Method, device and equipment for establishing sequence stratigraphic framework based on high-precision age stratigraphic body and medium

Abstract

The invention belongs to the technical field of oil and gas seismic exploration, and particularly discloses a sequence stratigraphic framework establishment method, device and equipment based on a high-precision age stratigraphic body and a medium. In the method, firstly, well-seismic comprehensive calibration is carried out by combining a well-connected sequence comparison result and a seismic data optimization processing result; then, on the basis of well-seismic comprehensive calibration, a sequence top and bottom grillwork model is constructed based on seed point tracking intelligent explanation; then, based on the sequence top and bottom framework model, age stratigraphic body tracking is carried out by adopting a plane wave decomposition method, and high-precision multi-layer interpretation is realized; and finally, constructing a high-precision sequence framework model by combining a well-connected sequence comparison result and high-precision multi-layer interpretation. According to the method, the sequence stratigraphic framework model which is finer, more accurate and higher in timeliness can be established, an important guiding effect is achieved for target conception and reservoir fine representation, and the method is suitable for sequence stratigraphic framework research in petroleum and natural gas seismic exploration.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of oil and gas seismic exploration, and relates to a method, device, equipment and medium for establishing a sequence stratigraphic framework based on a high-precision chronological stratigraphic body. BACKGROUND

[0002] The sequence stratigraphic framework is the basis of sequence stratigraphic research, can orderly incorporate the strata formed in the same period into the time-stratigraphic correlation framework of the relevant period, and perform isochronous stratigraphic correlation and stratigraphic superimposition pattern description. Since the strata marked in the time-stratigraphic framework must be formed in the same period, the sequence boundary with isochronous correlation significance is limited within a certain geological time interval, or in other words, the sequence stratigraphic unit in the sequence stratigraphic framework is an isochronous rock assemblage with a synchronous sedimentary evolution sequence defined by a physical interface with chronological significance. Therefore, the establishment of the sequence stratigraphic framework can effectively improve the regional stratigraphic correlation accuracy, thereby making more reasonable explanations for paleogeographic reconstruction, basin analysis and oil and gas geological evolution history, and providing a more reliable geological model for fine geological research such as favorable facies belt or block prediction and evaluation.

[0003] The conventional high-precision sequence stratigraphic framework construction is a small-range, high-precision stratigraphic framework guided by the high-resolution sequence stratigraphic theory represented by Cross (1993), which can realize the division of four to five sequence levels and plays a relatively positive guiding role in target conception and fine quantitative characterization of reservoirs. There are two main implementation methods: 1. On the basis of conventional fine interpretation based on the top and bottom boundaries of the target layer, the model-driven principle is used to realize the small layer division within the target layer by using the interpolation method parallel to the top and bottom or in equal proportions. 2. On the basis of conventional fine interpretation based on the top and bottom boundaries of the target layer, the data simulation driven by the occurrence of the seismic body is used to realize the small layer division within the target layer.

[0004] The above two methods for establishing a high-precision sequence stratigraphic framework still have the following shortcomings: 1. The overall efficiency is slow and the precision is low by using the traditional manual grid line interpretation scheme, which is difficult to adapt to the current fast-paced and efficient development needs; 2. The small layers interpolated based on the model-driven principle are difficult to truly reflect the strata sedimentary law, and even in low signal-to-noise ratio areas, there may be phenomena that violate the strata sedimentary law, and the precision is low; 3. Although the data-driven method can obtain fine information of the internal small layers, the conventional seismic data has serious tracking difficulties and poor tracking effect due to the control of the signal-to-noise ratio and the trend of the seismic occurrence. Based on the above research, the above methods are difficult to realize the establishment of a high-precision sequence stratigraphic framework.

[0005] Chinese invention patent application CN112505753B discloses a method for constructing a high-precision sequence lattice model driven by earthquakes. This method first interprets the seismic horizons and faults based on well-seismic calibration. Then, by combining the interpreted horizons, faults, well-surface strata, and seismic bodies, and under the constraints of the interpreted horizons and faults, the higher-level sequence strata above the well are interpolated using the attitude of the seismic bodies to achieve high-precision sequence horizon interpolation. This interpolation produces a fine sequence horizon that conforms to the stratigraphic distribution, thus realizing the construction of a high-precision sequence lattice model.

[0006] The aforementioned method only discloses the use of seismic body attitude-driven methods to interpolate the sub-layers within the higher-level sequence stratigraphy from the wellbore, thereby establishing a high-precision sequence stratigraphic framework. However, this method, which controls the quality of the seismic body's attitude, suffers from a significant decrease in stratigraphic tracking quality when well logging data is scarce or the conditions are not met in structurally complex areas. This leads to distortion of layers within the seismic stratigraphic body, severely impacting the accuracy of the high-precision sequence stratigraphic framework model.

[0007] The above analysis shows that the existing high-precision sequence stratigraphic framework model cannot meet the needs of high-precision reservoir description and exploration. There is an urgent need to establish a more refined, accurate and timely sequence stratigraphic framework model to improve the guidance for reservoir description and quantitative reservoir characterization, and to provide a more reliable geological model for subsequent target design and fine reservoir characterization. Summary of the Invention

[0008] The purpose of this invention is to provide a method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body, so as to establish a more refined, accurate and timely sequence stratigraphic framework model, which plays an important guiding role in target design and fine reservoir characterization.

