Frequency domain fusion gas-bearing prediction method and device

Abstract

The invention relates to the technical field of seismic signal processing and interpretation, and particularly discloses a frequency domain fusion gas-bearing prediction method and device, and the method comprises the steps: obtaining the logging and post-stack seismic data of a target region reservoir; de-noising processing is carried out on the post-stack seismic data to obtain seismic data with a high signal-to-noise ratio; performing time-frequency analysis on the logging well bypass to determine a fluid sensitive frequency band; performing multi-wavelet reconstruction of a fluid-sensitive frequency band on the seismic data with the high signal-to-noise ratio to obtain a reconstructed data volume; performing high-precision time-frequency analysis on the reconstructed data volume to obtain a gradient frequency attribute volume; and carrying out normalization processing and fusion on the gradient frequency attribute body to obtain a gas-bearing indicator factor data body. According to the method, the gas-bearing property is detected by adopting the thinking of frequency attribute fusion, more abnormalities caused by oil and gas can be represented, a more accurate target stratum gas-bearing property prediction result is finally obtained, and the method does not depend on logging data and seismic pre-stack data, is high in stability and strong in certainty, and still has applicability in a work area with fewer drilling samples.

Description

Technical Field

[0001] This invention relates to the field of seismic signal processing and interpretation technology, specifically to a frequency domain fusion method and apparatus for predicting gas content. Background Technology

[0002] As oil and gas exploration becomes increasingly difficult and in-depth, and with the scarcity of trapped structures, the search for hidden oil and gas resources is becoming more and more urgent, and seismic fluid identification technology is receiving increasing attention. Currently, gas-bearing detection is conducted domestically and internationally using three main approaches: 1. Based on the frequency-related attributes of post-stack seismic data, but multiple stackings at all angles can lead to the loss or blurring of lithological or hydrocarbon information in the data; 2. Pre-stack inversion, but the inherent ambiguity of low-frequency models and methods limits its accuracy in detecting gas-bearing properties; 3. AVO attributes, but conventional AVO attributes are not very sensitive to gas-bearing properties. Although pre-stack seismic data contains richer fluid information than post-stack seismic data, pre-stack gas-bearing prediction is often poorly performed due to the influence of signal-to-noise ratio and the quality of early acquired data, and requires high-fidelity processing of pre-stack gathers, whose data quality directly affects the calculation of AVO attribute results; furthermore, when measured shear wave data is lacking, shear wave velocity is crucial for pre-stack attribute analysis.

[0003] Meanwhile, due to the development of exploration and development, the demand for evaluation through concatenated reprocessing and interpretation of seismic data from different acquisition periods has been increasing in recent years. The varying acquisition years lead to inconsistent acquisition techniques and coverage times, significantly reducing the quality of pre-stack seismic data and thus affecting the applicability of conventional pre-stack gas-bearing detection techniques. Although post-stack seismic data has a higher signal-to-noise ratio than pre-stack data, current mainstream post-stack gas-bearing prediction techniques often exhibit varying sensitivities to hydrocarbons, and their relationship with hydrocarbons is unclear. There is an urgent need for a gas-bearing prediction method that can fully exploit the potential of relatively more stable post-stack seismic data. CN105388527B discloses a hydrocarbon detection method based on a complex domain matching pursuit algorithm, but this method focuses on saving computation time to achieve fast and efficient hydrocarbon detection processing; its detection accuracy is still affected by the quality of the original seismic data. CN108254783A discloses a post-stack seismic fluid identification method based on time-frequency analysis, but this method emphasizes enhancing the seismic response characteristics of the reservoir segment.

[0004] Seismic detection technology plays an increasingly important role in oil and gas exploration, particularly in lithology-structural traps. Reservoir gas-bearing prediction results are of significant guiding importance for the efficient exploration and development of gas reservoirs. The presence of fluids in a reservoir causes attenuation of seismic wave energy. Using frequency-related information from seismic data to determine the origin of oil and gas has been a long-standing practice. However, conventional pre-stack gas-bearing prediction techniques, such as pre-stack inversion and AVO attribute analysis, suffer from poor accuracy when seismic data exhibits uneven amplitude energy due to significant differences in coverage times. Furthermore, multiple overlays of the original data at all angles may lose or obscure information reflecting lithology or oil and gas content.

