Seismic data curvature picking method and device, computer device and storage medium

Abstract

The present application provides a kind of seismic data curvature picking method, device, computer equipment and storage medium, the method includes obtaining the stacked seismic data body as sample label data set, wherein sample label data set records the reflection waveform position information;According to the seed point information of zero offset position in sample label data set determined by reflection waveform position information, seed point set is obtained based on the seed point information of zero offset position;From sample label data set, extract the curvature curve of seismic data, and obtain the curvature curve label data set;Obtain the stacked seismic data body, extract the similar spectrum data from the stacked seismic data body according to the residual curvature range, establish the corresponding relationship between similar spectrum data and seed point set, and obtain the similar spectrum data label data set;Curvature curve label data set and similar spectrum data label data set are input into neural network for learning, and the neural network model for seismic data curvature picking is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of seismic exploration, and in particular to a seismic data curvature picking method and device, computer equipment and a storage medium. BACKGROUND

[0002] In the past decade, seismic exploration has completely entered the depth domain processing and interpretation field in terms of both theory and actual industrial processing flow, which is particularly important for depth domain processing of seismic data. For depth domain processing, the most important processing link is depth domain velocity modeling and depth domain migration imaging. Traditional depth domain velocity modeling is usually divided into two steps of near-surface velocity modeling and middle-deep layer velocity modeling. The middle-deep layer velocity modeling includes six steps of residual curvature calculation, dip scanning, seed point picking, residual curvature picking, ray tracing matrix building and matrix solving.

[0003] The specific steps of the traditional depth domain velocity modeling are as follows: first, prestack depth migration is performed by using an initial velocity model to obtain a migration gather and a profile; then, residual depth difference and reflection dip are picked up from the imaging gather to establish a tomography equation; then, a large sparse equation set is solved by using an iterative method through ray tracing to obtain a velocity correction amount; then, the initial velocity model is updated to perform a new round of iteration to obtain a high-precision velocity model. The picking of residual curvature affects the accurate estimation of residual time difference, and the accuracy of residual curvature picking greatly determines the accuracy of the tomography matrix built by ray tracing, which affects the accuracy of the depth domain velocity modeling of seismic exploration and the success of the current seismic exploration processing.

[0004] In the prior art, the residual curvature picking technology mainly depends on the relative relationship between the peaks and troughs of the seismic data phase axis to determine, but in a complex exploration area, the residual curvature picking is not in place and the accuracy of residual curvature picking is low in the case of phase axis intersection or weak seismic data energy. SUMMARY

[0005] Therefore, it is necessary to provide a seismic data curvature picking method, device, computer equipment and storage medium aiming at the above technical problems.

[0006] A seismic data curvature picking method comprises the following steps:

[0007] Obtain a prestacked seismic data volume as sample label data set, wherein the sample label data set records reflection waveform position information;

[0008] Determine seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and obtain a seed point set based on the seed point information of the zero offset position;

[0009] Extracting a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set;

[0010] Acquire a post-stack seismic data volume, extract similar spectrum data from the post-stack seismic data volume according to a residual curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set;

[0011] The curvature curve label data set and the similar spectrum data label data set are input into a neural network for learning to obtain a neural network model for seismic data curvature picking.

[0012] In one embodiment, the step of obtaining a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records the reflection waveform position information, comprises:

[0013] Obtain pre-stack seismic data volume;

[0014] Slicing the pre-stack seismic data volume according to the number of seismic traces and the number of sampling points to obtain a sliced ​​data volume;

[0015] A zero-offset seed point in the slice data volume is obtained along the sampling point direction in the 0th seismic trace direction, and a time-distance curve is produced using random velocity perturbation at a non-zero-offset seed point to generate the sample label data set.

[0016] In one embodiment, the step of obtaining a pre-stack seismic data volume as a sample label data set includes:

[0017] Acquire a pre-stack seismic data volume as the sample label data set;

[0018] Preprocessing the sample label data set to obtain a processed sample label data set;

[0019] The step of extracting the seismic data curvature curve from the sample label data set to obtain the curvature curve label data set includes:

[0020] The seismic data curvature curve is extracted from the processed sample label data set to obtain a curvature curve label data set.

[0021] In one embodiment, the step of preprocessing the sample label dataset to obtain a processed sample label dataset includes:

[0022] An amplitude value distribution range of a pre-stack seismic data volume is identified, and a truncation operation is performed on the pre-stack seismic data volume based on the amplitude value distribution range.

[0023] In one of the embodiments, the step of identifying the amplitude value distribution range of the stacked seismic data volume comprises:

[0024] The truncated seismic data volume is filtered and gray processed and normalized to obtain the processed sample label data set.

[0025] In one of the embodiments, the step of obtaining the stacked seismic data volume further comprises:

[0026] The remaining curvature range is determined according to the work area.

[0027] A seismic data curvature picking device comprises:

[0028] A sample label data set acquisition module is configured to acquire a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information.

[0029] A seed point set acquisition module is configured to determine seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and acquire a seed point set based on the seed point information of the zero offset position.

[0030] A curvature curve acquisition module is configured to extract a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set.

[0031] A similar spectrum data acquisition module is configured to acquire a stacked seismic data volume, extract similar spectrum data from the stacked seismic data volume according to a remaining curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set.

[0032] A model training module is configured to input the curvature curve label data set and the similar spectrum data label data set into a neural network for learning to obtain a neural network model for seismic data curvature picking.

[0033] In one of the embodiments, the sample label data set acquisition module comprises:

[0034] A pre-stack seismic data volume acquisition module is configured to acquire a pre-stack seismic data volume.

[0035] A pre-stack seismic data volume slicing module is configured to slice the pre-stack seismic data volume according to the number of seismic traces and the number of sampling points to obtain a sliced data volume.

[0036] A sample label data set generation module is configured to acquire a seed point of zero offset in the slice data volume in the 0th seismic trace direction, and generate the sample label data set by using random velocity disturbance to make a time-distance curve at a seed point of non-zero offset.

