Earthquake facies type prediction method and device, equipment and medium

Abstract

The embodiment of the invention discloses a seismic facies type prediction method and device, equipment and a medium. The method comprises the following steps: acquiring pre-stack seismic data of a target horizon as first target seismic data; performing Hilbert transform on the first target seismic data to obtain second target seismic data; performing joint sparse representation on the second target seismic data based on a pre-constructed joint dictionary to obtain a target optimization function; wherein the target optimization function is used for representing a sparse reconstruction error between the first target seismic data and the joint dictionary; and determining a target sparse reconstruction error based on a solving result of the target optimization function, and determining a seismic facies type of the first target seismic data according to the target sparse reconstruction error. According to the technical scheme, the seismic facies type is predicted based on the pre-stack seismic data and the pre-constructed joint dictionary, the influence of the lithology or physical property and other spatial changes on the seismic reflection amplitude intensity can be effectively relieved, and the seismic facies analysis precision is improved when the impedance difference is small.

Description

Technical Field

[0001] This invention relates to the field of petroleum exploration and development technology, and in particular to a method, apparatus, equipment and medium for predicting seismic facies types. Background Technology

[0002] Seismic facies analysis methods are mainly divided into two categories: supervised and unsupervised methods. Supervised analysis methods require prior acquisition of label information from well sites and areas identified by geological experts, and are therefore typically used in developed oilfields with abundant well logging data or a deep understanding of geological patterns. Unsupervised analysis methods do not require any label data, using only available seismic data, and are therefore the most widely used in seismic interpretation.

[0003] Terrestrial sediments are not only lithologically complex, but also exhibit varying wave impedance values ​​for the same lithology due to variations in porosity, resulting in small differences in wave impedance between different lithologies and even overlapping relationships. Changes in lithology or physical properties will cause variations in the amplitude and waveform of reflected seismic waves. Both of the aforementioned seismic facies analysis methods utilize post-stack seismic waveforms to classify or cluster seismic facies, while also taking advantage of the impedance differences between different lithologies and the seismic amplitude and waveform response characteristics generated by structures. However, when the impedance differences are small, these methods face significant challenges in predicting the distribution of dominant sedimentary facies, resulting in low accuracy. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and medium for predicting seismic facies types. Based on pre-stack seismic data and a pre-constructed joint dictionary, it predicts seismic facies types, which can effectively mitigate the influence of spatial variations in lithology or physical properties on the intensity of seismic reflection amplitude and improve the accuracy of seismic facies analysis when impedance differences are small.

[0005] According to one aspect of the present invention, a method for predicting seismic facies types is provided, the method comprising:

[0006] Prestack seismic data of the target layer are acquired as the first target seismic data;

[0007] The first target seismic data is subjected to a Hilbert transform to obtain the second target seismic data; wherein the second target seismic data includes real part data and imaginary part data;

[0008] The second target seismic data is subjected to joint sparse representation based on a pre-constructed joint dictionary to obtain a target optimization function; wherein, the target optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary;

[0009] The target sparse reconstruction error is determined based on the solution result of the target optimization function, and the seismic phase type of the first target seismic data is determined based on the target sparse reconstruction error.

[0010] According to another aspect of the present invention, a device for predicting seismic phase types is provided, comprising:

[0011] The first target seismic data determination module is used to acquire pre-stack seismic data of the target layer as the first target seismic data;

[0012] The second target seismic data determination module is used to perform a Hilbert transform on the first target seismic data to obtain the second target seismic data; wherein, the second target seismic data includes real part data and imaginary part data;

[0013] The objective optimization function determination module is used to perform joint sparse representation of the second target seismic data based on a pre-constructed joint dictionary to obtain an objective optimization function; wherein, the objective optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary;

[0014] The seismic facies type determination module is used to determine the target sparse reconstruction error based on the solution result of the target optimization function, and to determine the seismic facies type of the first target seismic data according to the target sparse reconstruction error.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the seismic phase type prediction method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the seismic phase type prediction method according to any embodiment of the present invention.

