Wave impedance inversion model training method, wave impedance inversion method and electronic device

Abstract

The present application relates to oil and gas geophysical exploration technology field, disclose a kind of wave impedance inversion model training method, wave impedance inversion method and electronic equipment, method includes according to poststack seismic data and each well's well logging data, by wideband Ricker wavelet constructs wideband wavelet library;According to target interval well logging data establishes well logging interpolation impedance model;According to well logging interpolation impedance model and wideband wavelet library, constructs noiseless wideband high-resolution data set and noisy wideband high-resolution data set;According to noiseless wideband high-resolution data set, noisy wideband high-resolution data set and well logging interpolation impedance model, constructs training sample set;The generative adversarial network for wave impedance inversion is constructed;Determine the objective function of wave impedance inversion;According to training sample set and objective function, the generative adversarial network is trained, obtains wave impedance inversion model.The present application can carry out high-resolution impedance inversion of thin reservoir, effectively improve the inversion accuracy of thin reservoir.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas geophysical exploration, and particularly relates to a wave impedance inversion model training method, a wave impedance inversion method and electronic equipment. BACKGROUND

[0002] The tight sandstone reservoir has obvious layering and the characteristic of thin reservoir thickness. With the deepening of oil exploration and development, the ability to identify thin layers by using wave impedance inversion has become an important means of reservoir prediction. Conventional wave impedance inversion methods include two categories: one is deterministic method, including sparse pulse inversion method; the other is statistical method, including statistical sampling of posterior distribution based on Bayesian theory framework.

[0003] Because the seismic data is band-limited, the wave impedance inversion has multiple solutions. Both the deterministic method and the statistical method need prior information to regularize the inversion process, and usually make some simple prior assumptions, such as sparsity assumption or minimum length assumption. However, these prior assumptions lead to excessive smoothing of the wave impedance inversion result, and the resolution of thin reservoirs is limited.

[0004] In recent years, the deep learning method has developed rapidly, providing a new idea for solving data-driven seismic inversion. It can directly learn the mapping function of seismic data and target parameters from the training data set end to end, and has achieved good results in some application examples. However, the data-driven deep learning method has a high degree of dependence on the training sample set, and needs to collect a large amount of observation data and high-resolution data pairs in advance to train the network.

[0005] Therefore, there is an urgent need for a wave impedance inversion model training method and a wave impedance inversion method to solve the above technical problems. SUMMARY

[0006] In view of the above problems, the present application provides a wave impedance inversion model training method, a wave impedance inversion method and electronic equipment, which can train a wave impedance inversion model capable of high-resolution impedance inversion of thin reservoirs without pre-collecting a large amount of observation data and high-resolution data pairs, thereby improving the inversion accuracy of thin reservoirs.

[0007] The present application provides a wave impedance inversion model training method, which comprises:

[0008] A wideband wavelet library is constructed from the post-stack seismic data and the logging data of each well by using wideband Ricker wavelets; wherein the wideband wavelet library includes wideband wavelets of each well and a comprehensive wideband wavelet;

[0009] Target layer logging data is obtained from the logging data of each well, and a logging interpolation impedance model is established according to the target layer logging data;

[0010] construct a noise-free wideband high-resolution data set and a noisy wideband high-resolution data set according to the well interpolation impedance model and the wideband wavelet library;

[0011] construct a training sample set for training the wave impedance inversion model according to the noise-free wideband high-resolution data set, the noisy wideband high-resolution data set and the well interpolation impedance model;

[0012] construct a generative adversarial network for wave impedance inversion according to the data structure characteristics of the training sample set and the post-stack seismic data;

[0013] determine a target function of the wave impedance inversion according to the preset wave impedance inversion requirement;

[0014] train the generative adversarial network according to the training sample set and the target function, and obtain the trained wave impedance inversion model.

[0015] Further, the wideband wavelet library is constructed by the wideband Ricker wavelet according to the post-stack seismic data and the well logging data of each well, including:

[0016] The wideband wavelet library is constructed by the wideband Ricker wavelet according to the post-stack seismic data and the well logging data of each well, including:

[0017] Further, the well interpolation impedance model is established according to the well logging data of the target layer, including:

[0018] The seismic horizon is obtained by horizon interpretation of the post-stack seismic data;

[0019] The stratigraphic framework is built according to the seismic horizon, and the well logging data of the target layer is interpolated along the stratigraphic framework by the Kriging method to obtain the well interpolation impedance model at the preset sampling time interval.

[0020] Further, the noise-free wideband high-resolution data set and the noisy wideband high-resolution data set are constructed according to the well interpolation impedance model and the wideband wavelet library, including:

[0021] The reflection coefficient model is generated according to the well interpolation impedance model, and each wavelet in the wideband wavelet library is respectively convolved with the reflection coefficient model to generate the noise-free wideband high-resolution data corresponding to each wavelet, so as to construct the noise-free wideband high-resolution data set;

[0022] The preset Gaussian noise is added to the noise-free wideband high-resolution data to generate the noisy wideband high-resolution data corresponding to each wavelet, so as to construct the noisy wideband high-resolution data set;

[0023] The preset Gaussian noise includes Gaussian noise having the same signal-to-noise ratio as the post-stack seismic data.

[0024] Furthermore, a training sample set is constructed based on the noise-free broadband high-resolution data set, the noisy broadband high-resolution data set, and the well logging interpolation impedance model, including:

[0025] Each piece of broadband high-resolution data in the noise-free broadband high-resolution data set and the noisy broadband high-resolution data set is normalized to obtain multiple input data, and the logging interpolation impedance model is used as label data to construct a training sample set.

[0026] The present invention also provides a wave impedance inversion method, the method comprising:

[0027] Normalizing the post-stack seismic data, and inputting the normalized post-stack seismic data into a preset wave impedance inversion model to perform wave impedance inversion and obtain wave impedance inversion results;

[0028] The preset wave impedance inversion model is obtained according to the above-mentioned wave impedance inversion model training method.

