Seismic wave detection method and system based on underground rock formation

Abstract

The present application relates to the field of seismic exploration, and relates to a kind of seismic wave detection method based on underground rock formation, comprising: carrying out seismic survey to the region to be measured, obtain seismic data set, in turn, seismic data set is carried out seismic wave noise removal, obtain denoising seismic data set;Pixel superposition scanning is carried out to denoising seismic data set, obtain scanning seismic data set, reflection point picking is carried out to scanning seismic data set, obtain picking positioning data;Initial velocity model is established according to picking positioning data, and interpretation horizon and impedance curve are generated according to picking positioning data;According to interpretation horizon and impedance curve, initial velocity model is carried out multistage inversion, and standard velocity model is obtained;According to standard velocity model, the region to be measured is carried out tomographic imaging, and standard seismic image is obtained, and rock layer identification is carried out to standard seismic image, and underground rock formation is obtained.The present application also proposes a kind of seismic wave detection system based on underground rock formation.The present application can improve the accuracy of underground rock formation detection.

Description

Technical Field

[0001] This invention relates to the field of seismic exploration technology, and in particular to a seismic wave detection method and system based on underground rock strata. Background Technology

[0002] The Earth's underground structure is of great significance for resource exploration, geological research and land use planning. Understanding the properties, depth and structure of underground rock strata can help determine information such as the distribution of underground resources, groundwater level and seismic activity. Therefore, it is necessary to use seismic waves to explore underground rock strata.

[0003] Existing methods for detecting underground rock strata are mostly seismic wave detection methods based on the simple reflection method. This method utilizes the reflection of seismic waves between different underground rock strata to obtain information about underground structures. Earthquake sources typically generate seismic waves at the Earth's surface. These waves penetrate underground, are reflected or refracted at the interfaces of different rock strata, and are then recorded by receivers on the surface. Analyzing this seismic data can reveal the location, depth, shape, and properties of different rock strata. However, in practical applications, seismic wave detection methods based on the simple reflection method have poor geological resolution, making it difficult to obtain effective seismic structures. Furthermore, they are greatly affected by the complexity of underground rock strata and geological noise, requiring manual data analysis, which may lead to poor accuracy when detecting underground rock strata. Summary of the Invention

[0004] This invention provides a seismic wave detection method and system based on underground rock strata, the main purpose of which is to solve the problem of poor accuracy when detecting underground rock strata.

[0005] To achieve the above objectives, the present invention provides a seismic wave detection method based on underground rock strata, comprising:

[0006] Seismic measurements are performed on the area to be measured to obtain a seismic dataset. The seismic dataset is then sequentially denoised to obtain a denoised seismic dataset. This sequential denoising process includes: selecting individual seismic data points from the seismic dataset as target seismic data, and calculating the corresponding seismic coefficient set for each target seismic data using the following seismic coefficient algorithm:

[0007]

[0008] Among them, w j (a, b) refers to the coefficient value of the j-th wave coefficient in the wave coefficient group under the scale parameter a and displacement parameter b, where a is the scale parameter, b is the displacement parameter, j is the index, ∞ represents infinity, and v j(t) refers to the signal value of the j-th seismic wave in the target seismic data at time number t, where t is the time number, g() is the conjugate function symbol, e is the Euler number, i is the imaginary symbol, ω0 is the preset frequency parameter, and d is the integral symbol. The seismic wave coefficient set is filtered for noise thresholding to obtain a denoised coefficient set. The denoised coefficient set is then transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm. All the denoised seismic data are then aggregated into a denoised seismic dataset.

[0009]

[0010] in, This refers to the signal value of the j-th seismic wave in the denoised seismic data at time number t. It refers to the coefficient value of the j-th wave coefficient in the noise reduction coefficient group under the scale parameter a and the displacement parameter b.

[0011] The denoised seismic dataset is pixel-overlay scanned to obtain a scanned seismic dataset, and reflection points are picked up from the scanned seismic dataset to obtain picking and positioning data.

[0012] An initial velocity model is established based on the picked-up positioning data, and the interpretation layer and impedance curve corresponding to the picked-up positioning data are generated using the initial velocity model.

[0013] Based on the interpretation level and the impedance curve, the initial velocity model is inverted at multiple levels to obtain the standard velocity model;

[0014] Tomographic imaging of the area to be measured is performed based on the standard velocity model to obtain a standard seismic image. Rock strata are then identified from the standard seismic image to obtain the underground rock strata.

[0015] Optionally, the step of performing pixel-overlay scanning on the denoised seismic dataset to obtain a scanned seismic dataset includes:

[0016] Each denoised seismic data point in the denoised seismic dataset is selected as the target denoised seismic data, and each seismic data point in the target denoised seismic data is selected as the target seismic data point.

[0017] The target seismic data points are locally superimposed using a preset local seismic window to obtain the target local superimposed value;

[0018] The target seismic data points are obtained by pixel scanning using the target local overlay value and the seismic local window;

[0019] The target scan data points corresponding to all target seismic data points in the target denoised seismic data are aggregated into scan seismic data, and the scan seismic data corresponding to all target denoised seismic data in the denoised seismic dataset are aggregated into scan seismic dataset.