[0037] A computer device comprises a memory and a processor, and the memory stores a computer program, wherein the processor implements the following steps when executing the computer program:

[0038] Acquire a stacked seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information;

[0039] Determine seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and acquire a seed point set based on the seed point information of the zero offset position;

[0040] Extract a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set;

[0041] Acquire a stacked seismic data volume, extract semblance spectrum data from the stacked seismic data volume according to a residual curvature range, establish a corresponding relationship between the semblance spectrum data and the seed point set, and obtain a semblance spectrum data label data set;

[0042] Input the curvature curve label data set and the semblance spectrum data label data set into a neural network for learning to obtain a neural network model for seismic data curvature picking.

[0043] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0044] Acquire a stacked seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information;

[0045] Determine seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and acquire a seed point set based on the seed point information of the zero offset position;

[0046] Extract a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set;

[0047] Acquire a stacked seismic data volume, extract semblance spectrum data from the stacked seismic data volume according to a residual curvature range, establish a corresponding relationship between the semblance spectrum data and the seed point set, and obtain a semblance spectrum data label data set;

[0048] The curvature curve label data set and the similar spectrum data label data set are input into a neural network for learning to obtain a neural network model for curvature picking of seismic data.

[0049] The curvature picking method, device, computer device and storage medium for seismic data, by extracting the curvature curve of the seismic data from the sample label data set, obtaining the curvature curve label set, and extracting the similar spectrum data from the stacked seismic data according to the residual curvature range, establishing the corresponding relationship between the similar spectrum data and the seed point set, and then inputting the curvature curve label data set and the similar spectrum data label data set into the neural network for learning, and then obtaining the neural network model capable of picking the curvature of the seismic data, during the training of the neural network model, through a large amount of data operation, the neural network model is gradually optimized, so that the neural network model can more accurately pick the curvature of the seismic data. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 A flowchart of the curvature picking method for seismic data in one embodiment;

[0051] Figure 2 A structural diagram of the curvature picking device for seismic data in one embodiment;

[0052] Figure 3 An internal structure diagram of the computer device in one embodiment;

[0053] Figure 4 A flowchart of the curvature picking method for seismic data in another embodiment;

[0054] Figure 5 A flowchart of the curvature picking method for seismic data in another embodiment;

[0055] Figure 6 An instance diagram of the seismic data sample set in one embodiment;

[0056] Figure 7 An example diagram of the preprocessed seismic data sample set in one embodiment;

[0057] Figure 8 Similar coefficient spectrum data obtained based on the seismic data in one embodiment;

[0058] Figure 9 An example diagram of the seismic data label set in one embodiment;

[0059] Figure 10 A schematic diagram of the neural network model in one embodiment;

[0060] Figure 11 A result diagram output by the neural network model in one embodiment;

[0061] Figure 12 A schematic diagram of the time-distance curve shape in an embodiment. DETAILED DESCRIPTION

[0062] For the purpose, technical solutions and advantages of the present application to be more clear, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0063] Embodiment One

[0064] In this embodiment, as shown in Figure 1 , a seismic data curvature picking method is provided, which comprises:

[0065] Step 110, acquiring a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information.

[0066] In this embodiment, a seismic data volume is acquired as sample data, which is a pre-stack gather data volume. The pre-stack seismic data volume contains the record of reflection waveforms. In this embodiment, the reflection waveform position corresponding to each reflection waveform is identified, and the identified seismic data volume is taken as label data, so as to obtain a sample label data set.

[0067] In the sample label data set, the sample data graph of the seismic data volume is as shown in Figure 6 , and the label data graph is as shown in Figure 9 .

[0068] Step 120, determining seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and acquiring a seed point set based on the seed point information of the zero offset position.

[0069] In this embodiment, the position of the first reflection waveform is the seed point of the zero offset position. In this step, the seed point information of the zero offset position in the sample label data set is determined according to the reflection waveform position information, so as to determine the position information of the first seed point. Based on the position information of the first seed point, the position information of other seed points can be acquired, so as to acquire various seed points in the sample label data set and obtain a seed point set.

[0070] Step 130, extracting a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set.

[0071] In this embodiment, the curvature curve label data set contains the position information of the seed point and the residual curvature corresponding to various seed points.

[0072] Step 140 , obtaining a post-stack seismic data volume, extracting similar spectrum data from the post-stack seismic data volume according to the residual curvature range, establishing a corresponding relationship between the similar spectrum data and the seed point set, and obtaining a similar spectrum data label data set.

[0073] In this embodiment, the similarity spectrum data includes the residual curvature of the seed point. By establishing a correspondence between the similarity spectrum data and the seed point set, the relationship between the seed point position information and the corresponding residual curvature can be obtained. Unlike step 130, the residual curvature in step 140 is extracted based on the residual curvature range, allowing a more targeted correspondence between the seed point position and the residual curvature based on the residual curvature range.

[0074] Step 150: Input the curvature curve label dataset and the similar spectrum data label dataset into a neural network for learning to obtain a neural network model for seismic data curvature picking.

[0075] In this embodiment, the curvature curve label dataset and the similar spectrum data label dataset are used as training sets and input into the neural network, the neural network is trained, the parameters of the neural network are determined, and the trained neural network is obtained. The trained neural network can be used to pick up the remaining curvature.

[0076] The above-mentioned seismic data curvature picking method extracts the seismic data curvature curve from the sample label data set to obtain the curvature curve label set, and extracts similar spectrum data from the stacked seismic data body according to the remaining curvature range, thereby establishing a corresponding relationship between the similar spectrum data and the seed point set, and then inputs the curvature curve label data set and the similar spectrum data label data set into the neural network for learning, thereby obtaining a neural network model capable of seismic data curvature picking. When training the neural network model, after a large amount of data calculation, the neural network model is gradually optimized so that the neural network model can pick up the seismic data curvature more accurately.

