Automatic surface wave inversion method and system

Abstract

The invention relates to the technical field of image processing and deep learning, in particular to an automatic surface wave inversion method and system, and the method comprises the steps: receiving original seismic information, sequentially carrying out the mean value removal, trend removal and spectrum whitening preprocessing of surface wave information, calculating a single-component noise cross-correlation function and a nine-component noise cross-correlation function based on the preprocessed data, and obtaining a single-component noise cross-correlation function and a nine-component noise cross-correlation function; and based on a time-frequency analysis method, calculating a surface wave group velocity and a velocity frequency dispersion energy diagram through the cross-correlation function, automatically extracting a frequency dispersion curve from the velocity frequency dispersion energy diagram by using an artificial intelligence frequency dispersion extraction method, screening frequency dispersion files of which effective frequency dispersion points and signal-to-noise ratios meet a preset standard, and calculating the surface wave group velocity and the velocity frequency dispersion energy diagram through the cross-correlation function. And inverting the parameters of the three-dimensional shear wave velocity model based on the underground interface constraint three-dimensional shear wave velocity inversion method to obtain a three-dimensional shear wave velocity model. The problems that in the prior art, efficiency is low, the operation process is complex, and the professional threshold is high are solved.

Description

Technical Field

[0001] This invention relates to the fields of image processing and deep learning technology, and in particular to an automated surface wave inversion method and system. Background Technology

[0002] Seismic data inversion presents challenges due to its large data volume and complex processing steps. Manual processing is time-consuming and labor-intensive, and the complex process increases the possibility of human error. These human factors can affect the accuracy and reliability of the inversion model. With the development of artificial intelligence, deep learning has achieved significant results in fields such as image recognition and speech recognition. Seismic wave inversion technology has also ushered in new development opportunities. By introducing automation and intelligent technologies, rapid processing and analysis of seismic wave data can be achieved, improving the efficiency and accuracy of inversion. Through preset algorithms and models, data processing, model construction, and parameter optimization can be completed automatically, reducing manual intervention.

[0003] However, existing manual processing methods suffer from inefficiencies due to large data volumes and complex processing steps. Furthermore, the complex procedures are prone to human error, and the experience level and fatigue of the operators directly affect the accuracy of the inversion results and the reliability of the model. In addition, the high professional threshold makes it difficult for non-professionals to quickly learn and apply the technology, limiting its widespread adoption. Summary of the Invention

[0004] The purpose of this invention is to provide an automated surface wave inversion method and system to solve the problems of low efficiency, complex operation process and high professional threshold in the existing technology.

[0005] To achieve the above objectives, the present invention provides an automated surface wave inversion method, comprising the following steps: S1. Receive the raw seismic information, and perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information in sequence. Based on the preprocessed data, calculate the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and superimpose them to generate the cross-correlation function. S2. Based on the time-frequency analysis method, the surface wave group velocity and velocity dispersion energy map are calculated through the cross-correlation function. The dispersion curve is automatically extracted from the velocity dispersion energy map using artificial intelligence dispersion extraction method. Effective dispersion points and dispersion files with signal-to-noise ratios that meet the preset standards are selected. S3. The three-dimensional shear wave velocity inversion method based on underground interface constraints is used to invert the parameters of the three-dimensional shear wave velocity model and obtain the three-dimensional shear wave velocity model.

[0006] In some embodiments of this application, in S1, the expression for calculating the single-component noise cross-correlation function is: ; in, The function is a single-component cross-correlation function, where x[n] and y[n] are both discrete-time signals. The maximum delay is set. This is the discrete-time index, corresponding to the nth sampling point of the signal. This represents the total number of sampling points.

[0007] In some embodiments of this application, in S1, the expression for calculating the nine-component noise cross-correlation function is as follows: ; in, This is a nine-component noise cross-correlation function, with components... Let be the cross-correlation function between component i and component j.