[0005] Based on this technical background, the present invention studies a frequency domain fusion gas content prediction method and apparatus. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a frequency domain fusion method and apparatus for predicting gas-bearing properties. Compared with existing technologies, this method uses frequency attribute fusion to detect gas-bearing properties, which can characterize more anomalies caused by oil and gas, and ultimately obtain more accurate gas-bearing property prediction results for the target layer. At the same time, it adopts a sensitive frequency band multi-wavelet reconstruction technology to highlight the seismic response sensitive to fluids. It does not rely on well logging data and pre-stack seismic data, has high stability and strong determinism, and is still applicable in work areas with few drilling samples.

[0007] To achieve the above objectives, a first aspect of the present invention provides a frequency domain fusion method for predicting gas content, comprising:

[0008] Acquire well logging and post-stack seismic data of the reservoir in the target area;

[0009] The post-stack seismic data is denoised to obtain high signal-to-noise ratio seismic data;

[0010] Time-frequency analysis of the well logging bypass channel is performed to determine the fluid-sensitive frequency band;

[0011] The high signal-to-noise ratio seismic data is reconstructed using multi-wavelet reconstruction in fluid-sensitive frequency bands to obtain the reconstructed data volume;

[0012] High-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume;

[0013] The gradient frequency attribute volume is normalized and fused to obtain the gas content indicator factor data volume.

[0014] A second aspect of the present invention provides a frequency domain fusion gas content prediction device, comprising:

[0015] The data acquisition module is used to acquire well logging and post-stack seismic data of the target area reservoir;

[0016] The denoising module is used to denoise the post-stack seismic data to obtain high signal-to-noise ratio seismic data.

[0017] The time-frequency analysis module is used to perform time-frequency analysis on the well logging bypass to determine the fluid-sensitive frequency band;

[0018] The wavelet reconstruction module is used to perform fluid-sensitive frequency band multi-wavelet reconstruction on the high signal-to-noise ratio seismic data to obtain the reconstructed data volume;

[0019] A high-precision time-frequency analysis module is used to perform high-precision time-frequency analysis on the reconstructed data volume to obtain a gradient frequency attribute volume.

[0020] The processing and fusion module is used to normalize and fuse the gradient frequency attribute volume to obtain the gas content indicator factor data volume.

[0021] A third aspect of the present invention provides an electronic device, the electronic device comprising:

[0022] Memory, which stores executable instructions;

[0023] A processor that executes the executable instructions in the memory to implement the frequency domain fusion gas content prediction method described in the first aspect.

[0024] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the frequency domain fusion gas content prediction method described in the first aspect.

[0025] The beneficial effects of this invention include:

[0026] The frequency domain fusion gas-bearing prediction method proposed in this invention, compared with the existing technology, uses the concept of frequency attribute fusion to detect gas-bearing, which can characterize more anomalies caused by oil and gas, and finally obtain more accurate gas-bearing prediction results for the target layer. At the same time, the use of sensitive frequency band multi-wavelet reconstruction technology highlights the seismic response sensitive to fluids. It does not rely on well logging data and pre-stack seismic data, has high stability and strong determinism, and is still applicable in work areas with few drilling samples.

[0027] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0028] The above and other objects, features and advantages of the present invention will become more apparent from the more detailed description of exemplary embodiments of the invention in conjunction with the accompanying drawings.

[0029] Figure 1 This is a flowchart illustrating the frequency domain fusion gas content prediction method proposed in this invention.

[0030] Figure 2 This is a flowchart illustrating a specific implementation of the frequency domain fusion gas content prediction method proposed in this invention.

[0031] Figure 3 This is a schematic diagram illustrating the determination of sensitive frequencies for gas wells and gas-water wells through time-frequency analysis in a specific embodiment of the frequency domain fusion gas content prediction method proposed in this invention.

[0032] Figure 4 This is a schematic diagram showing the comparison between the reconstructed multi-wavelet amplitude based on the fluid-sensitive frequency and the original amplitude in a specific implementation of the frequency domain fusion gas content prediction method proposed in this invention.

