Complex fracture hierarchical detection method, device and equipment based on frequency division fusion

Abstract

The application discloses a complex fracture grading detection method based on frequency division fusion, which comprises the following steps: acquiring single-frequency seismic data bodies of low, medium and high frequencies, then performing coherence calculation on the single-frequency seismic data bodies to obtain low, medium and high single-frequency fracture detection data bodies, then performing color fusion display on the single-frequency fracture detection data bodies, and finally performing fracture grading interpretation according to the color fusion display results and the meanings represented by the respective colors. Through the above method, accurate, intuitive and efficient identification of complex fractures is realized, and meanwhile, the identification results can be graded and interpreted. The method fully utilizes the response advantages of 3D seismic data frequency domain attributes to different scale fractures, combines the most advanced C3 coherence algorithm to perform fine edge detection and identify secondary fractures, and finally realizes grading detection of complex fractures, which is of great significance to fine exploration of mature exploration areas and well deployment in the later development stage.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration technology, and in particular to a method, apparatus and equipment for classifying and detecting complex fractures based on frequency division fusion. Background Technology

[0002] As oil and gas exploration becomes increasingly sophisticated, the industry has shifted from early exploration of large structural traps to exploration of concealed traps. Traps associated with low-level or secondary faults are important targets for exploration of concealed traps. In addition, low-level and secondary faults have a significant effect on reservoir modification. Therefore, the requirements for precise identification of low-level or secondary faults are becoming increasingly stringent.

[0003] However, because low-level or secondary faults have small vertical displacements, they do not exhibit obvious seismic axis displacement or torsion characteristics in most existing seismic data. Traditional methods for identifying seismic axis displacement cannot identify such faults.

[0004] Currently, mature fracture identification and interpretation technologies are mainly aimed at large fractures (with obvious faulting along the seismic axis) and can well meet the requirements for identifying and interpreting large fractures. However, there is still room for improvement in the interpretation and identification of secondary fractures (without obvious faulting along the seismic axis), and a technology capable of classifying and interpreting complex fractures has not yet been developed. The detailed interpretation of complex fractures, especially secondary fractures, is a crucial step in the exploration and risk assessment of hidden traps. Currently, there is no very effective method for classifying and identifying complex fractures, which restricts in-depth oil and gas exploration in areas with complex fracture development.

[0005] In view of this, in view of the difficulties in identifying and interpreting multi-level fractures, there is an urgent need to propose a complex fracture classification detection method based on frequency division fusion that can accurately, intuitively and efficiently identify secondary fractures that have not undergone obvious dislocation along the seismic axis, and facilitate clear and rapid interpretation of the identification results, so as to solve the above problems. Summary of the Invention

[0006] The purpose of this invention is to propose a method for the graded detection of complex fractures. This method utilizes Marr wavelet transform to first extract three single-frequency volumes (low, medium, and high frequencies); then, it employs the C3 algorithm based on feature structure to process each of the three single-frequency volumes, obtaining single-frequency fracture detection data volumes; finally, it fuses and displays the three low, medium, and high single-frequency fracture detection data volumes using CMY, achieving graded detection of complex fractures. This method can effectively identify complex fractures and display them in a graded manner, which is of great significance for the identification and risk assessment of secondary fracture-related traps.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0008] A method for graded detection of complex fractures based on frequency division fusion, the method comprising:

[0009] Acquire single-frequency seismic data volumes at three different frequencies: low, medium, and high.

[0010] Coherent calculations were performed on the single-frequency seismic data volume to obtain three single-frequency fault detection data volumes: low, medium, and high.

[0011] The single-frequency fracture detection data volume is displayed using color fusion.

[0012] Based on the color fusion display results, and combined with the meaning represented by each color, the fracture classification is interpreted to achieve graded detection of complex fractures.

[0013] As a further improvement of the present invention, the acquisition of three single-frequency seismic data volumes (low, medium, and high) includes:

[0014] Marr wavelet was used to simulate Ricker wavelet to divide the seismic data into frequencies, and single-frequency seismic data volumes of three different frequencies (low, medium, and high) were extracted.

[0015] As a further improvement of the present invention, the step of using Marr wavelet to simulate Ricker wavelet to perform frequency division of seismic data and extracting single-frequency seismic data volumes of three different frequencies (low, medium, and high) includes the following steps:

[0016] Obtain raw earthquake data;

[0017] The effective frequency band of the earthquake is determined based on the original seismic data;

[0018] The frequency division volume is calculated using Marr wavelet transform within the determined effective seismic frequency band;

[0019] Extract three different frequency single-frequency seismic data volumes (low, medium, and high) from the frequency-division volume.

[0020] As a further improvement of the present invention, the calculation of the frequency division body using Marr wavelet transform includes the following steps:

[0021] Obtain the time-frequency volume corresponding to the original seismic data;

[0022] A time-frequency attribute body is generated based on the time-frequency body;

[0023] The starting frequency, ending frequency, and frequency interval of the original seismic data are determined based on the time-frequency attribute volume.

[0024] A frequency divider is obtained based on the starting frequency, the ending frequency, and the frequency interval.

