A method and system for seismic data amplitude compensation

Abstract

The present application relates to a kind of seismic data amplitude compensation method and system.The method includes obtaining seismic data;The seismic data obtained is carried out surface consistency amplitude compensation, and the analysis is carried out to the stack section of the compensated seismic data, determines that energy lower area and energy higher area, at least one single shot data is extracted in energy lower area and energy higher area respectively, according to the frequency characteristics of each single shot data, according to the principle that the frequency characteristics trend of each single shot data in each frequency band is consistent, the frequency of seismic data is divided;The seismic data of each frequency band is respectively carried out surface consistency amplitude compensation.In the present application, the problem that the amplitude compensation effect is not obvious in prior art can be solved.

Description

Technical Field

[0001] This invention belongs to the field of seismic exploration technology, specifically relating to a method and system for seismic data amplitude compensation. Background Technology

[0002] Amplitude compensation processing in seismic data processing involves analyzing and processing the amplitude of seismic data to minimize differences in amplitude and frequency caused by factors such as spherical diffusion, absorption attenuation, excitation of strata, and non-geological factors during acquisition. This aims to maintain a relative balance in the amplitude energy of the seismic data while preserving its relative relationship, effectively reflecting the true state of the underground strata and eliminating the impact of energy unevenness on seismic data in subsequent processing.

[0003] Currently, the three most widely used methods for amplitude compensation are spherical diffusion compensation, surface consistency amplitude compensation, and Q-compensation. First, spherical diffusion compensation primarily uses the ratio T / V² of time T and the corresponding stratum's stacking velocity V to calculate a compensation factor. This compensation factor is then applied to the seismic data volume, effectively addressing the energy absorption and attenuation problem during seismic wave propagation, thus compensating the energy of the same single-shot seismic data to the same energy level. Second, surface consistency amplitude compensation statistically analyzes the energy values ​​of the seismic data across four dimensions: single-shot domain, common receiver domain, CMP domain, and common offset domain. The energy of all data is averaged, and this average is then used to automatically weight and adjust all single-shot data. Shots with higher energy have lower weights, and vice versa, adjusting the seismic data energy to the same energy level, achieving relative balance. Surface consistency amplitude compensation mainly addresses energy issues caused by surface factors, specifically the energy differences between shots and traces due to surface inconsistencies. Third, Q-compensation primarily compensates for data energy differences by calculating the Q value of the strata. There are many methods for determining the Q factor. Whether it's using amplitude characteristics in the time domain, applying amplitude spectrum, dominant frequency, and bandwidth in the frequency domain, statistically analyzing VSP data, or applying time-frequency spectrum analysis, all these methods extend and optimize the amplitude attributes of seismic data. The Q factor is calculated using the ratio of seismic attributes of the upper and lower reflecting layers, S(t1) / S(t2) (where S(t) is the seismic attribute of the reflecting layer). In other words, Q compensation primarily compensates for differences in seismic data energy over time. Compared to the effect of surface consistency amplitude compensation, Q compensation is mainly applied to compensate for the residual amplitude after surface consistency amplitude compensation. By combining these three methods in actual production, most energy problems related to seismic data can be solved.

[0004] However, during seismic data acquisition, significant differences in energy and spectrum arise due to the influence of near-surface strata, particularly in mountainous areas where near-surface conditions change rapidly. Surface uniformity amplitude compensation can effectively address these energy differences caused by near-surface strata. However, this method uses the energy values ​​of the dominant frequency bands in the seismic data. In other words, due to surface influences, high-frequency or low-frequency data from certain areas with poor single-shot excitation conditions suffer severe energy loss, while other frequency bands remain strong. Consequently, surface uniformity amplitude compensation fails to effectively compensate for the energy loss in frequency bands that are most severely affected by surface energy loss because it uses the dominant frequency band's compensation weighting coefficients. Spherical diffusion compensation only addresses the diffusion loss of seismic waves. Q-compensation methods, applied to individual seismic trace data volumes, theoretically address some spectral issues in the frequency and time domains, but their effectiveness in resolving lateral (spatial) problems is limited in practical applications.

[0005] Secondly, due to the influence of underground structures, the acquired seismic data often exhibits the following phenomenon: energy is weak at structural high points and strong at structural low points, somewhat like a convex and concave mirror; energy is dispersed at structural high points with lower frequencies, while energy is concentrated at structural low points with abundant high-frequency information. The seismic data obtained using the aforementioned compensation method still suffers from the problem of weak energy at structural high points and strong energy at structural low points, failing to fundamentally resolve the issue of significant energy distribution differences caused by underground structures.