[0008] In some embodiments of this application, in S2, automatically extracting the dispersion curve from the velocity dispersion energy map using an artificial intelligence dispersion extraction method includes: Calculate the loss function : ; Here, the loss function value measures the error between the model's prediction and the true label. P(c,v,T) is the model's predicted output, a three-dimensional distribution representing the probability / energy value predicted by the model at channel c, velocity v, and period T. L(c,v,T) is the true label, with the same dimensions as the predicted value P, representing the true probability / energy value at the corresponding position. These are normalization coefficients, where n is the maximum velocity, m is the minimum velocity, and bs is the starting point of the period. This is the end point of the cycle.

[0009] In some embodiments of this application, in S2, the expression for calculating the velocity dispersion energy map is: ; in, This is the output phase velocity probability image. For period One of the curves showing the greatest change from b to e. For points, The average phase velocity energy is calculated based on the phase velocity probability image P within the period T from b to e.

[0010] In some embodiments of this application, in S2, the effective frequency dispersion and signal-to-noise ratio conforming to preset standards include: Effective dispersion points > 5 and signal-to-noise ratio > 2, and the rate of decrease of effective dispersion points after reaching the maximum velocity point. .

[0011] In some embodiments of this application, in S3, the three-dimensional shear wave velocity inversion method based on subsurface interface constraints inverts the parameters of the three-dimensional shear wave velocity model to obtain the three-dimensional shear wave velocity model, including: The surface wave travel time of a forward-modeled ray path is calculated using the following formula: ; ; in, The length of each ray path segment, This represents the number of segments along the ray path AB. Let be the slowness of the p-th segment along the ray path AB, used to construct a two-dimensional sensitivity matrix in the horizontal direction. Angular frequency, These are linear interpolation coefficients. The slowness at the k-th grid point in the horizontal direction. The number of grid points in the horizontal direction. Let ω be the total travel time of a surface wave from the starting point A to the ending point B of the path when the angular frequency is ω. Based on the surface wave travel times obtained from forward modeling and the observed surface wave travel times, and with the constraint of the subsurface interface, the 3D surface wave travel time difference is inverted. The calculation formula is: ; in, For the surface wave travel time of the observed i-th ray path, This is the surface wave travel time of the i-th ray path calculated using forward modeling; The objective function is constructed based on the surface wave travel time difference and solved using the least squares method to obtain the objective function of the three-dimensional shear wave velocity model. The calculation formula is as follows: ; in, These are the parameter changes before and after the inversion. These are the theoretical observations predicted by the model. These are actual observation data. For regularization parameters; The objective function is solved using the least squares method. After multiple iterations, the objective function reaches the convergence condition. The final parameter m is then used as the parameter of the three-dimensional shear wave velocity model, and the three-dimensional shear wave velocity model is constructed.

[0012] In some embodiments of this application, an automated surface wave inversion system is also disclosed, comprising: The cross-correlation function calculation module is used to perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information. Based on the preprocessed data, it calculates the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and then superimposes them to generate the cross-correlation function. The dispersion extraction module is used to calculate the surface wave group velocity and velocity dispersion energy map based on the time-frequency analysis method and the cross-correlation function. It uses artificial intelligence dispersion extraction method to automatically extract the dispersion curve from the velocity dispersion energy map and select effective dispersion points and dispersion files with signal-to-noise ratio that meet the preset standards. The inversion module is used to invert the parameters of the three-dimensional shear wave velocity model based on the three-dimensional shear wave velocity inversion method constrained by the subsurface interface, and obtain the three-dimensional shear wave velocity model.

[0013] The advantages and beneficial effects of this invention compared to the prior art are: 1. This invention achieves rapid processing and analysis of seismic wave data through fully automated processing, reducing manual intervention and avoiding human error and subjective factors, thereby improving inversion efficiency and the accuracy of results.

[0014] 2. This invention employs an artificial intelligence-based dispersion extraction method and sets strict dispersion data screening conditions to ensure the quality of input data, prevent accuracy loss due to human factors, and improve the reliability of the inversion model.

