An irregular seismic observation method, device, equipment and medium

Abstract

The present application discloses a method, apparatus, device and medium for irregular seismic observation, relating to the technical field of seismic observation. The method includes: obtaining undersampled data in seismic data and constructing a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data; determining a selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determining the maximum Shannon entropy according to the selection probability; determining a target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determining a target undersampling operator according to the target undersampling pattern; and determining full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator, so as to complete irregular seismic observation. Through the above solution, it is possible to select a suitable undersampling operator during irregular seismic observation to optimize seismic sparse acquisition and reduce undersampling artifacts.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic observation, and particularly relates to a method, device, equipment and medium for irregular seismic observation. Background Art

[0002] In the field of seismic observation, in the process of obtaining the undersampling operator M, full random undersampling cannot control the maximum gap of the observation grid. The local random undersampling method controls the maximum gap size by introducing a deterministic variable, that is, the factor Q, but it also brings inconvenience. The factor Q requires that both K / Q and kQ / K are integers, which is difficult to meet in actual seismic exploration, especially in high-dimensional seismic exploration. Because assuming that each spatial dimension needs to meet these conditions, the computational complexity is too large. In summary, how to select a suitable undersampling operator during irregular seismic observation to optimize seismic sparse acquisition and reduce undersampling artifacts remains to be further solved. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for irregular seismic observation, which can select a suitable undersampling operator during irregular seismic observation to optimize seismic sparse acquisition and reduce undersampling artifacts. The specific scheme is as follows:

[0004] In a first aspect, the present application discloses a method for irregular seismic observation, including:

[0005] Obtaining undersampled data in seismic data and constructing a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data;

[0006] Determining a selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determining the maximum Shannon entropy according to the selection probability;

[0007] Determining a target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determining a target undersampling operator according to the target undersampling pattern;

[0008] Determining full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete irregular seismic observation.

[0009] Optionally, the constructing a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data includes:

[0010] Determining the number of observation points corresponding to the observation points in the observation range of the undersampled data, and determining the number of Gaussian kernels based on the number of observation points;

[0011] Superimpose the observation probabilities according to the sampling pattern of the observation points and the number of Gaussian kernels to obtain a probability density function, and construct a maximum Shannon entropy model based on the probability density function.

[0012] Optionally, the determining the selection probability according to the maximum Shannon entropy model and determining the maximum Shannon entropy according to the selection probability based on a preset genetic algorithm includes:

[0013] Construct a selection probability model according to the maximum Shannon entropy model and a preset constant;

[0014] Based on a preset genetic algorithm, perform selection operation, crossover operation, replication operation and mutation operation on the maximum Shannon entropy model by using the selection probability model to determine the maximum Shannon entropy.

[0015] Optionally, before constructing the selection probability model according to the maximum Shannon entropy model and a preset constant, it further includes:

[0016] Determine the Shannon entropy corresponding to each undersampling mode according to the maximum Shannon entropy model, and determine the preset constant according to the Shannon entropy corresponding to each undersampling mode.

[0017] Optionally, performing a crossover operation on the maximum Shannon entropy model includes:

[0018] Perform a crossover operation on the maximum Shannon entropy model according to a preset control crossover level parameter.

[0019] Optionally, the performing a crossover operation and a replication operation on the maximum Shannon entropy model includes:

[0020] Determine the Shannon entropy corresponding to each undersampling mode according to the maximum Shannon entropy model, and determine the control crossover probability parameter according to the Shannon entropy corresponding to each undersampling mode;

[0021] Perform a crossover operation and a replication operation on the maximum Shannon entropy model according to the control crossover probability parameter while keeping the parental individuals unchanged.

[0022] Optionally, performing a mutation operation on the maximum Shannon entropy model includes:

[0023] Select a preset number of offspring individuals and change the genes of the offspring individuals to complete the mutation operation on the maximum Shannon entropy model.