[0006] Therefore, current earthquake data amplitude compensation methods cannot address the influence of near-surface strata and underground structures, resulting in insignificant earthquake amplitude compensation effects. Summary of the Invention

[0007] This invention provides a method and system for earthquake data amplitude compensation, which solves the problem of insignificant amplitude compensation effect in the prior art.

[0008] To address the aforementioned technical problems, this invention provides a seismic data amplitude compensation method, comprising: 1) acquiring seismic data; 2) performing surface consistency amplitude compensation on the acquired seismic data, analyzing the superimposed profile of the compensated seismic data to identify low-energy and high-energy regions, extracting at least one shot data point from each of the low-energy and high-energy regions, and dividing the seismic data into frequencies based on the frequency characteristics of each shot data point, adhering to the principle of maintaining consistent frequency characteristic trends within each frequency band; 3) performing surface consistency amplitude compensation on the seismic data of each frequency band separately, and merging the compensated seismic data of each frequency band.

[0009] The beneficial effects of the above technical solution are as follows: Surface consistency amplitude compensation is performed on the acquired seismic data; the stacked profile of the compensated seismic data is analyzed to determine different energy regions; single-shot data from different energy regions are extracted; and the seismic data is frequency-divided according to the principle of maintaining consistent frequency characteristic trends among the single-shot data within each frequency band, based on the frequency characteristics of the extracted single-shot data. Therefore, compared with existing surface consistency amplitude compensation methods that do not divide frequencies, this invention can fully consider the energy differences between data from different frequency bands, and perform surface consistency amplitude compensation separately on the seismic data of each frequency band, enabling the energy within each frequency band to be consistent and improving the amplitude compensation effect. This effectively solves the problem of insignificant amplitude compensation effect caused by near-surface strata and underground structures.

[0010] Furthermore, in order to eliminate energy differences in the time direction, the present invention provides a method for earthquake data amplitude compensation, which further includes performing spherical diffusion compensation processing on the acquired earthquake data in step 1).

[0011] Furthermore, in order to remove significant noise generated during the acquisition process, this invention provides a seismic data amplitude compensation method, which also includes pre-stack noise suppression of the acquired seismic data before performing spherical diffusion compensation processing.

[0012] Furthermore, in order to better simplify the calculation, the present invention provides a seismic data amplitude compensation method, which also includes that the superimposed profile in step 2) is obtained by superimposing in an unweighted superposition method.

[0013] Furthermore, in order to ensure the effectiveness of subsequent amplitude compensation, the present invention provides a method for earthquake data amplitude compensation, which also includes a frequency band of 5-7 when performing frequency division.

[0014] Furthermore, in order to better perform frequency division, the present invention provides a method for earthquake data amplitude compensation, which also includes dividing the frequency bands in step 2) using bandpass filtering technology or wavelet transform technology.

[0015] Furthermore, in order to better ensure that the energy of each frequency band is basically consistent, the present invention provides a seismic data amplitude compensation method, which also includes comparing the energy of the seismic data of each frequency band after surface consistency amplitude compensation in step 3), determining whether the energy difference of each frequency band is greater than a set threshold, and if it is greater, then performing energy adjustment so that the energy difference of each frequency band is less than the set threshold.

[0016] Furthermore, in order to fully consider the data characteristics of the frequency band criticality, the present invention provides a seismic data amplitude compensation method, which also includes a transition band between adjacent frequency bands, the bandwidth of which is 4-5Hz.

[0017] Furthermore, in order to more accurately delineate energy regions, this invention provides a seismic data amplitude compensation method, which also includes obtaining the main reflection layer of the stacked profile, selecting multiple time windows from the main reflection layer and performing root mean square (RMS) processing to obtain the corresponding RMS amplitude values. If the RMS amplitude value of a certain time window is less than 1 / 2 of the RMS amplitude value of the time window with the largest amplitude value, then the time window is a region with weak energy. If the RMS amplitude value is greater than 4 / 5 of the RMS amplitude value of the time window with the largest amplitude value, then the time window is a region with high energy.