[0020] Optionally, the step of using the target local overlay value and the seismic local window to perform pixel scanning on the target seismic data points to obtain target scan data points includes:

[0021] The local stacked values ​​are stacked and tiled using the earthquake local window to obtain local tiled values;

[0022] The local scan value is obtained by subtracting the local tiling value from the local overlay value;

[0023] The target seismic data point is obtained by replacing the seismic data value with the local scan value.

[0024] Optionally, the step of picking reflection points from the scanned seismic dataset to obtain picking location data includes:

[0025] Each scanned seismic data point in the scanned seismic dataset is selected as the target scanned seismic data, and each seismic data point in the target scanned seismic data is selected as the target picking data point.

[0026] A picking window is generated based on the target picking data points, and the target picking data points within the picking window are calculated using the following extreme value picking formula:

[0027]

[0028] Where P refers to the target picked data point with coordinates (x, y, z), x refers to the horizontal axis coordinate of the target picked data point, y refers to the vertical axis coordinate of the target picked data point, z refers to the vertical axis coordinate of the target picked data point, Max is the maximum value sign, m1 refers to the horizontal half-length of the picking window, m2 refers to the vertical half-length of the picking window, and m3 refers to the vertical half-length of the picking window. This refers to the area within a picking window of size (2m1+1)(2m2+1)(2m3+1) centered on the target picking data point (x,y,z), where the coordinates are... The target picking data of the seismic data points, and The value of is in the range of (x-m1, x+m1). The value of is in the range of (y-m2, y+m2). The value of is in the range of (z-m3, z+m3);

[0029] The target picking data corresponding to all target scan seismic data in the scan seismic dataset is aggregated into picking and positioning data.

[0030] Optionally, generating the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model includes:

[0031] The initial velocity model is used to perform depth transformation on the pickup positioning data to obtain pickup depth data;

[0032] The picked depth data is subjected to layer calibration to obtain the interpreted layer;

[0033] Obtain the density data corresponding to the picking depth data, and extract the rock layer velocity data from the initial velocity model;

[0034] An impedance curve is generated based on the density data and the rock layer velocity data.

[0035] Optionally, the step of performing multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain the standard velocity model includes:

[0036] The initial velocity model is initially set using the aforementioned interpretation layer to obtain the interpretation velocity model;

[0037] Seismic forward modeling was performed using the aforementioned interpretation velocity model to obtain the analytical impedance curve;

[0038] Impedance matching is performed between the analytical impedance curve and the impedance curve to obtain the impedance difference.

[0039] The model parameters in the explained velocity model are iteratively modified based on the impedance difference to obtain the standard velocity model.

[0040] Optionally, the step of performing tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image includes:

[0041] Based on the standard velocity model, a forward simulation of the area to be measured is performed to obtain a simulated earthquake dataset.

[0042] The simulated earthquake dataset is overlaid and filtered for noise reduction to obtain corrected earthquake data;

[0043] The corrected seismic data were subjected to tomography to obtain a standard seismic image.

[0044] Optionally, the step of identifying subsurface rock layers from the standard seismic image includes:

[0045] Edge segmentation is performed on the standard seismic image to obtain a group of seismic rock strata blocks;

[0046] Multi-level feature extraction is performed on the earthquake rock strata block group to obtain rock strata feature groups;

[0047] The rock strata feature group is normalized to obtain the rock strata coding group;

[0048] The underground rock strata are obtained by identifying and mapping the rock strata coding group.

[0049] Optionally, the step of performing edge segmentation on the standard seismic image to obtain a seismic rock layer block group includes:

[0050] Edge detection is performed on the standard seismic image to obtain the primary rock layer edge group;

[0051] Connect the edge groups of the primary rock strata with line segments to obtain the outline of the primary rock strata;

[0052] Edge fitting is performed on the primary rock layer profile to obtain the secondary rock layer profile;

[0053] A rock layer contour mask is generated based on the secondary rock layer contour. The standard seismic image is then cropped using the rock layer contour mask to obtain an image of the rock layer of interest.

[0054] All images of rock strata of interest are compiled into a seismic rock strata tile group.

[0055] To address the aforementioned problems, the present invention also provides a seismic wave detection system based on underground rock strata, the system comprising:

[0056] The data denoising module is used to perform seismic measurements on the area to be measured, obtain a seismic dataset, and sequentially perform seismic wave denoising on the seismic dataset to obtain a denoised seismic dataset. The sequential seismic wave denoising of the seismic dataset to obtain the denoised seismic dataset includes: selecting seismic data points from the seismic dataset one by one as target seismic data, and calculating the seismic wave coefficient group corresponding to the target seismic data using the following seismic wave coefficient algorithm:

[0057]

[0058] Among them, w j (a, b) refers to the coefficient value of the j-th wave coefficient in the wave coefficient group under the scale parameter a and displacement parameter b, where a is the scale parameter, b is the displacement parameter, j is the index, ∞ represents infinity, and v j(t) refers to the signal value of the j-th seismic wave in the target seismic data at time number t, where t is the time number, g() is the conjugate function symbol, e is the Euler number, i is the imaginary symbol, ω0 is the preset frequency parameter, and d is the integral symbol. The seismic wave coefficient set is filtered for noise thresholding to obtain a denoised coefficient set. The denoised coefficient set is then transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm. All the denoised seismic data are then aggregated into a denoised seismic dataset.