[0024] In a second aspect, the present application discloses a seismic irregular observation device, including:

[0025] A Shannon entropy model construction module, configured to obtain undersampled data in seismic data and construct a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data;

[0026] The maximum Shannon entropy determination module is used to determine the selection probability based on a preset genetic algorithm according to the Shannon entropy maximum model, and determine the maximum Shannon entropy according to the selection probability;

[0027] The undersampling operator determination module is used to determine the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling pattern;

[0028] The full observation data determination module is used to determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator, so as to complete the seismic irregular observation.

[0029] In a third aspect, the present application discloses an electronic device, including:

[0030] A memory for storing a computer program;

[0031] A processor for executing the computer program to implement the steps of the seismic irregular observation method disclosed above.

[0032] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the seismic irregular observation method disclosed above are implemented.

[0033] When this application conducts irregular seismic observations, it first obtains the undersampled data in the seismic data and constructs a maximum Shannon entropy model based on the observation probability corresponding to the undersampled data. Based on a preset genetic algorithm, it determines the selection probability according to the maximum Shannon entropy model, and determines the maximum Shannon entropy according to the selection probability. According to the maximum Shannon entropy, it determines the target undersampling pattern corresponding to the maximum Shannon entropy, and determines the target undersampling operator according to the target undersampling pattern. According to the undersampled data and the target undersampling operator, it determines the full-observation data corresponding to the undersampled data to complete the irregular seismic observation. It can be seen that when this application conducts irregular seismic observations, it first obtains the undersampled data in the seismic data and constructs a maximum Shannon entropy model based on the observation probability corresponding to the undersampled data. Based on a preset genetic algorithm, it determines the selection probability according to the maximum Shannon entropy model, and determines the maximum Shannon entropy according to the selection probability. According to the maximum Shannon entropy, it determines the target undersampling pattern corresponding to the maximum Shannon entropy, and determines the target undersampling operator according to the target undersampling pattern. Further, according to the undersampled data and the target undersampling operator, it determines the full-observation data corresponding to the undersampled data to complete the irregular seismic observation. Thus, it can be seen that when this application conducts irregular seismic observations, it screens and optimizes the undersampling operator through the constructed maximum Shannon entropy model, and based on a preset genetic algorithm, it determines the selection probability and the maximum Shannon entropy according to the maximum Shannon entropy model, further determines the target undersampling operator corresponding to the maximum Shannon entropy, determines the full-observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator. The undersampling method optimized by the improved preset genetic algorithm has obvious effects in removing strong artifacts, can restore more details to make the undersampled reconstructed data smoother, and at the same time, there is less signal leakage in the undersampled reconstructed data. In summary, this application can select an appropriate undersampling operator when conducting irregular seismic observations to optimize seismic sparse acquisition and reduce undersampling artifacts. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0035] Figure 1 It is a flowchart of an irregular seismic observation method provided by this application;

[0036] Figure 2 It is a flowchart of a specific irregular seismic observation method provided by this application;

[0037] Figure 3Schematic diagram of the encoding process provided for this application;

[0038] Figure 4 Schematic diagram of the crossover operation workflow algorithm provided for this application;

[0039] Figure 5 Schematic diagram of the copy operation workflow algorithm provided for this application;

[0040] Figure 6 Schematic diagram of the copy operation workflow algorithm provided for this application;

[0041] Figure 7 Schematic diagram of the amplitude spectrum provided for this application;

[0042] Figure 8 Schematic diagram of the irregular observation area and different sparse observation systems provided for this application;

[0043] Figure 9 Schematic diagram of the reconstructed data of different sparse observation systems provided for this application;

[0044] Figure 10 Convergence curve graph of the reconstructed data of different observation methods provided for this application;

[0045] Figure 11 Schematic diagram of different undersampling modes of the field 4D data provided for this application;

[0046] Figure 12 Reconstructed four-dimensional field data graph under different undersampling methods provided for this application;

[0047] Figure 13 Graph of obtaining the slice data set at Y = 6 of the data provided for this application;

[0048] Figure 14 Schematic diagram of the reconstruction result of the slice at Y = 6 provided for this application;

[0049] Figure 15 Schematic diagram of the partially enlarged slice at Z = 60 provided for this application;

[0050] Figure 16 Schematic diagram of the structure of a seismic irregular observation device provided for this application;