[0029] The present invention also provides a wave impedance inversion model training device, the device comprising:

[0030] A broadband wavelet library construction module is used to construct a broadband wavelet library using broadband Ricker wavelets based on post-stack seismic data and well logging data of each well; wherein the broadband wavelet library includes broadband wavelets of each well and integrated broadband wavelets;

[0031] The well logging interpolation impedance model construction module is used to obtain the target layer logging data from the well logging data of each well and establish the well logging interpolation impedance model based on the target layer logging data;

[0032] A broadband high-resolution data generation module is used to construct noise-free broadband high-resolution data sets and noisy broadband high-resolution data sets based on the well logging interpolation impedance model and the broadband wavelet library;

[0033] A training sample set construction module is used to construct a training sample set for training a wave impedance inversion model based on a noise-free broadband high-resolution data set, a noisy broadband high-resolution data set, and a well logging interpolation impedance model;

[0034] A generative adversarial network construction module is used to construct a generative adversarial network for wave impedance inversion based on the data structure characteristics of the training sample set and post-stack seismic data;

[0035] An objective function determination module is used to determine the objective function of wave impedance inversion according to preset wave impedance inversion requirements;

[0036] The model training module is configured to train the generative adversarial network according to a training sample set and an objective function, and obtain a trained wave impedance inversion model.

[0037] The present application also provides a wave impedance inversion device, which comprises:

[0038] The normalization processing module is configured to perform normalization processing on the stacked seismic data.

[0039] The wave impedance inversion module is configured to input the normalized stacked seismic data into a preset wave impedance inversion model to perform wave impedance inversion, and obtain a wave impedance inversion result; wherein the preset wave impedance inversion model is obtained according to the wave impedance inversion model training method.

[0040] The present application also provides a computer readable storage medium storing a computer program, which, when executed by one or more processors, implements the steps of the above method.

[0041] The present application also provides an electronic device comprising a memory and one or more processors, wherein the memory stores a computer program, and when the computer program is executed by the one or more processors, the steps of the above method are performed.

[0042] The wave impedance inversion model training method, the wave impedance inversion method and the electronic device provided by the present application have at least the following beneficial effects:

[0043] (1) The wave impedance inversion model training method provided by the present application does not need to collect a large amount of observation data and high-resolution data pairs in advance during training, but only needs to construct a wideband wavelet library according to the stacked seismic data and the logging data of each well, and then obtain a noise-free wideband high-resolution data set, a noisy wideband high-resolution data set and a logging interpolation impedance model, so as to construct a training sample set to train the generative adversarial network and obtain a wave impedance inversion model with high precision.

[0044] (2) The wave impedance inversion method provided by the present application performs high-resolution impedance inversion on a thin reservoir according to the trained wave impedance inversion model, which can effectively improve the inversion accuracy of the thin reservoir. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0046] It should be further noted that, for the convenience of description, only the parts related to the present application are shown in the drawings. The drawings that form part of the present application are used to provide further understanding of the present application, the schematic embodiments in the present application and the description thereof are used to explain the present application and do not constitute undue limitation on the present application, in the drawings:

[0047] Figure 1 The wave impedance inversion model training method step flow chart provided in the embodiment one of the present application;

[0048] Figure 2 The wave impedance inversion model training device structure schematic diagram provided in the embodiment three of the present application;

[0049] Figure 3 The wave impedance inversion device structure schematic diagram provided in the embodiment four of the present application;

[0050] Figure 4 The wave impedance inversion model training and wave impedance inversion flow chart provided in the embodiment five of the present application;

[0051] Figure 5 The time domain post-stack seismic profile provided in the embodiment five of the present application;

[0052] Figure 6 The wideband Ricker wavelet waveform diagram provided in the embodiment five of the present application;

[0053] Figure 7 The wideband Ricker wavelet and the same main frequency conventional Ricker wavelet spectrum comparison diagram provided in the embodiment five of the present application;

[0054] Figure 8 The time domain well logging interpolation impedance model profile schematic diagram provided in the embodiment five of the present application;

[0055] Figure 9 The time domain reflection coefficient model profile schematic diagram provided in the embodiment five of the present application;

[0056] Figure 10 The time domain noise-free wideband high-resolution seismic profile schematic diagram provided in the embodiment five of the present application;

[0057] Figure 11 The preset generative adversarial network structure schematic diagram provided in the embodiment five of the present application;

[0058] Figure 12 The generative adversarial neural network structure training flow chart provided in the embodiment five of the present application;

[0059] Figure 13 The inversion impedance profile schematic diagram provided in the embodiment five of the present application obtained by using the method of the present application;

[0060] Figure 14 Fig. 6 is a schematic diagram of the inversion impedance profile obtained by using the conventional method provided in the embodiment five of the present application;

[0061] Figure 15 Fig. 8 is a schematic diagram of the electronic device structure provided in the embodiment eight of the present application;

[0062] Reference signs:

[0063] Figure 2 In the figure, 201 is a wideband wavelet library construction module, 202 is a well logging interpolation impedance model construction module, 203 is a wideband high resolution data generation module, 204 is a training sample set construction module, 205 is a generative adversarial network construction module, 206 is a target function determination module, and 207 is a model training module.

[0064] Figure 3 In the figure, 301 is a normalization processing module, and 302 is a wave impedance inversion module.

[0065] Figure 15 In the figure, 1500 is an electronic device, 1501 is a processor, 1502 is a communication bus, 1503 is a user interface, 1504 is a communication interface, and 1505 is a memory. DETAILED DESCRIPTION

[0066] The present application will be further described below with reference to the embodiments shown in the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and are not all the embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.

[0067] It should be noted that: unless otherwise specifically stated, the relative arrangement, numerical expressions and values of the components and steps set forth in these embodiments do not limit the scope of the present application.

[0068] It should also be understood that, for any component, data or structure mentioned in the embodiments of the present application, unless specifically limited or given a contrary implication by the context, it can be understood as one or more in general.

[0069] In addition, the term "and / or" in the present disclosure is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the character " / " in the present disclosure generally represents an "or" relationship between the front and rear associated objects.