[0025] As a further improvement of the present invention, the coherence calculation of the single-frequency seismic data volume includes:

[0026] The C3 coherent algorithm was used to calculate the single-frequency seismic data volumes of the three different frequencies: low, medium, and high.

[0027] As a further improvement of the present invention, the color fusion display of the single-frequency fracture detection data volume includes:

[0028] Normalization calculations were performed on the three single-frequency fracture detection data volumes of low, medium, and high frequencies, respectively.

[0029] The three normalized single-frequency fracture detection data volumes are used as light intensity volumes of cyan, magenta and yellow primary colors in the CMY fusion display, respectively, for CMY fusion display.

[0030] As a further improvement of the present invention, the normalization calculation of the three single-frequency fracture detection data volumes of low, medium and high includes the following steps:

[0031] Obtain the maximum value in the low-frequency fracture detection data volume;

[0032] Obtain the maximum value in the intermediate frequency fracture detection data volume;

[0033] Obtain the maximum value in the high-frequency fracture detection data volume;

[0034] Based on the three single-frequency fracture detection data volumes (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data volumes obtained, the normalized three single-frequency fracture detection data volumes are calculated.

[0035] As a further improvement of the present invention, the calculation formula for calculating the normalized three single-frequency fracture detection data bodies based on the three single-frequency fracture detection data bodies (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data bodies obtained are as follows:

[0036]

[0037] In the formula, V 归一化单频断裂检测体 This is the normalized single-frequency fracture detection data volume; V 单频断裂检测体 This is a single-frequency fracture detection data volume; This represents the maximum value in the single-frequency fracture detection data volume.

[0038] As a further improvement of the present invention, the step of interpreting the fracture grading based on the color fusion display results and the meaning represented by each color includes:

[0039] The cyan color indicates a larger-scale fault;

[0040] The magenta color indicates a medium-scale fault;

[0041] Yellow indicates smaller-scale fracture structures, which are small-scale faults or cracks.

[0042] The blue color represents the result of low- and mid-frequency fusion, reflecting fault information at larger and medium scales.

[0043] The green color represents the result of the fusion of low and high frequencies, reflecting both early faults with large displacements and later faults with smaller displacements.

[0044] The red color indicates the result of the fusion of mid-frequency and high-frequency signals, reflecting fractures or cracks with small breakpoints.

[0045] The black color is the result of the fusion of low, medium and high frequencies, indicating the development of large-scale faults, medium-scale faults and smaller-scale fracture structures.

[0046] The present invention also provides a complex fracture grading detection device based on frequency division fusion, the device comprising:

[0047] The acquisition unit is used to acquire single-frequency seismic data volumes of three different frequencies: low, medium, and high.

[0048] The calculation unit is used to perform coherent calculations on the single-frequency seismic data volume acquired by the acquisition unit to obtain three single-frequency fault detection data volumes: low, medium, and high.

[0049] A color fusion display unit is used to perform color fusion display on the single-frequency fracture detection data volume;

[0050] The fracture grading interpretation unit is used to interpret the fracture grading based on the color fusion display results and the meaning represented by each color.

[0051] As a further improvement of the present invention, the acquisition unit acquires single-frequency seismic data volumes of three different frequencies: low, medium, and high, including:

[0052] Marr wavelet was used to simulate Ricker wavelet to divide the seismic data into frequencies, and single-frequency seismic data volumes of three different frequencies (low, medium, and high) were extracted.

[0053] As a further improvement of the present invention, the acquisition unit uses Marr wavelet to simulate Ricker wavelet to perform frequency division on seismic data and extracts single-frequency seismic data volumes of three different frequencies (low, medium, and high), including the following steps:

[0054] Obtain raw earthquake data;

[0055] The effective frequency band of the earthquake is determined based on the original seismic data;

[0056] The frequency division volume is calculated using Marr wavelet transform within the determined effective seismic frequency band;

[0057] Extract three different frequency single-frequency seismic data volumes (low, medium, and high) from the frequency-division volume.

[0058] As a further improvement of the present invention, the acquisition unit calculates the frequency division body using Marr wavelet transform, including the following steps:

[0059] Obtain the time-frequency volume corresponding to the original seismic data;

[0060] A time-frequency attribute body is generated based on the time-frequency body;

[0061] The starting frequency, ending frequency, and frequency interval of the original seismic data are determined based on the time-frequency attribute volume.

[0062] A frequency divider is obtained based on the starting frequency, the ending frequency, and the frequency interval.

[0063] As a further improvement of the present invention, the computing unit performs coherent calculations on the acquired single-frequency seismic data volume to obtain three single-frequency fault detection data volumes: low, medium, and high frequency.

[0064] The C3 coherent algorithm was used to calculate the single-frequency seismic data volumes of the three different frequencies: low, medium, and high.

[0065] As a further improvement of the present invention, the color fusion display unit performs color fusion display on the single-frequency fracture detection data volume, including:

[0066] Normalization calculations were performed on the three single-frequency fracture detection data volumes of low, medium, and high frequencies, respectively.

[0067] The three normalized single-frequency fracture detection data volumes are used as light intensity volumes of cyan, magenta and yellow primary colors in the CMY fusion display, respectively, for CMY fusion display.