[0009] The second objective of this invention is to provide a sequence stratigraphic framework establishment device based on high-precision chronostratigraphic bodies for implementing the above-described method;

[0010] The third objective of this invention is to provide a terminal device that, when executing its own program, can implement a method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0011] The fourth objective of this invention is to provide a computer-readable storage medium for storing a corresponding computer program for a method of establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0012] To achieve the above objectives, the technical solution adopted by this invention is as follows:

[0013] A method for establishing a sequence stratigraphic framework based on high-precision chronostratigraphic bodies, comprising the following steps:

[0014] S1. Combining the results of well sequence comparison and the results of seismic data optimization, perform well-seismic comprehensive calibration;

[0015] In this step, single-well sequence stratigraphy is achieved by analyzing well logging data and core data. Then, sequence stratigraphy is correlated across wells based on the single-well sequence stratigraphy to construct a sequence stratigraphic correlation profile across wells and obtain the sequence stratigraphic correlation results across wells.

[0016] In this step, spatial scanning of the original 3D seismic data is performed, and dip steering information is used as the vertical control trend of noise. Denoising processing is carried out on the post-stack data volume to obtain the seismic data optimization processing results.

[0017] S2. Based on the comprehensive calibration of well seismic data, construct a sequence top and bottom grid model based on seed point tracking intelligent interpretation;

[0018] In this step, based on the integrated well-seismic calibration, and targeting the top and bottom boundaries of the target section, intelligent stratigraphic interpretation is carried out using the seed point impregnation method based on the acquisition of seed points in the stratigraphic framework, and a stratigraphic top and bottom framework model is constructed.

[0019] S3. Based on the sequence top-bottom framework model, plane wave decomposition method is used to trace the chronostratigraphic body and achieve high-precision multi-layer interpretation.

[0020] S4. Combining the results of well sequence correlation and high-precision multilayer interpretation, a high-precision sequence stratigraphic framework model is constructed, thus obtaining a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0021] As a limitation, the process of obtaining the seismic data optimization processing results in step S1 shall be performed in the following order:

[0022] S11. Calculate the tilt angle using 3D Fourier transform.

[0023] In this step, the original three-dimensional seismic data is spatially scanned, and the instantaneous frequency and instantaneous wavenumber information along the x and y directions are calculated using the 3D Fourier transform calculation formula, and then the stratum dip angle information is obtained.

[0024] During this process, the instantaneous frequencies along the x and y directions are calculated according to formula (1).

[0025]

[0026] in, Let u(x,t) represent the instantaneous phase, u(x,t) represent the seismic data to be processed, and t represent time. H (x,t) represents the Hilbert transform of u(x,t), where u and u H The derivative with respect to time can be obtained through Fourier transform;

[0027] According to formula (2), the instantaneous wavenumber k of u(x,t) in the x-direction is obtained. x ,

[0028]

[0029] Further, the apparent tilt angle of u(x,t) in the x-direction is obtained.

[0030] P x =k x / ω Equation (3)

[0031] Similarly, we obtain the apparent tilt angle P of u(x,y,t) in the y-direction. y ;

[0032] Finally, based on the above calculations and according to the directional definition relationship, the true dip angle θ of the formation is calculated.

[0033] S12. Seismic denoising based on dip angle constraint

[0034] In this step, structural analysis is introduced through anisotropic diffusion filtering theory, and the structural tensor E of the image is calculated according to formula (4). s ,

[0035]

[0036] Among them, u σ =G σ *u is obtained by convolving a Gaussian function with noise scale δ with a smoothed u; G s This represents the Gaussian kernel function with a s integration scale, used to calculate the mean directional field, where s > δ; Represents the gradient operator;

[0037] For E s Eigenvalue decomposition yields eigenvalues ​​λ1 and λ2, and their corresponding orthogonal eigenvectors ω1 and ω2. ω1 represents the direction of maximum gradient change, and ω2 represents the direction of minimum gradient change, i.e., the direction with a consistent structure.

[0038] Next, let the eigenvalues ​​be:

[0039] C1=α,λ1=λ2 Equation (5)

[0040]

[0041] Where 0 < α < 1, is used to control the diffusion intensity along the direction where the image gradient changes significantly; d represents the consistency parameter, d 2 =(λ1-λ2) 2 ;

[0042] In the isotropic region, C1 = C2, and the diffusion intensity is small in all directions; in the region where the image texture direction is highly consistent, the diffusion is mainly along the ω2 direction.

[0043] In the anisotropic region, the vertical information is denoised along the ω1 direction to enhance the continuity of the transverse seismic axis.

[0044] As a further limitation, step S2 includes the following steps:

[0045] S21, Grid Seed Point Pickup

[0046] Based on the integrated calibration of well seismic data, grid profiles with high signal-to-noise ratio and easy identification are selected for large-grid interpretation, and the points of these grid interpretations are used as seed points for data diffusion.

[0047] S22, Seed Implantation Algorithm Diffusion Propagation

[0048] Starting from the waveform characteristics of a seed point, the similarity characteristics of the seismic phase axis waveform are used to determine whether the pixels around the seed point are consistent with or similar to the seed point, so as to determine whether the pixel belongs to the filling area where the seed point is located.