[0070] It should also be understood that the description of the embodiments of the present application emphasizes the differences between the embodiments, and the same or similar parts can be referred to each other, and for the sake of brevity, they will not be repeated.

[0071] The following description of at least one exemplary embodiment is merely exemplary in nature and is in no way intended to limit the application or its application and uses.

[0072] The technology, methods, and apparatus known to those of ordinary skill in the relevant art(s) can not be discussed in detail herein, but should be considered as part of the specification for purposes of the present disclosure.

[0073] Embodiments of the application can be applied to terminal devices, computer systems, servers, and the like electronic devices, which can operate with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations that can be suitable for use with terminal devices, computer systems, servers, and the like electronic devices include, but are not limited to, personal computers, server computers, thin clients, thick clients, hand-held or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputers, mainframe computers, and distributed cloud computing technology environments that include any of the above systems, and the like.

[0074] Terminal devices, computer systems, servers, and the like electronic devices can be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules can include routines, programs, objects, components, logic, data structures, and the like, which perform particular tasks or implement particular abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules can be located in local or remote computer system storage media including storage devices.

[0075] Example One

[0076] In Embodiment One of the application, as shown in Figure 1 A wave impedance inversion model training method is provided, and the method specifically includes the following steps:

[0077] Step S101: Construct a wideband wavelet library according to the post-stack seismic data and the logging data of each well.

[0078] Step S102: Obtain target layer logging data from the logging data of each well, and establish a logging interpolation impedance model according to the target layer logging data.

[0079] Step S103: Construct a noise-free wideband high-resolution data set and a noisy wideband high-resolution data set according to the logging interpolation impedance model and the wideband wavelet library.

[0080] Step S104: constructing a training sample set for training the wave impedance inversion model according to the noise-free wideband high-resolution data set, the noisy wideband high-resolution data set, and the well logging interpolation impedance model.

[0081] Step S105: constructing a generative adversarial network for wave impedance inversion according to the training sample set and the data structure characteristics of the post-stack seismic data.

[0082] Step S106: determining a target function of the wave impedance inversion according to a preset wave impedance inversion requirement.

[0083] Step S107: training the generative adversarial network according to the training sample set and the target function, and obtaining a trained wave impedance inversion model.

[0084] Optionally, in step S101, the wideband wavelet library includes wideband wavelets of the wells and a comprehensive wideband wavelet.

[0085] According to the post-stack seismic data and the well logging data of the wells, the wideband wavelet library is constructed by wideband Ricker wavelets, including:

[0086] According to the post-stack seismic data and the well logging data of the wells, the wells are calibrated by wideband Ricker wavelets, the wideband wavelets of the wells are determined, and the comprehensive wideband wavelet is determined according to the wideband wavelets of the wells, so as to construct the wideband wavelet library.

[0087] Specifically, in an implementation manner, the comprehensive wideband wavelet determined according to the wideband wavelets of the wells can include:

[0088] According to the wideband wavelets of the wells, the average value of the wideband wavelets is calculated and obtained, and the average value of the wideband wavelets is taken as the comprehensive wideband wavelet.

[0089] More specifically, in step S101, the Ricker wavelet R(t) is known, the wideband Ricker wavelet Y(t) is synthesized by Ricker wavelets with different widths, and is defined as:

[0090]

[0091]

[0092] Wherein, q and p are the upper and lower limits of the integral of the parameter f0 in the Ricker wavelet R(t).

[0093] In step S101, the wideband Ricker wavelet library is constructed by the wideband Ricker wavelet, which is a good approximation of the impulse function, has a relatively narrow main lobe, a small side lobe amplitude, and a simple waveform. In the case of the same main lobe width, the actual resolution of the wideband Ricker wavelet is higher.

[0094] It should be understood that the stacked seismic data is stacked seismic data in the study area, and the logging data of each well is logging data of each well in the study area.

[0095] Optionally, in step S102, a logging interpolation impedance model is established according to the logging data of the target interval, including:

[0096] According to the stacked seismic data, horizon interpretation is performed to obtain seismic horizons.

[0097] According to the seismic horizons, a stratigraphic framework is built, and the logging data of the target interval is interpolated along the stratigraphic framework by using the Kriging method to obtain a logging interpolation impedance model under a preset sampling time interval.

[0098] It should be noted that the preset sampling time interval is not specially limited, and a skilled person can set it according to actual needs.

[0099] In one implementation, the preset sampling time interval can be set to 1 ms. That is, a logging interpolation impedance model with 1 ms sampling is obtained.

[0100] In addition, the target interval mentioned in step S102 refers to the interval where the tight sandstone reservoir is located.

[0101] Specifically, the impedance value at the i th wellbore is z (x i , the weight λ i of Kriging interpolation is set, and then the impedance at the interpolation point is:

[0102]

[0103] The above formula is the expression of the logging interpolation impedance model, wherein z * (x0) is the impedance at the interpolation point, and n is the total number of wells in the study area.

[0104] Optionally, in step S103, according to the logging interpolation impedance model and the wideband wavelet library, a noise-free wideband high-resolution data set and a noisy wideband high-resolution data set are constructed, including:

[0105] According to the logging interpolation impedance model, a reflection coefficient model is generated, and each wavelet in the wideband wavelet library is respectively convolved with the reflection coefficient model to generate noise-free wideband high-resolution data corresponding to each wavelet, so as to construct a noise-free wideband high-resolution data set;

[0106] In the noise-free wideband high-resolution data, a preset Gaussian noise is added to generate noisy wideband high-resolution data corresponding to each wavelet, so as to construct a noisy wideband high-resolution data set;

[0107] The preset Gaussian noise includes Gaussian noise with the same signal-to-noise ratio as the stacked seismic data.

[0108] In step S103, the reflection coefficient model is respectively combined with the wideband wavelet convolution of each well and the comprehensive wideband wavelet convolution to obtain noise-free wideband high-resolution data, the training samples are expanded and the diversity of the samples is improved. The noise-free wideband high-resolution data is added with Gaussian noise with the same signal-to-noise ratio as the original post-stack seismic data to obtain noisy wideband high-resolution data. The noisy wideband high-resolution data is used as input data in subsequent model training, thereby improving the robustness and generalization ability of the generative adversarial network.