[0068] As a further improvement of the present invention, the color fusion display unit performs normalization calculations on the low, medium, and high single-frequency fracture detection data volumes, including the following steps:

[0069] Obtain the maximum value in the low-frequency fracture detection data volume;

[0070] Obtain the maximum value in the intermediate frequency fracture detection data volume;

[0071] Obtain the maximum value in the high-frequency fracture detection data volume;

[0072] Based on the three single-frequency fracture detection data volumes (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data volumes obtained, the normalized three single-frequency fracture detection data volumes are calculated.

[0073] As a further improvement of the present invention, the color fusion display unit calculates the normalized three single-frequency fracture detection data bodies based on the low, medium, and high single-frequency fracture detection data bodies and the maximum values ​​corresponding to the three single-frequency fracture detection data bodies obtained, as follows:

[0074]

[0075] In the formula, V 归一化单频断裂检测体 This is the normalized single-frequency fracture detection data volume; V 单频断裂检测体 This is a single-frequency fracture detection data volume; This represents the maximum value in the single-frequency fracture detection data volume.

[0076] As a further improvement of the present invention, the fracture grading interpretation unit performs fracture grading interpretation based on the color fusion display results and the meaning represented by each color, including:

[0077] The cyan color indicates a larger-scale fault;

[0078] The magenta color indicates a medium-scale fault;

[0079] Yellow indicates smaller-scale fracture structures, which are small-scale faults or cracks.

[0080] The blue color represents the result of low- and mid-frequency fusion, reflecting fault information at larger and medium scales.

[0081] The green color represents the result of the fusion of low and high frequencies, reflecting both early faults with large displacements and later faults with smaller displacements.

[0082] The red color indicates the result of the fusion of mid-frequency and high-frequency signals, reflecting fractures or cracks with small breakpoints.

[0083] The black color is the result of the fusion of low, medium and high frequencies, indicating the development of large-scale faults, medium-scale faults and smaller-scale fracture structures.

[0084] This invention also provides a complex fracture grading detection device based on frequency division fusion, the device comprising a processor and a memory; wherein,

[0085] The memory is used to store machine-executable instructions;

[0086] The processor is used to read and execute machine-executable instructions stored in the memory to implement the aforementioned complex fracture grading detection method based on frequency division fusion.

[0087] The beneficial effects of this invention are:

[0088] This invention provides a method, apparatus, and equipment for classifying and detecting complex fractures based on frequency fusion. It acquires three single-frequency seismic data volumes (low, medium, and high frequencies), performs coherent calculations on these volumes to obtain three single-frequency fracture detection data volumes (low, medium, and high frequencies), and displays them using color fusion. Based on the color fusion results and the meaning represented by each color, fracture classification is interpreted. This method achieves accurate, intuitive, and efficient identification of complex fractures and facilitates the classification interpretation of the identification results. It fully utilizes the frequency domain attributes of 3D seismic data to respond to fractures at different scales, combines coherent algorithms to perform fine edge detection and identify secondary fractures, and ultimately achieves classified detection of complex fractures. This is of great significance for fine exploration in mature exploration areas and the deployment of subsequent development wells.

[0089] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0090] Figure 1 This is a flowchart of the complex fracture grading detection method based on frequency division fusion of the present invention;

[0091] Figure 2 The spectral distribution of Reckon wavelets with different dominant frequencies;

[0092] Figure 3 The extracted spectra of the Ricker wavelets with main frequencies of 20, 40, and 80 Hz;

[0093] Figure 4 To simulate the spectrum of different dominant frequency Ricker wavelets using Marr wavelet frequency division;

[0094] Figure 5 The waveform feature diagram of the Marr wavelet used in this invention;

[0095] Figure 6 The profile results of different single-frequency seismic data volumes extracted using Marr wavelet frequency division;

[0096] Figure 7 The computational interface for calculating the frequency division volume using Marr wavelet transform;

[0097] Figure 8 This is a graph showing the spectral analysis results of seismic data within the study block in Embodiment 1 of the present invention;

[0098] Figure 9 These are full-band and single-frequency body seismic profiles of the study area in Embodiment 1 of the present invention;

[0099] Figure 10 This is a planar distribution diagram of body fracture detection at different frequencies in the study area in Embodiment 1 of the present invention;

[0100] Figure 11 This is a cross-sectional view showing the CMY fusion fracture gradation in the study area of ​​Embodiment 1 of the present invention;

[0101] Figure 12 This is a plan view showing the fracture grading in the study area in Embodiment 1 of the present invention. Detailed Implementation

[0102] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0103] Please refer to Figure 1 As shown, this invention proposes a method for complex fault classification detection based on Marr wavelet transform to extract single-frequency volumes, perform C3 coherence calculation, and then use CMY fusion display to achieve rapid and efficient classification interpretation of complex faults. The principle of this method is based on the fact that seismic data volumes with different dominant frequencies reflect geological information at different subsurface scales. High-frequency seismic data volumes correspond to small-scale geological information and faults with small displacements; medium-frequency seismic data volumes correspond to medium-scale geological information and faults with small displacements; and low-frequency seismic data volumes correspond to large-scale geological information and faults with large displacements. Based on 3D seismic data and well distribution data, this method fully utilizes the high similarity between Marr wavelets and Ricker wavelets in the time and frequency domains. Marr wavelets are used to simulate Ricker wavelets to classify the seismic data into three single-frequency seismic volumes with physical meaning: low, medium, and high. Then, the C3 coherence algorithm, which has strong noise resistance based on feature structures, is used to perform coherence calculation on the series of frequency-divided seismic data volumes to obtain three single-frequency fault detection data volumes: low, medium, and high. Finally, CMY fusion display technology is used to classify and detect the faults.