[0018] To address the aforementioned technical problems, the present invention provides a seismic data amplitude compensation system, comprising: a memory and a processor, wherein the processor is configured to execute instructions stored in the memory to implement the aforementioned seismic data amplitude compensation method. Attached Figure Description

[0019] Figure 1 This is a flowchart of the seismic data amplitude compensation method of the present invention;

[0020] Figure 2 This is a schematic diagram of the data after noise suppression and spherical amplitude compensation according to the present invention;

[0021] Figure 3 It utilizes surface uniform amplitude compensation to process and overlay profiles;

[0022] Figure 4a yes Figure 3 Schematic diagram of the first single-shot data extraction of seismic data in the circled area on the right side of the middle;

[0023] Figure 4b yes Figure 3 Schematic diagram of seismic data extraction for the second single shot in the circled area on the left side of the middle;

[0024] Figure 4c yes Figure 3 Schematic diagram of the third single-shot data extraction from seismic data in medium-to-high energy regions;

[0025] Figure 5 This is a schematic diagram comparing the spectrum analysis of the first, second, and third single-shot data of the present invention.

[0026] Figure 6 This is a schematic diagram of the frequency band division method of the present invention;

[0027] Figure 7 This is a schematic diagram of the frequency band division of seismic data according to the present invention;

[0028] Figure 8 This is a schematic diagram of seismic data for each frequency band before and after frequency division according to the present invention;

[0029] Figure 9This is a schematic diagram of the spectrum analysis after surface uniform amplitude compensation for each frequency band of the present invention;

[0030] Figure 10 This is a schematic diagram of the seismic data across the entire frequency band and each frequency band after processing according to the present invention;

[0031] Figure 11 This is a schematic diagram of the spectrum analysis of each frequency band and the resulting data after processing by this invention;

[0032] Figure 12 This is a schematic diagram comparing the spectrum analysis before and after the processing of this invention;

[0033] Figure 13 This is a superimposed profile of the surface uniform amplitude compensation processing results of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and technical effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] Example of seismic data amplitude compensation method:

[0036] This embodiment provides a method for seismic data amplitude compensation. The method includes acquiring seismic data; performing surface consistency amplitude compensation on the acquired seismic data; analyzing the stacked profile of the compensated seismic data to identify low-energy and high-energy regions; extracting at least one shot data point from each of the low-energy and high-energy regions; dividing the seismic data into frequencies based on the frequency characteristics of each shot data point, adhering to the principle of maintaining consistent frequency characteristic trends within each frequency band; performing surface consistency amplitude compensation on the seismic data in each frequency band; and merging the compensated seismic data from each frequency band. The seismic data amplitude compensation method of this embodiment can solve the problem of insignificant amplitude compensation effects in existing technologies.

[0037] Figure 1 This is a flowchart of the seismic data amplitude compensation method of the present invention. The specific process is as follows:

[0038] Step 1: Obtain the seismic data from the original shot gather.

[0039] Specifically, in step one, the raw shot gather can come from seismic acquisition data. The seismic data can be the raw data from the seismic acquisition data.

[0040] Step 2: Perform spherical diffusion compensation processing on the seismic data.

[0041] Specifically, in step two, spherical diffusion compensation processing is performed on the seismic data to eliminate the influence of energy loss during the propagation of seismic waves, so that the seismic data reaches a relatively stable state within the same single shot and the energy difference is eliminated in the time direction.

[0042] In step two, as Figure 1 As shown, seismic data can be preprocessed before spherical diffusion compensation. Preprocessing can include noise suppression using pre-stack noise reduction techniques. In this case, noise suppression is performed on the seismic data. This can suppress high-energy noise such as surface waves, random noise, and linear interference in the seismic data, removing significant noise generated during acquisition. For example, performing noise suppression and spherical diffusion compensation on a specific 3D seismic dataset can yield results such as... Figure 2 The data diagram shown is based on... Figure 2 It can be seen that there are no prominent energy anomalies in the data.

[0043] Step 3: Perform frequency division processing on the data after spherical diffusion compensation to obtain multiple frequency bands.