[0109] The Gaussian noise is a noise generated by adding a random value with a normal distribution of zero mean and standard deviation to input data, and is a continuous probability distribution defined by a probability density function (PDF):

[0110]

[0111] Wherein, x is a random variable, sigma is a standard deviation, and u is a mean value. In the present application, u is 0.

[0112] Optionally, in step S104, a training sample set is constructed according to the noise-free wideband high-resolution data set, the noisy wideband high-resolution data set, and the well logging interpolation impedance model, including:

[0113] Each piece of wideband high-resolution data in the noise-free wideband high-resolution data set and the noisy wideband high-resolution data set is normalized to obtain a plurality of input data, and the well logging interpolation impedance model is used as label data to construct the training sample set.

[0114] Specifically, considering that the seismic data has positive and negative amplitude values, and that the negative parameters may cause node death and inaccurate training results in the convolution kernel operation and activation function during network convolution training, in order to facilitate neural network operation, the seismic data is preprocessed by normalizing the positive and negative amplitude values to the interval [0, 1] according to the application characteristics of deep learning. The normalization formula is:

[0115]

[0116] Wherein S and S norm represent the seismic data before and after normalization, respectively. max min are the maximum and minimum amplitude values of the seismic data, respectively.

[0117] According to the above normalization formula, the noise-free and noisy wideband high-resolution data are normalized as input data, and the well logging interpolation impedance model is used as label data to construct the training sample set of the impedance inversion neural network.

[0118] ​Optionally, in step S105, a generative adversarial network for wave impedance inversion is constructed, including constructing a generative adversarial network suitable for high-resolution impedance inversion of a thin reservoir.

[0119] Referring to Figure 11 The constructed generative adversarial network structure includes a generator and a discriminator, and both the generator and the discriminator are composed of a convolutional neural network. G(z) represents the estimated impedance generated by the generator, z(t) represents the real impedance, and s(t) represents the poststack seismic record. The generator is composed of a five-layer network, including two convolutional layers (CONV), two sub-sampling layers, and one fully-connected layer (Fully-Connect); and the discriminator is composed of a three-layer network, including one fully-connected layer and two transposed convolutional layers (CONV_tran). The network is first trained by the discriminator. The discriminator is essentially a binary classifier, which labels the input seismic impedance training sample as 1, i.e. the real impedance, and labels the output value of the initial generator as 0, i.e. the generated impedance. The generator is a convolutional neural network that only includes a forward propagation algorithm, and the network parameter update is transmitted by the discriminator.

[0120] For the generator, the preprocessed poststack seismic data and impedance data are extracted through the convolutional layer and the sub-sampling layer to obtain feature data. This process mainly obtains high-dimensional data features through convolution kernel operation. The convolutional layer calculation process is as follows:

[0121]

[0122] wherein, is a two-dimensional convolution operator; sigma is an activation function; W is a network weight parameter, which is initialized as a small random number; b is a network bias value, which is initialized as 0; and H is high-dimensional data output after convolution kernel operation. Through analysis of the characteristics of seismic data and experimental results verification, the activation function in the present application is preferably a ReLU function to improve the nonlinear fitting ability of the model.

[0123] sigma(x)=max(0,x)

[0124] wherein, x is an input feature value. The network weight parameter and the bias value are updated continuously during the continuous training process of the network. The dimension of the output data of the convolutional layer is high, which affects the training speed of the network and the accuracy of the training result. Therefore, by referring to the down-sampling process of an image, a sub-sampling layer is used to reduce the dimension of the high-dimensional feature data, and the feature data carrying position information and having strong correlation is retained. The feature data is transmitted to a fully-connected layer, and error discrimination is performed through a loss function.

[0125] Preferably, the loss function of the generative adversarial network in the present application adopts JS divergence, and the formula is:

[0126]

[0127] where P g (z) represents the probability distribution of the estimated impedance generated by the generator, P r (z) represents the probability distribution of the true impedance, and KL represents the KL divergence, and the formula is:

[0128]

[0129] For the discriminator, the network parameters are updated according to the loss function. When the generated sample and the real sample have a large deviation, the loss function can provide a large gradient, and the network updates the network parameters in the gradient descent direction, and the updated parameters are transmitted to the discriminator and the generator. With continuous training, the discrimination ability of the discriminator for the impedance authenticity is continuously enhanced, and finally an optimal discriminator is reached.

[0130] The training process of the generative adversarial network is cross-training, and the discriminator and the generator are continuously trained and optimized. The generator does not contain a back propagation algorithm, and its network parameter update depends on the error back propagation algorithm of the discriminator. After each parameter update of the generator, an estimated impedance is generated from a new set of stacked seismic records. The generator is similar to an inverse convolutional neural network, and the impedance input into the generator is converted into network data features through a fully connected layer. The specific form can be compared with the data features obtained after the impedance data in the discriminator pass through the convolutional layer. At this time, the data features obtained by the generator from the stacked seismic records are reconstructed through the deconvolutional layer, and the reconstructed estimated impedance G(z) is obtained after passing through two deconvolutional layers.

[0131] The specific formula of the deconvolutional layer is:

[0132] z = (d-1) x m + k

[0133] where z is the generated impedance; d is the feature value after the fully connected layer conversion; m is the sliding step, which is preferably 2 in the present application; and k represents the size of the convolution kernel, which is preferably 5 in the present application.

[0134] Optionally, in step S106, the preset wave impedance inversion requirement includes a thin reservoir high-resolution impedance inversion requirement. After the generative adversarial network framework is built, in order to realize high-resolution impedance inversion, the impedance generated by the generator needs to continuously approach the true impedance of the stacked seismic records:

[0135] J = min(G(z)-z) 2

[0136] where G(z) represents the estimated impedance generated by the generator.