[0104] The specific explanation is as follows:

[0105] (I) Extraction of single-frequency seismic data volume using Marr wavelet:

[0106] The Marr wavelet is the second derivative of the Gaussian function. The formula for the generating function of the Marr wavelet is as follows:

[0107] Time domain:

[0108] Frequency domain:

[0109] In the Marr wavelet time-domain expression, t is used... The substitution is essentially the expression for the Ricker wavelet; therefore, the Marr wavelet exhibits high similarity to the Ricker wavelet in both the frequency and time domains, possessing strong physical significance. Furthermore, the Marr wavelet is a real-number wavelet, making it simple and fast to compute. The Marr wavelet satisfies both the allowable conditions of wavelet transform and possesses good local properties. Therefore, the Marr wavelet can be used to simulate the Ricker wavelet for frequency division of seismic records. Figure 2 The image shows the spectral distribution of Ricker wavelets with different dominant frequencies, and further comparisons are made. Figure 3 and Figure 4 As can be seen, the spectra of Ricker wavelets with dominant frequencies of 20, 40, and 80 Hz are very similar to the spectra of Ricker wavelets with different dominant frequencies simulated using Marr wavelet frequency division. Therefore, using Marr wavelets to simulate Ricker wavelets of different frequencies for frequency division of seismic signals results in signals with clear physical meaning, which is not achieved by other wavelet frequency division methods.

[0110] like Figure 5 As shown, the Marr wavelet is simple to construct and faster than other wavelet transforms, even faster than the S-transform, while other matching pursuit methods are extremely slow. Wavelet transforms have the advantage of variable time windows, and their accuracy for low-frequency and high-frequency signals is far superior to the Short-Time Fourier Transform (STFT). The Marr wavelet can simulate the Ricker wavelet to divide the seismic record into frequencies; the frequency-divided record is the seismic response of the frequency-divided wavelet, equivalent to a deconvolution, effectively improving the resolution accuracy of geological bodies.

[0111] When using Marr wavelets for frequency division, if the division is strictly performed according to the harmonic relationship, it exhibits strict reversibility. Adding the divided signals together can recover the original signal with minimal computational error. Furthermore, wavelet transform for time-frequency analysis allows for analysis at different scales for different frequencies, demonstrating excellent frequency accuracy in the low-frequency range and good time-division capability in the high-frequency range. Because wavelet transform has a variable time window characteristic, both low-frequency and high-frequency information possess high reliability.

[0112] Wavelet transform-based seismic time-frequency analysis primarily generates discrete-frequency energy volumes, which are generated by sliding along a short time window to produce a series of discrete-frequency tuned-amplitude data. Vertically, this data volume is the same as conventional data volumes, consisting entirely of time, but each generated data volume contains only a single frequency component. First, spectral analysis is performed based on the original seismic data to determine the effective frequency band of the seismic event. Then, within the effective frequency band, a series of single-frequency seismic data volumes are generated, such as... Figure 6As shown in the figure, the profile results of different single-frequency seismic data volumes extracted using Marr wavelet frequency division can be seen. By comparing and optimizing different single-frequency seismic data volumes, three single-frequency seismic data volumes—low, medium, and high—that best reflect the faults in the study area are extracted. Then, the fault profile is tracked on these three single-frequency volumes.

[0113] The calculation process is as follows:

[0114] (1) Input the raw earthquake data to be calculated.

[0115] (2) Select the line range of the data volume to be calculated, output the time range and time type of the data volume, or input it according to the layer method and set the time of layer up and down drift.

[0116] (3) Wavelet transform calculation of frequency division body (wavelet):

[0117] Input the time-frequency volume, which is generated from the original seismic volume through time-frequency spectrum calculation;

[0118] A time-frequency attribute volume is generated based on this time-frequency volume;

[0119] When using the frequency spectrum attribute, enter the maximum / minimum frequency to be calculated. This parameter is given by the data volume parameter, which specifies the start and end of the frequency. The more sampling frequencies, the greater the computational load. For data with a small frequency band, it is best to make appropriate adjustments.

[0120] The frequency interval determines the number of frequency dividers generated. For example, inputting a starting frequency of 10, an ending frequency of 30, and a frequency interval of 10 will produce frequency dividers of 10Hz, 20Hz, and 30Hz. Figure 7 The image shows the calculation interface for wavelet transform to calculate the frequency division volume.