[0077] In one embodiment, the step of obtaining a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records the reflection waveform position information, includes: obtaining a pre-stack seismic data volume; slicing the pre-stack seismic data volume according to the number of seismic channels and the number of sampling points to obtain a sliced ​​data volume; obtaining a seed point with zero offset in the sliced ​​data volume in the direction of the 0th seismic channel along the sampling point direction, and using random velocity perturbation at the seed point with non-zero offset to produce a time-distance curve to generate the sample label data set.

[0078] In this embodiment, the number of seismic traces of the seismic data volume is N, and the number of sampling points is M. The seismic data volume is sliced according to the N seismic traces and the M sampling points to obtain a sliced data volume. The number of seed points of zero offset is obtained in the 0th seismic trace direction along the sampling point direction. A time-distance curve is generated by using a random velocity perturbation in the non-zero offset to generate a label data set.

[0079] In this embodiment, for the seismic reflection waveform sample data, a ricker wavelet is convolved with the label data to construct a sample label data set, which is calculated by the following formula (1):

[0080] T(n,m) = F(m,v,n) n∈[0,N-1],v∈[-25%,25%],m∈[0,M-1] (1)

[0081] wherein T is a function expression of a curvature curve existing in the label data, F is a rule for generating the curve, that is, the label data set is generated by perturbing the time-distance curve starting from the seed point n=0 to N-1 with a velocity perturbation range of-25% to 25%. Formula (1) is a calculation formula of the seismic exploration time-distance curve, and T(n,m) refers to a two-dimensional graph of all time-distance curves on the data volume corresponding to the size of the pre-stack seismic data slice. Each curve will generate different curvature changes according to the perturbation of v, and the specific calculation is as follows:

[0082]

[0083] wherein T(n,m) is a two-dimensional graph of all time-distance curves on the data volume corresponding to the size of the pre-stack seismic data slice. Each time-distance curve will generate different curvature changes according to the perturbation of v, x is the index of n, n corresponds to the offset X offset , h is the index of m, and m corresponds to the number of seed points. The two-dimensional graph of the time-distance curve is shown in Figure 12 , the number of seed points is m, and the time-distance curve generated by different V is shown in Figure 12 .

[0084] In one embodiment, the step of obtaining the pre-stack seismic data volume as the sample label data set comprises: obtaining the pre-stack seismic data volume as the sample label data set; and pre-processing the sample label data set to obtain a processed sample label data set. The step of extracting the seismic data curvature curve from the sample label data set to obtain a curvature curve label data set comprises: extracting the seismic data curvature curve from the processed sample label data set to obtain a curvature curve label data set. The pre-processed sample label data set is shown in Figure 7 .

[0085] In one embodiment, the step of preprocessing the sample label data set to obtain a processed sample label data set comprises: identifying an amplitude value distribution range of the prestacked seismic data volume, performing a clipping operation on the prestacked seismic data volume based on the amplitude value distribution range; performing filtering and grayscale processing, normalization operation on the clipped seismic data volume to obtain the processed sample label data set.

[0086] In this embodiment, the prestacked seismic data volume is identified, the amplitude value distribution range of the seismic data is determined, and the prestacked seismic data volume is clipped by the following formula (2) and formula (3):

[0087] D_1(n, m) = Clip(D(n, m)) n ∈ [0, N-1], m ∈ [0, M-1] (2)

[0088] Clip(D(n, m)) = if(D(n, m) > 0.90*Max(D(n, m))) then D(n, m)

[0089] = 0.90*Max(D(n, m)) (3)

[0090] Wherein, D(n, m) is seismic data, which is a prestacked seismic data trace set slice, n is a trace index, and m is a sampling index.

[0091] The clipped seismic data volume is filtered and grayscale processed and normalized by the following formula (4)

[0092] D_2(n, m) = SplitNorm(StructSmooth((D_1(n, m)))

[0093] n ∈ [0, N-1], m ∈ [0, M-1] (4)

[0094] Wherein, SplitNorm is a segmented normalization calculation along the sampling direction, and StructSmooth is a structure smoothing technology, which performs a filtering and smoothing operation along the same phase axis of the discontinuous seismic data phase axis.

[0095] In one embodiment, the step of obtaining the poststacked seismic data volume further comprises: determining the residual curvature range according to the work area. In this embodiment, in order to screen more residual curvatures that meet the situation of the work area, the residual curvature range is determined according to the requirements of the work area, so as to extract the corresponding similarity spectrum data.

[0096] In one embodiment, the steps of obtaining a post-stack seismic data volume, extracting similar spectrum data from the post-stack seismic data volume according to a residual curvature range, establishing a correspondence between the similar spectrum data and the seed point set, and obtaining a similar spectrum data label dataset include: scanning the post-stack seismic data volume according to the residual curvature range, calculating a similarity coefficient spectrum according to the following formula (5), and further extracting seismic features:

[0097] RD(g,m)=Semblance(D_2(n,m)) n∈[0,N-1], m∈[0,M-1], g∈[g1,g2] (5)

[0098] Among them, Semblance(D_2(n,m)) is the calculation method of seismic data similarity spectrum, g is the residual curvature scanning range;

[0099] The similarity spectrum is calculated using formula (6):

[0100]

[0101] Among them, T zx Indicates the offset is x halfoffset The remaining delay calculated at z0 is the time corresponding to zero bias, and γ is the residual curvature range.

[0102] Among them, the similarity spectrum calculated based on the seismic data volume is as follows Figure 8 shown.

[0103] In this embodiment, the residual curvature scanning range of equation (5) is [-0.5, 0.5].

[0104] In this embodiment, the seed point is selected by selecting the vertical depth position of the point with the maximum energy picked up on the similarity spectrum, thereby constructing a corresponding relationship between the similarity spectrum data and the seed point set.

[0105] In one embodiment, the neural network is based on a convolutional neural network autoencoder as the main body, and a 29-layer deep neural network model is built based on tensorflow 2.0. The deep neural network model built is as follows Figure 10 shown.