[0049] Based on the principle of discarding recursion, the eight-connectivity algorithm is used as the basis to expand in eight specific directions from the seed point to realize the propagation of the seed point within the work area, obtain the preliminary layer point, and obtain the layer information.

[0050] S22. Layer Phase Adjustment under Waveform Feature Control

[0051] By utilizing waveform characteristics to adjust the phase characteristics of the stratigraphic position, accurate positioning of the stratigraphic position can be achieved.

[0052] As a further refinement, step S3, guided by the theory of "predictive imaging", predicts the optimal slope of adjacent seismic data based on the principle of minimum error, extracts seismic images, and uses seed point control tracking to achieve high-precision multi-layer interpretation.

[0053] Specifically, the following steps are included:

[0054] S31. Determining the dip angle of local strata

[0055] Based on formula (7), the local slope of the seismic phase axis is estimated, and the local stratum dip angle is calculated.

[0056]

[0057] Where p(t,x) represents the wave field; ψ represents the slope of the spatial variation, which depends on the transformations of t and x;

[0058] Next, with the slope being constant, the equation represented by equation (7) is simplified to:

[0059] p(t,x)=f(t-ψx) Equation (8)

[0060] Where f(t) represents any waveform;

[0061] Then, during the seismic data dip angle scanning process, variance constraints are applied according to equation (9).

[0062] D(ψ)=E(ψ 2 )-E(ψ) 2 Equation (9)

[0063] Where D(ψ) represents the variance of the probability distribution, the slope ψ of the spatial variation is a random variable that follows a probability distribution, and E(ψ) represents the expected value of the random variable ψ.

[0064] Finally, the minimum variance is calculated based on the above results;

[0065] S32, Dip-propagation extraction of seismic imaging

[0066] Predictive imaging technology is used to predict seismic traces by utilizing the dip angle information of seismic phase axes in different orientations; a seismic trace is predicted using a simple recursive formula for adjacent seismic traces; the prediction of the m-th seismic trace from the 1st seismic trace is expressed by formula (10) as follows.

[0067] P 1,m =P m-1,m P m-2,m-1 ...P 2,3 P 1,2 Equation (10)

[0068] Among them, P m-1,m This represents a prediction operator, which means predicting the m-th seismic trace based on the (m-1)-th seismic trace;

[0069] Finally, by analogy, high-quality stratigraphic dip information can be predicted for the entire seismic space.

[0070] S33, Obtain the layer body

[0071] Seed points are used instead of seed seismic traces in the predicted imaging, serving as the starting point and reference point for horizon tracking. Under the control of stratigraphic dip information, the seed seismic traces are propagated in all directions. Once all seismic traces in the entire area have been tracked, a series of horizons are obtained, forming a horizon volume, thus achieving high-precision multi-layer interpretation.

[0072] As a further refinement, in step S4, based on step S3, according to stratigraphic sedimentary theory and geological phenomena, combined with the results of well sequence correlation, each stratum is given more sedimentary and chronostratigraphic meanings, and reordered to convert the stratum body into a high-precision sequence stratigraphic framework model, that is, to obtain a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0073] A sequence stratigraphic framework establishment device based on a high-precision chronostratigraphic body is used in the aforementioned method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body. The device includes:

[0074] The well-seismic integrated calibration module performs well-seismic integrated calibration by combining the results of well sequence comparison and the results of seismic data optimization.

[0075] The seed point tracking intelligent interpretation module constructs a sequence top-bottom grid model based on the well-seismic integrated calibration and seed point tracking intelligent interpretation.

[0076] The chronostratigraphic body tracing module, based on the sequence top-bottom framework model, uses the plane wave decomposition method to trace chronostratigraphic bodies, achieving high-precision multi-layer interpretation;

[0077] The high-precision sequence grid model construction module combines well-connected sequence comparison results and high-precision multilayer interpretation to construct a high-precision sequence grid model.

[0078] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0079] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement the above-described method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0080] The present invention, by adopting the above-described technical solution, achieves the following technical advancements compared to existing technologies:

[0081] (1) The method of the present invention can provide a more refined, accurate and timely sequence stratigraphic framework model to improve the guidance for reservoir description and quantitative reservoir characterization, and provide a more reliable geological model for subsequent target design and fine reservoir characterization; it can solve the problems of weak foundation, low accuracy and slow timeliness in the existing technology of the time sequence framework.

[0082] (2) This invention improves and innovates the conventional sequence stratigraphic framework model establishment process. By optimizing the original data and ensuring lateral continuity, it carries out AI intelligent interpretation based on the top and bottom layers of the target segment. At the same time, it innovates key links such as improving the internal sublayer tracking method by using plane wave decomposition technology. It establishes a more refined, accurate and timely sequence stratigraphic framework model, which plays an important guiding role in target conception and reservoir fine characterization.