[0121] (II) Calculation of single-frequency fracture detection data volume:

[0122] Frequency domain fracture detection technology has evolved from the first generation based on coherence calculation (based on cross-correlation) and the second generation based on similarity calculation (based on tilted superposition), to the third generation based on discontinuity detection of the intrinsic structure of the data covariance matrix. Let's assume that an analysis channel consisting of N sample points within a time window, and the data from J channels within the window (each channel being a unit of channel), are arranged together to form a data matrix:

[0123]

[0124] Among them, u ij This represents the amplitude value of the i-th seismic wave on the j-th track. The n-th row of matrix U is... Analyzing the nth amplitude value on each seismic trace within the body, the variance matrix of the nth sample is:

[0125]

[0126] Therefore, the covariance matrix formed by the covariances of each element in the channel within the coherence analysis window is:

[0127]

[0128] Suppose that the covariance matrix M has J eigenvalues ​​λ1, λ2, ..., λJ ordered from largest to smallest, and the corresponding eigenvectors are V1, V2, ..., VJ. Then the formula for calculating the coherence volume based on the feature structure is:

[0129]

[0130] In the formula, λj represents the eigenvalues ​​of the covariance matrix of the seismic data within the calculation window. Theoretically, the third-generation algorithm C3 based on eigenvalues ​​is superior to the second-generation algorithm C2 based on multi-channel cross-correlation and the algorithm C1 based on two-channel cross-correlation. This is mainly because when the signal level is higher than the added Gaussian noise level, the noise can be removed from the coherence calculation, and this method can provide the best lateral resolution in noisy data. Therefore, C3 has great application potential due to its strong noise resistance and relatively high resolution, but its computational cost is relatively high.

[0131] Frequency domain fracture detection technology utilizes multichannel similarity to transform 3D data volumes into correlation coefficient data volumes. Through refined fracture detection processing of the 3D data volume, the similarity of local seismic waveforms is compared (points with lower coherence values ​​correlate better with discontinuities in reflected wave waveforms). This allows for a more intuitive identification of geological phenomena related to fractures, cracks, sedimentary facies, lithological changes, and even fluid variations. As a state-of-the-art coherent algorithm, frequency domain fracture detection technology can construct a covariance matrix from multichannel seismic data and apply multichannel eigenvalue decomposition techniques to determine the correlation between multichannel data. Its computational characteristics include high resolution based on coherent calculations of 3D seismic data volumes, and it is a coherent algorithm with dip and azimuth angles. On horizontal slices of the fracture detection data volume, it can reveal geological phenomena such as faults, lithological body edges, and unconformities, providing valuable information for solving specific problems in oil and gas exploration.

[0132] The calculation process is as follows:

[0133] (1) Input data volume. You can choose a conventional seismic data volume or an attribute data volume, such as instantaneous phase or colored inversion data volume. Among them, the three single-frequency seismic data volumes of low, medium and high selected from the series of single-frequency volumes obtained in step (1) are named V respectively. 低频体 V 中频体 and V 高频体 , which serves as the input data for fracture detection.

[0134] (2) Select the input azimuth data and dip data volume of the formation.

[0135] (3) Select the main frequency for fracture detection. If no main frequency is entered, perform conventional full-frequency coherent processing. If different main frequencies are entered, perform coherent processing according to different main frequencies.

[0136] (4) Select the time window width for coherent calculation. The higher the main frequency, the narrower the time window. Generally, it is 1 / 2 to 1 of the wavelength of the seismic data after frequency division.

[0137] (5) Select the method for calculating coherence, namely the third-generation coherence with stratigraphic dip angle. Calculate the plane width.

[0138] (6) Output single-frequency fracture detection data volume V 低频体-断裂检测体 V 中频体-断裂检测体 and V 高频体-断裂检 Body measurement.

[0139] (III) CMY fusion for fracture grading:

[0140] The three colors CMY represent the low, medium, and high frequency ranges, respectively, where C represents cyan, M represents magenta, and Y represents yellow.

[0141] The principle of CMY color fusion is to normalize the three input values ​​V1, V2, and V3 to a value range of 0-255, and then assign them to three different primary colors with corresponding light intensities A1, A2, and A3, respectively. Mixing A1, A2, and A3 yields a new color. The primary colors for CMY fusion are cyan (C), magenta (M), and yellow (Y), with light intensities Ac, Am, and Ay, respectively. The resulting accent color is black (all three primary colors have a light intensity of 255). CMY fusion of low, medium, and high single-frequency fracture detection data volumes provides excellent differentiation and characterization of complex fractures.

[0142] The low-frequency, mid-frequency, and high-frequency fracture detection data volumes (V) obtained in step (II) 低频体 - 断裂检测体 V 中频体-断裂检测体 and V 高频体-断裂检测体 The values ​​were normalized to the range of 0-255 to obtain three normalized single-frequency fracture detection data volumes: V 低频体--断裂检测体归一化 V 中频体-断裂检测体归一化 and V 高频体-断裂检测体归一化 The normalization formula is as follows:

[0143]

[0144] in, This refers to the maximum value calculated in a single-frequency fracture detection data volume.