[0051] Figure 17 Schematic diagram of the structure of an electronic device provided for this application. Detailed implementation manner

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] In the field of seismic observation, in the process of obtaining the undersampling operator M, fully random undersampling cannot control the maximum gap of the observation grid. The local random undersampling method controls the maximum gap size by introducing a deterministic variable, namely factor Q, but it also brings inconvenience at the same time. Factor Q requires that both K / Q and kQ / K are integers, which is difficult to meet in actual seismic exploration, especially in high-dimensional seismic exploration. Because assuming that each spatial dimension needs to meet these conditions, the computational amount is too large. For this reason, the present application provides a seismic irregular observation method that can select a suitable undersampling operator during seismic irregular observation to optimize seismic sparse acquisition and reduce undersampling artifacts.

[0054] An embodiment of the present invention discloses a seismic irregular observation method. Refer to Figure 1 As shown, the method includes:

[0055] Step S11: Obtain the undersampled data in the seismic data and construct a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data.

[0056] In this embodiment, the undersampled data in the seismic data is obtained and a maximum Shannon entropy model is constructed according to the observation probability corresponding to the undersampled data. Specifically, first, the undersampled data in the seismic data is obtained, and further, a maximum Shannon entropy model is constructed according to the observation probability corresponding to the undersampled data. Considering seismic sparse acquisition as a random system, for an ideal seismic acquisition of an unknown subsurface structure, it should have good ergodicity. Therefore, the maximum Shannon entropy model is used to quantitatively evaluate the undersampling pattern. Through the above technical solution, a maximum Shannon entropy model is constructed according to the observation probability corresponding to the undersampled data.

[0057] Step S12: Determine the selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determine the maximum Shannon entropy according to the selection probability.

[0058] In this embodiment, the selection probability is determined based on a preset genetic algorithm according to the maximum Shannon entropy model. Specifically, an improved preset genetic algorithm is used to determine the selection probability according to the maximum Shannon entropy model, and the maximum Shannon entropy is determined according to the selection probability. It can be understood that when the Shannon entropy is the largest, the corresponding undersampling mode is the obtained undersampling mode. Through the above technical solution, the maximum Shannon entropy is determined based on the preset genetic algorithm according to the maximum Shannon entropy model, so as to subsequently determine the target undersampling mode corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling mode.

[0059] Step S13: Determine the target undersampling mode corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling mode.

[0060] In this embodiment, the target undersampling mode corresponding to the maximum Shannon entropy is determined according to the maximum Shannon entropy, and the target undersampling operator is determined according to the target undersampling mode. Specifically, the maximum Shannon entropy is determined according to the selection probability, and further, the target undersampling mode corresponding to the maximum Shannon entropy is determined according to the maximum Shannon entropy, and the target undersampling operator corresponding to the target undersampling mode is obtained. Through the above technical solution, the target undersampling mode corresponding to the maximum Shannon entropy is determined according to the maximum Shannon entropy, and the target undersampling operator is determined according to the target undersampling mode, so as to determine a suitable undersampling operator, and determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete seismic irregular observation.

[0061] Step S14: Determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete seismic irregular observation.

[0062] In this embodiment, the full observation data corresponding to the undersampled data is determined according to the undersampled data and the target undersampling operator to complete seismic irregular observation. Specifically, the full observation data corresponding to the undersampled data is further determined according to the undersampled data and the target undersampling operator, and seismic irregular observation is completed. It can be understood that this solution can be used in both full random undersampling and local random undersampling. Through the above technical solution, the undersampling method optimized by the improved preset genetic algorithm has obvious effects in removing strong artifacts, can restore more details to make the undersampled reconstructed data smoother, and at the same time has less signal leakage in the undersampled reconstructed data. In summary, this application can select a suitable undersampling operator to optimize seismic sparse acquisition and reduce undersampling artifacts when performing seismic irregular observation.