[0083] (3) Chronostratigraphic tracking is mainly divided into two categories: model-driven and data-driven. Most existing technologies are based on model-driven methods, which use interpolation methods with stratigraphic slices parallel to the top and bottom or at equal proportions. This method has no requirements for the quality of seismic data and is easy to implement, but it cannot truly reflect the stratigraphic depositional patterns. In fact, it may even violate the stratigraphic depositional patterns in areas with low signal-to-noise ratios, resulting in low overall accuracy. With the advancement of technology, data-driven isochronous layer analysis methods are being developed. This method can truly reflect the characteristics of isochronous interfaces of syn-sedimentary geological bodies. Currently, most data-driven algorithms mainly rely on seismic dip data. This method first performs dip scanning on the original data to extract stratigraphic dip bodies, and then calculates the slope of each data volume based on the dip body to complete high-precision layer tracking. This method is suitable for seismic data with small variations in layer thickness, high signal-to-noise ratios, or good continuity of seismic phase axes. While the data is suitable and can yield good results in seismic stratigraphic analysis, if the quality of the seismic data or the geological structure characteristics fail to meet the above conditions, the quality of stratigraphic tracking will significantly decrease, leading to distortion of each layer in the seismic stratigraphic body and a significant reduction in its continuity. Furthermore, because this method is implemented in two steps, the tracking process takes twice the time of the original data volume, resulting in low timeliness. This invention employs a plane wave decomposition method based on the nonlinear seismic waveform matching principle of dynamic programming theory. By calculating the optimal slope of adjacent seismic data according to the principle of minimizing the error between predicted data and seismic data, and guided by the theory of "predictive imaging," seismic images are extracted. Seed point control tracking is used to achieve high-precision multi-layer interpretation. This is a completely new algorithm that directly applies the in-phase axis texture and morphology of the original seismic data for tracking, effectively improving the tracking quality in low signal-to-noise areas and doubling the timeliness.

[0084] This invention belongs to the field of seismic exploration technology for oil and gas. It can establish a more refined, accurate and timely sequence stratigraphic framework model, which plays an important guiding role in target design and fine reservoir characterization. Attached Figure Description

[0085] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0086] In the attached diagram:

[0087] Figure 1 This is a flowchart of Embodiment 1 of the present invention;

[0088] Figure 2 These are cross-sectional views showing the effects of tilt-guided optimization before and after the process in Embodiment 1 of the present invention.

[0089] in, Figure 2 a is a cross-sectional view showing the effect of the tilt-guided optimization process before the process.

[0090] Figure 2 b is the cross-sectional view after optimization based on tilt angle guidance;

[0091] Figure 3 This is a diagram illustrating the effect of predictive imaging guiding high-precision small-layer interpretation in Embodiment 1 of the present invention.

[0092] in, Figure 3 a represents the prediction model diagram;

[0093] Figure 3 b is the extracted seismic imaging result image;

[0094] Figure 3 c is a high-precision small-layer interpretation diagram.

[0095] Figure 4 This is a diagram showing the high-precision stratigraphic layer tracking effect based on the plane wave decomposition method in Embodiment 1 of the present invention. Detailed Implementation

[0096] The preferred embodiments of the present invention will now be described with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the scope of the invention.

[0097] Example 1: A method for establishing a sequence stratigraphic framework based on high-precision chronostratigraphic bodies

[0098] The examples were applied to the study of a high-precision sequence stratigraphic framework model system of the Jurassic strata in a basin in western my country, such as... Figure 1 As shown, proceed in the order of steps S1, S2, S3, and S4.

[0099] S1. Combining the results of well sequence comparison and the results of seismic data optimization, perform well-seismic comprehensive calibration;

[0100] In this step, single-well sequence stratigraphy is achieved by analyzing well logging data and core data. Then, sequence stratigraphy is correlated across wells based on the single-well sequence stratigraphy to construct a sequence stratigraphic correlation profile across wells and obtain the sequence stratigraphic correlation results across wells.

[0101] In this step, spatial scanning of the original 3D seismic data is performed, and dip steering information is used as the vertical control trend for noise. Denoising processing is then carried out on the post-stack data volume to obtain the optimized seismic data processing results.

[0102] Different sedimentary environments, due to differences in source material, hydrodynamic conditions, and water depth, inevitably result in different sedimentary assemblages and sequence characteristics (positive cycles, reverse cycles, and massive formations). By analyzing the differences in well logging and core data, it is possible to divide and compare single-well and interconnected well sequences, providing a sequence correlation scheme for seismic interpretation.

[0103] By analyzing the spectrum of the target layer, the spectral characteristics of the study area were clarified. A theoretical Ricker wavelet consistent with the dominant frequency of the seismic data was selected for calibration, preliminarily clarifying the correspondence between well logging and seismic data. Simultaneously, during the synthetic record calibration process, the matching relationship between the seismic record and wellbore traces was precisely calibrated based on the seismic data waveforms and marker layers. On this basis, the main window containing the target layer was selected, and the wellbore trace seismic wavelet was extracted for refined reservoir-based calibration. After single-well calibration, multi-well synthetic record comparisons were conducted to ensure geological and seismic consistency, thus achieving a combination of geology and seismic analysis.

[0104] S2. Based on the comprehensive calibration of well seismic data, construct a sequence top and bottom grid model based on seed point tracking intelligent interpretation;

[0105] In this step, based on the integrated well-seismic calibration, and targeting the top and bottom boundaries of the target section, intelligent stratigraphic interpretation is carried out using the seed point impregnation method based on the acquisition of grid seed points, and a sequence top and bottom grid model is constructed.

[0106] S3. Based on the sequence top-bottom framework model, plane wave decomposition method is used to trace the chronostratigraphic body and achieve high-precision multi-layer interpretation.