[0020] The technical solution of this invention involves acquiring pre-stack seismic data of a target layer as first target seismic data; performing a Hilbert transform on the first target seismic data to obtain second target seismic data; wherein the second target seismic data includes real and imaginary data; performing joint sparse characterization on the second target seismic data based on a pre-constructed joint dictionary to obtain a target optimization function; wherein the target optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary; determining the target sparse reconstruction error based on the solution result of the target optimization function; and determining the seismic facies type of the first target seismic data based on the target sparse reconstruction error. This technical solution, based on pre-stack seismic data and a pre-constructed joint dictionary to predict seismic facies types, can effectively mitigate the influence of spatial variations in lithology or physical properties on the intensity of seismic reflection amplitude, and improve the accuracy of seismic facies analysis when impedance differences are small.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a method for predicting seismic facies types according to Embodiment 1 of the present invention;

[0024] Figure 2 This is a schematic diagram of the signal before and after the Hilbert transform according to Embodiment 1 of the present invention;

[0025] Figure 3 This is a histogram of the wave impedance distribution of sandstone and mudstone according to Embodiment 1 of the present invention;

[0026] Figure 4 This is a schematic diagram of a seismic profile through a well according to Embodiment 1 of the present invention;

[0027] Figure 5 This is a comparison diagram of seismic facies analysis results provided in Embodiment 1 of the present invention;

[0028] Figure 6 This is a flowchart of a method for predicting seismic phase types according to Embodiment 2 of the present invention;

[0029] Figure 7This is a schematic diagram of the structure of a seismic phase type prediction device provided in Embodiment 3 of the present invention;

[0030] Figure 8 This is a schematic diagram of the structure of an electronic device that implements a method for predicting seismic phase types according to an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first," "second," "target," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] Example 1

[0034] Figure 1 This is a flowchart of a seismic facies type prediction method provided in Embodiment 1 of the present invention. This embodiment is applicable to the accurate prediction of seismic facies types from seismic data. The method can be executed by a seismic facies type prediction device, which can be implemented in hardware and / or software and can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:

[0035] S110: Acquire pre-stack seismic data of the target layer as the first target seismic data.

[0036] In this embodiment, firstly, based on the target stratigraphic information obtained from drilling, and combined with the dominant frequency of seismic data, usable pre-stack seismic data is extracted along the target stratigraphic region within a determined analysis window size as the first target seismic data. The target stratigraphic region can be preset according to actual application needs; this embodiment does not impose specific limitations on it. It should be noted that, compared to post-stack seismic data, pre-stack seismic data carries richer information, thus being more advantageous for predicting seismic facies types.

[0037] S120: Perform a Hilbert transform on the first target seismic data to obtain the second target seismic data.

[0038] The second target seismic data includes real and imaginary data. In this embodiment, after determining the first target seismic data, a Hilbert transform is performed on the first target seismic data to obtain the second target seismic data. The Hilbert transform is an important tool in digital signal processing for signal analysis; it can perform a 90° phase shift on the original signal. The specific formula is as follows: Where s(t) represents the original signal and h(t) represents the response function. This represents the signal after the original signal has undergone the Hilbert transform. Specifically, after the Hilbert transform, the real part of the original signal represents its amplitude, while the imaginary part represents its phase. The Hilbert transform separates the amplitude and phase of the original signal, transforming a one-dimensional signal into a two-dimensional signal, combining the real and complex domains to aid in analyzing the true nature of the signal. Therefore, the second target seismic data obtained after the Hilbert transform includes both real and imaginary data.

[0039] Figure 2 This is a schematic diagram of the signal before and after the Hilbert transform, provided in Embodiment 1 of the present invention. Figure 2 (a) is the original signal. Figure 2 (b) is the signal after the Hilbert transform. For example... Figure 2 As shown, the original signal is at zero in the target stratum at the dashed line. After the Hilbert transform, the original signal is phase-shifted by 90° and is at its maximum value, making it more distinguishable. Therefore, it can effectively solve the zero-crossing problem, that is, effectively reduce the noise influence at the zero point.

[0040] S130: Based on the pre-constructed joint dictionary, the second target seismic data is jointly sparsely represented to obtain the target optimization function.

[0041] The objective optimization function characterizes the sparse reconstruction error between the first target seismic data and the joint dictionary. In this embodiment, after obtaining the second target seismic data, a joint sparse representation can be performed on the second target seismic data based on the pre-constructed joint dictionary to obtain the objective optimization function. The sparse representation can be understood as a signal y∈R M×L The dictionary D can be represented by a linear weighted sum of a few atoms. Here, D is an M×L matrix, where each unit column vector of length M is called an atom, and L represents the number of atoms in the dictionary. L >> M, meaning the number of atoms is much greater than the atomic dimension, thus ensuring the dictionary's overcompleteness; that is, all stratigraphic reflection structure features contained in the training seismic data can be included in the atoms of a complete dictionary. The objective optimization function can be used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary. For example, the objective optimization function can be expressed as:

[0042]

[0043]