[0033] Figure 5 This is a schematic diagram illustrating the principle of calculating the high-frequency attenuation gradient and the low-frequency increase gradient based on the frequency domain fusion gas content prediction method proposed in this invention, in a specific implementation of the frequency domain fusion method.

[0034] Figure 6 This is a schematic diagram of the gas-bearing indicator factor profile based on the multi-wavelet reconstructed frequency attribute volume fusion in a specific embodiment of the frequency domain fusion gas-bearing prediction method proposed in this invention.

[0035] Figure 7 This diagram illustrates a comparison between the gas content prediction results and traditional post-stack gas content prediction techniques in a specific embodiment of the frequency domain fusion gas content prediction method proposed in this invention. Detailed Implementation

[0036] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention are described below, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0037] This invention provides a frequency domain fusion method for predicting gas content, such as... Figure 1 As shown, it includes:

[0038] Acquire well logging and post-stack seismic data of the reservoir in the target area;

[0039] Denoising the post-stack seismic data yields high signal-to-noise ratio seismic data.

[0040] Time-frequency analysis of the well logging bypass channel is performed to determine the fluid-sensitive frequency band;

[0041] Multi-wavelet reconstruction of high signal-to-noise ratio seismic data in fluid-sensitive frequency bands yields the reconstructed data volume;

[0042] High-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume;

[0043] The gradient frequency attribute volume is normalized and fused to obtain the gas content indicator factor data volume.

[0044] Compared with existing technologies, the method of this invention uses frequency attribute fusion to detect gas content, which can characterize more anomalies caused by oil and gas, and ultimately obtain more accurate gas content prediction results for the target layer. At the same time, the use of sensitive frequency band multi-wavelet reconstruction technology highlights the seismic response sensitive to fluids. It does not rely on well logging data and pre-stack seismic data, has high stability and strong determinism, and is still applicable in work areas with few drilling samples.

[0045] According to the present invention, the denoising process employs a principal component filtering denoising method controlled by tilt angle tendency;

[0046] The filtering and denoising method does not change the spectral information and preserves the wave group characteristics to reduce the impact of the acquisition footprint.

[0047] According to the present invention, the well logging bypass is for gas wells and gas-water wells.

[0048] According to the present invention, fluid-sensitive multi-wavelet reconstruction is performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume comprising:

[0049] Multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction were performed on high signal-to-noise ratio seismic data to obtain the reconstructed data volume.

[0050] According to the present invention, high-precision time-frequency analysis is performed on the reconstructed data volume to obtain a gradient frequency attribute volume, including:

[0051] High-precision time-frequency analysis was performed on the reconstructed data volume to obtain the frequency attribute volumes of low-frequency increase and high-frequency decay gradients.

[0052] Preferably, the multi-wavelet seismic trace model used for multi-wavelet reconstruction of high signal-to-noise ratio seismic data is as follows:

[0053]

[0054] Where S represents earthquake data, W i For the i-th sub-wave, R i Let be the i-th reflection coefficient, and N be the noise.

[0055] According to the present invention, multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction are performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume comprising:

[0056] High signal-to-noise ratio seismic data are subjected to multi-wavelet spectral decomposition, and then the spectra of each wavelet are algebraically superimposed to obtain the spectrum of the seismic trace within the calculation window.

[0057] By selecting fluid-sensitive frequency bands, a single primary seismic wavelet is screened to obtain the reconstructed data volume.

[0058] The present invention will be described in more detail below through embodiments.

[0059] Example 1:

[0060] like Figure 2 As shown in the figure, this embodiment proposes a frequency domain fusion method for gas-bearing prediction. First, post-stack seismic data is collected; second, the post-stack data is denoised to obtain a data volume with a high signal-to-noise ratio; third, the frequency bands sensitive to fluid are determined by well logging side-channel seismic time-frequency analysis, and multi-wavelet reconstruction is performed to obtain seismic data that highlights fluid response characteristics; next, high-precision time-frequency analysis is performed on the preprocessed data, selecting two frequency attributes that characterize fluid information: low-frequency enhancement gradient and high-frequency attenuation gradient; finally, the two frequency attribute data volumes are normalized and fused to obtain the final gas-bearing indicator factor.