[0137] In order to make the value of J in the formula as close to 0 as possible, according to the requirements of high-resolution impedance inversion of thin reservoirs, in an implementation, the objective function V of the generative adversarial network is set as:

[0138] min G max D V[D,G]=E x [lgD(z)]+E′ w [lg(1-D(G(z′)))]

[0139] wherein D(z) represents an optimal discriminator, and D(G(z')) represents the discrimination result of the optimal discriminator on the estimated impedance generated by the generator. The generative adversarial network mainly learns the probability distribution characteristics of the data, and the objective function is used to measure the degree of coincidence between the generated impedance and the real impedance. By analyzing the objective function, for a fixed generator G, the discriminator D needs to continuously improve the discrimination ability, that is, maximize the cross entropy between them, so as to accurately distinguish the difference between the generated sample and the real sample; for a fixed discriminator D, the accuracy of the generated sample generated by the generator G needs to be continuously improved, so that D cannot distinguish the difference between them. The two are constantly in confrontation and game in the learning process.

[0140] Optionally, in step S107, the generative adversarial network is trained according to the training sample set and the objective function, including: inputting the training sample set into the constructed generative adversarial network structure, starting network training, continuously updating network parameters, and obtaining a network model suitable for high-resolution impedance inversion of thin reservoirs.

[0141] The specific implementation method is as follows:

[0142] The training sample set is input into the constructed generative adversarial network structure, and network training is started. The whole training process of the generative adversarial network is alternately performed by the generator and the discriminator. Each training will alternately input the well logging interpolation impedance model in the training set and the estimated impedance generated by the generator into the discriminator network, extract data features through the convolution layer, and then reduce the dimension of the feature data through the down-sampling layer to prevent overfitting. Finally, 32 feature map sets are output. Taking the feature map set as the discrimination condition, the well logging interpolation impedance model feature is marked as 1 (real impedance), and the generated impedance is discriminated as an intermediate value between 0 (pseudo impedance) and 1 (real impedance) through the loss function error. If the determination result is greater than 0.5, it is considered that the generated impedance is the real impedance corresponding to the post-stack seismic data.

[0143] During the training process of the entire network, the estimated impedance output by the generator and the real impedance are constantly matched in the discriminator, the parameters of the generator and the discriminator are updated according to the objective function, and through the game between the generator and the discriminator, the estimated impedance is constantly converging to the real impedance. When the generative adversarial network reaches Nash equilibrium, that is, when the discriminator cannot distinguish the estimated impedance from the real impedance, the result impedance generated by the generator in the last iteration is considered as the real impedance in the corresponding stacked seismic data.

[0144] The wave impedance inversion model training method provided in the embodiment does not need to collect a large amount of observation data and high-resolution data pairs in advance during training, and only needs to construct a wideband wavelet library according to the stacked seismic data and the logging data of each well, and then obtain a noise-free wideband high-resolution data set, a noisy wideband high-resolution data set and a logging interpolation impedance model, so as to construct a training sample set to train the generative adversarial network and obtain a wave impedance inversion model with high precision.

[0145] Example Two

[0146] In the second embodiment of the present application, a wave impedance inversion method is provided, and the method specifically comprises:

[0147] The stacked seismic data is normalized, and the normalized stacked seismic data is input into a preset wave impedance inversion model to perform wave impedance inversion, and a wave impedance inversion result is obtained.

[0148] The preset wave impedance inversion model is obtained according to the wave impedance inversion model training method in the first embodiment.

[0149] In the embodiment, based on the trained network model, high-resolution impedance inversion is performed on actual stacked seismic data. The seismic data is also normalized and preprocessed, so that the value range is adjusted to the interval [0, 1], avoiding the influence of negative amplitude of the seismic record on the operation of the convolution kernel and the activation function. The preprocessed seismic record is input into the previously trained neural network model to obtain the corresponding impedance inversion result.

[0150] The wave impedance inversion method provided in the embodiment performs high-resolution impedance inversion on a thin reservoir based on the wave impedance inversion model obtained by the wave impedance inversion model training method provided in the first embodiment, and can effectively improve the inversion precision of the thin reservoir.

[0151] Example Three

[0152] In the third embodiment of the present application, a wave impedance inversion model training device is provided, as shown in Figure 2 The device comprises:

[0153] The wideband wavelet library construction module 201 is configured to construct a wideband wavelet library from the post-stack seismic data and the logging data of the wells by using wideband Ricker wavelets; wherein the wideband wavelet library comprises the wideband wavelets of the wells and a comprehensive wideband wavelet.

[0154] The logging interpolation impedance model construction module 202 is configured to obtain target layer logging data from the logging data of the wells and to establish a logging interpolation impedance model according to the target layer logging data.

[0155] The wideband high-resolution data generation module 203 is configured to construct a noise-free wideband high-resolution data set and a noisy wideband high-resolution data set according to the logging interpolation impedance model and the wideband wavelet library.

[0156] The training sample set construction module 204 is configured to construct a training sample set for training the wave impedance inversion model according to the noise-free wideband high-resolution data set, the noisy wideband high-resolution data set and the logging interpolation impedance model.

[0157] The generative adversarial network construction module 205 is configured to construct a generative adversarial network for wave impedance inversion according to the training sample set and the data structure features of the post-stack seismic data.

[0158] The objective function determination module 206 is configured to determine an objective function of the wave impedance inversion according to a preset wave impedance inversion requirement.

[0159] The model training module 207 is configured to train the generative adversarial network according to the training sample set and the objective function to obtain the wave impedance inversion model.

[0160] Optionally, the wideband wavelet library construction module 201 constructs the wideband wavelet library from the post-stack seismic data and the logging data of the wells by using wideband Ricker wavelets, which comprises:

[0161] The well-seismic calibration is performed on the wells by using the wideband Ricker wavelets according to the post-stack seismic data and the logging data of the wells to determine the wideband wavelets of the wells, and the comprehensive wideband wavelet is determined according to the wideband wavelets of the wells to construct the wideband wavelet library.

[0162] Optionally, the logging interpolation impedance model construction module 202 establishes the logging interpolation impedance model according to the target layer logging data, which comprises:

[0163] The seismic horizon is obtained by horizon interpretation according to the post-stack seismic data;

[0164] The formation framework is built according to the seismic horizon, and the logging interpolation impedance model at a preset sampling time interval is obtained by interpolating the target layer logging data along the formation framework by using the Kriging method.