[0044] Among them, D real and D imag These represent the real number field dictionary and the corresponding imaginary number field dictionary in the union dictionary (the real number field dictionary and the imaginary number field dictionary have the same number of atoms); Y real and Y imag Let represent the real and imaginary parts of the second target seismic data, respectively; ||·||0 is the l0 norm, used to represent the number of non-zero elements in the vector; K is a constant representing the sparsity; α is the sparsity coefficient (a multi-dimensional vector), and the dimension of α is equal to the number of atoms in the real domain dictionary (or imaginary domain dictionary); i represents the label corresponding to each element in α. It should be noted that the size of K can be preset according to actual needs, and the size of K can be used to limit the maximum number of iterations and the number of non-zero elements in α.

[0045] In this embodiment, optionally, the construction process of the joint dictionary includes: acquiring pre-stack seismic data of at least two candidate horizons as first sample seismic data; performing a Hilbert transform on the first sample seismic data to obtain second sample seismic data; performing joint sparse representation on the second sample seismic data to obtain a sample optimization function, and iteratively solving the sample optimization function based on the second sample seismic data; stopping the iteration if a preset termination condition is met during the iterative solution process, and determining a candidate joint dictionary based on the final iterative solution result; wherein, the candidate joint dictionary includes a real domain dictionary and its corresponding imaginary domain dictionary; performing clustering processing on the candidate joint dictionary to obtain at least two joint sub-dictionaries, and determining the joint dictionary based on the at least two joint sub-dictionaries; wherein, each joint sub-dictionary corresponds to a different seismic facies type.

[0046] In this embodiment, a joint dictionary needs to be constructed in advance. Specifically, firstly, pre-stack seismic data from at least two candidate horizons are acquired as the first sample seismic data. The determination method for the first sample seismic data can refer to the first target seismic data. Then, the first sample seismic data is subjected to a Hilbert transform according to the Hilbert transform formula mentioned above to obtain the second sample seismic data. Finally, the second sample seismic data is subjected to joint sparse representation to obtain the sample optimization function. The sample optimization function can be expressed as:

[0047]

[0048]

[0049] Among them, X real and X imag These represent the real and imaginary data from the second sample of seismic data. It's important to note that the sample optimization function ensures that the real and imaginary parts of each sample pair share the same sparse representation, thus introducing a correlation constraint between the real and imaginary parts. This joint sparse representation method can be called a two-box sparse representation. Therefore, the joint dictionary obtained after training is no longer independent but interconnected.

[0050] After determining the sample optimization function, iterative solutions can be performed based on the second sample seismic data. For example, the Orthogonal Matching Pursuit (OMP) algorithm based on a greedy criterion can be used to solve the sample optimization function. The OMP algorithm calculates α iteratively. In each iteration, an atom from the dictionary D that best matches the current residual signal is selected, thereby updating α to continuously approximate the second sample seismic data X. Where D = [D...] real D imag ], X = [X real ,X imag The residual signal refers to X in the sample optimization function. real -D reak *α and X imag -D imag *α.

[0051] This scheme employs a dictionary learning method based on Hilbert transform. The core of this method lies in the differences between pre-stack seismic data in the real and imaginary domains. In the real domain, amplitude characteristics are preserved, while in the imaginary domain, there is a 90° phase shift. Therefore, a joint dictionary of real and imaginary data can be learned from these two spaces, including a real domain dictionary D. real and the dictionary of imaginary fields D imag .

[0052] In this embodiment, optionally, iteratively solving the sample optimization function based on the second sample seismic data includes: determining initial sample data according to the second sample seismic data, and determining the initial sample data as the initial dictionary; wherein, the initial sample data includes real-domain data and imaginary-domain data, and the initial dictionary includes an initial real-domain dictionary and an initial imaginary-domain dictionary; determining the initial sparse coefficients of the sample optimization function according to the initial dictionary; updating the initial dictionary according to the initial sparse coefficients to obtain a candidate dictionary; determining candidate sparse coefficients of the sample optimization function according to the candidate dictionary, and updating the candidate dictionary according to the candidate sparse coefficients.