[0059]

[0060] in, This refers to the signal value of the j-th seismic wave in the denoised seismic data at time number t. It refers to the coefficient value of the j-th wave coefficient in the noise reduction coefficient group under the scale parameter a and the displacement parameter b.

[0061] The reflection picking module is used to perform pixel overlay scanning on the noise-reduced seismic dataset to obtain a scanned seismic dataset, and to pick up reflection points on the scanned seismic dataset to obtain picking and positioning data.

[0062] The model building module is used to establish an initial velocity model based on the picked-up positioning data, and to generate the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model;

[0063] The model training module is used to perform multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain a standard velocity model.

[0064] The rock strata identification module is used to perform tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image, and to identify the underground rock strata on the standard seismic image.

[0065] This invention provides a denoised seismic dataset by sequentially performing seismic wave denoising on the seismic dataset. This process suppresses high-frequency noise in the seismic data, preserves the seismic signal, and improves the time and frequency domain resolution of the seismic signal. It also allows for better processing of the sublinear features in the signal, thus improving the accuracy of the seismic data. By performing pixel-overlay scanning on the denoised seismic dataset, clutter data in the seismic data can be effectively filtered out, achieving feature enhancement. By picking reflection points from the scanned seismic dataset, picking location data can be obtained, which can further extract extreme points in the seismic data, enhancing the texture features of the seismic data. By establishing an initial velocity model based on the picking location data, and using the initial velocity model to generate the interpretation horizon and impedance curve corresponding to the picking location data, the interpretation horizon and impedance curve of the area to be measured can be simulated using the initial velocity model and the picking location data, thereby facilitating subsequent training and updating of the initial velocity model.

[0066] By performing multi-level inversion on the initial velocity model based on the interpreted stratigraphic level and the impedance curve, the velocity model can be iteratively updated by combining the impedance curves of the actual seismic data and the predicted impedance curves. This allows the velocity model to more accurately reflect the geological structure information of the area under test, thereby improving the accuracy of subsequent rock strata identification. By performing tomographic imaging on the area under test based on the standard velocity model, standard seismic images are obtained. Rock strata identification is then performed on these standard seismic images to identify the underground rock strata. Machine learning algorithms can be used to segment and identify the rock strata in the obtained standard seismic images, improving the efficiency of rock strata detection and the accuracy of rock strata classification. Therefore, the seismic wave detection method and system based on underground rock strata proposed in this invention can solve the problem of poor accuracy in detecting underground rock strata. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating a seismic wave detection method based on underground rock strata according to an embodiment of the present invention.

[0068] Figure 2 This is a schematic diagram of the process for generating a scanned seismic dataset according to an embodiment of the present invention;

[0069] Figure 3 This is a schematic diagram of the process for extracting and interpreting layer positions and impedance curves according to an embodiment of the present invention;

[0070] Figure 4 A functional block diagram of a seismic wave detection system based on underground rock strata provided in an embodiment of the present invention;

[0071] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0072] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0073] This application provides a seismic wave detection method based on underground rock strata. The execution entity of the seismic wave detection method based on underground rock strata includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the seismic wave detection method based on underground rock strata can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0074] Reference Figure 1 The diagram shown is a flowchart illustrating a seismic wave detection method based on underground rock strata according to an embodiment of the present invention. In this embodiment, the seismic wave detection method based on underground rock strata includes:

[0075] S1. Perform seismic measurements on the area to be measured to obtain a seismic dataset. Then, perform seismic wave denoising on the seismic dataset to obtain a denoised seismic dataset.

[0076] In this embodiment of the invention, each earthquake data in the earthquake dataset refers to earthquake data obtained by conducting a seismic wave detection on the area to be measured. The earthquake data refers to the reflection record data of earthquake waves, which can be seismic wave data represented using time profiles or depth profiles.

[0077] In this embodiment of the invention, the step of sequentially performing seismic wave denoising on the seismic dataset to obtain a denoised seismic dataset includes:

[0078] Seismic data points are selected one by one from the earthquake dataset as target seismic data, and the seismic wave coefficient set corresponding to the target seismic data is calculated using the following seismic wave coefficient algorithm:

[0079]

[0080] Among them, w j (a, b) refers to the coefficient value of the j-th wave coefficient in the wave coefficient group under the scale parameter a and displacement parameter b, where a is the scale parameter, b is the displacement parameter, j is the index, ∞ represents infinity, and vj (t) refers to the signal value of the j-th seismic wave in the target seismic data at time number t, where t is the time number, g() is the conjugate function symbol, e is the Euler number, i is the imaginary number symbol, ω0 is the preset frequency parameter, and d is the integral symbol.

[0081] The vibration wave coefficient set is subjected to noise threshold filtering to obtain a noise reduction coefficient set;

[0082] The denoised coefficient set is transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm, and all the denoised seismic data are then aggregated into a denoised seismic dataset:

[0083]

[0084] in, This refers to the signal value of the j-th seismic wave in the denoised seismic data at time number t. It refers to the coefficient value of the j-th wave coefficient in the noise reduction coefficient group under the scale parameter a and displacement parameter b.