[0044] In step three, as Figure 1 As shown, the frequency division processing method specifically includes: performing surface consistency amplitude compensation on the data after spherical diffusion compensation processing; analyzing the stacked profile of the compensated seismic data to determine low-energy regions (i.e., weak-energy regions) and high-energy regions (i.e., high-energy regions); extracting single-shot data from the high-energy and weak-energy regions respectively; performing spectral analysis and comparison on the extracted single-shot data; and formulating a frequency division scheme based on the comparison results. Specifically, surface consistency amplitude compensation can eliminate spatial differences in seismic waves, and the data after surface consistency amplitude compensation (e.g., the compensated profile) is stacked without weighting. This simplifies the calculation. Analyzing the stacked data, based on the stacked seismic data (e.g., the stacked profile), regions in the seismic data that cannot be effectively compensated by surface consistency amplitude compensation (i.e., weak-energy regions) and high-energy regions are identified. At least one single-shot data is extracted from the regions that cannot be effectively compensated, and at least one single-shot data is extracted from the high-energy regions. The spectra of single-shot data in the low-energy region are compared with those in the high-energy region. The missing spectral components of the effective frequency bands (in seismic data processing, frequency components with amplitudes above -24dB are defined as effective frequencies) of each single-shot data in the low-energy region are examined compared with the effective frequency bands of the single-shot data in the high-energy region. A frequency division scheme is then formulated based on the spectral characteristics of each single-shot data.

[0045] In this embodiment, low-energy and high-energy regions can be determined based on the amplitude of the primary reflection layer in the stacked seismic data. Specifically, the primary reflection layer in the seismic data (stacked profile) is selected as the evaluation horizon. The primary reflection layer can be one or more strata. Multiple time windows are selected from the primary reflection layer for root-mean-square (RMS) amplitude statistical analysis to obtain the RMS amplitude value. The length of each time window is similar to the stratum thickness, and the width of the time window is a predetermined number of seismic traces. For example, the predetermined number could be 100 traces. If there is no significant abrupt change in the amplitude values ​​along the same stratum, and the RMS amplitude value of a selected time window is less than half of the RMS amplitude value of the time window with the largest amplitude value, it is considered a region where surface consistency amplitude compensation cannot effectively compensate (i.e., a low-energy region); if the RMS amplitude value is greater than four-fifths of the RMS amplitude value of the time window with the largest amplitude value, it is considered a high-energy region. This allows for more accurate division of different energy regions. Furthermore, if multiple consecutive weak or strong energy windows are connected, these multiple consecutive weak or strong energy windows can be regarded as a large weak or strong energy window. This can reduce the amount of data processing.

[0046] In some embodiments, the high-energy region can be a region that has been effectively compensated through surface uniform amplitude compensation, or it can be selected from regions with better energy in the seismic acquisition data.

[0047] In step three, the frequency division scheme based on the spectral characteristics of each individual shot's data includes: dividing the spectrum according to the spectral characteristics (i.e., spectral features) of the individual shot data extracted from low-energy regions and high-energy regions. Spectral features can refer to spectral trend. Specifically, the boundary frequencies for dividing the frequency bands are determined by whether the spectral trend of the individual shot data extracted from low-energy regions and high-energy regions is consistent. Consistent spectral trend refers to the consistency of the changing trends of the spectral features and the consistency of the division of dominant frequency bands. Consistent changing trends mainly include upward trends, fluctuating trends, and downward trends. Each changing trend can be further subdivided according to the actual situation; for example, a downward trend can be divided into rapid downward and general downward trends, thus enabling a better determination of the frequency division scheme. In this case, the number of frequency bands can be determined based on the boundary frequencies. Adjacent frequency bands are separated by the boundary frequencies.

[0048] In this embodiment, the number of frequency bands needs to be determined based on the processing requirements and characteristics of the seismic data in different regions. Generally, it is divided into 5-7 frequency bands. The number of frequency bands (i.e., frequency zones) should not be too many or too few. If the number of frequency bands is too small (i.e., the frequency range of each band is too large), for frequency zones with rapidly changing energy levels and large spectral fluctuations, the effect of compensating for missing frequency band energy cannot be achieved in subsequent processing, and the spectral change trend of different data within each frequency band cannot be guaranteed to be basically consistent. If the number of frequency bands is too large, it is easy to generate a large amount of data after division, increasing the computational load. In addition, too many frequency bands can easily lead to a too small frequency range, which can easily cause distortion when using some frequency division processing. The frequency range is generally selected as one octave. Furthermore, if there are spectral characteristic differences in each single-shot data within a certain frequency band (possibly due to missing effective frequency band spectrum), the frequency band with spectral characteristic differences can be further divided into multiple bands. Considering the computational and storage burdens, as well as the insufficient fidelity and data distortion issues caused by using frequency division methods, the number of frequency bands with different spectral characteristics should not be excessive. For frequency bands outside the effective frequency range, they can be divided into a single frequency band, which will not be processed during the overall process. In this case, considering both the missing spectral portion of the effective frequency band and the processing requirements, better frequency division can be achieved.