[0145] Three data volumes (V) normalized to the 0-255 range 低频体--断裂检测体归一化 V 中频体-断裂检测体归一化 and V 高频体-断裂检测体归一化 Ac, Am, and Ay, respectively, serve as the light intensities of the three primary colors—cyan, magenta, and yellow—when CMY is blended for display, i.e., Ac = V 低频体-断裂检测体归一化 Am = V 中频体-断裂检测体归一化 And Ay = V 高频体-断裂检测体归一化 The assigned values ​​Ac, Am, and Ay are the coefficients of the CMY fusion display basis function, ultimately used to display the CMY values. Finally, by mixing Ac, Am, and Ay, CMY fusion display can be performed.

[0146] Based on the CMY fusion results, the colors are interpreted according to their respective meanings. According to the CMY fusion principle, cyan represents larger-scale faults; magenta represents medium-scale faults; yellow represents small-scale faults or fractures; blue is the result of low- and mid-frequency fusion, reflecting information on larger and medium-scale faults; green is the result of low- and high-frequency fusion, reflecting large- and small-scale faults; red is the result of mid- and high-frequency fusion, reflecting fractures or fractures with smaller displacements; black is the result of low-, mid-, and high-frequency fusion, representing the development of large-scale, mid-scale, and small-scale faults (or fractures).

[0147] The present invention also provides a complex fault grading detection device based on frequency division fusion. The device includes an acquisition unit for acquiring single-frequency seismic data volumes of three different frequencies (low, medium, and high), a calculation unit for performing coherence calculations on the single-frequency seismic data volumes acquired by the acquisition unit, a color fusion display unit for color fusion display of the single-frequency fault detection data volumes, and a fault grading interpretation unit for interpreting the fault grading based on the color fusion display results and the meaning represented by each color.

[0148] The present invention also provides a complex fracture grading detection device based on frequency division fusion, the device including a processor and a memory; wherein, the memory is used to store machine-executable instructions; the processor is used to read and execute the machine-executable instructions stored in the memory to realize the aforementioned complex fracture grading detection method based on frequency division fusion.

[0149] The following detailed description of the application of the complex fracture grading detection method based on frequency division fusion of the present invention is provided in conjunction with specific embodiments.

[0150] Example 1

[0151] The present invention was used to perform graded detection of complex faults in the Ecuadorian slope zone, achieving ideal results and providing a new solution for fine exploration of the study area.

[0152] The example is located in a block in South America, where deep layers exhibit NNE-trending rift structures and extensional faults. The faults develop along near-north-south trending low-convex zones, differing from common depression-controlling faults and representing regulating faults within a "double-fault" structure. In the middle and shallow layers, monoclines and compressive faults develop with a "high in the northwest, low in the southeast" pattern. The overall faults dip eastward, exhibit strong segmentation, and are arranged in an en echelon pattern, with linear structures developed on both sides.

[0153] The cross-section shows a normal fault at the bottom and a reverse fault at the top, indicating a normal-reverse fault evolution. Before the Cretaceous, it was a tensional normal fault, but it reactivated under compression at the end of the Cretaceous, forming a reverse fault with complex fracture characteristics. The cross-section mostly shows a vertical fault profile. It is difficult to accurately characterize the fault using conventional full-band seismic data, and conventional coherent methods have poor fault identification performance on a plane. Therefore, a targeted frequency domain fault detection technique using CMY fusion to identify multi-level faults is used to address this problem.

[0154] Spectral analysis results of seismic data within the instance block, such as Figure 8 As shown, its dominant frequency is 31.5Hz, and its effective frequency band is 5-60Hz. The seismic data was decomposed into three single-frequency volumes: 20Hz, 30Hz, and 40Hz.

[0155] Comparative analysis revealed that different single-frequency data could reflect seismic data information from different perspectives. The 20Hz single-frequency volume mainly reflected large-scale stratigraphic information, while some small stratigraphic details were ignored; the 40Hz single-frequency volume seemed to have a higher visual resolution than the original data, but the accuracy of fault identification was not significantly improved. This is because excessively increasing the frequency may lead to a decrease in the signal-to-noise ratio; the 30Hz single-frequency volume significantly improved the accuracy of fault identification. Figure 9 The image shows seismic profiles of the study area across the entire frequency band and at different frequencies in single-frequency bodies. The faults on the original earthquake on the left are rather chaotic, with virtually no visible fault features. However, on the 30Hz single-frequency body, the distribution of several small faults is very clear, which greatly helps in interpreting the stratigraphic structure of these small faults.

[0156] The C3 coherent algorithm was used to calculate the fracture detection data volumes for the three single-frequency data volumes mentioned above. For the primary research target layer, slices along the layer were extracted for each single-frequency fracture detection volume. For example... Figure 10 As shown, the comparison reveals that the overall fault plane distribution characteristics exhibited by coherent and conventional coherent data in different frequency bands are quite similar, but each has its own distinct detailed features. The 20Hz single-frequency attribute mainly shows some larger faults with longer extensions; the 30Hz single-frequency attribute, due to its proximity to the dominant frequency of the original seismic data, is quite similar to the original data and mainly reflects medium-scale faults; the 40Hz single-frequency attribute reflects richer fault development characteristics and provides a clearer depiction of the details of smaller-scale faults.

