Horizon interpretation method and device based on uncertainty quantization, equipment and medium

Abstract

The invention provides a horizon interpretation method and device based on uncertainty quantization, equipment and a medium, and belongs to the technical field of seismic data processing. The method comprises the following steps: acquiring a first seismic profile image, wherein the first seismic profile image comprises seismic waves of a plurality of seismic channels; processing the first seismic section image by adopting any horizon tracking method to obtain a horizon tracking result of the first seismic section image; determining an uncertainty parameter of each horizon based on the distribution condition of each horizon in the first seismic profile image in the horizon tracking result; and processing the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image. According to the scheme, the accuracy of the horizon tracking result can be known more accurately, more accurate next-step processing is carried out based on the horizon tracking result, and the horizon interpretation accuracy is improved.

Description

Technical Field

[0001] The present application relates to the technical field of seismic data processing, and in particular to a layer interpretation method, device, equipment and medium based on uncertainty quantification. Background Art

[0002] Seismic data interpretation is an important part of seismic exploration, and layer tracing is an important step in seismic data interpretation. The accuracy of layer tracing results has an important impact on subsequent structural interpretation and reservoir prediction. Summary of the invention

[0003] The embodiment of the present application provides a layer interpretation method, device, equipment and medium based on uncertainty quantification, which can more accurately understand the accuracy of the layer tracking result, perform more accurate next step processing based on the layer tracking result, and improve the accuracy of layer interpretation. The technical solution is as follows:

[0004] On the one hand, a horizon interpretation method based on uncertainty quantification is provided, the method comprising:

[0005] Acquire a first seismic profile image, wherein the first seismic profile image includes seismic waves of a plurality of seismic traces;

[0006] Using any layer tracing method to process the first seismic section image, to obtain a layer tracing result of the first seismic section image, wherein the layer tracing result indicates a position of at least one layer in the first seismic section image;

[0007] Determine an uncertainty parameter of each layer based on the distribution of each layer in the layer tracing result in the first seismic profile image, wherein the uncertainty parameter is used to indicate the degree of uncertainty of the layer obtained by processing the first seismic profile image using the layer tracing method;

[0008] The layer tracking result is processed based on the uncertainty parameter of each layer in the first seismic profile image.

[0009] In a possible implementation, determining the uncertainty parameter of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image includes:

[0010] Based on the distribution of each layer in the layer tracking result in the first seismic profile image, the mean and variance of each layer are determined.

[0011] In a possible implementation, the processing of the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image includes:

[0012] Determining a confidence interval for each horizon based on a mean and a variance of each horizon in the first seismic image;

[0013] For each layer in the layer tracking result, the points that do not belong to the confidence interval of the layer are deleted.

[0014] In a possible implementation, determining the uncertainty parameter of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image includes:

[0015] Sampling the layer tracking results to obtain layer samples of each layer;

[0016] Based on the layer samples of each layer, using Bayesian theory, a first-order moment estimation and a second-order moment estimation of each layer sample are obtained;

[0017] The first-order moment estimate of the layer samples of the layer is determined as the mean of the layer, and the second-order moment estimate of the layer samples of the layer is determined as the variance of the layer.

[0018] In a possible implementation, the first-order moment estimation and the second-order moment estimation of each layer sample are obtained based on the layer sample of each layer by using Bayesian theory, including:

[0019] Acquire first relationship data corresponding to the Monte Carlo integral, where the first relationship data is used to represent the relationship between the layer samples of each layer, the probability density function of each layer and the mean value of each layer;

[0020] Based on the Bayesian estimation theory and a first assumption, the first relationship data is converted into second relationship data, the first assumption being that the layers and the bombardment recording points in the first seismic profile image are independent and identically distributed, and the second relationship data is used to represent the relationship between the layer sample, the conditional probability density function, the layer function and the mean of each layer;

[0021] Based on a second assumption, the second relationship data is simplified to obtain third relationship data, wherein the second assumption is that the conditional probability density function of the layer is a function;

[0022] Based on the third relationship data, the Bayesian statistical theory is used to determine the first-order moment estimate and the second-order moment estimate of the layer samples of each layer.

[0023] In a possible implementation, the processing of the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image includes:

[0024] When the uncertainty parameter of any layer in the first seismic profile is greater than a first threshold, the layer is deleted from the layer tracking result, or the layer tracking result is discarded.