[0063] It can be seen that when performing irregular seismic observations in this embodiment, the undersampled data in the seismic data is first obtained, and the maximum Shannon entropy model is constructed according to the observation probability corresponding to the undersampled data. Based on a preset genetic algorithm, the selection probability is determined according to the maximum Shannon entropy model, and the maximum Shannon entropy is determined according to the selection probability. The target undersampling pattern corresponding to the maximum Shannon entropy is determined according to the maximum Shannon entropy, and the target undersampling operator is determined according to the target undersampling pattern. The full observation data corresponding to the undersampled data is determined according to the undersampled data and the target undersampling operator, so as to complete the irregular seismic observation. It can be seen that when performing irregular seismic observations in this application, the undersampled data in the seismic data is first obtained, and the maximum Shannon entropy model is constructed according to the observation probability corresponding to the undersampled data. Based on a preset genetic algorithm, the selection probability is determined according to the maximum Shannon entropy model, and the maximum Shannon entropy is determined according to the selection probability. The target undersampling pattern corresponding to the maximum Shannon entropy is determined according to the maximum Shannon entropy, and the target undersampling operator is determined according to the target undersampling pattern. Further, the full observation data corresponding to the undersampled data is determined according to the undersampled data and the target undersampling operator, so as to complete the irregular seismic observation. Thus, it can be seen that when performing irregular seismic observations in this application, the undersampling operator is screened and optimized through the constructed maximum Shannon entropy model, and the selection probability is determined and the maximum Shannon entropy is determined based on a preset genetic algorithm according to the maximum Shannon entropy model. Further, the target undersampling operator corresponding to the maximum Shannon entropy is determined. The full observation data corresponding to the undersampled data is determined according to the undersampled data and the target undersampling operator. The undersampling method optimized by the improved preset genetic algorithm has obvious effects in removing strong artifacts, can restore more details to make the undersampled reconstructed data smoother, and at the same time has less signal leakage in the undersampled reconstructed data. In summary, this application can select an appropriate undersampling operator when performing irregular seismic observations to optimize seismic sparse acquisition and reduce undersampling artifacts.

[0064] See Figure 2 As shown, this embodiment of the present invention discloses a specific irregular seismic observation method. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution.

[0065] Step S21: Obtain the undersampled data in the seismic data and construct the maximum Shannon entropy model according to the observation probability corresponding to the undersampled data.

[0066] In this embodiment, obtaining the undersampled data in the seismic data and constructing a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data includes: determining the number of observation points corresponding to the observation points in the observation range of the undersampled data, and determining the number of Gaussian kernels based on the number of observation points; superimposing the observation probability according to the sampling pattern of the observation points and the number of Gaussian kernels to obtain a probability density function, and constructing a maximum Shannon entropy model based on the probability density function. Specifically, regarding seismic sparse acquisition as a random system, then for the ideal seismic acquisition of unknown subsurface structures, it should have good ergodicity. Therefore, the maximum Shannon entropy model is used to quantitatively evaluate the undersampling pattern. Assuming that the seismic undersampling problem is 1 + L dimensions, that is, 1 time dimension plus L spatial dimensions, in order to avoid symbol clutter, the time coordinate is ignored, and the maximum Shannon entropy model is expressed as:

[0067]

[0068] where H represents the Shannon entropy; p(x) = p(x1,…,x L ) represents the observed probability density function; X1,…,X L represents the observation range under the L spatial coordinates; p(x) represents the observation probability of each point in the region; the points with larger median values in p(x) indicate that the probability of their being observed is larger. For a specific undersampling pattern P with k observation points, its probability density function is estimated by superimposing k Gaussian kernels as follows:

[0069]

[0070] where, represents the estimated probability density function, describing the most likely undersampling pattern P generated. G i (x) represents that the Gaussian kernel located at the i-th observation grid point is:

[0071]

[0072] where x(i) = [x(i) 1 ,…,x(i) L represents the position of the i-th observation grid point; σ l represents the mean square deviation that controls the shape of the Gaussian kernel on the l-th spatial coordinate; α represents the scale of the amplitude of the normalized Gaussian kernel G i (x).

[0073] Step S22: Construct a selection probability model according to the maximum Shannon entropy model and a preset constant, and perform selection operation, crossover operation, replication operation, and mutation operation on the maximum Shannon entropy model by using the preset genetic algorithm to determine the maximum Shannon entropy.