[0157] To address the differences in fault resolution across different scales caused by various single-frequency fault detection methods, and to effectively utilize seismic frequency information and rationally display the dominant frequency of each sample point, cyan, magenta, and yellow are used to represent the frequency-divided fault detection information at 20Hz, 30Hz, and 40Hz, respectively. The results of the frequency-divided energy comparison are then displayed using color overlay.

[0158] like Figure 11 and Figure 12 The diagram shows the cross-sectional and planar views of the CMY-fused fracture classification in the study area. According to the CMY fusion principle, blue represents the result of low-frequency and mid-frequency fusion; low-frequency reflects larger-scale information. Since the fractures in this study area generally have small displacements, the blue distribution is relatively limited. Green represents the result of low- and high-frequency fusion, reflecting both early faults with larger displacements and later faults with smaller displacements. Red represents the result of mid- and high-frequency fusion, reflecting fractures or cracks with smaller displacements. The small fractures depicted are almost invisible in conventional coherence properties, indicating that higher-frequency band-based fracture detection can significantly improve the identification of small faults.

[0159] In summary, the method, apparatus, and equipment for grading and detecting complex fractures based on frequency fusion provided by this invention acquires three single-frequency seismic data volumes at low, medium, and high frequencies. Coherent calculations are performed on these single-frequency seismic data volumes to obtain three single-frequency fracture detection data volumes. These data volumes are then displayed using color fusion, and fracture grading is interpreted based on the meaning represented by each color. This achieves accurate, intuitive, and efficient identification of complex fractures, and facilitates grading interpretation of the identification results. The method fully utilizes the frequency domain attributes of 3D seismic data to respond to fractures at different scales, and combines this with the state-of-the-art C3 algorithm for fine edge detection and identification of secondary fractures, ultimately achieving graded detection of complex fractures. This is of great significance for fine exploration in mature exploration areas and the deployment of subsequent development wells.

[0160] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended 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 described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for graded detection of complex fractures based on frequency division fusion, the method comprising: Acquire single-frequency seismic data volumes at three different frequencies: low, medium, and high. Coherent calculations were performed on the single-frequency seismic data volume to obtain three single-frequency fault detection data volumes: low, medium, and high. The single-frequency fracture detection data volume is displayed using color fusion. Based on the color fusion display results, and combined with the meaning represented by each color, the fracture classification is interpreted to achieve the classification detection of complex fractures. The acquisition of single-frequency seismic data volumes at low, medium, and high frequencies includes: Marr wavelet was used to simulate Ricker wavelet to divide the seismic data into frequencies and extract single-frequency seismic data volumes of three different frequencies: low, medium and high. The color fusion display of the single-frequency fracture detection data volume includes: Normalization calculations were performed on the three single-frequency fracture detection data volumes of low, medium, and high frequencies, respectively. The three normalized single-frequency fracture detection data volumes are used as light intensity volumes of cyan, magenta and yellow primary colors in the CMY fusion display, respectively, and then CMY fusion display is performed. The normalization calculation for the three single-frequency fracture detection data volumes (low, medium, and high) includes the following steps: Obtain the maximum value in the low-frequency fracture detection data volume; Obtain the maximum value in the intermediate frequency fracture detection data volume; Obtain the maximum value in the high-frequency fracture detection data volume; Based on the three single-frequency fracture detection data volumes (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data volumes obtained, the normalized three single-frequency fracture detection data volumes are calculated. The formula for calculating the normalized three single-frequency fracture detection data volumes based on the three single-frequency fracture detection data volumes (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data volumes is as follows: ； In the formula, This is the normalized single-frequency fracture detection data volume; This is a single-frequency fracture detection data volume; This represents the maximum value in the single-frequency fracture detection data volume.

2. The method for graded detection of complex fractures based on frequency division fusion according to claim 1, wherein, The method of using Marr wavelet to simulate Ricker wavelet to perform frequency division on seismic data and extract single-frequency seismic data volumes of low, medium and high frequencies includes the following steps: Obtain raw earthquake data; The effective frequency band of the earthquake is determined based on the original seismic data; The frequency division volume is calculated using Marr wavelet transform within the determined effective seismic frequency band; Extract three different frequency single-frequency seismic data volumes (low, medium, and high) from the frequency-division volume.

3. The method for graded detection of complex fractures based on frequency division fusion according to claim 2, wherein, The calculation of the frequency division body using Marr wavelet transform includes the following steps: Obtain the time-frequency volume corresponding to the original seismic data; A time-frequency attribute body is generated based on the time-frequency body; The start frequency, end frequency, and frequency interval of the original seismic data are determined based on the time-frequency attribute volume. The frequency divider is obtained based on the starting frequency, the ending frequency, and the frequency interval.

4. The method for graded detection of complex fractures based on frequency division fusion according to claim 1, wherein, The coherence calculation of the single-frequency seismic data volume includes: The C3 coherent algorithm was used to calculate the single-frequency seismic data volumes of the three different frequencies: low, medium, and high.