[0049] For the purposes of this embodiment, Figure 2 This is a schematic diagram of the data after noise suppression and spherical amplitude compensation according to the present invention; Figure 3 It utilizes surface uniform amplitude compensation to process and overlay profiles; Figure 4a yes Figure 3 Schematic diagram of the first single-shot data extraction of seismic data in the circled area on the right side of the middle; Figure 4b yes Figure 3 Schematic diagram of seismic data extraction for the second single shot in the circled area on the left side of the middle; Figure 4c yes Figure 3 Schematic diagram of the third single-shot data extraction from seismic data in medium-to-high energy regions; Figure 5 This is a schematic diagram comparing the spectrum analysis of the first, second, and third single-shot data of the present invention. Figure 6 This is a schematic diagram of the frequency band allocation method of the present invention. (Regarding...) Figure 2 By performing surface consistency amplitude compensation and overlay processing on the data, we can obtain, for example... Figure 3 The superimposed cross-section shown. Based on Figure 3 The overlay profile of the seismic data shows that after surface uniformity amplitude compensation, there are still two parts of energy (e.g., Figure 3 The two circular areas in the image are lowered, resulting in unclear imaging. From Figure 3Single-shot data were extracted from the two low-energy circular regions and the high-energy region, respectively. Among them, in... Figure 3 The first single-shot data extracted from the right-hand circle region can be represented by F1. The second single-shot data extracted from the left-hand circle region can be represented by F2, and the third single-shot data extracted from the high-energy region can be represented by F. The first single-shot data F1 (e.g.) Figure 4a As shown), the second single-shot data F2 (as shown) Figure 4b (as shown) and the third single-shot data F (as shown) Figure 4c Spectral analysis and comparison were performed on each of the following (as shown). Figure 5 The horizontal axis represents frequency, and the vertical axis represents amplitude. Based on... Figure 5 Spectral analysis comparison shows that, compared with the third single-shot data F, the first single-shot data F1 and the second single-shot data F2 have varying degrees of missing low-frequency, mid-high-frequency and high-frequency components.

[0050] Frequency bands are divided based on the spectral characteristics of the three data sets: the first single-shot data F1, the second single-shot data F2, and the third single-shot data F. For example... Figure 6 The frequency band division diagram shown is as follows: Figure 6 The horizontal axis represents frequency, and the vertical axis represents amplitude. The determination is based on whether the spectral characteristics of the three data sets (first shot data F1, second shot data F2, and third shot data F) within each frequency band remain approximately consistent. Figure 2 The seismic data is divided into boundary frequencies f1, f2, f3, and f4. Based on these boundary frequencies, the data is then... Figure 2 The seismic data (i.e., the full frequency band) is divided into five frequency bands: 0-f1, f1-f2, f2-f3, f3-f4, and f4-fn (fn being the maximum frequency). For example... Figure 6 As shown, the spectra of the first single-shot data F1, the second single-shot data F2, and the third single-shot data F within the 0-f1 frequency band basically maintain the same upward trend; the f1-f2 frequency band is the main dominant frequency band for the first single-shot data F1, the second single-shot data F2, and the third single-shot data F; the spectra of the first single-shot data F1, the second single-shot data F2, and the third single-shot data F within the f2-f3 frequency band show a similar fluctuating trend; the spectra of the first single-shot data F1, the second single-shot data F2, and the third single-shot data F within the f3-f4 frequency band show a rapid downward trend.

[0051] In this embodiment, 0-f1, f1-f2, f2-f3, and f3-f4 represent the effective frequency band data, while f4-fn represent data outside the effective frequency band. F4-fn is not processed in subsequent processing, thus reducing the computational burden. Furthermore, the frequency bands outside the effective frequency band (f4-fn) do not need to be further subdivided. Generally, according to the regulations for effective frequencies in seismic processing, the highest effective frequency is around 85Hz. However, considering statistical errors, the effective frequency band division point for seismic data is set at frequency f4, with a buffer zone between frequency f4 and the specified 85Hz. f4 can be around 100Hz.

[0052] Frequency bands with differing spectral characteristics can be further subdivided. For example, within the f3-f4 band, due to differences in the spectral characteristics of the first single-shot data F1, the second single-shot data F2, and the third single-shot data F, the f3-f4 band can be further divided into multiple sub-bands. However, considering the computational and storage burden, as well as the insufficient fidelity and data distortion issues resulting from using frequency division methods, the number of sub-bands in the f3-f4 band should not be excessive.