[0074] In this embodiment, a selection probability model is constructed based on the maximum Shannon entropy model and a preset constant. The maximum Shannon entropy is determined by performing selection, crossover, replication, and mutation operations on the maximum Shannon entropy model using the preset genetic algorithm, including: determining the Shannon entropy corresponding to each undersampling pattern according to the maximum Shannon entropy model, and determining the preset constant according to the Shannon entropy corresponding to each undersampling pattern. Performing a crossover operation on the maximum Shannon entropy model according to the preset control crossover level parameter. Moreover, determining the Shannon entropy corresponding to each undersampling pattern according to the maximum Shannon entropy model, and determining the control crossover probability parameter according to the Shannon entropy corresponding to each undersampling pattern; performing crossover and replication operations on the maximum Shannon entropy model according to the control crossover probability parameter while keeping the parental individuals unchanged. Selecting a preset number of offspring individuals and changing the genes of the offspring individuals to complete the mutation operation on the maximum Shannon entropy model. Specifically, the schematic diagram of the encoding process is as shown in Figure 3 shown. To solve the problem of how to place k 1s on K grids, that is, the undersampling process, to maximize the above Shannon entropy model, an improved genetic algorithm is proposed. An individual in the preset genetic algorithm can be represented by a vector composed of k 1s (representing the observed grids) and K - k 0s (representing the unobserved grids) for the encoding process. The genetic algorithm consists of four parts: selection, crossover, replication, and mutation, and all are probabilistic decisions based on fitness. The fitness of each individual will be evaluated through the Shannon entropy model. Individuals with higher fitness have a higher chance of being inherited by the offspring, and the selection probability is:

[0075]

[0076] In the formula and H j represent the selection probability and Shannon entropy of the jth individual; c represents a constant, and its value ranges between 0 and the minimum value of H1,…,H N to balance the probability differences of individuals.

[0077] In this embodiment, the schematic diagram of the crossover operation workflow algorithm is as shown in Figure 4 shown, where λ is a parameter controlling the crossover level and can accelerate the convergence speed in the later stage of the algorithm. λ decreases from k / 2 to 0 with the number of iterations; the schematic diagram of the replication operation workflow algorithm is as shown in Figure 5 shown. The replication operation keeps the parental individuals unchanged. Usually, replication and crossover work together. Among them, 0 ≤ γ f ≤ γ m ≤ γ c ≤ 1 are three parameters controlling the crossover probability. H max 、H min 、H median are H1,…,H NThe maximum value, minimum value, and median; the schematic diagram of the replication operation workflow algorithm is as Figure 6 shown. The last part is mutation. In this step, a few offspring individuals are selected and part of their genes are changed. The mutation probability in the algorithm is low because a high mutation probability will cause the genetic algorithm to degenerate into a random search algorithm, where P robmut is the mutation probability and μ is the mutation rate.

[0078] Step S23: Determine the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling pattern.

[0079] Step S24: Determine the full-observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete seismic irregular observation.