[0015] 3. This invention automatically completes data processing, dispersion extraction, and model construction through preset algorithms and models, enabling non-professionals to quickly get started and reducing the reliance of staff on geophysical expertise and experience, which is conducive to the promotion and application of the technology.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a flowchart of an automated surface wave inversion method according to an embodiment of the present invention; Figure 2 This is a structural diagram of an automated surface wave inversion system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the cross-correlation function extracted by the cross-correlation function calculation module in an embodiment of the present invention; Figure 4 The dispersion extraction module of this embodiment extracts the time-frequency diagram of dispersion. Figure 5 This is a cross-sectional slice obtained by the inversion module in an embodiment of the present invention. Detailed Implementation

[0018] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," and "connect" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0019] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] like Figure 1 As shown, this invention provides an automated surface wave inversion method, comprising the following steps: S1. Receive the raw seismic information, and perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information in sequence. Based on the preprocessed data, calculate the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and superimpose them to generate the cross-correlation function. S2. Based on the time-frequency analysis method, the surface wave group velocity and velocity dispersion energy map are calculated through the cross-correlation function. The dispersion curve is automatically extracted from the velocity dispersion energy map using artificial intelligence dispersion extraction method. Effective dispersion points and dispersion files with signal-to-noise ratios that meet the preset standards are selected. S3. The three-dimensional shear wave velocity inversion method based on underground interface constraints is used to invert the parameters of the three-dimensional shear wave velocity model and obtain the three-dimensional shear wave velocity model.

[0021] This invention achieves rapid processing and analysis of seismic wave data through fully automated processing, reducing manual intervention and avoiding human error and subjective factors, thereby improving inversion efficiency and the accuracy of results.

[0022] In some embodiments of this application, in S1, the expression for calculating the single-component noise cross-correlation function is: ; in, The function is a single-component cross-correlation function, where x[n] and y[n] are both discrete-time signals. The maximum delay is set. This is the discrete-time index, corresponding to the nth sampling point of the signal. This represents the total number of sampling points.

[0023] In some embodiments of this application, in S1, the expression for calculating the nine-component noise cross-correlation function is as follows: ; in, This is a nine-component noise cross-correlation function, with components... Let be the cross-correlation function between component i and component j.

[0024] In some embodiments of this application, S2 automatically extracts the loss function in dispersion. The formula for calculation is: ; Here, the loss function value measures the error between the model's prediction and the true label. P(c,v,T) is the model's predicted output, a three-dimensional distribution representing the model's predicted probability / energy value at channel c, velocity v, and period T. L(c,v,T) is the true label, with the same dimensions as the predicted value P, representing the true probability / energy value at the corresponding location. These are normalization coefficients, where n is the maximum velocity, m is the minimum velocity, and b is the normalization coefficient. s As the starting point of the cycle, e s This is the end point of the cycle.

[0025] In some embodiments of this application, in S2, the expression for calculating the velocity dispersion energy map is: ; in, This is the output phase velocity probability image. For period One of the curves showing the greatest change from b to e. For points, The average phase velocity energy is calculated based on the phase velocity probability image P within the period T from b to e.

[0026] In some embodiments of this application, in S2, the effective frequency dispersion and signal-to-noise ratio conforming to preset standards include: Effective dispersion points > 5 and signal-to-noise ratio > 2, and the rate of decrease of effective dispersion points after reaching the maximum velocity point. .

[0027] This invention employs an artificial intelligence-based dispersion extraction method and sets strict dispersion data screening conditions to ensure the quality of input data, prevent accuracy loss due to human factors, and improve the reliability of the inversion model.

[0028] In some embodiments of this application, in S3, the three-dimensional shear wave velocity inversion method based on subsurface interface constraints inverts the parameters of the three-dimensional shear wave velocity model to obtain the three-dimensional shear wave velocity model, including: The surface wave travel time of a forward-modeled ray path is calculated using the following formula: ; ; in, The length of each ray path segment, This represents the number of segments along the ray path AB. Let be the slowness of the p-th segment along the ray path AB, used to construct a two-dimensional sensitivity matrix in the horizontal direction. Angular frequency, These are linear interpolation coefficients. The slowness at the k-th grid point in the horizontal direction. The number of grid points in the horizontal direction. Let ω be the total travel time of a surface wave from the starting point A to the ending point B of the path. Based on the surface wave travel times obtained from forward modeling and the observed surface wave travel times, and with the constraint of the subsurface interface, the 3D surface wave travel time difference is inverted. The calculation formula is: ; in, For the surface wave travel time of the observed i-th ray path, This is the surface wave travel time of the i-th ray path calculated using forward modeling; The objective function is constructed based on the surface wave travel time difference and solved using the least squares method to obtain the objective function of the three-dimensional shear wave velocity model. The calculation formula is as follows: ; in, These are the parameter changes before and after the inversion. These are the theoretical observations predicted by the model. These are actual observation data. For regularization parameters; The objective function is solved using the least squares method. After multiple iterations, the objective function reaches the convergence condition. The final parameter m is then used as the parameter of the three-dimensional shear wave velocity model, and the three-dimensional shear wave velocity model is constructed.