[0107] S4. Combining the results of well sequence correlation and high-precision multilayer interpretation, a high-precision sequence stratigraphic framework model is constructed, thus obtaining a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0108] Specifically, in step S1, since the dip steering body with spatial information has a good characterization effect on the vertical changes of strata, it can significantly reduce random disturbances. Therefore, using the dip steering information as the vertical control trend of noise and carrying out denoising processing on the post-stack data volume is beneficial to improving attribute calculation and target detection capabilities. The process of obtaining the seismic data optimization processing results is carried out in the following steps:

[0109] S11. Calculate the tilt angle using 3D Fourier transform.

[0110] In this step, the original three-dimensional seismic data is spatially scanned, and the instantaneous frequency and instantaneous wavenumber information along the x and y directions are calculated using the 3D Fourier transform calculation formula, and then the stratum dip angle information is obtained.

[0111] During this process, the instantaneous frequencies along the x and y directions are calculated according to formula (1).

[0112]

[0113] in, Let u(x,t) represent the instantaneous phase, u(x,t) represent the seismic data to be processed, and t represent time. H (x,t) represents the Hilbert transform of u(x,t), where u and u H The derivative with respect to time can be obtained through Fourier transform;

[0114] According to formula (2), the instantaneous wavenumber k of u(x,t) in the x-direction is obtained. x ,

[0115]

[0116] Further, the apparent tilt angle of u(x,t) in the x-direction is obtained.

[0117] P x =k x / ω, Equation (3)

[0118] Similarly, we obtain the apparent tilt angle P of u(x,y,t) in the y-direction. y ;

[0119] Finally, based on the above calculations and according to the directional definition relationship, the true dip angle θ of the formation is calculated.

[0120] S12. Seismic denoising based on dip angle constraint

[0121] To ensure reliable and accurate denoising, the tilt guide body is used as the noise control trend, and denoising is performed along the vertical information of the structure. In this step, structural analysis is introduced through anisotropic diffusion filtering theory, and the structural tensor E of the image is calculated according to formula (4). s ,

[0122]

[0123] Among them, u σ =G σ *u is obtained by convolving a Gaussian function with noise scale δ with a smoothed u; G s This represents the Gaussian kernel function with a s integration scale, used to calculate the mean directional field; generally, s > δ. Represents the gradient operator;

[0124] For E s Eigenvalue decomposition yields eigenvalues ​​λ1 and λ2, and their corresponding orthogonal eigenvectors ω1 and ω2. ω1 represents the direction of maximum gradient change, and ω2 represents the direction of minimum gradient change, i.e., the direction with a consistent structure.

[0125] To ensure that the diffusion direction follows the structural direction, the eigenvalues ​​of the diffusion tensor should be consistent with those of the structure tensor. Let the eigenvalues ​​be:

[0126] C1=α,λ1=λ2 Equation (5)

[0127]

[0128] Where 0 < α < 1, is used to control the diffusion intensity along the direction where the image gradient changes significantly; d represents the consistency parameter, d 2 =(λ1-λ2) 2 ;

[0129] In isotropic regions, C1 = C2, and the diffusion intensity is relatively small in all directions; in regions with strong image texture direction consistency, diffusion mainly occurs along the ω2 direction; in anisotropic regions, denoising of vertical information is performed along the ω1 direction to enhance the continuity of lateral seismic axes. This is beneficial for AI-driven automatic layer interpretation and high-precision chronograph tracking. Figure 2 The image shown is a cross-sectional view of the optimization process based on tilt angle guidance in this embodiment; wherein, Figure 2 a is a cross-sectional view showing the effect of the tilt-guided optimization process before the process. Figure 2 b is the profile effect diagram after dip-guided optimization. It can be seen from the figure that the original seismic data is affected by vertical linear noise and has poor lateral continuity. The seismic data optimized by dip-guided optimization has significantly reduced vertical information and enhanced lateral continuity, which is beneficial to the interpretation of sequence stratigraphic framework.

[0130] Further, step S2 includes the following steps:

[0131] S21, Grid Seed Point Pickup

[0132] Based on the integrated calibration of well and seismic data, grid profiles with high signal-to-noise ratio and easy identification are selected for large-grid interpretation, and the points interpreted from these grid profiles are used as seed points for data diffusion; laying the foundation for subsequent stratigraphic tracking.

[0133] S22, Seed Implantation Algorithm Diffusion Propagation

[0134] Starting from the waveform characteristics of a seed point, the system determines whether the pixels around the seed point are consistent with or similar to the seed point based on the similarity characteristics of the seismic phase axis waveform, thus determining whether the pixel belongs to the filling area where the seed point is located. Based on the principle of discarding recursion, the system adopts an eight-connected algorithm to expand in eight specific directions from the seed point as the center, effectively avoiding problems such as stack overflow in automatic stratigraphic interpretation, meeting the requirements of efficiency and minimum propagation path of the stratigraphic tracking method, realizing the propagation of the seed point within the work area, obtaining the preliminary stratigraphic point, and acquiring stratigraphic information.