[0085] In detail, the step of performing noise threshold filtering on the shock wave coefficient group to obtain a noise reduction coefficient group means removing coefficients in the shock wave coefficient group that are greater than a preset noise threshold from the shock wave coefficient group to obtain a noise reduction coefficient group.

[0086] In detail, by using the seismic wave coefficient algorithm to calculate the seismic wave coefficient set corresponding to the target seismic data, multi-scale analysis of the target seismic data can be achieved, frequency noise can be suppressed, and the time-domain and frequency-domain characteristics of the signal can be located more accurately, thereby improving the signal resolution.

[0087] In this embodiment of the invention, by sequentially performing seismic wave denoising on the seismic dataset, a denoised seismic dataset is obtained. This can suppress high-frequency noise in the seismic data, preserve the seismic signal, and improve the time and frequency domain resolution of the seismic signal, allowing for better processing of the sublinear features in the signal and improving the accuracy of the seismic data.

[0088] S2. Perform pixel overlay scanning on the noise-reduced seismic dataset to obtain a scanned seismic dataset. Then, pick up the reflection points on the scanned seismic dataset to obtain the pickup and positioning data.

[0089] In this embodiment of the invention, each scanned seismic data in the scanned seismic dataset corresponds to a seismic profile image of each denoised seismic data in the denoised seismic dataset after stacking.

[0090] In this embodiment of the invention, reference is made to Figure 2 As shown, the step of performing pixel-overlay scanning on the denoised seismic dataset to obtain a scanned seismic dataset includes:

[0091] S21. Select the denoised seismic data in the denoised seismic dataset one by one as the target denoised seismic data, and select the seismic data points in the target denoised seismic data one by one as the target seismic data points;

[0092] S22. Using a preset local seismic window, the target seismic data points are locally superimposed to obtain the target local superimposed value;

[0093] S23. Using the target local overlay value and the seismic local window, perform pixel scanning on the target seismic data points to obtain target scan data points;

[0094] S24. Collect all target scan data points corresponding to all target seismic data points in the target denoised seismic data into scan seismic data, and collect all scan seismic data corresponding to all target denoised seismic data in the denoised seismic dataset into scan seismic dataset.

[0095] In detail, the seismic data points refer to the pixels in the imaging data of the target denoised seismic data. These pixels represent discrete regions of the subsurface structure, similar to pixels in an image. Seismic data points can be understood as a discrete spatial and depth grid used to describe changes in the subsurface structure.

[0096] Specifically, the step of using a preset local seismic window to locally overlay the target seismic data points to obtain the target local overlay value means using the local seismic window to obtain the absolute average value of the seismic data points of all seismic data points with the target seismic data point as the midpoint, and using the absolute average value as the target local overlay value. The local seismic window is a time-domain window.

[0097] Specifically, the step of performing pixel scanning on the target seismic data points using the target local overlay value and the seismic local window to obtain target scan data points includes:

[0098] The local stacked values ​​are stacked and tiled using the earthquake local window to obtain local tiled values;

[0099] The local scan value is obtained by subtracting the local tiling value from the local overlay value;

[0100] The target seismic data point is obtained by replacing the seismic data value with the local scan value.

[0101] In this embodiment of the invention, the step of overlaying and tiling the local seismic window based on the local overlay value means replacing the seismic data values ​​of all seismic data points with the target seismic data point as the center point with the local overlay value to obtain the overlay local window.

[0102] Specifically, the step of picking reflection points from the scanned seismic dataset to obtain picking and positioning data includes:

[0103] Each scanned seismic data point in the scanned seismic dataset is selected as the target scanned seismic data, and each seismic data point in the target scanned seismic data is selected as the target picking data point.

[0104] A picking window is generated based on the target picking data points, and the target picking data points within the picking window are calculated using the following extreme value picking formula:

[0105]

[0106] Where P refers to the target picked data point with coordinates (x, y, z), x refers to the horizontal axis coordinate of the target picked data point, y refers to the vertical axis coordinate of the target picked data point, z refers to the vertical axis coordinate of the target picked data point, Max is the maximum value sign, m1 refers to the horizontal half-length of the picking window, m2 refers to the vertical half-length of the picking window, and m3 refers to the vertical half-length of the picking window. This refers to the area within a picking window of size (2m1+1)(2m2+1)(2m3+1) centered on the target picking data point (x,y,z), where the coordinates are... The target picking data of the seismic data points, and The value of is in the range of (x-m1, x+m1). The value of is in the range of (y-m2, y+m2). The value of is in the range of (z-m3, z+m3);

[0107] The target picking data corresponding to all target scan seismic data in the scan seismic dataset is aggregated into picking and positioning data.

[0108] In detail, by using the extreme value picking formula to calculate the target picking data point within the picking window, extreme value picking of the reflection point can be achieved, enhancing the seismic wave texture of the seismic data, thereby improving the accuracy of subsequent rock strata detection.