[0165] Optionally, the wideband high-resolution data generation module 203 constructs a noise-free wideband high-resolution data set and a noisy wideband high-resolution data set according to the well logging interpolation impedance model and the wideband wavelet library, including:

[0166] According to the well logging interpolation impedance model, a reflection coefficient model is generated, and each wavelet in the wideband wavelet library is respectively convolved with the reflection coefficient model to generate noise-free wideband high-resolution data corresponding to each wavelet, so as to construct a noise-free wideband high-resolution data set.

[0167] In the noise-free wideband high-resolution data, a preset Gaussian noise is added to generate noisy wideband high-resolution data corresponding to each wavelet, so as to construct a noisy wideband high-resolution data set.

[0168] The preset Gaussian noise includes Gaussian noise with the same signal-to-noise ratio as the post-stack seismic data.

[0169] Optionally, the training sample set construction module 204 constructs a training sample set according to the noise-free wideband high-resolution data set, the noisy wideband high-resolution data set, and the well logging interpolation impedance model, including:

[0170] Each piece of wideband high-resolution data in the noise-free wideband high-resolution data set and the noisy wideband high-resolution data set is respectively normalized to obtain a plurality of input data, and the well logging interpolation impedance model is taken as label data to construct a training sample set.

[0171] The wave impedance inversion model training device provided in the embodiment does not need to collect a large amount of observation data and high-resolution data pairs in advance during training, but only needs to construct a wideband wavelet library according to post-stack seismic data and well logging data of each well, and then obtain a noise-free wideband high-resolution data set, a noisy wideband high-resolution data set, and a well logging interpolation impedance model, so as to construct a training sample set to train the generative adversarial network and obtain a wave impedance inversion model with high precision.

[0172] Example Four

[0173] In the fourth embodiment of the present application, a wave impedance inversion device is provided, as shown in Figure 3 The device includes:

[0174] The normalization processing module 301 is configured to normalize the post-stack seismic data.

[0175] The wave impedance inversion module 302 is configured to input the normalized post-stack seismic data into a preset wave impedance inversion model to perform wave impedance inversion and obtain a wave impedance inversion result. The preset wave impedance inversion model is obtained according to the wave impedance inversion model training method in the first embodiment.

[0176] The wave impedance inversion method provided by the embodiment can effectively improve the inversion accuracy of the thin reservoir by performing high-resolution impedance inversion on the thin reservoir based on the wave impedance inversion model obtained by the wave impedance inversion model training method provided in Embodiment One.

[0177] Example Five

[0178] In Embodiment Five of the present application, the wave impedance inversion model training method and the wave impedance inversion method in the above embodiments of the present application are used to train a wave impedance inversion model for the tight sandstone reservoir in the Sichuan Basin in Southwest China, and the wave impedance inversion model is used to perform high-resolution impedance inversion on the tight sandstone reservoir in the Sichuan Basin in Southwest China. The specific process is shown in Figure 4

[0179] Figure 5 The post-stack seismic profile in Embodiment Five shows that the time domain range is 1.5s to 2.0s, the length is 0.5s, and the tight sandstone reservoir is located in the range from above the strong wave trough at 1.8s to 1.5s, showing a thin-layered seismic response feature. Therefore, the method provided in the above embodiments of the present application is used to perform high-resolution impedance inversion on the thin reservoir.

[0180] In step (1), based on the wideband Ricker wavelet, the well-to-seismic calibration is performed on each well in the study area according to the post-stack seismic data of the study area and the well logging data of each well in the study area, the corresponding wideband wavelet of each well is determined, and the comprehensive wideband wavelet suitable for the study area is obtained to form a wideband wavelet library.

[0181] The wideband Ricker wavelet waveform is shown in Figure 6 The main lobe of the wideband Ricker wavelet is relatively narrow, the side lobe amplitude is small, the waveform is simple, and the resolution is high. Figure 7 The frequency spectrum comparison chart of the wideband Ricker wavelet and the conventional Ricker wavelet with the same main frequency shows that the high-frequency band of the wideband Ricker wavelet is wider than that of the conventional Ricker wavelet.

[0182] In step (2), first, the horizon interpretation is performed on the actual post-stack seismic data, and the stratigraphic framework is built; then, the well logging data of the target layer is interpolated along the stratigraphic framework by using the Kriging method to obtain a 1ms-sampled well logging interpolation impedance model, as shown in Figure 8

[0183] In step (3), the corresponding reflection coefficient model is generated based on the well logging interpolation impedance model. In order to expand the training samples and improve the diversity of the samples, the reflection coefficient model is respectively convolved with each well wavelet and with the comprehensive wavelet to obtain noise-free wideband high-resolution data. In order to improve the robustness and generalization ability of the adversarial network and evaluate the signal-to-noise ratio of the original post-stack seismic data, the Gaussian noise with the corresponding signal-to-noise ratio is added to the noise-free wideband high-resolution data to obtain the noisy wideband high-resolution data.​​

[0184] Figure 9 For Figure 8 the purpose layer interval reflectivity model generated by the logging interpolation model, Figure 10 is the reflectivity model and the noise-free wideband high-resolution data obtained by the comprehensive wideband wavelet convolution extracted in step (1), the resolution of the data is higher Figure 5 than the actual seismic data.

[0185] In step (4), because the seismic data has positive and negative amplitude values, the negative parameters may cause node death and inaccurate training results during the convolution kernel operation and the influence of the activation function in the network convolution training process. In order to facilitate neural network operation, referring to the application characteristics of deep learning, the positive and negative amplitude values are normalized to the interval. The normalized wideband high-resolution seismic data is used as input data, and the logging interpolation impedance model is used as label data to construct the training sample set of the impedance inversion neural network.

