B, and the data obtained by performing phase consistency processing on the original data D in the overlapping area is represented as O D (t).

4. The method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraint according to claim 3, wherein In step 4, based on the completion of step 3, calculate the data C and the data O after phase consistency processing respectively D (t) of the amplitude spectra S of each trace within the time window C (f,i) and Sort the amplitudes of the amplitude spectra of all traces at a certain frequency from small to large, eliminate one-fourth of the maximum values and one-fourth of the minimum values, calculate the mean of the remaining part, and perform this processing for all frequencies to obtain the data C and the data O D (t) of the statistical amplitude spectrum and The amplitude spectrum matching operator P(f) is 5. The method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraint according to claim 4, wherein In step 5, based on the completion of step 4, for the result O B (t) after phase consistency processing, amplitude spectrum consistency processing is performed using the amplitude spectrum matching operator P(f). Among them, represents the amplitude spectrum of data O B (t), represents the phase spectrum of data O B (t), Q B (t) represents the final result after the amplitude spectrum consistency processing.

6. The method for processing the coherence of continuous data based on multi-channel statistics and signal-to-noise ratio constraint according to claim 5, wherein In step 5, according to Q B (t), the output of the coherent data consistency processing result with multi-trace statistics and signal-to-noise ratio constraints is obtained.

Images