[0109] In this embodiment of the invention, by performing pixel-overlay scanning on the denoised seismic dataset, clutter data in the seismic data can be effectively filtered out, achieving feature enhancement. By picking up reflection points on the scanned seismic dataset to obtain picking and positioning data, extreme points in the seismic data can be further extracted, enhancing the texture features of the seismic data.

[0110] S3. Establish an initial velocity model based on the picked-up positioning data, and use the initial velocity model to generate the interpretation layer and impedance curve corresponding to the picked-up positioning data.

[0111] In this embodiment of the invention, the initial velocity model is used to represent the spatial distribution of velocity in the underground medium of the area to be measured. It is usually represented in the form of a grid or depth layer. Each grid cell or depth layer contains velocity values. The initial velocity model is used to describe the velocity distribution of different underground rock layers and can be used to simulate seismic wave propagation, generate seismic profile images, and perform underground structure analysis.

[0112] In this embodiment of the invention, establishing an initial velocity model based on the picked-up positioning data refers to establishing a relationship between velocity and depth based on the depth information in the picked-up positioning data, constructing a model framework by selecting a linear or exponential function, and replacing the parameters of the constructed model framework based on the picked-up positioning data to obtain the initial velocity model.

[0113] In this embodiment of the invention, reference is made to Figure 3 As shown, the step of generating the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model includes:

[0114] S31. The initial velocity model is used to perform depth transformation on the pickup positioning data to obtain pickup depth data;

[0115] S32. Perform layer calibration on the picked depth data to obtain the interpretation layer;

[0116] S33. Obtain the density data corresponding to the picking depth data, and extract the rock layer velocity data from the initial velocity model;

[0117] S34. Generate an impedance curve based on the density data and the rock layer velocity data.

[0118] In detail, the picked depth data can be stratified according to geological information such as preset geological profiles, rock layer descriptions or drilling data to obtain the interpreted strata. The density data refers to the density information of different rock layers. The generation of impedance curves based on the density data and the rock layer velocity data refers to multiplying the density data and the rock layer velocity data to obtain impedance values, and then stitching all the impedance values ​​together to form an impedance curve.

[0119] In this embodiment of the invention, by establishing an initial velocity model based on the picked-up positioning data, and using the initial velocity model to generate the interpretation layer and impedance curve corresponding to the picked-up positioning data, the interpretation layer and impedance curve of the area to be tested can be simulated using the initial velocity model and the picked-up positioning data, thereby facilitating subsequent training and updating of the initial velocity model.

[0120] S4. Perform multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain the standard velocity model.

[0121] In this embodiment of the invention, the standard velocity model is an iterative model of the initial velocity model. The standard velocity model can converge the initial velocity model to a more accurate underground structure, thereby providing a more accurate description of the rock strata.

[0122] In this embodiment of the invention, the step of performing multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain the standard velocity model includes:

[0123] The initial velocity model is initially set using the aforementioned interpretation layer to obtain the interpretation velocity model;

[0124] Seismic forward modeling was performed using the aforementioned interpretation velocity model to obtain the analytical impedance curve;

[0125] Impedance matching is performed between the analytical impedance curve and the impedance curve to obtain the impedance difference.

[0126] The model parameters in the explained velocity model are iteratively modified based on the impedance difference to obtain the standard velocity model.

[0127] In detail, the initial setting of the initial velocity model using the interpretation layer refers to constraining the initial setting of the initial velocity model using the interpretation layer, thus limiting the search space for inversion; the model parameters in the interpretation velocity model can be iteratively modified using a full waveform inversion algorithm or a quasi-Newton algorithm based on the impedance difference to obtain a standard velocity model.

[0128] In this embodiment of the invention, by performing multi-level inversion on the initial velocity model based on the interpreted horizon and the impedance curve, the velocity model can be iteratively updated by combining the impedance curve of the actual seismic data and the predicted impedance curve. This allows the velocity model to more accurately reflect the geological structure information of the area under test, thereby improving the accuracy of subsequent rock strata identification.

[0129] S5. Perform tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image, and perform rock layer identification on the standard seismic image to obtain the underground rock layers.

[0130] In this embodiment of the invention, the standard seismic image refers to the seismic data profile image of the area to be measured after noise reduction, inversion and other operations.

[0131] In this embodiment of the invention, the step of performing tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image includes:

[0132] Based on the standard velocity model, a forward simulation of the area to be measured is performed to obtain a simulated earthquake dataset.

[0133] The simulated earthquake dataset is overlaid and filtered for noise reduction to obtain corrected earthquake data;

[0134] The corrected seismic data were subjected to tomography to obtain a standard seismic image.

[0135] In detail, the forward simulation is a geophysical modeling method used to simulate how seismic waves propagate from underground sources to the Earth's surface and their interaction with underground rock strata during propagation. The corrected seismic data can be tomographically analyzed using reverse time migration or inverse scattering imaging algorithms to obtain standard seismic images.

[0136] Specifically, the process of identifying subsurface rock layers from the standard seismic image includes:

[0137] Edge segmentation is performed on the standard seismic image to obtain a group of seismic rock strata blocks;

[0138] Multi-level feature extraction is performed on the earthquake rock strata block group to obtain rock strata feature groups;

[0139] The rock strata feature group is normalized to obtain the rock strata coding group;

[0140] The underground rock strata are obtained by identifying and mapping the rock strata coding group.