[0080] In this embodiment, when the diagonal elements of the undersampling operator M contain 0, it will cause artifacts in the result. Therefore, the strength of the artifacts can be an important basis for judging the quality of the processing technology. In a synthetic irregular observation area, chaotic mapping random undersampling, mixed congruence undersampling, and the improved genetic algorithm optimized undersampling method are used for observation respectively. The amplitude spectra of the results are as Figure 7 shown, where (a)(d) are chaotic mapping undersampling, (b)(e) are mixed congruence undersampling, and (c)(f) are the undersampling optimized by the improved genetic algorithm. From the Figure 7 results, it can be clearly seen that there are still strong artifacts remaining in the traditional methods. In contrast, the undersampling method optimized by the improved genetic algorithm does better in removing strong artifacts. To prove the superiority of this technology, the processing effects of this technology and traditional technologies on two test data sets are shown next. The first processed data set is synthetic 3D seismic data, and the second processed data set is field 4D seismic data. For the first synthetic 3D seismic data set, the full-sampled data consists of 100×100 traces, with 400 time samples per trace. The observation area is as Figure 8 shown. The seismic data is observed by chaotic mapping random undersampling, mixed congruence random undersampling, and the proposed improved GA optimized undersampling respectively. The reconstructed data is as Figure 9 shown, where (d)(e)(f) are the reconstructed data of (a)(b)(c), and (g)(h)(i) are the corresponding reconstructions. The reconstruction convergence trends under the three modes are as Figure 10 shown. It can be seen that the reconstruction error caused by the improved GA optimized undersampling method is smaller. For the second field 4D seismic data set, it consists of one time scale and three spatial scales (200×10×50). The observation area is as Figure 11As shown in (a), only spatial scale undersampling is performed in this technology. The seismic data are observed by chaotic mapping random undersampling and improved genetic algorithm optimized undersampling respectively. The fully sampled data, chaotic mapping random undersampled data, and improved genetic algorithm optimized undersampled data are as shown in Figure 11 (b), (c), and (d). The reconstructed data of the two undersampling modes are as shown in Figure 12 (a) and (b), Figure 12 (c) and (d) are the reconstruction errors. Comparing the reconstructed data of the two undersampling modes Figure 12 at the arrow positions in (a) and (b), it can be clearly seen that the improved genetic algorithm optimized undersampling method proposed in this technology can recover more details. The slice dataset is obtained at Y = 6 of the data as shown in Figure 13 , and the reconstructed data of the two undersampling methods are as shown in Figure 14 (a) and (b), Figure 14 (c) and (d) are the corresponding reconstruction errors. Figure 15 is Figure 14 the enlarged view of a partial slice at Y = 6 and Z = 60. It can be clearly seen in the area circled by the ellipse in Figure 15 (a) and (b) that the data reconstructed by the improved genetic algorithm optimized undersampling is smoother, and Figure 15 at the positions pointed by the arrows in (c) and (d), it can be seen that the signal leakage of the data reconstructed by the improved genetic algorithm optimized undersampling is less.

[0081] It can be seen that in this embodiment, by constructing the maximum Shannon entropy model to screen and optimize the undersampling operator, and based on the preset genetic algorithm, the maximum Shannon entropy is determined according to the Shannon entropy model, and then the most suitable target undersampling operator is further determined. According to the undersampled data and the target undersampling operator, the fully observed data corresponding to the undersampled data is determined. The undersampling method optimized by the improved preset genetic algorithm has obvious effects in removing strong artifacts, can recover more details to make the reconstructed data of undersampling smoother, and at the same time, the signal leakage of the reconstructed data of undersampling is less. In summary, this application can select a suitable undersampling operator to optimize seismic sparse acquisition and reduce undersampling artifacts when performing seismic irregular observations.

[0082] When this application performs seismic irregular observations, it first obtains the undersampled data in the seismic data and calculates the Shannon entropy value of this undersampling mode according to the observation probability corresponding to the undersampled data, using the Shannon entropy as the objective function; improves the traditional genetic algorithm, determines the selection probability of this undersampling mode in the genetic algorithm according to the Shannon entropy value, and designs a crossover probability that gradually decreases with the iteration of the algorithm to ensure the calculation efficiency of the algorithm; iterates the algorithm to obtain the undersampling mode corresponding to the maximum Shannon entropy value to obtain the undersampling operator; determines the fully observed data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete the seismic irregular observation.

[0083] See Figure 16 As shown, an irregular seismic observation device according to an embodiment of the present application includes:

[0084] A Shannon entropy model construction module 11, configured to obtain undersampled data in seismic data and construct a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data;

[0085] A maximum Shannon entropy determination module 12, configured to determine a selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determine the maximum Shannon entropy according to the selection probability;

[0086] An undersampling operator determination module 13, configured to determine a target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine a target undersampling operator according to the target undersampling pattern;

[0087] A full observation data determination module 14, configured to determine full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator, so as to complete irregular seismic observation.