[0053] Exemplarily, assume that the first sample seismic data is represented as X = [x1, x2, x3... x n ∈ R k×m , after performing the Hilbert transform on X, the second sample seismic data X real and X imag are obtained. Randomly select k1 (k1 < k) columns from X real as the initial real-domain dictionary D 0_real , and correspondingly select k1 columns from X imag as the initial imaginary-domain dictionary D 0_imag , thus obtaining the initial dictionary D0 = [D 0_real , D 0_imag . Meanwhile, preset the maximum number of iterations and the reconstruction error threshold for dictionary learning (set according to actual application requirements). First, fix the initial dictionary D0, and use the OMP algorithm to obtain the sparse coefficients α i (i = 1…n) of each sample, thereby obtaining the initial sparse coefficients A0 = [α 1 , α 2 , α 3 …α n , then fix A0 and update D0 column by column to obtain a candidate dictionary D1. Furthermore, fix D1 to calculate the candidate sparse coefficients A1, and then fix A1 to update D1 column by column to obtain the next candidate dictionary D2. It should be noted that when updating the l-th atom d l of the dictionary, the other atoms in the dictionary remain fixed, but all non-zero sparse coefficients A l,i , i ∈ ω l = {j|1 ≤ j ≤ L, A ij ≠0} corresponding to the atom d l will be updated.

[0054] Iterate in the above manner, and judge whether the preset end condition is satisfied after each iteration. Among them, the preset end condition can be set as the sparse reconstruction error being less than the reconstruction error threshold or reaching the maximum number of iterations. Among them, the sparse reconstruction error refers to the If the preset termination condition is met, the iteration stops, and the final iterative solution is used as the candidate joint dictionary D. μ =[D μ_real D μ_imag ], where D μ_real and D μ_imag These represent the real number field dictionary and its corresponding imaginary number field dictionary in the candidate union dictionary, respectively.

[0055] To avoid information redundancy in the candidate joint dictionary, it can be further clustered to obtain at least two joint sub-dictionaries, and the joint dictionary is determined based on these two sub-dictionaries. Each joint sub-dictionary corresponds to a different seismic facies type. For example, seismic facies types can include river channels, oceans, and swamps. For instance, the SOM (self-organization of mapping) algorithm can be used to cluster the candidate joint dictionary. The SOM model contains only one input layer and one competition layer. The core idea of ​​SOM is to adopt a competitive learning strategy, where each neuron in the competition layer competes with each other for the right to update weights. Following this idea, each iteration updates one neuron and its neighboring neurons. SOM only needs to calculate the similarity between the sparse coefficient α and each corresponding w to find the winning neuron, thus obtaining the category corresponding to α. The similarity can be measured using cosine similarity, thus the optimization function of SOM clustering can be obtained as follows:

[0056]

[0057] Where j is the number of the winning neuron, w j Let T be the feature vector of the neuron, and T denote the transpose. The implementation process of the SOM algorithm is as follows: First, parameter preprocessing is performed, including parameter initialization and normalization. Specifically, from the candidate joint dictionary D... μ Randomly select k atoms d random As the initial weight of the competition layer, i.e. w j =d random and for d random Normalize the element values ​​to obtain the unit vector Simultaneously set the initial learning rate η0, the maximum number of iterations ε, and the learning rate threshold δ. Then calculate the unit neuron vector. And find the winning neuron Next, obtain the updated neighborhood radius of the winning neuron, find other neurons that fall within the updated neighborhood radius, and calculate the distance r between them and the winning neuron. i =||w i -w j 2. Then update the learning rate. And based on the updated learning rate η i Update the competition layer weights Finally, determine whether the maximum number of iterations ε has been reached or whether η is satisfied. i If the value is less than δ, the loop stops and the result of the last iteration is taken as the clustering result; otherwise, the loop continues until the stopping condition is met. Thus, at least two joint sub-dictionaries can be obtained, and a joint dictionary can be derived from these at least two joint sub-dictionaries.

[0058] In this embodiment, optionally, joint sparse representation of the second target seismic data is performed based on a pre-constructed joint dictionary to obtain the target optimization function, including: performing joint sparse representation of the second target seismic data based on at least two joint sub-dictionaries in the joint dictionary to obtain at least two target optimization functions.

[0059] In this embodiment, since the constructed joint dictionary includes at least two joint sub-dictionaries, when performing joint sparse representation of the second target seismic data based on the pre-constructed joint dictionary, it is necessary to perform joint sparse representation of the second target seismic data according to each joint sub-dictionary in the joint dictionary, thereby obtaining at least two target optimization functions.

[0060] S140, determine the target sparse reconstruction error based on the solution of the target optimization function, and determine the seismic phase type of the first target seismic data based on the target sparse reconstruction error.