[0141] In detail, multi-level feature extraction can be performed on the seismic rock strata block group using algorithms such as gray-level co-occurrence matrix method, Gabor filtering algorithm, and gray-level histogram to obtain rock strata feature group. The rock strata feature group can be normalized using fully connected layer or global pooling layer to obtain rock strata coding group. The rock strata coding group can be identified and mapped using a preset rock strata coding library to obtain underground rock strata. The rock strata coding library contains the correspondence between various rock strata codes and rock strata types. The rock strata coding library is trained by a preset convolutional neural network model.

[0142] Specifically, the step of performing edge segmentation on the standard seismic image to obtain a seismic rock layer block group includes:

[0143] Edge detection is performed on the standard seismic image to obtain the primary rock layer edge group;

[0144] Connect the edge groups of the primary rock strata with line segments to obtain the outline of the primary rock strata;

[0145] Edge fitting is performed on the primary rock layer profile to obtain the secondary rock layer profile;

[0146] A rock layer contour mask is generated based on the secondary rock layer contour. The standard seismic image is then cropped using the rock layer contour mask to obtain an image of the rock layer of interest.

[0147] All images of rock strata of interest are compiled into a seismic rock strata tile group.

[0148] In detail, the standard seismic image can be edge-detected using threshold segmentation, edge segmentation, or watershed segmentation algorithms to obtain primary rock layer edge groups. The primary rock layer contours can be edge-fitted using the Hough transform algorithm or active contour model to obtain secondary rock layer contours.

[0149] In this embodiment of the invention, a standard seismic image is obtained by performing tomographic imaging on the area to be measured according to the standard velocity model. The rock strata are then identified from the standard seismic image to obtain the underground rock strata. Machine learning algorithms can be used to segment and identify the rock strata in the obtained standard seismic image, which improves the efficiency of rock strata detection and the accuracy of rock strata classification.

[0150] This invention provides a denoised seismic dataset by sequentially performing seismic wave denoising on the seismic dataset. This process suppresses high-frequency noise in the seismic data, preserves the seismic signal, and improves the time and frequency domain resolution of the seismic signal. It also allows for better processing of the sublinear features in the signal, thus improving the accuracy of the seismic data. By performing pixel-overlay scanning on the denoised seismic dataset, clutter data in the seismic data can be effectively filtered out, achieving feature enhancement. By picking reflection points from the scanned seismic dataset, picking location data can be obtained, which can further extract extreme points in the seismic data, enhancing the texture features of the seismic data. By establishing an initial velocity model based on the picking location data, and using the initial velocity model to generate the interpretation horizon and impedance curve corresponding to the picking location data, the interpretation horizon and impedance curve of the area to be measured can be simulated using the initial velocity model and the picking location data, thereby facilitating subsequent training and updating of the initial velocity model.

[0151] By performing multi-level inversion on the initial velocity model based on the interpreted stratigraphic level and the impedance curve, the velocity model can be iteratively updated by combining the impedance curves of the actual seismic data and the predicted impedance curves. This allows the velocity model to more accurately reflect the geological structure information of the area under test, thereby improving the accuracy of subsequent rock strata identification. By performing tomographic imaging on the area under test based on the standard velocity model, standard seismic images are obtained. Rock strata identification is then performed on these standard seismic images to identify the underground rock strata. Machine learning algorithms can be used to segment and identify the rock strata in the obtained standard seismic images, improving the efficiency of rock strata detection and the accuracy of rock strata classification. Therefore, the seismic wave detection method based on underground rock strata proposed in this invention can solve the problem of poor accuracy in detecting underground rock strata.

[0152] like Figure 4 The diagram shown is a functional block diagram of a seismic wave detection system based on underground rock strata provided in an embodiment of the present invention.

[0153] The seismic wave detection system 100 based on underground rock strata described in this invention can be installed in an electronic device. Depending on the functions implemented, the seismic wave detection system 100 may include a data noise reduction module 101, a reflection acquisition module 102, a model building module 103, a model training module 104, and a rock strata identification module 105. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0154] In this embodiment, the functions of each module / unit are as follows:

[0155] The data denoising module 101 is used to perform seismic measurements on the area to be measured, obtain a seismic dataset, and sequentially perform seismic wave denoising on the seismic dataset to obtain a denoised seismic dataset. The sequential seismic wave denoising of the seismic dataset to obtain the denoised seismic dataset includes: selecting seismic data from the seismic dataset one by one as target seismic data, and calculating the seismic wave coefficient group corresponding to the target seismic data using the following seismic wave coefficient algorithm:

[0156]

[0157] Among them, w j (a, b) refers to the coefficient value of the j-th wave coefficient in the wave coefficient group under the scale parameter a and displacement parameter b, where a is the scale parameter, b is the displacement parameter, j is the index, ∞ represents infinity, and v j(t) refers to the signal value of the j-th seismic wave in the target seismic data at time number t, where t is the time number, g() is the conjugate function symbol, e is the Euler number, i is the imaginary symbol, ω0 is the preset frequency parameter, and d is the integral symbol. The seismic wave coefficient set is filtered for noise thresholding to obtain a denoised coefficient set. The denoised coefficient set is then transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm. All the denoised seismic data are then aggregated into a denoised seismic dataset.