[0106] In the embodiment, considering that the main feature of the stacked seismic data volume or the stacked seismic data volume as a whole is its spatial position information, a self-encoder based on a convolutional neural network is adopted as the main body of the technology, and a 29-layer deep neural network model is built based on tensorflow 2.0. The deep neural network model includes seven convolutional up-sampling blocks and seven convolutional down-sampling blocks, wherein each convolutional block is composed of two convolutional layers, a BN layer, a down-sampling / up-sampling layer, and a LeakyRelu activation function layer. The convolutional kernels adopted by the two convolutional layers are 7x1 and 1x7, so as to obtain a larger spatial receptive field and reduce the amount of calculation.

[0107] In an embodiment, the step of inputting the curvature curve label data set and the similar spectrum data label data set into a neural network for learning to obtain a neural network model for seismic data curvature picking includes: establishing a training set by using the curvature curve label data set and the similar spectrum data label data set, and performing distributed training on the neural network; wherein the error measure of the neural network takes the RMSE error functional based on L2 regularization as the objective function of the neural network, the parameter initialization mode of the convolutional layer in the neural network adopts Xavier initialization, the L2 regularization with a coefficient of 0.01 is set for the parameters of the convolutional layer, and the Adam algorithm is adopted as the optimization algorithm.

[0108] In the embodiment, the Adam algorithm is a first-order optimization algorithm that can replace the traditional stochastic gradient descent process. It designs independent adaptive learning rates for different parameters by calculating the first moment estimate and the second moment estimate of the gradient, and then iteratively updates the neural network weights. The Adam algorithm has better convergence than the traditional stochastic gradient descent algorithm.

[0109] In the embodiment, the RMSE functional expression is shown in formula (7):

[0110]

[0111] wherein E(x) is the objective function of the neural network, x is the model parameter of the neural network, m is the selected small batch quantity, y i (x) is the actual label data, is the predicted data, and γ is a regularization parameter. Further, the regularization parameter γ is 0.01.

[0112] In the embodiment, after obtaining the neural network model for seismic data curvature picking, the parameters of the trained neural network are saved to the local disk and deployed in the special cluster for seismic data curvature picking to form a reliable seismic data curvature picking technology.

[0113] It should be understood that, although Figure 1The steps in the flowchart are shown in sequence according to the arrows, but the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated herein, there is no strict order limitation for the execution of the steps, and the steps can be executed in other orders. Moreover, Figure 1 At least one part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the sub-steps or stages is not necessarily sequential, but can be alternately executed with at least one part of other steps or sub-steps or stages of other steps.

[0114] Embodiment Two

[0115] In this embodiment, as shown in Figure 4 A seismic data curvature picking method is provided, comprising:

[0116] Step 310, establishing a seismic data volume as sample data, wherein a large number of reflection waveform records exist, and simultaneously establishing an accurate reflection waveform position data volume corresponding to the reflection waveforms as label data, thereby establishing a sample label data set.

[0117] In this embodiment, a sample label data set is established in which the seismic curvature data corresponds to the actual curvature position: specifically, the following method is used. First, for a seismic data slice with N seismic traces and M sampling points, a number of seed points with zero offset distance are established in the 0th seismic trace direction along the sampling point direction, and a time-distance curve generated by random velocity perturbation is used for non-zero offset distance to generate a label data set. For seismic reflection waveform sample data, a ricker wavelet is convolved with the label data to construct a sample label data set, and the specific formula is shown in equation (1):

[0118] T(n, m) = F(m, v, n) n∈[0, N-1], v∈[-25%, 25%], m∈[0, M-1] (1)

[0119] Wherein, T is the functional expression of a curvature curve existing in the label data, and F is the rule for generating the curve, that is, a time-distance curve generated by perturbation of seed points from n=0 to n=N-1 with a velocity perturbation range of -25% to 25%. Equation (1) is a calculation formula for a time-distance curve in seismic exploration, and T(n, m) refers to a two-dimensional graph of all time-distance curves on the entire data volume corresponding to the size of the pre-stack seismic data slice. Each curve will generate different curvature changes according to the perturbation of v, and the specific calculation is as follows:

[0120]

[0121] Wherein, T(n, m) is a two-dimensional graph of all time-distance curves on the whole data body corresponding to the size of the pre-stack seismic data slice, each time-distance curve will produce different curvature changes according to the disturbance of v, x is the index of n, n corresponds to the offset X offset , h is the index of m, m corresponds to the number of seed points. The two-dimensional graph of the time-distance curve is shown in Figure 12 , the seed points are m, and the time-distance curve forms produced by different V are shown in Figure 12 .

[0122] Step 320, preprocessing the established sample label data set.

[0123] In this embodiment, the seismic data body is preprocessed to the sampling standard and numerical size that can be input by the deep learning, and the seed point information at the zero offset position is extracted.

[0124] In this embodiment, the seismic data body is preprocessed as follows: first, input the seismic data body data, determine the amplitude value distribution range of the seismic data, and perform truncation operation on the input seismic data according to the following formula:

[0125] D_1(n, m) = Clip(D(n, m)) n∈[0, N-1], m∈[0, M-1]

[0126] Clip(D(n, m)) = if(D(n, m) > 0.90*Max(D(n, m))) then D(n, m)

[0127] = 0.90*Max(D(n, m))

[0128] Wherein, D(n, m) is seismic data, which is a pre-stack seismic data trace slice, n is the trace index, and m is the sampling index.

[0129] The truncated seismic data is filtered and processed by gray scale and normalization operation, and the following formula is used

[0130] D_2(n, m) = SplitNorm(StructSmooth((D_1(n, m)))

[0131] n∈[0, N-1], m∈[0, M-1]

[0132] Wherein SplitNorm adopts segmented normalization calculation along the sampling direction, and StructSmooth adopts construction filtering technology to perform filtering and smoothing operation along the trend of the seismic data event.

[0133] Step 330, output the seismic data volume, determine the remaining curvature range according to the work area requirement, extract the seismic data semblance spectrum data, and establish the data set corresponding to the semblance spectrum and the seed point.