[0061] The specific steps of this method are as follows:

[0062] I. Obtain post-stack seismic data and related well data of the target reservoir;

[0063] Second, the principal component filtering denoising method controlled by tilt angle is selected to denoise the original data volume. This technique can maintain wave group characteristics, reduce the influence of acquisition footprint, and improve the signal-to-noise ratio without changing the spectral information.

[0064] 3. Perform time-frequency analysis on the wellbore bypass of gas wells and gas-water wells to determine the frequency bands that are sensitive to fluids and perform multi-wavelet reconstruction to obtain seismic data volumes that highlight the characteristics of fluid seismic response.

[0065] Conventional seismic trace model:

[0066] S(t) = W(t) * R(t) + N(t);

[0067] The multi-wavelet seismic trace model is as follows:

[0068]

[0069] In conventional seismic trace models, the wavelet W(t) does not change with time (depth). In multi-wavelet seismic trace models, there may be multiple wavelets W(t). It's possible that a specific stratum (corresponding to the reflection coefficient of a local sequence) corresponds to a single wavelet. During wavelet propagation, the shape of the wavelet changes due to energy diffusion, ground filtering, multiples, and interference. Therefore, multi-wavelet models are theoretically more realistic. Strata with different physical properties, such as reservoirs and non-reservoirs, and hydrocarbon-bearing and non-hydrocarbon-bearing reservoirs, have different seismic responses, and the seismic wavelets undergo different modifications as they pass through different strata, resulting in different shape changes. Therefore, some conventional reservoir and hydrocarbon prediction methods based on convolutional models using a single seismic wavelet have certain limitations. On the one hand, this approach may lose valuable information about reservoirs and hydrocarbon content; on the other hand, it may introduce false information. After decomposing multi-wavelet seismic traces into a set of wavelets, the wavelets are classified and screened based on their dominant frequencies to extract wavelet information relevant to the research objective, i.e., "wavelet reconstruction." The spectrum of multi-wavelet seismic traces is calculated by finding the wavelets contained within a time window (containing one wavelet means containing one wavelet recording point; the zero-phase wavelet recording point is at zero), and then algebraically superimposing the spectra of each wavelet to obtain the spectrum of the seismic trace within the calculation time window. Then, by selecting frequency bands sensitive to fluids, the single seismic response (wavelet) is screened for wavelet reconstruction to obtain seismic data highlighting fluid seismic responses.

[0070] IV. Perform high-precision time-frequency analysis on multi-wavelet reconstructed seismic data volume based on matching pursuit technology to obtain frequency attribute volumes of low-frequency increasing gradient and high-frequency attenuation gradient.

[0071] Normally, the attenuation of seismic waves during propagation is due to factors such as spherical diffusion and scattering caused by the propagation characteristics of seismic waves, and also due to the absorption effect of the formation medium, which absorbs the energy of the seismic waves and converts it into heat energy. This attenuation includes both energy and frequency attenuation, which can reflect the intrinsic properties of the underground medium. When the formation is saturated with oil and gas, the attenuation rate of high-frequency energy of seismic waves is much greater than that of low-frequency energy. This is the "high-frequency attenuation" and "relative increase in low frequency" characteristics that are usually used in post-stack seismic data to detect oil and gas in the target layer.

[0072] The frequencies corresponding to 65% and 85% energy mainly reflect changes in seismic wave frequency. When the reservoir has well-developed pores and is saturated with oil and gas, the high-frequency energy attenuation in the seismic wave is greater than that of the low-frequency energy attenuation. By extracting the attenuation gradient attribute at the high-frequency end, the oil and gas characteristics of the reservoir can be indirectly detected. The attenuation gradient (low frequency) is calculated by the increase in slope after the low-frequency energy increases. The calculation method is the same, only the parameters are changed to 15% and 35%.