[0088] It can be seen that in this embodiment

[0089] When performing irregular seismic observations, first obtain the undersampled data in the seismic data and construct a maximum Shannon entropy model based on the observation probability corresponding to the undersampled data. Determine the selection probability based on the preset genetic algorithm according to the maximum Shannon entropy model, and determine the maximum Shannon entropy according to the selection probability. Determine the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling pattern. Determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete the irregular seismic observation. It can be seen that when performing irregular seismic observations in this application, first obtain the undersampled data in the seismic data and construct a maximum Shannon entropy model based on the observation probability corresponding to the undersampled data. Determine the selection probability based on the preset genetic algorithm according to the maximum Shannon entropy model, and determine the maximum Shannon entropy according to the selection probability. Determine the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine the target undersampling operator according to the target undersampling pattern. Further, determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete the irregular seismic observation. Thus, it can be seen that when performing irregular seismic observations in this application, the undersampling operator is screened and optimized through the constructed maximum Shannon entropy model, and the selection probability is determined and the maximum Shannon entropy is determined based on the preset genetic algorithm according to the maximum Shannon entropy model. Further, the target undersampling operator corresponding to the maximum Shannon entropy is determined. The full observation data corresponding to the undersampled data is determined according to the undersampled data and the target undersampling operator. The undersampling method optimized by the improved preset genetic algorithm has obvious effects in removing strong artifacts, can restore more details to make the undersampled reconstructed data smoother, and at the same time has less signal leakage in the undersampled reconstructed data. In summary, this application can select a suitable undersampling operator when performing irregular seismic observations to optimize seismic sparse acquisition and reduce undersampling artifacts.

[0090] In some specific embodiments, the Shannon entropy model construction module 11 specifically includes:

[0091] An observation point number determination unit, configured to determine the number of observation points corresponding to the observation points in the observation range of the undersampled data, and determine the number of Gaussian kernels based on the number of observation points;

[0092] A model construction unit, configured to superimpose the observation probabilities according to the sampling pattern of the observation points and the number of Gaussian kernels to obtain a probability density function, and construct a maximum Shannon entropy model based on the probability density function.

[0093] In some specific embodiments, the maximum Shannon entropy determination module 12 specifically includes:

[0094] A probability model construction unit for constructing a selection probability model according to the maximum Shannon entropy model and a preset constant;

[0095] An algorithm call unit for performing selection operation, crossover operation, replication operation and mutation operation on the maximum Shannon entropy model by using the selection probability model based on a preset genetic algorithm to determine the maximum Shannon entropy.

[0096] In some specific embodiments, the seismic irregular observation device further includes:

[0097] A preset constant selection module for determining the Shannon entropy corresponding to each undersampling mode according to the maximum Shannon entropy model and determining the preset constant according to the Shannon entropy corresponding to each undersampling mode.

[0098] In some specific embodiments, the algorithm call unit is specifically configured to: perform a crossover operation on the maximum Shannon entropy model according to a preset control crossover level parameter.

[0099] In some specific embodiments, the algorithm call unit is specifically configured to: determine the Shannon entropy corresponding to each undersampling mode according to the maximum Shannon entropy model and determine a control crossover probability parameter according to the Shannon entropy corresponding to each undersampling mode; perform a crossover operation and a replication operation on the maximum Shannon entropy model according to the control crossover probability parameter while keeping the parental individuals unchanged.

[0100] In some specific embodiments, the algorithm call unit is specifically configured to: select a preset number of offspring individuals and change the genes of the offspring individuals to complete the mutation operation on the maximum Shannon entropy model.

[0101] Figure 17 Shown is an electronic device 20 provided by an embodiment of the present application. The electronic device 20 may specifically further include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the seismic irregular observation method disclosed in any of the foregoing embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0102] In this embodiment, the power supply 23 is used to provide voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows is any communication protocol applicable to the technical solution of the present application, and no specific limitation is imposed on it here; the input / output interface 25 is used to obtain external input data or output data to the outside, and its specific interface type can be selected according to specific application requirements, and no specific limitation is made here.

[0103] In addition, as a carrier for storing resources, the memory 22 can be a read-only memory, a random access memory, a magnetic disk, an optical disc, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc. The storage method can be temporary storage or permanent storage.