[0134] In the embodiment, the input seismic data volume is scanned according to the selected remaining curvature range of the work area, the semblance coefficient spectrum is calculated, the seismic characteristics are further extracted, and the following formula is used for calculation:

[0135] RD(g,m) = Semblance(D_2(n,m)) n∈[0,N-1],m∈[0,M-1],g∈[g1,g2]

[0136] Wherein, Semblance(D_2(n,m)) is the calculation method of the semblance spectrum of the seismic data, g is the remaining curvature scanning range, and the sample label data set range is [-0.5, 0.5]. The semblance spectrum is calculated by using the following formula:

[0137]

[0138] Wherein, T zx represents the residual delay amount calculated at the offset distance x halfoffset , T z0 is the time corresponding to zero offset, and gamma is the residual curvature range.

[0139] In the embodiment, the seed point is selected, the point with the maximum energy in the longitudinal depth position picked up on the semblance spectrum is selected, and the one-to-one corresponding data set of the semblance spectrum and the seed point is constructed.

[0140] Step 340, a neural network model based on deep learning is constructed.

[0141] In the embodiment, considering that the seismic data volume is as a whole, the main feature is the spatial position information of the seismic data volume, therefore, the autoencoder based on the convolutional neural network is used as the main body, and the 29-layer deep neural network model is built based on tensorflow 2.0. The deep neural network model includes seven convolutional up-sampling blocks and seven convolutional down-sampling blocks, wherein each convolutional block is composed of two convolutional layers, a BN layer, a down-sampling / up-sampling layer and a LeakyRelu activation function layer. Two convolutional layers use convolution kernels of 7x1 and 1x7 to obtain a larger spatial receptive field and reduce the calculation amount.

[0142] Step 350, the neural network model constructed in step 340 is distributedly trained by using the training set established in steps 320 and 330.

[0143] In the embodiment, the error metric of the training of the neural network model is based on the RMSE error functional of L2 regularization as the objective function of the neural network model, the parameter initialization method of the convolution layer is Xavier initialization, the L2 regularization with a coefficient of 0.01 is set in the convolution layer parameters, and the optimization algorithm of the neural network model is the Adam algorithm. The Adam algorithm is a first-order optimization algorithm that can replace the traditional stochastic gradient descent process, which designs independent adaptive learning rates for different parameters by calculating the first and second moment estimates of the gradient, and then iteratively updates the neural network model weights, which has better convergence than the traditional stochastic gradient descent algorithm.

[0144] In the embodiment, the RMSE functional expression is as follows:

[0145]

[0146] Wherein E(x) is the objective function, x is the neural network model parameter, m is the selected small batch quantity, y i (x) is the actual label data, is the predicted data, and γ is the regularization parameter. The regularization parameter γ is 0.01.

[0147] Step 360, save the parameters of the trained network model to the local disk, and deploy it to the special cluster of the seismic data curvature picking to form a reliable seismic data curvature picking technology.

[0148] The key of the application lies in the training of the neural network model and the saving and loading of the neural network model. The application uses GPU computing resources to train the neural network model according to the requirements of step 350, saves the parameters of the neural network model to the local after the training is completed, and then uploads the model to the processing cluster, which can use CPU resources to infer the seismic data to be processed, and complete the popularization and application of the application technology.

[0149] Embodiment three

[0150] In the embodiment, as shown in Figure 5 A seismic data curvature picking method is provided, which comprises: generating a label data set according to the number of seed points in the sampling direction and the random speed disturbance, convolving the label data set with a ricker wavelet to establish a sample data set, preprocessing the sample data set, establishing a seismic data curvature curve sample label data set, calculating a similarity coefficient spectrum using different range residual curvatures, establishing a similarity spectrum and a seed point sample label data set, building a convolutional neural network and training, obtaining deep learning model parameters and deploying a cluster. The test data is input into the deep learning model, and the prediction result is output. As shown in Figure 11As shown, the processing result of the deep learning model on the actual material seismic data volume can be seen, the position picking of the seismic data volume by the application is relatively accurate, and the requirement of the deep domain modeling can be met.

[0151] The seismic data curvature picking method of the application is of great significance to the deep domain seismic data processing by establishing an accurate deep domain velocity model. The picking of the residual curvature of the seismic data in the deep domain velocity modeling link is a big problem in this link. The traditional curvature picking technology based on the wave peak and wave trough cannot adapt to the complex exploration area. The application uses the big data and the artificial intelligence technology based on image recognition gradually applied in the industry in recent years, builds a residual curvature picking system based on deep learning, realizes the accurate extraction of the position of the seismic data volume, and further provides the accuracy guarantee for the subsequent link of the deep domain velocity modeling of the seismic exploration, and solves the problem of inaccurate deep domain seismic velocity modeling in the complex surface and underground geological structure area such as the mountain front zone and the alluvial fan.

[0152] Embodiment four

[0153] In this embodiment, as shown in Figure 2 The application provides a seismic data curvature picking device, comprising:

[0154] The sample label data set acquisition module 210 is configured to acquire a stacked seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information.

[0155] The seed point set acquisition module 220 is configured to determine seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and acquire a seed point set based on the seed point information of the zero offset position.

[0156] The curvature curve acquisition module 230 is configured to extract a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set.

[0157] The similar spectrum data acquisition module 240 is configured to acquire a stacked seismic data volume, extract similar spectrum data from the stacked seismic data volume according to a residual curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set.

[0158] The model training module 250 is configured to input the curvature curve label data set and the similar spectrum data label data set into a neural network for learning to obtain a neural network model for seismic data curvature picking.

[0159] The sample label data set acquisition module 210 is configured to acquire a stacked seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information.

[0160] In this embodiment, a seismic data volume is obtained as sample data, which is a pre-stack gather data volume. The pre-stack seismic data volume contains records of reflection waveforms. In this embodiment, the reflection waveform positions corresponding to the reflection waveforms are identified, and the identified seismic data volume is taken as label data, thereby obtaining a sample label data set.

[0161] The seed point set acquisition module 220 is configured to determine seed point information of a zero-offset position in the sample label data set according to the reflection waveform position information, and acquire a seed point set based on the seed point information of the zero-offset position.