[0029] This invention automatically completes data processing, dispersion extraction, and model building through preset algorithms and models, enabling non-professionals to quickly get started and reducing the reliance of staff on geophysical expertise and experience, which is conducive to the promotion and application of the technology.

[0030] In some embodiments of this application, an automated surface wave inversion system is also disclosed, such as Figure 2 As shown, it includes: The cross-correlation function calculation module is used to perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information. Based on the preprocessed data, it calculates the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and then superimposes them to generate the cross-correlation function. The dispersion extraction module is used to calculate the surface wave group velocity and velocity dispersion energy map based on the time-frequency analysis method and the cross-correlation function. It uses artificial intelligence dispersion extraction method to automatically extract the dispersion curve from the velocity dispersion energy map and select effective dispersion points and dispersion files with signal-to-noise ratio that meet the preset standards. The inversion module is used to invert the parameters of the three-dimensional shear wave velocity model based on the three-dimensional shear wave velocity inversion method constrained by the subsurface interface, and obtain the three-dimensional shear wave velocity model.

[0031] The following experimental verification is conducted in conjunction with the embodiments.

[0032] The cross-correlation function calculation module prepares the raw data and runs the system script. The system preprocesses the raw data, including removing the mean, trend, and spectral whitening. Using the processed data, it calculates the single / nine-component noise cross-correlation function and then superimposes them to generate the cross-correlation spectrum. After deleting the autocorrelation file, the data is input to the dispersion extraction module. The cross-correlation function extracted by the cross-correlation function calculation module is as follows: Figure 3 As shown.

[0033] The dispersion extraction module, based on time-frequency analysis, calculates the surface wave group velocity and velocity dispersion energy map using a cross-correlation function. Then, it automatically extracts the dispersion curve from the dispersion energy map using an AI-based dispersion extraction method. After obtaining the dispersion files, the module filters them, retaining only high-quality files with a signal-to-noise ratio greater than 2, more than five valid data points, and a velocity decrease after reaching its maximum velocity, ensuring the reliability of the inversion. The time-frequency map of the dispersion extracted by the dispersion extraction module is shown below. Figure 4 As shown.

[0034] The inversion module receives the filtered files and converts them into a format suitable for inversion; the 3D shear wave velocity inversion method based on subsurface interface constraints inverts the 3D shear wave velocity structural model. The cross-sectional slices obtained by the inversion module after inversion are shown below. Figure 5 As shown.

[0035] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. In case of any inconsistency, the meaning set forth in this specification or derived from the content described herein shall prevail. Furthermore, the terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. An automated surface wave inversion method, characterized in that, Includes the following steps: S1. Receive the raw seismic information, and perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information in sequence. Based on the preprocessed data, calculate the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and superimpose them to generate the cross-correlation function. S2. Based on the time-frequency analysis method, the surface wave group velocity and velocity dispersion energy map are calculated through the cross-correlation function. The dispersion curve is automatically extracted from the velocity dispersion energy map using artificial intelligence dispersion extraction method. Effective dispersion points and dispersion files with signal-to-noise ratios that meet the preset standards are selected. S3. The three-dimensional shear wave velocity inversion method based on underground interface constraints is used to invert the parameters of the three-dimensional shear wave velocity model and obtain the three-dimensional shear wave velocity model.

2. The automated surface wave inversion method according to claim 1, characterized in that, In S1, the expression for calculating the cross-correlation function of single-component noise is: ； in, The function is a single-component cross-correlation function, where x[n] and y[n] are both discrete-time signals. The maximum delay is set. This is the discrete-time index, corresponding to the nth sampling point of the signal. This represents the total number of sampling points.