[0061] In this embodiment, after obtaining the objective optimization function, the objective optimization function can be solved, and the objective sparse reconstruction error can be determined based on the solution result. Optionally, determining the objective sparse reconstruction error based on the solution result of the objective optimization function includes: solving at least two objective optimization functions separately, and determining at least two objective sparse reconstruction errors based on the solution results. In this embodiment, it is necessary to solve the at least two obtained objective optimization functions separately, and determine the corresponding objective sparse reconstruction error based on the solution result of each objective optimization function.

[0062] In this embodiment, after determining the target sparse reconstruction error, the seismic phase type of the first target seismic data can be determined based on the target sparse reconstruction error. Optionally, determining the seismic phase type of the first target seismic data based on the target sparse reconstruction error includes: using the joint sub-dictionary corresponding to the minimum value among at least two target sparse reconstruction errors as the target joint sub-dictionary; and determining the seismic phase type of the first target seismic data based on the seismic phase type corresponding to the target joint sub-dictionary.

[0063] In this embodiment, each objective optimization function corresponds to an objective sparse reconstruction error. The objective sparse reconstruction error corresponding to the joint sub-dictionary with the smallest value indicates that the seismic facies type of the first objective seismic data is closest to that of the joint sub-dictionary. At this point, it can be determined that the first objective seismic data belongs to the seismic facies category to which the joint sub-dictionary belongs. Therefore, when determining the seismic facies type of the first objective seismic data, the joint sub-dictionary corresponding to the minimum value among the objective sparse reconstruction errors can be selected as the objective joint sub-dictionary, and the seismic facies type corresponding to the objective joint sub-dictionary can be used as the seismic facies type of the first objective seismic data.

[0064] In this embodiment, data from the Jurassic Shaximiao Formation in a certain area of ​​the Sichuan Basin are used to verify the effectiveness of this scheme. The upper section of the Shaximiao Formation (i.e., the Sha-2 member) is dominated by shallow-water deltaic-fluvial facies deposits, with numerous river channels. Five wells within the seismic work area have logging data analysis showing that the acoustic impedance values ​​of the sandstone and mudstone overlap, with peak values ​​differing by only 400 m / s*g / cm². 3 ,like Figure 3 As shown. Figure 4 This is a schematic diagram of a seismic profile through a well, provided in Embodiment 1 of the present invention. Figure 4 The well logging curves are superimposed; the left curve represents GR (natural gamma ray), and the right curve represents P-wave velocity. The location of the connecting well profile can be found in [reference needed]. Figure 5 . Figure 5 This is a comparison diagram of seismic facies analysis results provided in Embodiment 1 of the present invention. Wherein, Figure 5 (a) Seismic facies analysis results obtained using the traditional SOM method. Figure 5 (b) The results of seismic facies analysis obtained using this scheme.

[0065] Using this method, the optimal number of categories (i.e., the optimal clustering result) for the above test area can be determined to be 10 categories. This method is used to perform a 30ms window on the target layer (…). Figure 4 Seismic facies analysis was performed on pre-stack seismic data (middle and upper horizon lines), and the results are shown in […]. Figure 5 (b) It was found that the river channel features were significantly different from the background, and multiple river channels in different directions could be quickly and clearly identified. Figure 5 (a) shows the seismic facies analysis results obtained using the traditional SOM method. Compared with the seismic facies analysis results obtained by this scheme, there is more background noise, and the river channel features appear more blurred due to the influence of noise.

[0066] The technical solution of this invention involves acquiring pre-stack seismic data of a target layer as first target seismic data; performing a Hilbert transform on the first target seismic data to obtain second target seismic data; wherein the second target seismic data includes real and imaginary data; performing joint sparse characterization on the second target seismic data based on a pre-constructed joint dictionary to obtain a target optimization function; wherein the target optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary; determining the target sparse reconstruction error based on the solution result of the target optimization function; and determining the seismic facies type of the first target seismic data based on the target sparse reconstruction error. This technical solution, based on pre-stack seismic data and a pre-constructed joint dictionary to predict seismic facies types, can effectively mitigate the influence of spatial variations in lithology or physical properties on the intensity of seismic reflection amplitude, and improve the accuracy of seismic facies analysis when impedance differences are small.

[0067] In this embodiment, optionally, before performing Hilbert transform on the first sample seismic data, the method further includes: performing data preprocessing on the first sample seismic data; wherein, data preprocessing includes data denoising and outlier removal; correspondingly, performing Hilbert transform on the first sample seismic data includes: performing Hilbert transform on the first sample seismic data after data preprocessing.