[0162] In this embodiment, the position of the first reflection waveform is the seed point of the zero-offset position. In this step, the seed point information of the zero-offset position in the sample label data set is determined according to the reflection waveform position information, thereby determining the position information of the first seed point. Based on the position information of the first seed point, the position information of other seed points can be acquired, thereby acquiring various seed points in the sample label data set, and obtaining a seed point set.

[0163] The curvature curve acquisition module 230 is configured to extract a seismic data curvature curve from the sample label data set, and obtain a curvature curve label data set.

[0164] In this embodiment, the curvature curve label data set contains the position information of the seed points and the residual curvatures corresponding to the various seed points.

[0165] The similar spectrum data acquisition module 240 is configured to acquire a post-stack seismic data volume, extract similar spectrum data from the post-stack seismic data volume according to a residual curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set.

[0166] In this embodiment, the similar spectrum data contains the residual curvatures of the seed points, and the corresponding relationship between the similar spectrum data and the seed point set can be established to obtain the relationship between the position information of the seed points and the corresponding residual curvatures. Unlike the curvature curve acquisition module 230, the residual curvatures of the similar spectrum data acquisition module 240 are extracted according to the residual curvature range, and the corresponding relationship between the seed point positions and the residual curvatures can be obtained more specifically according to the residual curvature range.

[0167] The model training module 250 is configured to input the curvature curve label data set and the similar spectrum data label data set into a neural network for learning, and obtain a neural network model for seismic data curvature picking.

[0168] In the embodiment, the curvature curve label data set and the similar spectrum data label data set are taken as the training set, input to the neural network, the neural network is trained, the parameters of the neural network are determined, and the trained neural network is obtained. The trained neural network can be used for picking up residual curvature.

[0169] In one embodiment, the sample label data set acquisition module 210 includes,

[0170] The pre-stack seismic data volume acquisition module 211 is configured to acquire a pre-stack seismic data volume.

[0171] The pre-stack seismic data volume slicing module 212 is configured to slice the pre-stack seismic data volume according to the number of seismic traces and the number of sampling points to obtain a sliced data volume.

[0172] The sample label data set generation module 213 is configured to acquire seed points of zero offset in the sliced data volume along the sampling point direction in the 0th seismic trace direction, and generate a time-distance curve by using random velocity perturbation for seed points of non-zero offset to generate the sample label data set.

[0173] In the embodiment, the number of seismic traces of the seismic data volume is N, and the number of sampling points is M. The seismic data volume is sliced according to the number of N seismic traces and the number of M sampling points to obtain a sliced data volume. The number of seed points of zero offset is acquired along the sampling point direction in the 0th seismic trace direction, and a time-distance curve generated by using random velocity perturbation for non-zero offset is used to generate a label data set.

[0174] In the embodiment, for the seismic reflection waveform sample data, a ricker wavelet is convolved with the label data to construct a sample label data set, which is calculated by the following formula (1):

[0175] T(n, m) = F(m, v, n) n∈[0, N-1], v∈[-25%, 25%], m∈[0, M-1] (1)

[0176] Wherein, T is a function expression of a curvature curve existing in the label data, and F is a rule for generating the curve, that is, the label data set is generated by perturbing a time-distance curve of seed points of n=0 to N-1 with a velocity perturbation range of-25% to 25%.

[0177] In one embodiment, the sample label data set acquisition module 210 is configured to acquire a pre-stack seismic data volume as the sample label data set, pre-process the sample label data set to obtain a processed sample label data set, and extract a seismic data curvature curve from the processed sample label data set to obtain a curvature curve label data set.

[0178] In one embodiment, the sample label data set acquisition module 210 is configured to identify an amplitude value distribution range of the pre-stack seismic data volume, perform a clipping operation on the pre-stack seismic data volume based on the amplitude value distribution range, perform filtering and grayscale processing and normalization operation on the clipped seismic data volume to obtain the processed sample label data set.

[0179] In this embodiment, the pre-stack seismic data volume is identified, the amplitude value distribution range of the seismic data is determined, and the pre-stack seismic data volume is clipped by the following formula (2) and formula (3):

[0180] D_1(n, m) = Clip(D(n, m)) n∈[0, N-1], m∈[0, M-1] (2)

[0181] Clip(D(n, m)) = if(D(n, m) > 0.90*Max(D(n, m))) then D(n, m)

[0182] = 0.90*Max(D(n, m)) (3)

[0183] Wherein, D(n, m) is seismic data, which is a pre-stack seismic data trace set slice, n is a trace index, and m is a sampling index.

[0184] The clipped seismic data volume is filtered and grayscale processed and normalized, and the following formula (4) is used:

[0185] D_2(n, m) = SplitNorm(StructSmooth((D_1(n, m)))

[0186] n∈[0, N-1], m∈[0, M-1] (4)

[0187] Wherein, SplitNorm is a segmented normalization calculation along the sampling direction, and StructSmooth is a structure smoothing technology, which performs a filtering and smoothing operation along the trend of the seismic data event to smooth the discontinuous seismic data event.

[0188] In one embodiment, the similar spectrum data acquisition module is further configured to determine the residual curvature range according to the work area. In this embodiment, in order to screen more residual curvatures that meet the situation of the work area, the residual curvature range is determined according to the requirements of the work area, so as to extract the corresponding similar spectrum data.

[0189] In one embodiment, the similar spectrum data acquisition module 240 is configured to scan the stacked seismic data volume according to the residual curvature range, calculate the similarity coefficient spectrum according to the following formula (5), and further extract the seismic features:

[0190] RD(g, m) = Semblance(D_2(n, m)) n∈[0, N-1], m∈[0, M-1], g∈[g1, g2] (5)

[0191] wherein, Semblance(D_2(n, m)) is a seismic data semblance spectrum calculation method, g is a residual curvature scanning range;

[0192] The semblance spectrum is calculated by using formula (6):

[0193]

[0194] wherein, T zx represents the residual delay amount calculated at the offset distance x halfoffset , T z0 is the time corresponding to zero offset, and g is the residual curvature range.