[0073] V. After normalization, the frequency attribute data is fused to restore the true spectral characteristics and obtain the final gas content indicator data:

[0074] Actual geological formations are typically non-perfectly elastic, anisotropic, or non-homogeneous media with continuously varying physical properties. When seismic waves pass through such media, within the effective frequency band of seismic exploration, the viscoelastic properties of the bottom layer absorb high-frequency components much more readily than other frequency components. That is, when seismic waves pass through formations with significant viscoelasticity filled with fluid, the dispersion phenomenon becomes more pronounced. By effectively fusing the two frequency attribute volumes on the basis of normalization, the characteristics of fluid-bearing formations are highlighted, and the results are consistent with known wells.

[0075] In this embodiment, Figure 3 This is a schematic diagram illustrating the determination of sensitive frequencies through time-frequency analysis of gas wells and gas-water wells in the actual work area. A comparison of wavelet time-frequency energy for different drilling target layers is shown: for pure gas wells, the wavelet time-frequency energy is 10-37Hz, with a centroid frequency of approximately 22Hz; for gas-water co-layer well 1, the wavelet time-frequency energy is 6-50Hz, with a centroid frequency of approximately 27Hz; and for gas-water co-layer well 2, the wavelet time-frequency energy is 4-36Hz, with a centroid frequency of approximately 26Hz. Therefore, a 10-22Hz wavelet is preferred for seismic trace reconstruction to identify gas content. Figure 4 This is a schematic diagram comparing the well profile based on fluid-sensitive frequency multi-wavelet reconstruction and the original amplitude; the seismic response at the top of the gas-bearing section of the gas well exhibits strong amplitude characteristics, while the gas-water co-layer well exhibits medium to weak amplitude characteristics; Figure 5 This is a schematic diagram of the gas-bearing indicator factor profile based on the fusion of frequency attribute volume of multi-wavelet reconstruction. Figure 6 This is a schematic diagram of the gas-bearing indicator profile based on the multi-wavelet reconstructed frequency attribute volume fusion. From the well-connected profile, the "high-frequency attenuation" and "low-frequency relative increase" attributes show slight differences in characterization in gas wells and gas-water co-layer wells. To obtain more accurate fluid detection results, this study delves into the spectral feature information and constructs a post-stack time-frequency fusion gas-bearing indicator based on the normalized high-frequency and low-frequency attenuation gradient data volume. From the well-connected profile, the reconstructed gas-bearing indicator integrates the gradient features of "high-frequency attenuation" and "low-frequency relative enhancement," which is more conducive to accurately predicting the gas content of the target layer. Figure 7 This is a comparison of the gas-bearing prediction results of this technology with those of traditional post-stack gas-bearing prediction technologies; a) Gas-bearing prediction results of this method; b) Gas-bearing prediction results of traditional high-frequency attenuation gradient; c) Gas-bearing prediction results of multi-wavelet reconstruction data volume amplitude attribute. Compared with the other two post-stack gas-bearing technologies, this technology not only matches the actual drilled gas wells and gas-water wells, but also shows better consistency with the fluid characteristics and structural morphology around gas-water wells in the eastern part of the study target area. In the northwestern part of the study area, the gas-water separation amplitude attribute is more obvious and reasonable. In addition, combined with structural characteristics, the predicted gas-bearing range is generally located in a relatively high position within the favorable facies zone. The prediction results play a positive role in the subsequent trap effectiveness analysis.

[0076] The final results in this embodiment are consistent with the actual drilling results, which verifies the rationality and effectiveness of the present invention. It can effectively guide the selection of the optimal target location in the work area and can be promoted and applied in other work areas to promote the efficient exploration and development of oil and gas.

[0077] Example 2:

[0078] This embodiment provides a frequency domain fusion method for predicting gas content, such as... Figure 1 As shown, it includes:

[0079] Acquire well logging and post-stack seismic data of the reservoir in the target area;

[0080] Denoising the post-stack seismic data yields high signal-to-noise ratio seismic data.