[0104] Among them, the operating system 221 is used to manage and control each hardware device on the electronic device 20 and the computer program 222, which can be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program that can be used to complete the seismic irregular observation method executed by the electronic device 20 disclosed in any of the foregoing embodiments, the computer program 222 can further include computer programs that can be used to complete other specific tasks.

[0105] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the seismic irregular observation method disclosed above is implemented. For the specific steps of this method, reference can be made to the corresponding content disclosed in the foregoing embodiments, and details will not be repeated here.

[0106] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0107] The above has introduced in detail a seismic irregular observation method, device, equipment and medium provided by the present invention. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An irregular earthquake observation method, characterized in that, Including: Obtaining the undersampled data in the seismic data and constructing a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data; Determining a selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determining the maximum Shannon entropy according to the selection probability; Determining the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determining a target undersampling operator according to the target undersampling pattern; Determining the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete irregular seismic observations.

2. The seismic irregular observation method according to claim 1, wherein The constructing of the maximum Shannon entropy model according to the observation probability corresponding to the undersampled data includes: Determining the number of observation points corresponding to the observation points in the observation range of the undersampled data, and determining the number of Gaussian kernels based on the number of observation points; Superposing the observation probabilities according to the sampling pattern of the observation points and the number of Gaussian kernels to obtain a probability density function, and constructing a maximum Shannon entropy model based on the probability density function.

3. The seismic irregular observation method according to claim 1, characterized in that The determining of the selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model and determining the maximum Shannon entropy according to the selection probability includes: Constructing a selection probability model according to the maximum Shannon entropy model and a preset constant; Based on a preset genetic algorithm, performing selection operation, crossover operation, replication operation and mutation operation on the maximum Shannon entropy model by using the selection probability model to determine the maximum Shannon entropy.

4. The seismic irregular observation method according to claim 3, wherein Before the constructing of the selection probability model according to the maximum Shannon entropy model and a preset constant, it further includes: Determining the Shannon entropy corresponding to each undersampling pattern according to the maximum Shannon entropy model, and determining the preset constant according to the Shannon entropy corresponding to each undersampling pattern.

5. The seismic irregular observation method according to claim 3, characterized in that, Performing a crossover operation on the maximum Shannon entropy model includes: Performing a crossover operation on the maximum Shannon entropy model according to a preset control crossover level parameter.

6. The seismic irregular observation method according to claim 3, characterized in that, Performing a crossover operation and a replication operation on the maximum Shannon entropy model includes: Determining the Shannon entropy corresponding to each undersampling pattern according to the maximum Shannon entropy model, and determining a control crossover probability parameter according to the Shannon entropy corresponding to each undersampling pattern; Performing a crossover operation and a replication operation on the maximum Shannon entropy model according to the control crossover probability parameter while keeping the parent individuals unchanged.

7. The seismic irregular observation method according to claim 3, characterized in that Performing a mutation operation on the maximum Shannon entropy model includes: Selecting a preset number of offspring individuals and changing the genes of the offspring individuals to complete the mutation operation on the maximum Shannon entropy model.

8. An irregular seismic observation device, characterized in that, Including: A Shannon entropy model construction module, configured to obtain the undersampled data in the seismic data and construct a maximum Shannon entropy model according to the observation probability corresponding to the undersampled data; A maximum Shannon entropy determination module, configured to determine a selection probability based on a preset genetic algorithm according to the maximum Shannon entropy model, and determine the maximum Shannon entropy according to the selection probability; An undersampling operator determination module, configured to determine the target undersampling pattern corresponding to the maximum Shannon entropy according to the maximum Shannon entropy, and determine a target undersampling operator according to the target undersampling pattern; A full observation data determination module, configured to determine the full observation data corresponding to the undersampled data according to the undersampled data and the target undersampling operator to complete irregular seismic observations.

9. An electronic device, characterized in that, Including: A memory for storing a computer program; A processor for executing the computer program to implement the steps of the seismic irregular observation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, For storing a computer program; wherein, when the computer program is executed by a processor, the steps of the seismic irregular observation method according to any one of claims 1 to 7 are implemented.

Images