[0195] In this embodiment, the residual curvature scanning range of formula (5) is [-0.5, 0.5].

[0196] In this embodiment, for the selection of seed points, the point with the maximum energy picked up on the semblance spectrum is selected in the vertical depth position, so as to construct the corresponding relationship between the semblance spectrum data and the seed point set.

[0197] In an embodiment, the neural network is a convolutional neural network-based autoencoder as the main body, and a 29-layer deep neural network model is built based on tensorflow 2.0.

[0198] In this embodiment, considering that the most important feature of the stacked seismic data volume or the stacked seismic data volume as a whole is the spatial position information, a convolutional neural network-based autoencoder is adopted as the main body of the present technology, and a 29-layer deep neural network model is built based on tensorflow 2.0. The deep neural network model includes seven convolutional up-sampling blocks and seven convolutional down-sampling blocks, wherein each convolutional block is composed of two convolutional layers, a BN layer, a down-sampling / up-sampling layer and a LeakyRelu activation function layer. The convolutional kernels adopted by the two convolutional layers are 7x1 and 1x7, so as to obtain a larger spatial receptive field and reduce the amount of calculation.

[0199] In an embodiment, the model training module 250 is configured to establish a training set by using the curvature curve label data set and the semblance spectrum data label data set, and to perform distributed training on the neural network; wherein the error measurement of the neural network takes the RMSE error functional based on L2 regularization as the objective function of the neural network, the parameter initialization mode of the convolutional layer in the neural network adopts Xavier initialization, the L2 regularization with a coefficient of 0.01 is set for the parameters of the convolutional layer, and the Adam algorithm is adopted as the optimization algorithm.

[0200] In this embodiment, the Adam algorithm is a first-order optimization algorithm that can replace the traditional stochastic gradient descent process, which designs independent adaptive learning rates for different parameters by calculating the first and second moment estimates of the gradient, and then iteratively updates the neural network weights, which has better convergence than the traditional stochastic gradient descent algorithm.

[0201] In this embodiment, the RMSE functional expression is shown in equation (7):

[0202]

[0203] where E(x) is the objective function of the neural network, x is the model parameter of the neural network, m is the selected small batch quantity, y i (x) is the actual label data, is the predicted data, and γ is the regularization parameter. Further, the regularization parameter γ is 0.01.

[0204] In this embodiment, after obtaining the neural network model for seismic data curvature picking, the parameters of the trained neural network are saved to the local disk and deployed to the dedicated cluster for seismic data curvature picking, forming a reliable seismic data curvature picking technology.

[0205] The specific limitations of the seismic data curvature picking device can be referred to the limitations of the seismic data curvature picking method in the above, which will not be repeated here. Each unit in the above seismic data curvature picking device can be realized by software, hardware and combinations thereof, in whole or in part. The above each unit can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each unit.

[0206] Embodiment five

[0207] In this embodiment, a computer device is provided. Its internal structure diagram can be as shown in Figure 3As shown in the figure. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system and a computer program, and the non-volatile storage medium is deployed with a database for storing prestack seismic data volume and stacked seismic data volume. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices deployed with application software. The computer program is executed by the processor to implement a seismic data curvature picking method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0208] Those skilled in the art can understand that, Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0209] In one embodiment, a computer device is provided, comprising a memory storing a computer program and a processor executing the computer program to implement the following steps:

[0210] Step 110, obtaining a prestack seismic data volume as sample label data set, wherein the sample label data set records reflection waveform position information.

[0211] In this embodiment, a seismic data volume is obtained as sample data, which is a prestack gather data volume. The prestack seismic data volume contains the record of reflection waveform. In this embodiment, the reflection waveform position corresponding to the reflection waveform is identified, and the identified seismic data volume is used as label data, thereby obtaining a sample label data set.

[0212] Step 120, determining seed point information of a zero offset position in the sample label data set according to the reflection waveform position information, and obtaining a seed point set based on the seed point information of the zero offset position.

[0213] In this embodiment, the position of the first reflection waveform is the seed point at the zero-offset position. In this step, the seed point information at the zero-offset position in the sample label data set is determined according to the reflection waveform position information, so as to determine the position information of the first seed point. Based on the position information of the first seed point, the position information of other seed points can be obtained, so as to obtain various seed points in the sample label data set, and obtain the seed point set.

[0214] In step 130, a curvature curve of seismic data is extracted from the sample label data set, and a curvature curve label data set is obtained.

[0215] In this embodiment, the curvature curve label data set contains the position information of the seed point and the residual curvature corresponding to each seed point.

[0216] In step 140, a stacked seismic data volume is obtained, the semblance spectrum data is extracted from the stacked seismic data volume according to the residual curvature range, the correspondence between the semblance spectrum data and the seed point set is established, and a semblance spectrum data label data set is obtained.

[0217] In this embodiment, the semblance spectrum data contains the residual curvature of the seed point, and the correspondence between the semblance spectrum data and the seed point set is established, so that the relationship between the position information of the seed point and the corresponding residual curvature can be obtained. Different from step 130, the residual curvature in step 140 is extracted according to the residual curvature range, so that the corresponding relationship between the seed point position and the residual curvature can be obtained more specifically.

[0218] In step 150, the curvature curve label data set and the semblance spectrum data label data set are input into a neural network for learning, and a neural network model for picking up the curvature of seismic data is obtained.

[0219] In this embodiment, the curvature curve label data set and the semblance spectrum data label data set are input into the neural network as a training set, the neural network is trained, the parameters of the neural network are determined, and the trained neural network is obtained. The trained neural network can be used to pick up the residual curvature.

[0220] Embodiment six

[0221] In this embodiment, a computer readable storage medium is provided, and a computer program is stored on the computer readable storage medium. The computer program is executed by a processor to implement the following steps:

[0222] In step 110, a pre-stacked seismic data volume is obtained as a sample label data set, wherein the sample label data set records reflection waveform position information.