[0053] In other embodiments, the seismic data obtained in step one or the noise-suppressed seismic data can be directly subjected to surface uniform amplitude compensation and superposition processing in sequence to obtain a frequency division scheme.

[0054] In step three, after determining the boundary frequencies, the seismic data processed in step two is divided into multiple frequency bands using seismic data frequency division techniques. The entire seismic data is full-band data. Each frequency band can be considered a sub-band of the full-band data. The seismic data in each frequency band is the sub-band data. Seismic data frequency division techniques can be, for example, bandpass filtering or wavelet transform. This allows for better frequency division. In other embodiments, the seismic data obtained in step one or the noise-suppressed seismic data can also be directly divided into multiple frequency bands.

[0055] In other embodiments, frequency bands are divided based on boundary frequencies. However, since the spectral characteristics of single-shot data extracted from low-energy regions and high-energy regions near the boundary frequency (division point) differ significantly, a transition band needs to be considered when dividing frequency bands. This ensures that the data characteristics at the frequency band boundaries are fully taken into account. The transition band exists near the boundary frequency between adjacent frequency bands. Specifically, the transition band is a frequency band within a defined range centered on the boundary frequency. Taking frequency f1 and a defined range of Δf as an example, when dividing frequency bands, frequency bands are first divided with frequency f1 as the boundary, and then a transition band is divided from the two frequency bands bounded by frequency f1. The frequency range of the transition band (the overlapping part of the two frequency bands) is from f1-Δf to f1+Δf. The bandwidth of the transition band is 2Δf. Here, f1-Δf is the boundary frequency between the transition band and the smaller frequency band of the two frequency bands. f1+Δf is the boundary frequency between the transition band and the larger frequency band of the two frequency bands. 2Δf is generally within 4-5 Hz. If the frequency band of a transition zone with significant differences in spectral characteristics is too large, the transition zone can be divided into frequency bands. This increases the number of frequency bands that can be divided.

[0056] For the purposes of this embodiment, Figure 7 This is a schematic diagram of the frequency band division of seismic data according to the present invention; Figure 8 This is a schematic diagram of seismic data for each frequency band before and after frequency division according to the present invention. Figure 7 The horizontal axis represents frequency, and the vertical axis represents amplitude. Figure 8 The diagram, from left to right, shows seismic data before and after frequency division into five bands: band 1, band 2, band 3, band 4, and band 5. Based on the above frequency division scheme, the following data was obtained: Figure 6 The data volumes of the five frequency bands shown are based on Figure 7 and Figure 8 It is known that the energy difference between the first and fourth frequency bands within the effective frequency band is greater than the energy difference between the second and third frequency bands. On the other hand, the energy of the fifth frequency band is far below that of the effective frequency band and is therefore disregarded. In this case, only the data from the first to fourth frequency bands are processed, while the data from the fifth frequency band remains unchanged.

[0057] Step 4: Perform surface uniform amplitude compensation on the seismic data of each frequency band.

[0058] In step four, as Figure 1As shown, surface consistency amplitude compensation is performed on the sub-band data within each of the defined frequency bands. This effectively avoids the situation where, when performing surface consistency amplitude compensation on data that has undergone spherical diffusion compensation, only the high-energy frequency band data is statistically analyzed, and the energy of that frequency band is used to represent the energy value of all frequency band data.

[0059] Specifically, in step four, energy statistics are performed on the data within each frequency band in the shot domain, receiver domain, offset domain, and CMP domain. The corresponding compensation weight ratio for each frequency band is calculated. Based on the corresponding compensation weight ratio for each frequency band, surface consistency amplitude compensation processing is performed on each frequency band. This ensures that the seismic data in each frequency band achieves a relative energy balance within its respective frequency band (i.e., the energy of the data in each frequency band remains consistent within a certain range), without any abnormal energy. This avoids the problem that when the seismic data quality is high, the energy differences between frequency bands are small, and compensation using the same energy value results in minimal energy differences between frequency bands. Conversely, when the seismic data quality is poor, the energy differences between frequency bands are large, and compensation using the same weight ratio leads to weaker energy frequency bands not being effectively compensated.

[0060] In this embodiment, no surface consistency compensation is performed on frequency bands outside the effective frequency band. This reduces the computational burden.

[0061] In some examples, the frequency bands processed in step four can be directly merged to form full-frequency seismic data.

[0062] Step 5: Analyze and process the data after surface uniform amplitude compensation to obtain the compensated target shot focus.