[0025] In a possible implementation, the adopting any layer tracing method to process the first seismic profile image to obtain a layer tracing result of the first seismic profile image includes:

[0026] Using a layer tracing model to perform layer tracing on the first seismic section image to obtain a layer tracing result of the first seismic section image; or,

[0027] Based on the first seismic profile image, a conventional dynamic time warping algorithm is used to process the first seismic profile image to obtain a horizon tracing result of the first seismic profile image; or,

[0028] A layer tracing model is used to extract features of the first seismic profile image. A feature map of the first seismic profile image is processed using a traditional dynamic time warping algorithm based on the feature map to obtain a layer tracing result of the first seismic profile image.

[0029] In a possible implementation, the layer tracking model is used to extract features from the first seismic profile image, and the feature map of the first seismic profile image includes:

[0030] Using the layer tracking model, extracting features from the first seismic profile image to obtain a first feature map;

[0031] Sampling the first feature map to obtain a second feature map;

[0032] Repeat the process of inputting the currently obtained feature map into the layer tracking model, extracting features from the input feature map using the layer tracking model, outputting the feature map obtained by this feature extraction, and sampling the feature map obtained by this feature extraction until a termination condition is met.

[0033] In a possible implementation, the method further includes:

[0034] After obtaining the layer tracing results of the plurality of continuous seismic section images, determining the layer position distribution function based on the layer tracing results of the plurality of continuous seismic section images;

[0035] Based on the layer position distribution function, the layer in the next seismic section image of the plurality of continuous seismic section images is predicted.

[0036] On the other hand, a horizon interpretation device based on uncertainty quantification is provided, the device comprising:

[0037] An acquisition module, configured to acquire a first seismic profile image, wherein the first seismic profile image includes seismic waves of a plurality of seismic traces;

[0038] A tracking module, configured to process the first seismic profile image by using any layer tracking method to obtain a layer tracking result of the first seismic profile image, wherein the layer tracking result indicates a position of at least one layer in the first seismic profile image;

[0039] A determination module, configured to determine an uncertainty parameter of each layer based on the distribution of each layer in the layer tracing result in the first seismic profile image, wherein the uncertainty parameter is used to indicate the degree of uncertainty of the layer obtained by processing the first seismic profile image using the layer tracing method;

[0040] A processing module is used to process the layer tracking result based on the uncertainty parameter of each layer in the first seismic profile image.

[0041] In a possible implementation, the determination module is used to determine the mean and variance of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image.

[0042] In one possible implementation, the processing module is used to determine the confidence interval of each layer based on the mean and variance of each layer in the first seismic image; for each layer in the layer tracking result, the points that do not belong to the confidence interval of the layer are deleted.

[0043] In a possible implementation, the determination module is used to sample the layer tracking results to obtain layer samples for each layer; based on the layer samples for each layer, the Bayesian theory is used to obtain the first-order moment estimate and the second-order moment estimate of each layer sample; the first-order moment estimate of the layer samples for the layer is determined as the mean of the layer, and the second-order moment estimate of the layer samples for the layer is determined as the variance of the layer.

[0044] In a possible implementation, the determination module is used to obtain first relationship data corresponding to the Monte Carlo integral, the first relationship data being used to represent the relationship between the stratigraphic samples of each layer, the probability density function of each layer and the mean of each layer; based on the Bayesian estimation theory and the first assumption, the first relationship data is converted into second relationship data, the first assumption being that the layers and the shelling record points in the first seismic profile image are independent and identically distributed, the second relationship data being used to represent the relationship between the stratigraphic samples, the conditional probability density function, the stratigraphic function and the mean of each layer; based on the second assumption, the second relationship data is simplified to obtain third relationship data, the second assumption being that the conditional probability density function of the layer is a function; based on the third relationship data, the Bayesian statistical theory is used to determine the first-order moment estimate and the second-order moment estimate of the stratigraphic samples of each layer.

[0045] In a possible implementation, the processing module is used to delete any layer in the layer tracking result, or discard the layer tracking result, when the uncertainty parameter of any layer in the first seismic profile is greater than a first threshold.

[0046] In a possible implementation, the tracking module is used to use a layer tracking model to perform layer tracking on the first seismic profile image to obtain a layer tracking result of the first seismic profile image; or,

[0047] The tracking module is used to obtain a horizon tracking result of the first seismic profile image by processing the first seismic profile image using a traditional dynamic time warping algorithm; or

[0048] The tracking module is used to extract features from the first seismic profile image using a layer tracking model. The feature map of the first seismic profile image is processed using a traditional dynamic time warping algorithm based on the feature map to obtain a layer tracking result of the first seismic profile image.