[0068] In this embodiment, to reduce the impact of outlier data and noise on data quality, the first sample seismic data can be preprocessed before performing Hilbert transform. This preprocessing includes data denoising and outlier removal, thereby eliminating outlier data in the first sample seismic data. The data after outlier removal is then subjected to noise reduction processing, which effectively improves data quality. For example, filtering methods (such as low-pass or high-pass filtering) can be used for data preprocessing.

[0069] This solution, through such a setup, can reduce the impact of outliers and noise on data quality by preprocessing data, thereby effectively improving data quality.

[0070] Example 2

[0071] Figure 6 This is a flowchart of a method for predicting seismic facies types according to Embodiment 2 of the present invention. This embodiment is an optimization based on the above embodiment.

[0072] like Figure 6 As shown, the method in this embodiment specifically includes the following steps:

[0073] S210: Obtain pre-stack seismic data of the target layer as the first target seismic data.

[0074] S220, perform Hilbert transform on the first target seismic data to obtain the second target seismic data.

[0075] The second target seismic data includes real data and imaginary data.

[0076] S230, based on at least two joint sub-dictionaries in the joint dictionary, performs joint sparse representation on the second target seismic data to obtain at least two objective optimization functions.

[0077] The objective optimization function characterizes the sparse reconstruction error between the first target seismic data and the joint dictionary. The joint dictionary comprises at least two joint sub-dictionaries, each corresponding to a different seismic facies type.

[0078] S240, solve at least two objective optimization functions respectively, and determine at least two objective sparse reconstruction errors based on the solution results.

[0079] S250, take the joint sub-dictionary corresponding to the minimum value among at least two target sparse reconstruction errors as the target joint sub-dictionary.

[0080] S260, determine the seismic phase type of the first target seismic data based on the seismic phase type corresponding to the target joint sub-dictionary.

[0081] The technical solution of this invention, based on pre-stack seismic data and a pre-constructed joint dictionary to predict seismic facies types, can effectively mitigate the influence of spatial variations in lithology or physical properties on the intensity of seismic reflection amplitude, and improve the accuracy of seismic facies analysis when impedance differences are small. Furthermore, determining the seismic facies type of the first target seismic data based on the target sparse reconstruction error corresponding to at least two joint sub-dictionaries in the joint dictionary can effectively improve the efficiency of seismic facies analysis.

[0082] Example 3

[0083] Figure 7 This is a schematic diagram of a seismic facies type prediction device provided in Embodiment 3 of the present invention. This device can execute the seismic facies type prediction method provided in any embodiment of the present invention, and possesses the corresponding functional modules and beneficial effects for executing the method. For example... Figure 7 As shown, the device includes:

[0084] The first target seismic data determination module 310 is used to acquire pre-stack seismic data of the target layer as the first target seismic data;

[0085] The second target seismic data determination module 320 is used to perform a Hilbert transform on the first target seismic data to obtain second target seismic data; wherein, the second target seismic data includes real part data and imaginary part data;

[0086] The objective optimization function determination module 330 is used to perform joint sparse characterization on the second target seismic data based on a pre-built joint dictionary to obtain an objective optimization function; wherein, the objective optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary;

[0087] The seismic facies type determination module 340 is used to determine the target sparse reconstruction error based on the solution result of the target optimization function, and to determine the seismic facies type of the first target seismic data according to the target sparse reconstruction error.

[0088] Optionally, the device further includes:

[0089] The first sample seismic data determination module is used to acquire pre-stack seismic data from at least two candidate horizons as the first sample seismic data.

[0090] The second sample seismic data determination module is used to perform Hilbert transform on the first sample seismic data to obtain the second sample seismic data.

[0091] The sample optimization function iterative solution module is used to perform joint sparse characterization on the second sample seismic data to obtain the sample optimization function, and to iteratively solve the sample optimization function based on the second sample seismic data;

[0092] The candidate joint dictionary determination module is used to stop iterating if a preset termination condition is met during the iterative solution process, and to determine the candidate joint dictionary based on the final iterative solution result; wherein, the candidate joint dictionary includes a real number field dictionary and its corresponding imaginary number field dictionary;

[0093] The joint dictionary determination module is used to perform clustering processing on the candidate joint dictionaries to obtain at least two joint sub-dictionaries, and determine the joint dictionary based on the at least two joint sub-dictionaries; wherein each joint sub-dictionary corresponds to a different seismic facies type.

[0094] Optionally, the device further includes:

[0095] The data preprocessing module is used to preprocess the first sample seismic data before performing Hilbert transform on the first sample seismic data; wherein, the data preprocessing includes data denoising and outlier removal;

[0096] Accordingly, the second sample seismic data determination module is also used for:

[0097] Hilbert transform was performed on the first sample of seismic data after data preprocessing.