[0186] In step (5), according to the data structure characteristics of the training sample set and the actual post-stack seismic data, a generative adversarial network structure suitable for thin reservoir high-resolution impedance inversion is constructed, including a generator and a discriminator, as shown in Figure 11 The generator is composed of five layers of network, including two layers of convolution layer (CONV), two layers of subsampling layer (Sub-Sampling) and one layer of fully connected layer (Fully-Connect); the discriminator is composed of three layers of network, including one layer of fully connected layer and two layers of deconvolution layer (CONV_tran).

[0187] In step (6), according to the requirements of thin reservoir high-resolution impedance inversion, the objective function of inversion is set, which is used to measure the degree of agreement between the generated impedance and the real impedance. Through continuous training of the network and continuous adjustment of the parameters, the impedance generated by the generator can continuously approach the real impedance of the post-stack seismic record.

[0188] In step (7), the training sample set is input into the constructed generative adversarial network structure, and the network training is started. The training process of the whole generative adversarial network is alternately performed by the generator and the discriminator. The estimated impedance output by the generator and the real impedance are continuously matched in the discriminator. Through the game between the generator and the discriminator, the network parameters are continuously updated, so that the estimated impedance continuously converges to the real impedance, as shown in Figure 12 .

[0189] When the generative adversarial network reaches the Nash equilibrium, that is, the moment when the discriminator cannot distinguish between the estimated impedance and the real impedance, the result impedance generated by the generator in the last iteration is considered as the real impedance in the corresponding post-stack seismic record, and at this time, the network model suitable for thin reservoir high-resolution impedance inversion is obtained.

[0190] In step (8), based on the network model obtained by training, high-resolution impedance inversion is performed on the actual post-stack seismic data as shown in the figure. Figure 5 Similarly, the seismic data is normalized and preprocessed, so that the value range is adjusted to the interval, and the negative amplitude of the seismic record is avoided to affect the operation of the convolution kernel and the activation function.

[0191] The preprocessed seismic record is input into the neural network model trained in the foregoing, and the corresponding impedance inversion result is obtained, as shown in the figure. Figure 13

[0192] Figure 14 For the conventional post-stack wave impedance inversion result, it can be seen from the comparison with Figure 13 and Figure 14 It can be seen that the thin reservoir high-resolution impedance inversion method based on the generative adversarial network has higher resolution than the conventional post-stack impedance inversion method, and can better distinguish the thin dense sandstone reservoir in details, and is more consistent with the actual drilling.

[0193] The embodiment is aimed at the characteristics of the thin thickness of the dense sandstone reservoir, uses a wideband Ricker wavelet for well-seismic calibration, establishes a wideband wavelet library suitable for the study area, and obtains wideband high-resolution data by folding the reflection coefficients of the well logging interpolation impedance model, so as to use the wideband high-resolution data as a training sample set, train the generative adversarial network, form the thin reservoir high-resolution impedance inversion method based on the generative adversarial network, and thus improve the inversion accuracy of the thin reservoir, and provide strong data support for the oil and gas geophysical exploration of the thin dense sandstone reservoir in the study area.

[0194] Example Six

[0195] In the embodiment six of the present application, a computer program product is also provided, which includes a computer program or instructions, and when the computer program or instructions are executed by a processor, all or part of the steps of the method described in the above embodiments are realized.

[0196] Further, the computer program product can include one or more computer executable components configured to perform the embodiments when the program is run; the computer program product can also include a computer program tangibly embodied on a computer readable medium, and the computer program includes program codes for executing any method in the embodiments of the present application. In such embodiments, the computer program can be downloaded and installed from a network by a communication part, and / or installed from a detachable medium.​

[0197] Example Seven

[0198] In the seventh embodiment of the present application, a computer readable storage medium is also provided, which stores a computer program that, when executed by one or more processors, implements all or part of the steps of the method described in the above embodiments.

[0199] The various functional units in the embodiments of the present application can be integrated in one processing unit, or exist separately as individual physical units, or two or more units are integrated in one unit. When the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a computer readable storage medium.

[0200] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read only memory (EPROM), an optical fiber, a portable compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0201] Example Eight

[0202] In the eighth embodiment of the present application, an electronic device 1500 is also provided, which can be a mobile phone, a computer or a tablet computer, etc. Figure 15 The schematic diagram of the composition structure of the electronic device provided in the embodiments of the present application is as follows: Figure 15As shown, the electronic device 1500 includes at least one processor 1501, at least one communication bus 1502, a user interface 1503, at least one external communication interface 1504, and a memory 1505. The communication bus 1502 is configured to enable connection and communication between these components. The user interface 1503 can include a display screen, and the external communication interface 1504 can include a standard wired interface and a wireless interface. The memory 1505 stores a computer program, and the memory 1505 and the one or more processors 1501 are in communication connection with each other. When the computer program is executed by the one or more processors, the processor 1501 is configured to execute the computer program stored in the memory to implement all or part of the steps of the method in the above embodiments.

[0203] The processor can be an Application Specific Integrated Cricuit (ASIC), a Digital Signal Processor (DSP), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic elements, which is configured to execute all or part of the steps of the method described in the above embodiments. The present embodiment will not be repeated here.

[0204] The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0205] The wave impedance inversion model training method, the wave impedance inversion method and the electronic equipment provided by the application do not need to collect a large amount of observation data and high-resolution data pairs in advance during training, only need to construct a wideband wavelet library according to the post-stack seismic data and the logging data of each well, and then obtain a noise-free wideband high-resolution data set, a noisy wideband high-resolution data set and a logging interpolation impedance model, so as to construct a training sample set to train a generative adversarial network and obtain a wave impedance inversion model with high precision; and the wave impedance inversion model obtained by training is used for high-resolution impedance inversion of a thin reservoir, so that the inversion precision of the thin reservoir can be effectively improved.

[0206] The above describes the basic principles of the application in combination with specific embodiments, but it should be pointed out that the advantages, advantages and effects mentioned in the application are only examples and not limitations, and these advantages, advantages and effects cannot be considered as the must-haves of each embodiment of the disclosure. In addition, the above specific details are only for the purpose of example and for the purpose of understanding, and are not limited to the above specific details. The application must be implemented using the above specific details.