[0049] In one possible implementation, the tracking module is used to use the layer tracking model to perform feature extraction on the first seismic profile image to obtain a first feature map; sample the first feature map to obtain a second feature map; repeatedly input the currently obtained feature map into the layer tracking model, use the layer tracking model to perform feature extraction on the input feature map, output the feature map obtained by this feature extraction, and sample the feature map obtained by this feature extraction until a termination condition is met.

[0050] In a possible implementation manner, the device further includes:

[0051] A prediction module is used to determine a layer position distribution function based on the layer tracing results of a plurality of consecutive seismic section images after obtaining the layer tracing results of the plurality of consecutive seismic section images; and predict the layer in the next seismic section image of the plurality of consecutive seismic section images based on the layer position distribution function.

[0052] On the other hand, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the layer interpretation method based on uncertainty quantification as described in any of the above implementations.

[0053] On the other hand, a computer-readable storage medium is provided, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor to implement the layer interpretation method based on uncertainty quantification as described in any of the above implementations.

[0054] On the other hand, a computer program product is provided, which includes at least one program code, and the at least one program code is loaded and executed by a processor to implement the layer interpretation method based on uncertainty quantification as described in any of the above implementations.

[0055] The beneficial effects of the technical solution provided by the embodiments of the present application include at least:

[0056] The embodiment of the present application provides a stratigraphic interpretation method based on uncertainty quantification. After obtaining the stratigraphic tracing result, the uncertainty parameter of the stratigraphic tracing result is determined so as to more accurately understand the accuracy of the stratigraphic tracing result, and perform more accurate next step processing based on the stratigraphic tracing result to improve the accuracy of the stratigraphic interpretation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0058] Figure 1 It is a flow chart of a horizon interpretation method based on uncertainty quantification provided in an embodiment of the present application;

[0059] Figure 2 is a schematic diagram of a seismic profile image provided in an embodiment of the present application;

[0060] Figure 3is a schematic diagram of a formation tracing result provided in an embodiment of the present application;

[0061] Figure 4 It is a flow chart of a horizon interpretation method based on uncertainty quantification provided in an embodiment of the present application;

[0062] Figure 5 It is a flow chart based on uncertainty quantification of layer tracking results provided by an embodiment of the present application;

[0063] Figure 6 is a flow chart of a formation prediction method provided in an embodiment of the present application;

[0064] Figure 7 It is a structural schematic diagram of a horizon interpretation device based on uncertainty quantification provided in an embodiment of the present application;

[0065] Figure 8 is a schematic diagram of the structure of a terminal provided in an embodiment of the present application;

[0066] Fig. 9 It is a structural diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0068] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0069] Figure 1 is a flow chart of a layer interpretation method based on uncertainty quantification provided in an embodiment of the present application. The embodiment of the present application is illustrated by taking a computer device as an example of an execution subject. Figure 1 , the method comprising:

[0070] 101. A computer device acquires a first seismic profile image, where the first seismic profile image includes seismic waves of multiple seismic traces.

[0071] The first seismic profile image includes seismic waves of multiple seismic channels. In some embodiments, the horizontal axis of the first seismic profile image is the seismic channel, and the vertical axis is time. The time corresponding to the seismic wave can be used to represent the depth of the stratum. Therefore, by marking the stratum in the first seismic profile image, the depth corresponding to the stratum can be determined based on the time corresponding to the stratum. In other embodiments, the horizontal axis of the first seismic profile image is time, and the vertical axis is the seismic channel. The embodiments of the present application do not limit the first seismic profile image. For example, Figure 2 A seismic cross-section image is shown, which includes seismic waves of multiple seismic traces.

[0072] 102. The computer device processes the first seismic profile image using any layer tracing method to obtain a layer tracing result of the first seismic profile image, wherein the layer tracing result indicates a position of at least one layer in the first seismic profile image.

[0073] The horizon interpretation based on uncertainty quantification provided in the embodiments of the present application can use any horizon tracking method to process the first seismic profile image, and the embodiments of the present application do not limit this. In some embodiments, the computer device uses only one horizon tracking method to process the first seismic profile image to obtain a horizon tracking result. In other embodiments, the computer device uses multiple horizon tracking methods to process the first seismic profile image respectively to obtain multiple horizon tracking results. Subsequently, the most accurate horizon tracking result can be determined from the multiple horizon tracking results.