[0098] Optionally, the sample optimization function iterative solution module is used for:

[0099] Initial sample data is determined based on the second sample seismic data, and the initial sample data is used as an initial dictionary; wherein, the initial sample data includes real number field data and imaginary number field data, and the initial dictionary includes an initial real number field dictionary and an initial imaginary number field dictionary;

[0100] The initial sparse coefficients of the sample optimization function are determined based on the initial dictionary;

[0101] The initial dictionary is updated based on the initial sparsity coefficients to obtain a candidate dictionary;

[0102] Candidate sparse coefficients of the sample optimization function are determined based on the candidate dictionary, and the candidate dictionary is updated based on the candidate sparse coefficients.

[0103] Optionally, the objective optimization function determination module 330 is used for:

[0104] Based on at least two joint sub-dictionaries in the joint dictionary, joint sparse representations are performed on the second target seismic data to obtain at least two objective optimization functions.

[0105] Optionally, the seismic phase type determination module 340 is used for:

[0106] Solve the at least two objective optimization functions respectively, and determine the at least two objective sparse reconstruction errors based on the solution results.

[0107] Optionally, the seismic phase type determination module 340 is further configured to:

[0108] The joint sub-dictionary corresponding to the minimum value among the at least two target sparse reconstruction errors is taken as the target joint sub-dictionary;

[0109] The seismic phase type of the first target seismic data is determined based on the seismic phase type corresponding to the target joint sub-dictionary.

[0110] The seismic facies type prediction device provided in this embodiment of the invention can execute the seismic facies type prediction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0111] Example 4

[0112] Figure 8A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0113] like Figure 8 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0114] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0115] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for predicting seismic facies types.

[0116] In some embodiments, the seismic facies type prediction method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the seismic facies type prediction method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the seismic facies type prediction method by any other suitable means (e.g., by means of firmware).

[0117] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0118] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0119] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0120] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0121] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0122] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0123] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0124] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for predicting seismic facies types, characterized in that, The method includes: Prestack seismic data of the target layer are acquired as the first target seismic data; The first target seismic data is subjected to a Hilbert transform to obtain the second target seismic data; wherein the second target seismic data includes real part data and imaginary part data; The second target seismic data is subjected to joint sparse representation based on a pre-constructed joint dictionary to obtain a target optimization function; wherein, the target optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary; The target sparse reconstruction error is determined based on the solution result of the target optimization function, and the seismic phase type of the first target seismic data is determined based on the target sparse reconstruction error.

2. The method according to claim 1, characterized in that, The process of constructing the union dictionary includes: Acquire pre-stack seismic data from at least two candidate horizons as the first sample seismic data; Perform a Hilbert transform on the first sample of seismic data to obtain the second sample of seismic data; The second sample seismic data is subjected to joint sparse characterization to obtain the sample optimization function, and the sample optimization function is solved iteratively based on the second sample seismic data; If a preset termination condition is met during the iterative solution process, the iteration stops, and a candidate joint dictionary is determined based on the final iterative solution result; wherein, the candidate joint dictionary includes a real number field dictionary and its corresponding imaginary number field dictionary; The candidate joint dictionary is clustered to obtain at least two joint sub-dictionaries, and the joint dictionary is determined based on the at least two joint sub-dictionaries; wherein each joint sub-dictionary corresponds to a different seismic facies type.

3. The method according to claim 2, characterized in that, Before performing the Hilbert transform on the first sample of seismic data, the following steps are also included: The first sample of seismic data is preprocessed; wherein, the data preprocessing includes data denoising and outlier removal; Accordingly, a Hilbert transform is performed on the first sample seismic data, including: Hilbert transform was performed on the first sample of seismic data after data preprocessing.

4. The method according to claim 2, characterized in that, The sample optimization function is iteratively solved based on the second sample seismic data, including: Initial sample data is determined based on the second sample seismic data, and the initial sample data is used as an initial dictionary; wherein, the initial sample data includes real number field data and imaginary number field data, and the initial dictionary includes an initial real number field dictionary and an initial imaginary number field dictionary; The initial sparse coefficients of the sample optimization function are determined based on the initial dictionary; The initial dictionary is updated based on the initial sparsity coefficients to obtain a candidate dictionary; Candidate sparse coefficients of the sample optimization function are determined based on the candidate dictionary, and the candidate dictionary is updated based on the candidate sparse coefficients.