[0135] S22. Layer Phase Adjustment under Waveform Feature Control

[0136] Due to the propagation of seed points, the obtained stratigraphic information is relative to adjacent seismic traces, inevitably containing errors and causing deviations in the actual stratigraphic positions. Therefore, it is necessary to adjust for this phase deviation. Since the seismic phase axis is closely related to the seismic waveform, the waveform characteristics are used to adjust the phase characteristics of the stratigraphic positions to achieve accurate repositioning of the stratigraphic points.

[0137] Step S3, the implementation of the plane wave decomposition algorithm tracking, includes three key steps: S31, estimating the local slope of the seismic phase axis based on the minimum variance algorithm and calculating the local stratum dip angle; S32, guided by the "predictive imaging" theory, using dip propagation to predict and extract seismic images; S33, selecting the "seed point contamination method" for efficient propagation within the work area to achieve high-precision multi-layer interpretation.

[0138] Specifically, step S3 includes the following steps:

[0139] S31. Determining the dip angle of local strata

[0140] Based on formula (7), the local slope of the seismic phase axis is estimated, and the local stratum dip angle is calculated.

[0141]

[0142] Where p(t,x) represents the wave field; ψ is the slope representing the spatial variation, which depends on the transformations of t and x;

[0143] Next, with the slope being constant, the equation represented by equation (7) is simplified to:

[0144] p(t,x)=f(t-ψx) Equation (8)

[0145] Here, f(t) represents an arbitrary waveform; and the slope ψ, representing spatial variation, propagates outwards along the seismic trace trajectory. During this movement, to ensure its rationality, variance is used to measure the degree of dispersion of random variables between traces.

[0146] In probability distributions, variance is used to calculate the difference between each variable and the population mean, representing the distance between a variable and its expected value. In this embodiment, during the seismic data dip angle scanning process, variance is used for constraint according to equation (9).

[0147] D(ψ)=E(ψ 2 )-E(ψ) 2 Equation (9)

[0148] Wherein, the variance D(ψ) of the probability distribution and the slope ψ of the spatial variation are random variables that follow a probability distribution, and E(ψ) is the expected value of the random variable ψ.

[0149] Finally, the minimum variance is calculated based on the above results;

[0150] With the same sample size, a larger variance indicates a greater degree of dispersion and poorer stability of the random variable; conversely, a smaller variance indicates a smaller degree of dispersion and better stability. Since the waveforms of adjacent seismic traces are essentially similar, minimum variance is used as a constraint during seismic dip angle scanning to ensure smooth spatial changes in the dip angle. In 3D seismic data, when the dip angle moves within a certain range of the target area, the center points of the seismic phase axes can be flattened into a straight line, and the variance can be calculated. The dip angle corresponding to the minimum variance can then be considered as the dip angle of adjacent seismic traces at that point.

[0151] S32, Dip-propagation extraction of seismic imaging

[0152] Predictive imaging technology is used to predict seismic traces by utilizing the dip angle information of seismic phase axes in different orientations; a seismic trace is predicted using a simple recursive formula for adjacent seismic traces; the k-th seismic trace is predicted from the 1st seismic trace, expressed by formula (10) as follows:

[0153] P 1,m =P m-1,m P m-2,m-1 ...P 2,3 P 1,2 Equation (10)

[0154] Among them, P m-1,m This represents a prediction operator, which means predicting the m-th seismic trace based on the (m-1)-th seismic trace;

[0155] Finally, by analogy, high-quality stratigraphic dip information can be predicted for the entire seismic space.

[0156] like Figure 3 The image shown is an illustration of the effect of predictive imaging guiding high-precision small-layer interpretation in this embodiment; wherein, Figure 3 a represents the prediction model diagram; Figure 3 b is the extracted seismic imaging result image;Figure 3 c shows the high-precision sub-layer interpretation effect. As can be seen from the figure, the seismic imaging effect extracted using the optimal slope method based on adjacent seismic data is basically consistent with the original prediction model image, reflecting the overall contour information of the predicted model, such as dipping strata, unconformities, and faults; then along... Figure 3 b. Extract the seismic imaging effect map for tracking, and use formula (10) to predict the imaging operator to extract the layer information.

[0157] S33, Obtain the layer body

[0158] To further extract high-precision chronostratigraphic bodies, seed points are used instead of seed seismic traces in the predicted imaging, serving as the starting and reference points for stratigraphic tracking. Under the control of stratigraphic dip information, the seed seismic traces are propagated outwards. Once all seismic traces in the entire area have been tracked, a series of stratigraphic levels are obtained, forming a stratigraphic body, thus achieving high-precision multi-layer interpretation. For example... Figure 4 The figure shown is a high-precision stratigraphic layer tracking effect diagram based on the plane wave decomposition method in this embodiment. It can be seen from the figure that the tracked layers can reflect the stratigraphic morphology of the original seismic data, the lateral continuity of the stratigraphic volume is good, and it can better characterize the stratigraphic information of inclined strata and unconformities.

[0159] In this embodiment, in step S4, based on step S3, according to stratigraphic sedimentary theory and geological phenomena, combined with the results of well sequence correlation, each stratum is given more sedimentary and chronostratigraphic meanings, and reordered to convert the stratum body into a high-precision sequence stratigraphic framework model, that is, to obtain a sequence stratigraphic framework based on a high-precision chronostratigraphic body.