[0081] Time-frequency analysis of the well logging bypass channel is performed to determine the fluid-sensitive frequency band;

[0082] Multi-wavelet reconstruction of high signal-to-noise ratio seismic data in fluid-sensitive frequency bands yields the reconstructed data volume;

[0083] High-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume;

[0084] The gradient frequency attribute volume is normalized and fused to obtain the gas content indicator factor data volume;

[0085] In this embodiment, the denoising process uses a principal component filtering denoising method controlled by tilt angle tendency;

[0086] The filtering and denoising method does not change the spectral information and preserves the wave group characteristics to reduce the impact of acquisition footprints;

[0087] In this embodiment, the logging bypass is for gas wells and gas-water wells;

[0088] In this embodiment, fluid-sensitive multi-wavelet reconstruction is performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume including:

[0089] Multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction were performed on high signal-to-noise ratio seismic data to obtain the reconstructed data volume;

[0090] In this embodiment, high-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume, which includes:

[0091] High-precision time-frequency analysis was performed on the reconstructed data volume to obtain the frequency attribute volumes of low-frequency increase and high-frequency decay gradients, respectively.

[0092] In this embodiment, the multi-wavelet seismic trace model used for multi-wavelet reconstruction of high signal-to-noise ratio seismic data is as follows:

[0093]

[0094] Where S represents earthquake data, W i For the i-th sub-wave, R i Let be the i-th reflection coefficient, and N be the noise.

[0095] In this embodiment, multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction are performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume including:

[0096] High signal-to-noise ratio seismic data are subjected to multi-wavelet spectral decomposition, and then the spectra of each wavelet are algebraically superimposed to obtain the spectrum of the seismic trace within the calculation window.

[0097] By selecting fluid-sensitive frequency bands, a single primary seismic wavelet is screened to obtain the reconstructed data volume.

[0098] Example 3:

[0099] This embodiment provides a frequency domain fusion gas content prediction device, including:

[0100] The data acquisition module is used to acquire well logging and post-stack seismic data of the target area reservoir;

[0101] The denoising module is used to denoise post-stack seismic data to obtain high signal-to-noise ratio seismic data;

[0102] The time-frequency analysis module is used to perform time-frequency analysis on the well logging bypass to determine the fluid-sensitive frequency band;

[0103] The wavelet reconstruction module is used to perform fluid-sensitive multi-wavelet reconstruction on high signal-to-noise ratio seismic data to obtain the reconstructed data volume.

[0104] The high-precision time-frequency analysis module is used to perform high-precision time-frequency analysis on the reconstructed data volume to obtain the gradient frequency attribute volume;

[0105] The processing and fusion module is used to normalize and fuse the gradient frequency attribute volume to obtain the gas content indicator factor data volume.

[0106] In this embodiment, the denoising process uses a principal component filtering denoising method controlled by tilt angle tendency;

[0107] The filtering and denoising method does not change the spectral information and preserves the wave group characteristics to reduce the impact of acquisition footprints;

[0108] In this embodiment, the logging bypass is for gas wells and gas-water wells;

[0109] In this embodiment, fluid-sensitive multi-wavelet reconstruction is performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume including:

[0110] Multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction were performed on high signal-to-noise ratio seismic data to obtain the reconstructed data volume;

[0111] In this embodiment, high-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume, which includes:

[0112] High-precision time-frequency analysis was performed on the reconstructed data volume to obtain the frequency attribute volumes of low-frequency increase and high-frequency decay gradients, respectively.

[0113] In this embodiment, the multi-wavelet seismic trace model used for multi-wavelet reconstruction of high signal-to-noise ratio seismic data is as follows:

[0114]

[0115] Where S represents earthquake data, W i For the i-th sub-wave, R i Let be the i-th reflection coefficient, and N be the noise.

[0116] In this embodiment, multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction are performed on high signal-to-noise ratio seismic data to obtain a reconstructed data volume including:

[0117] High signal-to-noise ratio seismic data are subjected to multi-wavelet spectral decomposition, and then the spectra of each wavelet are algebraically superimposed to obtain the spectrum of the seismic trace within the calculation window.

[0118] By selecting fluid-sensitive frequency bands, a single primary seismic wavelet is screened to obtain the reconstructed data volume.

[0119] Example 4:

[0120] This invention provides an electronic device including a memory and a processor, comprising:

[0121] Memory, which stores executable instructions;

[0122] The processor executes executable instructions in memory to implement a frequency-domain fusion gas content prediction method.

[0123] This memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0124] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of the invention, the processor is used to execute computer-readable instructions stored in the memory.