[0063] In step five, the analysis and processing can utilize spectral analysis methods to compare and adjust the energy of seismic data across different frequency bands. Specifically, statistical analysis of the overall energy level is performed on each frequency band after surface uniform amplitude compensation. The statistical analysis checks whether the energy difference between frequency bands exceeds a set threshold. If the energy difference between the seismic data of each frequency band exceeds the set threshold, energy adjustment is performed to achieve a relative balance in the energy of the data volume of each frequency band (i.e., the energy difference between each frequency band is less than the set threshold). During adjustment, adjustments are made to a few frequency bands whose energy is significantly different from that of other frequency bands. For example, if five frequency bands are divided, and four of them have relatively consistent energy, while one frequency band has energy lower or higher than the energy of the other four, then this one frequency band is adjusted. If there is no energy difference in the data of each frequency band (i.e., the energy is basically consistent, meeting the standard of a high-quality single-shot band), then no energy adjustment is performed on any frequency band. This allows for a better consistency in energy across the frequency bands.

[0064] In this embodiment, the data of each frequency band that needs to be adjusted can be multiplied by the corresponding coefficient.

[0065] In some embodiments, the adjustment can be made manually. For this embodiment, Figure 9 This is a schematic diagram of the spectrum analysis after surface uniform amplitude compensation for each frequency band of the present invention. Figure 9 The horizontal axis represents frequency, and the vertical axis represents amplitude. Based on... Figure 9 It can be seen that the energy of the data in each frequency band is basically the same.

[0066] In this embodiment, for the transition band between adjacent frequency bands, the transition band needs to be adjusted twice according to the respective adjustment methods of the two frequency bands in which the transition band is located. After adjustment, the final energy of the transition band is obtained by weighting or averaging. In this way, the original characteristics of the seismic data can be preserved.

[0067] In step five, the energy-adjusted frequency bands and the unadjusted frequency bands are merged to form a target shot gather with a full frequency range. This maintains the integrity of the acquired seismic data volume. Specifically, as... Figure 1 As shown, the energy-adjusted frequency band and the non-energy-adjusted frequency band can be merged according to their pre-division correspondence using wavelet transform inverse transform processing to obtain a complete frequency band of compensated shot gather (i.e., target shot gather), thus completing the amplitude compensation work. In some embodiments, the merging processing method can also be superposition processing.

[0068] For the purposes of this embodiment, Figure 10 This is a schematic diagram of the seismic data across the entire frequency band and each frequency band after processing according to the present invention. Figure 11 This is a schematic diagram of the spectral analysis of each frequency band and the resulting data after processing according to this invention. Figure 10 The diagrams, from left to right, show the processed full-band, first-band, second-band, third-band, fourth-band, and fifth-band seismic data. Figure 11 and Figure 12 The horizontal axis represents frequency, and the vertical axis represents amplitude. Based on Figure 10 and Figure 11 It can be seen that after the frequency bands that need energy adjustment are aligned, they are merged into a full-band data, realizing five-dimensional surface consistency amplitude compensation processing. Figure 12 This is a schematic diagram comparing the spectrum analysis before and after the processing of this invention. Figure 13 This is a superimposed profile of the surface uniform amplitude compensation processing results of the present invention. Based on Figure 12It is evident that the seismic data before processing had significant deficiencies in both high and low frequency energy. After processing, the energy across all effective frequency bands became essentially consistent, achieving an ideal spectral shape (in seismic data processing, the ideal spectral shape is generally considered to approximate a trapezoidal shape). Figure 3 and Figure 13 The area marked with a circle on the right side of the two cross-sections represents a structural high point. The area marked with a circle on the left side represents a non-structural high point. Based on Figure 3 and Figure 13 It can be seen that the marked area on the right side of both cross-sections has been significantly improved, and the marked area on the left side (energy problems caused by surface factors) has also been significantly improved.