[0207] The block diagrams of the devices, apparatuses, equipment and systems involved in the application are only illustrative examples and are not intended to require or imply that the connection, arrangement and configuration shown in the block diagrams must be used. As those skilled in the art will recognize, these devices, apparatuses, equipment and systems can be connected, arranged and configured in any way. Words such as "include", "contain", "have" and the like are open-ended words, which mean "including but not limited to", and can be used interchangeably.

[0208] It should also be noted that in the devices, equipment and methods of the disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions of the application.

[0209] The above description of the disclosed aspects is provided so that any person skilled in the art can make or use the application. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the application. Therefore, the application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0210] The above description has been given for the purpose of example and description. Furthermore, this description is not intended to limit the embodiments of the application to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain modifications, alterations, changes, additions and subcombinations thereof.

Claims

1. A wave impedance inversion model training method, characterized in that: The method comprises: Based on the post-stack seismic data and the logging data of each well, a broadband wavelet library is constructed by using broadband Ricker wavelets; wherein the broadband wavelet library includes the broadband wavelets of each well and the integrated broadband wavelet; Acquire target layer logging data from the logging data of each well, and establish a logging interpolation impedance model based on the target layer logging data; Constructing a noise-free broadband high-resolution data set and a noisy broadband high-resolution data set according to the well logging interpolation impedance model and the broadband wavelet library; Constructing a training sample set for training a wave impedance inversion model based on the noise-free broadband high-resolution data set, the noisy broadband high-resolution data set, and the well logging interpolation impedance model; Constructing a generative adversarial network for wave impedance inversion based on the training sample set and the data structure characteristics of the post-stack seismic data; Determine the objective function of wave impedance inversion according to the preset wave impedance inversion requirements; The generative adversarial network is trained according to the training sample set and the objective function to obtain a trained wave impedance inversion model.

2. The wave impedance inversion model training method according to claim 1, characterized in that: The broadband wavelet library is constructed by using broadband Ricker wavelets based on the post-stack seismic data and the logging data of each well, including: Based on the post-stack seismic data and the logging data of each well, each well is calibrated by broadband Ricker wavelet to determine the broadband wavelet of each well. Based on the broadband wavelet of each well, a comprehensive broadband wavelet is determined to construct the broadband wavelet library.

3. The wave impedance inversion model training method according to claim 1, characterized in that: The step of establishing a well logging interpolation impedance model based on the well logging data of the target layer includes: Performing horizon interpretation based on the post-stack seismic data to obtain seismic horizons; A stratigraphic framework is constructed according to the seismic horizon, and the well logging data of the target layer is interpolated along the stratigraphic framework using the Kriging method to obtain a well logging interpolation impedance model at a preset sampling time interval.

4. The wave impedance inversion model training method according to claim 1, characterized in that: The step of constructing a noise-free broadband high-resolution data set and a noisy broadband high-resolution data set based on the well logging interpolation impedance model and the broadband wavelet library includes: generating a reflection coefficient model according to the well logging interpolation impedance model, and convolving each wavelet in the broadband wavelet library with the reflection coefficient model to generate noise-free broadband high-resolution data corresponding to each wavelet, so as to construct the noise-free broadband high-resolution data set; Adding preset Gaussian noise to the noise-free broadband high-resolution data to generate noisy broadband high-resolution data corresponding to each wavelet, so as to construct the noisy broadband high-resolution data set; The preset Gaussian noise includes Gaussian noise having the same signal-to-noise ratio as the post-stack seismic data.

5. The wave impedance inversion model training method according to claim 1, characterized in that: The constructing of a training sample set according to the noise-free broadband high-resolution data set, the noisy broadband high-resolution data set, and the well logging interpolation impedance model comprises: Each piece of broadband high-resolution data in the noise-free broadband high-resolution data set and the noisy broadband high-resolution data set is normalized to obtain multiple input data, and the well logging interpolation impedance model is used as label data to construct a training sample set.

6. A wave impedance inversion method, characterized in that: The method comprises: Normalizing the post-stack seismic data, and inputting the normalized post-stack seismic data into a preset wave impedance inversion model to perform wave impedance inversion and obtain wave impedance inversion results; Wherein, the preset wave impedance inversion model is obtained according to the method described in any one of claims 1 to 5.

7. A wave impedance inversion model training device, characterized in that: The device comprises: A broadband wavelet library construction module is used to construct a broadband wavelet library using broadband Ricker wavelets based on post-stack seismic data and well logging data of each well; wherein the broadband wavelet library includes broadband wavelets of each well and integrated broadband wavelets; A well logging interpolation impedance model building module is used to obtain target layer logging data from the well logging data of each well, and to build a well logging interpolation impedance model based on the target layer logging data; A broadband high-resolution data generation module is used to construct a noise-free broadband high-resolution data set and a noisy broadband high-resolution data set based on the well logging interpolation impedance model and the broadband wavelet library; A training sample set construction module is used to construct a training sample set for training a wave impedance inversion model based on the noise-free broadband high-resolution data set, the noisy broadband high-resolution data set, and the well logging interpolation impedance model; A generative adversarial network construction module is used to construct a generative adversarial network for wave impedance inversion based on the data structure characteristics of the training sample set and the post-stack seismic data; An objective function determination module is used to determine the objective function of wave impedance inversion according to preset wave impedance inversion requirements; The model training module is used to train the generative adversarial network according to the training sample set and the objective function to obtain a trained wave impedance inversion model.

8. A wave impedance inversion device, characterized in that: The device comprises: Normalization processing module, used to perform normalization processing on post-stack seismic data; The wave impedance inversion module is used to input the normalized post-stack seismic data into a preset wave impedance inversion model to perform wave impedance inversion and obtain wave impedance inversion results; wherein, the preset wave impedance inversion model is obtained according to the method described in any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that The computer program stored in the computer-readable storage medium, when executed by one or more processors, implements the steps of the method according to any one of claims 1 to 6.

10. An electronic device, characterized in that: The method comprises a memory and one or more processors, wherein a computer program is stored in the memory, and when the computer program is executed by the one or more processors, the steps of the method according to any one of claims 1 to 6 are performed.

Images