[0158]

[0159] in, This refers to the signal value of the j-th seismic wave in the denoised seismic data at time number t. It refers to the coefficient value of the j-th wave coefficient in the noise reduction coefficient group under the scale parameter a and the displacement parameter b.

[0160] The reflection picking module 102 is used to perform pixel overlay scanning on the noise-reduced seismic dataset to obtain a scanned seismic dataset, and to pick up reflection points on the scanned seismic dataset to obtain picking and positioning data.

[0161] The model building module 103 is used to establish an initial velocity model based on the picked-up positioning data, and to generate the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model.

[0162] The model training module 104 is used to perform multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain a standard velocity model.

[0163] The rock strata identification module 105 is used to perform tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image, and to perform rock strata identification on the standard seismic image to obtain the underground rock strata.

[0164] In detail, the modules of the seismic wave detection system 100 based on underground rock strata described in this embodiment of the invention employ the same methods as described above. Figures 1 to 3 The method uses the same techniques as the seismic wave detection method based on underground rock strata described above, and can produce the same technical effects, so it will not be repeated here.

[0165] In the several embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0166] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0168] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0169] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0170] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems described in the system embodiments may also be implemented by a single unit or system through software or hardware. Terms such as "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. 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 be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A seismic wave detection method based on underground rock strata, characterized in that, The method includes: Seismic measurements are performed on the area to be measured to obtain a seismic dataset. The seismic dataset is then sequentially denoised to obtain a denoised seismic dataset. This sequential denoising process includes: selecting individual seismic data points from the seismic dataset as target seismic data, and calculating the corresponding seismic coefficient set for each target seismic data using the following seismic coefficient algorithm: ； in, It refers to the first in the group of seismic wave coefficients. Individual wave coefficients in scale parameters Displacement parameters The coefficient value below, For scale parameters, For displacement parameters, For serial number, The symbol for infinity. This refers to the first [number] seismic data in the target seismic data. The seismic waves are in time sequence number The signal value below, It is a time sequence number. It is the symbol for the conjugate function. It is the Euler number. It is the symbol for an imaginary number. These are preset frequency parameters. The integral sign is used; noise threshold filtering is applied to the seismic wave coefficient set to obtain a denoised coefficient set; the denoised coefficient set is transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm, and all the denoised seismic data are aggregated into a denoised seismic dataset: ； in, This refers to the first [number] in the noise-reduced seismic data. The seismic waves are in time sequence number The signal value below, It refers to the first in the noise reduction coefficient group. Individual wave coefficients in scale parameters Displacement parameters The coefficient value below; The denoised seismic dataset is subjected to pixel overlay scanning to obtain a scanned seismic dataset. Reflection points are picked up from the scanned seismic dataset to obtain picking and positioning data. The denoised seismic dataset is subjected to pixel-overlay scanning to obtain a scanned seismic dataset, including: Each denoised seismic data point in the denoised seismic dataset is selected as the target denoised seismic data, and each seismic data point in the target denoised seismic data is selected as the target seismic data point. The target seismic data points are locally superimposed using a preset local seismic window to obtain the target local superimposed value; The target seismic data points are obtained by pixel scanning using the target local overlay value and the seismic local window; The target scan data points corresponding to all target seismic data points in the target denoised seismic data are aggregated into scan seismic data, and the scan seismic data corresponding to all target denoised seismic data in the denoised seismic dataset are aggregated into scan seismic dataset. The step of performing pixel scanning on the target seismic data points using the target local overlay value and the seismic local window to obtain target scan data points includes: The local stacked values ​​of the target are averaged and tiled using the earthquake local window to obtain the local tiled values; The local scan value is obtained by subtracting the local tiled value from all the original seismic data within the local seismic window; The target seismic data point is obtained by replacing the seismic data value with the local scan value. An initial velocity model is established based on the picked-up positioning data, and the interpretation layer and impedance curve corresponding to the picked-up positioning data are generated using the initial velocity model. Based on the interpretation level and the impedance curve, the initial velocity model is inverted at multiple levels to obtain the standard velocity model; Tomographic imaging of the area to be measured is performed based on the standard velocity model to obtain a standard seismic image. Rock strata are then identified from the standard seismic image to obtain the underground rock strata.

2. The seismic wave detection method based on underground rock strata as described in claim 1, characterized in that, The step of picking reflection points from the scanned seismic dataset to obtain picking and positioning data includes: Each scanned seismic data point in the scanned seismic dataset is selected as the target scanned seismic data, and each seismic data point in the target scanned seismic data is selected as the target picking data point. A picking window is generated based on the target picking data points, and the target picking data points within the picking window are calculated using the following extreme value picking formula: ； in, This refers to the coordinates of the target data point being picked up. The target picking data described below, This refers to the x-axis coordinate of the target data point. This refers to the vertical axis coordinate of the target data point. This refers to the ordinate of the target data point. It is the sign of the maximum value. This refers to the horizontal axis half-length of the pickup window. This refers to the vertical axis half-length of the pickup window. This refers to the vertical axis of the pickup window, which is half the length of the pickup window. This refers to picking data points with the target. The size of the center picking window is The coordinates within the range are The target picking data of the seismic data points, and The range of values ​​is within between, The range of values ​​is within between, The range of values ​​is within between; The target picking data corresponding to all target scan seismic data in the scan seismic dataset is aggregated into picking and positioning data.