5. The method for graded detection of complex fractures based on frequency division fusion according to claim 1, wherein, The interpretation of the fracture grading based on the color fusion display results and the meaning represented by each color includes: The cyan color indicates a larger-scale fault; The magenta color indicates a medium-scale fault; Yellow indicates smaller-scale fracture structures, which are small-scale faults or cracks. The blue color represents the result of low- and mid-frequency fusion, reflecting fault information at larger and medium scales. The green color represents the result of the fusion of low and high frequencies, reflecting both early faults with large displacements and later faults with smaller displacements. The red color indicates the result of the fusion of mid-frequency and high-frequency signals, reflecting fractures or cracks with small breakpoints. The black color is the result of the fusion of low, medium and high frequencies, indicating the development of large-scale faults, medium-scale faults and smaller-scale fracture structures.

6. A complex fracture grading detection device based on frequency division fusion, the device comprising: The acquisition unit is used to acquire single-frequency seismic data volumes of three different frequencies: low, medium, and high. The calculation unit is used to perform coherent calculations on the single-frequency seismic data volume acquired by the acquisition unit to obtain three single-frequency fault detection data volumes: low, medium, and high. A color fusion display unit is used to perform color fusion display on the single-frequency fracture detection data volume; The fracture grading interpretation unit is used to interpret the fracture grading based on the color fusion display results and the meaning represented by each color. The acquisition unit acquires single-frequency seismic data volumes of three different frequencies: low, medium, and high. Marr wavelet was used to simulate Ricker wavelet to divide the seismic data into frequencies and extract single-frequency seismic data volumes of three different frequencies: low, medium and high. The color fusion display unit performs color fusion display on the single-frequency fracture detection data volume, including: Normalization calculations were performed on the three single-frequency fracture detection data volumes of low, medium, and high frequencies, respectively. The three normalized single-frequency fracture detection data volumes are used as light intensity volumes of cyan, magenta and yellow primary colors in the CMY fusion display, respectively, and then CMY fusion display is performed. The color fusion display unit performs normalization calculations on the low, medium, and high single-frequency fracture detection data volumes, including the following steps: Obtain the maximum value in the low-frequency fracture detection data volume; Obtain the maximum value in the intermediate frequency fracture detection data volume; Obtain the maximum value in the high-frequency fracture detection data volume; Based on the three single-frequency fracture detection data volumes (low, medium, and high) and the maximum values ​​corresponding to the three single-frequency fracture detection data volumes obtained, the normalized three single-frequency fracture detection data volumes are calculated. The color fusion display unit calculates the normalized three single-frequency fracture detection data bodies based on the low, medium, and high single-frequency fracture detection data bodies and the maximum values ​​corresponding to the three single-frequency fracture detection data bodies obtained, using the following formula: ； In the formula, This is the normalized single-frequency fracture detection data volume; This is a single-frequency fracture detection data volume; This represents the maximum value in the single-frequency fracture detection data volume.

7. The complex fracture grading detection device based on frequency division fusion according to claim 6, wherein, The acquisition unit uses Marr wavelet to simulate Ricker wavelet to divide the seismic data into frequencies and extracts single-frequency seismic data volumes of three different frequencies (low, medium, and high), including the following steps: Obtain raw earthquake data; The effective frequency band of the earthquake is determined based on the original seismic data; The frequency division volume is calculated using Marr wavelet transform within the determined effective seismic frequency band; Extract three different frequency single-frequency seismic data volumes (low, medium, and high) from the frequency-division volume.

8. The complex fracture grading detection device based on frequency division fusion according to claim 7, wherein, The acquisition unit calculates the frequency division body using Marr wavelet transform, including the following steps: Obtain the time-frequency volume corresponding to the original seismic data; A time-frequency attribute body is generated based on the time-frequency body; The start frequency, end frequency, and frequency interval of the original seismic data are determined based on the time-frequency attribute volume. The frequency divider is obtained based on the starting frequency, the ending frequency, and the frequency interval.

9. The complex fracture grading detection device based on frequency division fusion according to claim 6, wherein, The computing unit performs coherent calculations on the acquired single-frequency seismic data volume to obtain three single-frequency fault detection data volumes: low, medium, and high. The C3 coherent algorithm was used to calculate the single-frequency seismic data volumes of the three different frequencies: low, medium, and high.

10. The complex fracture grading detection device based on frequency division fusion according to claim 6, wherein, The fracture grading interpretation unit interprets the fracture grading based on the color fusion display results and the meaning represented by each color, including: The cyan color indicates a larger-scale fault; The magenta color indicates a medium-scale fault; Yellow indicates smaller-scale fracture structures, which are small-scale faults or cracks. The blue color represents the result of low- and mid-frequency fusion, reflecting fault information at larger and medium scales. The green color represents the result of the fusion of low and high frequencies, reflecting both early faults with large displacements and later faults with smaller displacements. The red color indicates the result of the fusion of mid-frequency and high-frequency signals, reflecting fractures or cracks with small breakpoints. The black color is the result of the fusion of low, medium and high frequencies, indicating the development of large-scale faults, medium-scale faults and smaller-scale fracture structures.

11. A complex fracture grading detection device based on frequency division fusion, the device comprising a processor and a memory; wherein, The memory is used to store machine-executable instructions; The processor is configured to read and execute machine-executable instructions stored in the memory to implement the method as described in any one of claims 1 to 5.

Images