3. The automated surface wave inversion method according to claim 2, characterized in that, In S1, the expression for calculating the nine-component noise cross-correlation function is as follows: ； in, This is a nine-component noise cross-correlation function, with components... Let be the cross-correlation function between component i and component j.

4. The automated surface wave inversion method according to claim 3, characterized in that, In step S2, the automatic extraction of dispersion curves from the velocity dispersion energy map using artificial intelligence dispersion extraction methods includes: Calculate the loss function : ； Here, the loss function value measures the error between the model's prediction and the true label. P(c,v,T) is the model's predicted output, a three-dimensional distribution representing the probability / energy value predicted by the model at channel c, velocity v, and period T. L(c,v,T) is the true label, with the same dimensions as the predicted value P, representing the true probability / energy value at the corresponding position. These are normalization coefficients, where n is the maximum velocity, m is the minimum velocity, and b is the normalization coefficient. s As the starting point of the cycle, This is the end point of the cycle.

5. The automated surface wave inversion method according to claim 4, characterized in that, In S2, the expression for calculating the velocity dispersion energy map is: ； in, This is the output phase velocity probability image. For period One of the curves showing the greatest change from b to e. For points, The average phase velocity energy is calculated based on the phase velocity probability image P within the period T from b to e.

6. The automated surface wave inversion method according to claim 5, characterized in that, In step S2, the effective frequency dispersion point and signal-to-noise ratio meet the preset standards, including: Effective dispersion points > 5 and signal-to-noise ratio > 2, and the rate of decrease of effective dispersion points after reaching the maximum velocity point. .

7. The automated surface wave inversion method according to claim 6, characterized in that, In step S3, the three-dimensional shear wave velocity inversion method based on subsurface interface constraints inverts the parameters of the three-dimensional shear wave velocity model, resulting in the three-dimensional shear wave velocity model including: The surface wave travel time of a forward-modeled ray path is calculated using the following formula: ； ； in, The length of each ray path segment, This represents the number of segments along the ray path AB. Let be the slowness of the p-th segment along the ray path AB, used to construct a two-dimensional sensitivity matrix in the horizontal direction. Angular frequency, These are linear interpolation coefficients. The slowness at the k-th grid point in the horizontal direction. The number of grid points in the horizontal direction. Let ω be the total travel time of a surface wave from the starting point A to the ending point B of the path when the angular frequency is ω. Based on the surface wave travel times obtained from forward modeling and the observed surface wave travel times, and with the constraint of the subsurface interface, the 3D surface wave travel time difference is inverted. The calculation formula is: ； in, For the surface wave travel time of the observed i-th ray path, This is the surface wave travel time of the i-th ray path calculated using forward modeling; The objective function is constructed based on the surface wave travel time difference and solved using the least squares method to obtain the objective function of the three-dimensional shear wave velocity model. The calculation formula is as follows: ； in, These are the parameter changes before and after the inversion. These are the theoretical observations predicted by the model. These are actual observation data. For regularization parameters; The objective function is solved using the least squares method. After multiple iterations, the objective function reaches the convergence condition. The final parameter m is then used as the parameter of the three-dimensional shear wave velocity model, and the three-dimensional shear wave velocity model is constructed.

8. An automated surface wave inversion system, characterized in that, include: The cross-correlation function calculation module is used to perform mean removal, trend removal, and spectral whitening preprocessing on the surface wave information. Based on the preprocessed data, it calculates the single-component noise cross-correlation function and the nine-component noise cross-correlation function, and then superimposes them to generate the cross-correlation function. The dispersion extraction module is used to calculate the surface wave group velocity and velocity dispersion energy map based on the time-frequency analysis method and the cross-correlation function. It uses artificial intelligence dispersion extraction method to automatically extract the dispersion curve from the velocity dispersion energy map and select effective dispersion points and dispersion files with signal-to-noise ratio that meet the preset standards. The inversion module is used to invert the parameters of the three-dimensional shear wave velocity model based on the three-dimensional shear wave velocity inversion method constrained by the subsurface interface, and obtain the three-dimensional shear wave velocity model.

Images