[0125] Those skilled in the art should understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this invention.

[0126] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0127] Example 5:

[0128] This invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a frequency domain fusion gas content prediction method.

[0129] A computer-readable storage medium according to embodiments of the present invention stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods described in the foregoing embodiments of the present invention are performed.

[0130] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0131] The frequency domain fusion gas-bearing prediction method proposed in the embodiments of the present invention, compared with the prior art, uses the concept of frequency attribute fusion to detect gas-bearing, which can characterize more anomalies caused by oil and gas, and finally obtain more accurate gas-bearing prediction results for the target layer. At the same time, the use of sensitive frequency band multi-wavelet reconstruction technology highlights the seismic response sensitive to fluids. It does not rely on well logging data and pre-stack seismic data, has high stability and strong determinism, and is still applicable in work areas with few drilling samples.

[0132] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A frequency domain fusion method for predicting gas content, characterized in that, include: Acquire well logging and post-stack seismic data of the reservoir in the target area; The post-stack seismic data is denoised to obtain high signal-to-noise ratio seismic data; Time-frequency analysis of the well logging bypass channel is performed to determine the fluid-sensitive frequency band; The high signal-to-noise ratio seismic data is reconstructed using multi-wavelet reconstruction in fluid-sensitive frequency bands to obtain the reconstructed data volume; High-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume; The gradient frequency attribute volume is normalized and fused to obtain the gas content indicator factor data volume.

2. The method according to claim 1, characterized in that, The denoising process employs a principal component filtering denoising method controlled by tilt angle tendency. The filtering and denoising method described above does not change the spectral information and preserves the wave group characteristics to reduce the impact of the acquisition footprint.

3. The method according to claim 1, characterized in that, The well logging access channels are for gas wells and gas-water wells.

4. The method according to claim 1, characterized in that, Fluid-sensitive multi-wavelet reconstruction of the high signal-to-noise ratio seismic data was performed to obtain a reconstructed data volume comprising: The high signal-to-noise ratio seismic data is subjected to multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction to obtain the reconstructed data volume.

5. The method according to claim 1, characterized in that, High-precision time-frequency analysis is performed on the reconstructed data volume to obtain the gradient frequency attribute volume, which includes: High-precision time-frequency analysis was performed on the reconstructed data volume to obtain the low-frequency increase and high-frequency decay gradient frequency attribute volumes, respectively.

6. The method according to claim 1, characterized in that, The multi-wavelet seismic trace model used for multi-wavelet reconstruction of the high signal-to-noise ratio seismic data is as follows: Where S represents earthquake data, W i For the i-th sub-wave, R i Let be the i-th reflection coefficient, and N be the noise.

7. The method according to claim 4, characterized in that, The high signal-to-noise ratio seismic data is subjected to multi-wavelet decomposition and fluid-sensitive frequency band seismic reconstruction to obtain a reconstructed data volume including: The high signal-to-noise ratio seismic data is subjected to multi-wavelet spectral decomposition, and then the spectra of each wavelet are algebraically superimposed to obtain the spectrum of the seismic trace within the calculation window. By selecting fluid-sensitive frequency bands, a single primary seismic wavelet is screened to obtain the reconstructed data volume.

8. A frequency domain fusion gas content prediction device, characterized in that, include: The data acquisition module is used to acquire well logging and post-stack seismic data of the target area reservoir; The denoising module is used to denoise the post-stack seismic data to obtain high signal-to-noise ratio seismic data. The time-frequency analysis module is used to perform time-frequency analysis on the well logging bypass to determine the fluid-sensitive frequency band; The wavelet reconstruction module is used to perform fluid-sensitive frequency band multi-wavelet reconstruction on the high signal-to-noise ratio seismic data to obtain the reconstructed data volume; A high-precision time-frequency analysis module is used to perform high-precision time-frequency analysis on the reconstructed data volume to obtain a gradient frequency attribute volume. The processing and fusion module is used to normalize and fuse the gradient frequency attribute volume to obtain the gas content indicator factor data volume.

9. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement the frequency domain fusion gas content prediction method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the frequency domain fusion gas content prediction method according to any one of claims 1-7.

Images