[0069] The seismic data amplitude compensation method in this embodiment performs surface-consistent amplitude compensation on the acquired seismic data, analyzes the stacked profile of the compensated seismic data to identify different energy regions, extracts single-shot data from different energy regions, and, based on the frequency characteristics of the extracted single-shot data, divides the seismic data into frequencies according to the principle of maintaining consistent frequency characteristic trends among the single-shot data within each frequency band, fully considering the energy differences between different frequency data. Spherical diffusion compensation is used to introduce the time dimension into the surface-consistent amplitude compensation process. By adjusting the energy of data in the same frequency band, the energy loss in certain frequency bands caused during the acquisition and propagation processes is compensated, ensuring energy consistency within each frequency band. Furthermore, energy adjustment between frequency bands further ensures data energy consistency across different frequency bands, ultimately achieving overall data energy consistency and realizing the purpose of surface-consistent amplitude compensation. This method effectively compensates for seismic data energy across five dimensions: shot point, receiver point, CMP, offset, and frequency domain. It achieves good energy consistency and effectively addresses the excitation differences (i.e., missing high-frequency and low-frequency energy) caused by rapid changes in near-surface strata conditions during seismic data acquisition. Simultaneously, it significantly improves the phenomenon where seismic data exhibits weak energy and low frequencies at structural high points and strong energy and abundant high-frequency information at structural low points due to subsurface structural influences. After processing, the energy across all effective frequency bands is essentially consistent, improving amplitude compensation and achieving an ideal spectral profile. Significant improvement is observed at structural high points, and the problem of energy issues at non-structural high points (energy problems caused by surface factors) is also significantly improved. Therefore, it effectively solves the problem of insignificant amplitude compensation caused by near-surface strata and subsurface structures.

[0070] Example of a seismic data amplitude compensation system:

[0071] This embodiment discloses a seismic data amplitude compensation system. The seismic data amplitude compensation system of this embodiment enables the implementation of the seismic data amplitude compensation method described in the method embodiment of this invention.

[0072] In this embodiment, the seismic data amplitude compensation system includes a processor and a memory. The processor executes instructions stored in the memory to implement the seismic data amplitude compensation method in the method embodiment of the present invention. This seismic data amplitude compensation method has been described in detail in the above method embodiment. Those skilled in the art can generate corresponding computer instructions based on this seismic data amplitude compensation method to obtain the seismic data amplitude compensation system; further details are omitted here. The memory stores the computer instructions generated according to the seismic data amplitude compensation method.

[0073] The seismic data amplitude compensation system based on this embodiment can solve the problem of insignificant amplitude compensation effect in the prior art.

Claims

1. A method of seismic data amplitude compensation, characterized in that, The method comprises: 1) acquiring seismic data; 2) performing surface consistency amplitude compensation on the acquired seismic data, analyzing the stacked section of the compensated seismic data, determining a lower energy area and a higher energy area, extracting at least one single shot data in the lower energy area and the higher energy area respectively, and performing frequency division on the seismic data according to the frequency characteristics of each single shot data and the principle that the frequency characteristics of each single shot data in each frequency band are consistent; acquiring a main reflection layer of the stacked section, selecting multiple time windows from the main reflection layer to obtain corresponding root mean square amplitudes by root mean square processing, if the root mean square amplitude of a certain time window is less than 1 / 2 of the root mean square amplitude of the maximum amplitude window, then the time window is a lower energy area, and if the root mean square amplitude is greater than 4 / 5 of the root mean square amplitude of the maximum amplitude window, then the time window is a higher energy area; 3) performing surface consistency amplitude compensation on the seismic data of each frequency band respectively, and merging the compensated seismic data of each frequency band; Comparing the energy of the seismic data of each frequency band after surface consistency amplitude compensation in step 3), determining whether the energy difference of each frequency band is greater than a set threshold, if yes, adjusting the energy to make the energy difference of each frequency band less than the set threshold.

2. The seismic data amplitude compensation method of claim 1, wherein, In step 1), the acquired seismic data is also subjected to spherical diffusion compensation processing.

3. The seismic data amplitude compensation method of claim 2, wherein, Before the spherical diffusion compensation processing, the acquired seismic data is also subjected to pre-stack noise suppression.

4. The seismic data amplitude compensation method of claim 1, wherein, The stacked section in step 2) is obtained by unweighted stacking.

5. The seismic data amplitude compensation method of claim 1, wherein, The number of frequency bands during frequency division is 5-7.

6. The seismic data amplitude compensation method of claim 1, wherein, In step 2), the frequency bands are divided by using band-pass filtering technology or wavelet transform technology.

7. The seismic data amplitude compensation method of claim 1, wherein, There is a transition band between adjacent frequency bands, and the bandwidth of the transition band is 4-5 Hz.

8. A seismic data amplitude compensation system characterized by, The method comprises: a memory and a processor, the processor is used to execute instructions stored in the memory to realize the seismic data amplitude compensation method in any one of claims 1-7.

Images