[0160] This embodiment leverages the abundant drilling and logging data, as well as two-width and one-height seismic data, in the study area to fully explore seismic structural information. It utilizes seismic waveform control to conduct AI-powered intelligent interpretation and high-precision chronostratigraphic body tracing to establish a high-precision sequence stratigraphic framework model. Based on the improved accuracy of the sequence stratigraphic framework model, its lateral distribution is made more reliable, providing a stronger technical foundation for high-precision reservoir prediction and characterization in the oil and gas exploration field.

[0161] Example 2: A sequence stratigraphic framework establishment device based on high-precision chronostratigraphic bodies

[0162] This embodiment is used to implement Embodiment 1, and includes:

[0163] The well-seismic integrated calibration module performs well-seismic integrated calibration by combining the results of well sequence comparison and the results of seismic data optimization.

[0164] The seed point tracking intelligent interpretation module constructs a sequence top-bottom grid model based on the well-seismic integrated calibration and seed point tracking intelligent interpretation.

[0165] The chronostratigraphic body tracing module, based on the sequence top-bottom framework model, uses the plane wave decomposition method to trace chronostratigraphic bodies, achieving high-precision multi-layer interpretation;

[0166] The high-precision sequence grid model construction module combines well-connected sequence comparison results and high-precision multilayer interpretation to construct a high-precision sequence grid model.

[0167] Example 3: A terminal device

[0168] This embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the sequence stratigraphic framework establishment method based on a high-precision chronostratigraphic body as described in Embodiment 1.

[0169] Example 4: A computer-readable storage medium

[0170] The computer-readable storage medium of this embodiment stores a computer program that, when executed by a processor, is used to implement the sequence stratigraphic framework establishment method based on a high-precision chronostratigraphic body as described in Embodiment 1.

[0171] The computer-readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a computer-readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the computer-readable storage medium. Of course, the computer-readable storage medium can also be a component of the processor. The processor and the computer-readable storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the ASIC can reside in a user equipment. Of course, the processor and the computer-readable storage medium can also exist as discrete components in a communication device. Specifically, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk, etc. The storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

Claims

1. A method for establishing a sequence stratigraphic framework based on a high-precision chronostratigraphic body, characterized in that, The following steps are performed in sequence: S1, based on the well-to-well sequence correlation results and the seismic data optimization processing results, well-to-seismic comprehensive calibration is performed; In this step, single-well sequence division is realized by analyzing well logging data and core data, then well-to-well sequence correlation is performed according to the single-well sequence division, a well-to-well sequence stratigraphic correlation profile is constructed, and well-to-well sequence correlation results are obtained; In this step, by spatial scanning of the original 3D seismic data, the dip angle guide information is used as the noise vertical control trend, and the denoising processing is carried out on the post-stack data volume, to obtain the seismic data optimization processing results; S2, based on the well-to-seismic comprehensive calibration, a sequence top and bottom framework model is constructed based on seed point tracking intelligent interpretation; In this step, based on the well-to-seismic comprehensive calibration, for the top and bottom boundaries of the target layer, on the basis of the grid seed point picking, the seed point dyeing method is used to carry out intelligent horizon interpretation, and a sequence top and bottom framework model is constructed; S3, based on the sequence top and bottom framework model, a plane wave decomposition method is used for age stratigraphic body tracking, to realize high-precision multi-layer interpretation; S4, combined with the well-to-well sequence correlation results and the high-precision multi-layer interpretation, a high-precision sequence framework model is constructed, that is, a sequence stratigraphic framework based on high-precision age stratigraphic bodies is obtained.

2. The method according to claim 1, wherein, In step S1, the process of obtaining the seismic data optimization processing results is performed in the following steps in sequence: S11, calculate the dip angle volume using 3D Fourier transform In this step, by spatial scanning of the original 3D seismic data, the 3D Fourier transform formula is used to calculate the instantaneous frequency and instantaneous wave number information along the x and y directions, and then the stratigraphic dip angle information is obtained; In this process, the instantaneous frequency along the x and y directions is calculated according to formula (1) wherein, denotes the instantaneous phase, u(x, t) denotes the seismic data to be processed, t denotes time, u H (x, t) denotes the Hilbert transform of u(x, t), u and u H The derivative with respect to time can be implemented by Fourier transform; From equation (2) we obtain the instantaneous wave number k in the x direction of u(x,t) x , Further, the apparent dip angle of u(x, t) in the x direction is obtained P x = k x / ω Equation (3) Similarly, the apparent dip angle P of u(x, y, t) in the y direction is obtained as y ; Finally, based on the above calculation, the true stratigraphic dip angle θ is calculated according to the direction definition relationship; S12, seismic denoising processing based on dip angle constraint In this step, by anisotropic diffusion filtering theory, structure analysis is introduced, and the structure tensor E of the image is calculated according to formula (4) s , where u σ = G σ * u, obtained by convolving u with a Gaussian function of noise scale δ and smoothing adjustment; G s represents a Gaussian kernel function of integration scale s, used to calculate the average direction field, s > δ; represents a gradient operator; E s Eigenvalue decomposition is performed to obtain eigenvalues λ1and λ2, and their corresponding orthogonal eigenvectors ω1and ω2, ω1represents the direction with the largest gradient change, and ω2represents the direction with the smallest gradient change, i.e., the direction with consistent structure. Then, let the eigenvalue be: C1=α, λ1=λ2 formula (5) where 0 < a < 1 is used to control the diffusion strength of the direction with larger variation along the image gradient direction; d represents a consistency parameter, d 2 = (λ1- λ2) 2 ; In the isotropic region, C1=C2, the diffusion intensity along each direction is small; in the region where the image texture direction consistency is strong, the diffusion is mainly along the ω2 direction; In the anisotropic region, the vertical information is denoised along the ω1 direction, and the continuity of the horizontal seismic homodromic axis is enhanced.