3. The seismic wave detection method based on underground rock strata as described in claim 1, characterized in that, The step of generating the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model includes: The initial velocity model is used to perform depth transformation on the pickup positioning data to obtain pickup depth data; The picked depth data is subjected to layer calibration to obtain the interpreted layer; Obtain the density data corresponding to the picking depth data, and extract the rock layer velocity data from the initial velocity model; An impedance curve is generated based on the density data and the rock layer velocity data.

4. The seismic wave detection method based on underground rock strata as described in claim 1, characterized in that, The step of performing multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain the standard velocity model includes: The initial velocity model is initially set using the aforementioned interpretation layer to obtain the interpretation velocity model; Seismic forward modeling was performed using the aforementioned interpretation velocity model to obtain the analytical impedance curve; Impedance matching is performed between the analytical impedance curve and the impedance curve to obtain the impedance difference; The model parameters in the explained velocity model are iteratively modified based on the impedance difference to obtain the standard velocity model.

5. The seismic wave detection method based on underground rock strata as described in claim 1, characterized in that, The step of performing tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image includes: Based on the standard velocity model, a forward simulation of the area to be measured is performed to obtain a simulated earthquake dataset. The simulated earthquake dataset is overlaid and filtered for noise reduction to obtain corrected earthquake data; The corrected seismic data were subjected to tomography to obtain a standard seismic image.

6. The seismic wave detection method based on underground rock strata as described in claim 1, characterized in that, The process of identifying underground rock strata from the standard seismic image includes: Edge segmentation is performed on the standard seismic image to obtain a seismic rock layer block group; Multi-level feature extraction is performed on the earthquake rock strata block group to obtain rock strata feature groups; The rock strata feature group is normalized to obtain the rock strata coding group; The underground rock strata are obtained by identifying and mapping the rock strata coding group.

7. The seismic wave detection method based on underground rock strata as described in claim 6, characterized in that, The process of edge segmentation of the standard seismic image to obtain a seismic rock layer block group includes: Edge detection is performed on the standard seismic image to obtain the primary rock layer edge group; Connect the edge groups of the primary rock strata with line segments to obtain the outline of the primary rock strata; Edge fitting is performed on the primary rock layer profile to obtain the secondary rock layer profile; A rock layer contour mask is generated based on the secondary rock layer contour. The standard seismic image is then cropped using the rock layer contour mask to obtain an image of the rock layer of interest. All images of rock strata of interest are compiled into a seismic rock strata tile group.

8. A seismic wave detection system based on underground rock strata, characterized in that, The system includes: The data denoising module is used to perform seismic measurements on the area to be measured, obtain a seismic dataset, and sequentially perform seismic wave denoising on the seismic dataset to obtain a denoised seismic dataset. The sequential seismic wave denoising of the seismic dataset to obtain the denoised seismic dataset includes: selecting seismic data points from the seismic dataset one by one as target seismic data, and calculating the seismic wave coefficient group corresponding to the target seismic data using the following seismic wave coefficient algorithm: ； in, It refers to the first in the group of seismic wave coefficients. Individual wave coefficients in scale parameters Displacement parameters The coefficient value below, For scale parameters, For displacement parameters, For serial number, The symbol for infinity. This refers to the first [number] seismic data in the target seismic data. The seismic waves are in time sequence number The signal value below, It is a time sequence number. It is the symbol for the conjugate function. It is the Euler number. It is the symbol for an imaginary number. These are preset frequency parameters. The integral sign is used; noise threshold filtering is applied to the seismic wave coefficient set to obtain a denoised coefficient set; the denoised coefficient set is transformed into denoised seismic data using the following inverse seismic wave coefficient algorithm, and all the denoised seismic data are aggregated into a denoised seismic dataset: ； in, This refers to the first [number] in the noise-reduced seismic data. The seismic waves are in time sequence number The signal value below, It refers to the first in the noise reduction coefficient group. Individual wave coefficients in scale parameters Displacement parameters The coefficient value below; The reflection picking module is used to perform local window overlay scanning processing on the noise-reduced seismic dataset to obtain a scanned seismic dataset, and to pick up reflection points on the scanned seismic dataset to obtain picking and positioning data. The model building module is used to establish an initial velocity model based on the picked-up positioning data, and to generate the interpretation layer and impedance curve corresponding to the picked-up positioning data using the initial velocity model. The model training module is used to perform multi-level inversion on the initial velocity model based on the interpretation level and the impedance curve to obtain a standard velocity model. The rock strata identification module is used to perform tomographic imaging on the area to be measured according to the standard velocity model to obtain a standard seismic image, and to identify the underground rock strata on the standard seismic image.

Images