[0223] In this embodiment, a seismic data volume is obtained as sample data, and the seismic data volume is a pre-stack gather data volume. The pre-stack seismic data volume contains records of reflection waveforms. In this embodiment, the reflection waveform positions corresponding to the reflection waveforms are identified, and the identified seismic data volume is taken as label data, so as to obtain a sample label data set.

[0224] In step 120, seed point information of a zero-offset position in the sample label data set is determined according to the reflection waveform position information, and a seed point set is obtained based on the seed point information of the zero-offset position.

[0225] In this embodiment, the position of the first reflection waveform is the seed point of the zero-offset position. In this step, the seed point information of the zero-offset position in the sample label data set is determined according to the reflection waveform position information, so as to determine the position information of the first seed point. Based on the position information of the first seed point, the position information of other seed points can be obtained, so as to obtain various seed points in the sample label data set, and a seed point set is obtained.

[0226] In step 130, a curvature curve of seismic data is extracted from the sample label data set, and a curvature curve label data set is obtained.

[0227] In this embodiment, the curvature curve label data set contains the position information of the seed points and the residual curvatures corresponding to the seed points.

[0228] In step 140, a stacked seismic data volume is obtained, similar spectrum data is extracted from the stacked seismic data volume according to a residual curvature range, a corresponding relationship between the similar spectrum data and the seed point set is established, and a similar spectrum data label data set is obtained.

[0229] In this embodiment, the similar spectrum data contains the residual curvatures of the seed points, and the corresponding relationship between the similar spectrum data and the seed point set is established, so as to obtain the relationship between the position information of the seed points and the corresponding residual curvatures. Different from step 130, the residual curvatures in step 140 are extracted according to the residual curvature range, and the corresponding relationship between the seed point positions and the residual curvatures can be obtained more specifically according to the residual curvature range.

[0230] In step 150, the curvature curve label data set and the similar spectrum data label data set are input into a neural network for learning, and a neural network model for picking up seismic data curvatures is obtained.

[0231] In this embodiment, the curvature curve label data set and the similar spectrum data label data set are taken as a training set, and are input into a neural network. The neural network is trained, parameters of the neural network are determined, a trained neural network is obtained, and the trained neural network can be used to pick up residual curvatures.

[0232] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, databases, or other media in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0233] Any combination of the technical features of the above embodiments can be made, and in order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0234] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of the patent protection of the present application should be subject to the appended claims.

Claims

1. A method for picking curvature of seismic data, characterized in that: include: Acquire a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records reflection waveform position information; Determining seed point information at a zero-offset position in the sample label data set according to the reflected waveform position information, and acquiring a seed point set based on the seed point information at the zero-offset position; Extracting a seismic data curvature curve from the sample label data set to obtain a curvature curve label data set; Acquire a post-stack seismic data volume, extract similar spectrum data from the post-stack seismic data volume according to a residual curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set; The curvature curve label data set and the similar spectrum data label data set are input into a neural network for learning to obtain a neural network model for seismic data curvature picking.

2. The method according to claim 1, characterized in that The step of obtaining a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records the reflection waveform position information, comprises: Obtain pre-stack seismic data volume; Slicing the pre-stack seismic data volume according to the number of seismic traces and the number of sampling points to obtain a sliced ​​data volume; A zero-offset seed point in the slice data volume is obtained along the sampling point direction in the 0th seismic trace direction, and a time-distance curve is produced using random velocity perturbation at a non-zero-offset seed point to generate the sample label data set.

3. The method according to claim 1, characterized in that The step of obtaining a pre-stack seismic data volume as a sample label data set includes: Acquire a pre-stack seismic data volume as the sample label data set; Preprocessing the sample label data set to obtain a processed sample label data set; The step of extracting the seismic data curvature curve from the sample label data set to obtain the curvature curve label data set includes: The seismic data curvature curve is extracted from the processed sample label data set to obtain a curvature curve label data set.

4. The method according to claim 3, characterized in that The step of preprocessing the sample label data set to obtain a processed sample label data set includes: An amplitude value distribution range of a pre-stack seismic data volume is identified, and a truncation operation is performed on the pre-stack seismic data volume based on the amplitude value distribution range.

5. The method according to claim 4, characterized in that The identifying of the amplitude value distribution range of the pre-stack seismic data volume and performing a truncation operation on the pre-stack seismic data volume based on the amplitude value distribution range includes: The truncated seismic data volume is filtered, gray-scale processed, and normalized to obtain the processed sample label data set.

6. The method according to claim 1, characterized in that Before the step of obtaining the post-stack seismic data volume, the following steps are further included: The remaining curvature range is determined according to the work area.

7. A seismic data curvature picking device, characterized in that: include: A sample label data set acquisition module is used to acquire a pre-stack seismic data volume as a sample label data set, wherein the sample label data set records the reflection waveform position information; A seed point set acquisition module is configured to determine the seed point information of the zero offset position in the sample label data set according to the reflected waveform position information, and acquire a seed point set based on the seed point information of the zero offset position; Curvature curve acquisition module: used for extracting the seismic data curvature curve from the sample label data set to obtain a curvature curve label data set; Similar spectrum data acquisition module: used to obtain a post-stack seismic data volume, extract similar spectrum data from the post-stack seismic data volume according to the residual curvature range, establish a corresponding relationship between the similar spectrum data and the seed point set, and obtain a similar spectrum data label data set; Model training module: used to input the curvature curve label data set and the similar spectrum data label data set into the neural network for learning, so as to obtain a neural network model for seismic data curvature picking.

8. The device according to claim 7, characterized in that The sample label data set acquisition module includes: A pre-stack seismic data volume acquisition module is used to acquire pre-stack seismic data volumes; A pre-stack seismic data volume slicing module is used to slice the pre-stack seismic data volume according to the number of seismic traces and the number of sampling points to obtain a sliced ​​data volume; The sample label data set generation module is used to obtain the zero-offset seed point in the slice data volume along the sampling point direction in the 0th seismic trace direction, and use random velocity perturbation to create a time-distance curve at the non-zero offset seed point to generate the sample label data set.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images