5. The method according to any one of claims 2-4, characterized in that, Based on a pre-constructed joint dictionary, the second target seismic data is jointly sparsely represented to obtain the target optimization function, including: Based on at least two joint sub-dictionaries in the joint dictionary, joint sparse representations are performed on the second target seismic data to obtain at least two objective optimization functions.

6. The method according to claim 5, characterized in that, The objective sparse reconstruction error is determined based on the solution results of the objective optimization function, including: Solve the at least two objective optimization functions respectively, and determine the at least two objective sparse reconstruction errors based on the solution results.

7. The method according to claim 6, characterized in that, Determining the seismic facies type of the first target seismic data based on the target sparse reconstruction error includes: The joint sub-dictionary corresponding to the minimum value among the at least two target sparse reconstruction errors is taken as the target joint sub-dictionary; The seismic phase type of the first target seismic data is determined based on the seismic phase type corresponding to the target joint sub-dictionary.

8. A device for predicting seismic phase types, characterized in that, The device includes: The first target seismic data determination module is used to acquire pre-stack seismic data of the target layer as the first target seismic data; The second target seismic data determination module is used to perform a Hilbert transform on the first target seismic data to obtain the second target seismic data; wherein, the second target seismic data includes real part data and imaginary part data; The objective optimization function determination module is used to perform joint sparse representation of the second target seismic data based on a pre-constructed joint dictionary to obtain an objective optimization function; wherein, the objective optimization function is used to characterize the sparse reconstruction error between the first target seismic data and the joint dictionary; The seismic facies type determination module is used to determine the target sparse reconstruction error based on the solution result of the target optimization function, and to determine the seismic facies type of the first target seismic data according to the target sparse reconstruction error.

9. The apparatus according to claim 8, characterized in that, The device further includes: The first sample seismic data determination module is used to acquire pre-stack seismic data from at least two candidate horizons as the first sample seismic data. The second sample seismic data determination module is used to perform Hilbert transform on the first sample seismic data to obtain the second sample seismic data. The sample optimization function iterative solution module is used to perform joint sparse characterization on the second sample seismic data to obtain the sample optimization function, and to iteratively solve the sample optimization function based on the second sample seismic data; The candidate joint dictionary determination module is used to stop iterating if a preset termination condition is met during the iterative solution process, and to determine the candidate joint dictionary based on the final iterative solution result; wherein, the candidate joint dictionary includes a real number field dictionary and its corresponding imaginary number field dictionary; The joint dictionary determination module is used to perform clustering processing on the candidate joint dictionaries to obtain at least two joint sub-dictionaries, and determine the joint dictionary based on the at least two joint sub-dictionaries; wherein each joint sub-dictionary corresponds to a different seismic facies type.

10. The apparatus according to claim 9, characterized in that, The device further includes: The data preprocessing module is used to preprocess the first sample seismic data before performing Hilbert transform on the first sample seismic data; wherein, the data preprocessing includes data denoising and outlier removal; Accordingly, the second sample seismic data determination module is also used for: Hilbert transform was performed on the first sample of seismic data after data preprocessing.

11. The apparatus according to claim 9, characterized in that, The sample optimization function iterative solution module is used for: Initial sample data is determined based on the second sample seismic data, and the initial sample data is used as an initial dictionary; wherein, the initial sample data includes real number field data and imaginary number field data, and the initial dictionary includes an initial real number field dictionary and an initial imaginary number field dictionary; The initial sparse coefficients of the sample optimization function are determined based on the initial dictionary; The initial dictionary is updated based on the initial sparsity coefficients to obtain a candidate dictionary; Candidate sparse coefficients of the sample optimization function are determined based on the candidate dictionary, and the candidate dictionary is updated based on the candidate sparse coefficients.

12. The apparatus according to any one of claims 9-11, characterized in that, The objective optimization function determination module is used for: Based on at least two joint sub-dictionaries in the joint dictionary, joint sparse representations are performed on the second target seismic data to obtain at least two objective optimization functions.

13. The apparatus according to claim 12, characterized in that, The seismic phase type determination module is used for: Solve the at least two objective optimization functions respectively, and determine the at least two objective sparse reconstruction errors based on the solution results.

14. The apparatus according to claim 13, characterized in that, The seismic facies type determination module is also used for: The joint sub-dictionary corresponding to the minimum value among the at least two target sparse reconstruction errors is taken as the target joint sub-dictionary; The seismic phase type of the first target seismic data is determined based on the seismic phase type corresponding to the target joint sub-dictionary.

15. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the seismic facies type prediction method according to any one of claims 1-7.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method for predicting seismic phase types according to any one of claims 1-7.

Images