3. The method according to claim 2, wherein, Step S2 includes the following steps: S21, grid seed point picking Based on well-to-seismic comprehensive calibration, select the grid profile with high signal-to-noise ratio and easy to identify for large grid interpretation, and use these grid interpretation points as seed points for data diffusion; S22, seed dyeing algorithm diffusion and propagation Starting from the waveform characteristics of a seed point, whether the pixel points around the seed point are consistent or similar to the seed point is determined according to the waveform similarity characteristics of the seismic event, to determine whether the pixel points belong to the filling area of the seed point; Based on the discard recursion principle, the seed point is expanded in 8 specific directions based on the 8-connected algorithm, to realize the propagation of the seed point in the work area, obtain the preliminary horizon points, and obtain the horizon information; S22, horizon phase adjustment under the control of waveform characteristics The phase characteristics of the horizon points are adjusted by using the waveform characteristics to realize accurate homing of the horizon points.

4. The method according to claim 3, wherein, Step S3, based on the seismic data, the optimal slope of the adjacent seismic data is predicted according to the error minimum principle guided by the "predictive imaging" theory, the seismic imaging is extracted, the seed point control tracking is adopted, and the high-precision multi-layer interpretation is realized. Specifically, the following steps are included: S31, local stratigraphic dip angle calculation According to formula (7), the local slope of the seismic event is estimated, and the local stratigraphic dip angle is calculated Wherein, p(t, x) represents the wave field; ψ represents the spatially varying slope, which depends on the transformation of t and x; Then, in the case of constant slope, the equation represented by formula (7) is simplified as p(t, x) = f(t-ψx) formula (8) Wherein, f(t) represents an arbitrary waveform; Then, in the stratigraphic dip angle scanning process of the seismic data, the variance is used for constraint according to formula (9) D(ψ) = E(ψ 2 ) - E(ψ) 2 Equation (9) Wherein, D(ψ) represents the variance of the probability distribution, and the spatially varying slope ψ is a random variable subject to the probability distribution, and E(ψ) represents the expected value of the random variable ψ; Finally, the minimum variance is calculated according to the above results; S32, dip angle propagation to extract seismic imaging The "predictive imaging" technology is adopted to realize seismic trace prediction by using the dip angle information of seismic events in different directions; a seismic trace is predicted by using a simple adjacent seismic trace recursive formula; the mth seismic trace is predicted from the first seismic trace, which is represented by formula (10) P 1,m = P m-1,m , P m-2,m-1 ,... P 2,3 , P 1,2 Equation (10) where P m-1,m denotes a prediction operator, which means predicting the mth trace from the (m-1)th trace. Finally, by analogy, the prediction of high-quality stratigraphic dip angle information of the entire seismic space is realized; S33, obtain horizon volume The seed point is used to replace the seed seismic trace in predictive imaging as the starting point and reference point of horizon tracking, and the seed seismic trace is propagated to the surrounding under the control of stratigraphic dip angle information; after all seismic traces in the whole area are tracked, a series of horizons are obtained, a horizon volume is formed, and high-precision multi-layer interpretation is realized.

5. The method according to claim 4, wherein, In step S4, on the basis of step S3, according to the stratigraphic deposition theory and geological phenomena, combined with the well-to-well sequence correlation results, more sedimentary and chronological stratigraphic meanings are given to each horizon, and the horizons are reordered, the horizon volume is converted into a high-precision sequence stratigraphic framework model, and a sequence stratigraphic framework based on a high-precision chronological stratigraphic volume is obtained.

6. An apparatus for establishing a sequence stratigraphic framework based on high-precision chronostratigraphic volume, which is used to implement the method for establishing a sequence stratigraphic framework based on high-precision chronostratigraphic volume according to any one of claims 1-5, characterized in that, The device comprises: a well-seismic comprehensive calibration module which, combined with the well-to-well sequence correlation results and the seismic data optimization processing results, performs well-seismic comprehensive calibration; a seed point tracking intelligent interpretation module which, on the basis of the well-seismic comprehensive calibration, constructs a sequence top and bottom framework model based on seed point tracking intelligent interpretation; a chronological stratigraphic volume tracking module which, based on the sequence top and bottom framework model, performs chronological stratigraphic volume tracking by using a plane wave decomposition method, and realizes high-precision multi-layer interpretation; a high-precision sequence framework model construction module which, combined with the well-to-well sequence correlation results and the high-precision multi-layer interpretation, constructs a high-precision sequence framework model.

7. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the sequence stratigraphic framework establishment method based on a high-precision chronological stratigraphic volume according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer program is stored in the computer readable storage medium, and the computer program is executed by the processor to realize the sequence stratigraphic framework establishment method based on a high-precision chronological stratigraphic volume according to any one of claims 1 to 5.

Images