[0074] The layer tracking method is a method for extracting the phase axis in the seismic profile image. Therefore, the position of the layer in the layer tracking result is the position of the phase axis. The layer tracking result is as follows: Figure 3 shown.

[0075] 103. The computer device determines the uncertainty parameter of each layer based on the distribution of each layer in the layer tracing result in the first seismic profile image. The uncertainty parameter is used to indicate the degree of uncertainty of the layer obtained by processing the first seismic profile image using the layer tracing method.

[0076] The layer tracking technology is the technology of extracting the event axis in the seismic profile image, that is, the layer is the event axis in the seismic profile image. Therefore, the more concentrated the distribution of each layer in the first seismic profile image is, the smaller the uncertainty of the layer is, and the higher the accuracy of the layer is. The more dispersed the distribution of each layer in the first seismic profile image is, the greater the uncertainty of the layer is, and the lower the accuracy of the layer is.

[0077] 104. The computer device processes the layer tracking result based on the uncertainty parameter of each layer in the first seismic profile image.

[0078] Since the uncertainty parameter of a layer indicates the degree of uncertainty of the layer, the higher the uncertainty degree of the layer, the lower the accuracy of the layer; the lower the uncertainty degree of the layer, the higher the accuracy of the layer. Therefore, the uncertainty parameter of the layer can indicate the accuracy of the layer, and subsequently, when further processing is performed based on the layer tracking result, the layer tracking result can be referred to. The embodiment of the present application does not limit "the computer device processes the layer tracking result based on the uncertainty parameter of each layer in the first seismic profile image", and only exemplifies it with the following two embodiments.

[0079] In a possible implementation, the uncertainty parameters of the layer include the mean and variance of the layer, and the confidence interval of the layer can be determined based on the mean and variance of the layer, and the layer tracking result can be corrected based on the confidence interval. The computer device processes the layer tracking result based on the uncertainty parameters of each layer in the first seismic profile image, including: determining the confidence interval of each layer based on the mean and variance of each layer in the first seismic image; for each layer in the layer tracking result, deleting the points that do not belong to the confidence interval of the layer.

[0080] In another possible implementation, since the uncertainty parameter of the layer can indicate the accuracy of the layer, it can be determined based on the uncertainty parameter of the layer whether to perform the next step based on the layer tracking result. In some embodiments, the computer device processes the layer tracking result based on the uncertainty parameter of each layer in the first seismic profile image, including: when the uncertainty parameter of any layer in the first seismic profile is greater than the first threshold, the layer is deleted from the layer tracking result, or the layer tracking result is discarded.

[0081] The first threshold value may be any value, and the embodiment of the present application does not limit the first threshold value. Optionally, the first threshold value is an empirical value.

[0082] The embodiments of the present application only take the above two embodiments as examples to exemplify "processing the layer tracking results based on the uncertainty parameters of each layer in the first seismic profile image". Of course, other possible implementation methods can also be used to implement "processing the layer tracking results based on the uncertainty parameters of each layer in the first seismic profile image". For example, multiple layer tracking methods are used to process the first seismic profile image respectively to obtain multiple layer tracking results, and the layer tracking result with the lower degree of uncertainty is determined as the final layer tracking result.

[0083] The layer interpretation method based on uncertainty quantification provided in the embodiment of the present application will determine the uncertainty parameters of the layer tracing results after obtaining the layer tracing results, so as to more accurately understand the accuracy of the layer tracing results, and perform more accurate next step processing based on the layer tracing results, thereby improving the accuracy of the layer interpretation.

[0084] In a possible implementation, based on the distribution of each layer in the layer tracking result in the first seismic profile image, the uncertainty parameter of each layer is determined, including:

[0085] Based on the distribution of each layer in the first seismic profile image in the layer tracing result, the mean and variance of each layer are determined.

[0086] In a possible implementation, the horizon tracking result is processed based on the uncertainty parameter of each horizon in the first seismic profile image, including:

[0087] Determining a confidence interval for each horizon based on the mean and variance of each horizon in the first seismic image;

[0088] For each horizon in the horizon tracking results, the points that do not belong to the confidence interval of the horizon are deleted.

[0089] In a possible implementation, based on the distribution of each layer in the layer tracking result in the first seismic profile image, the uncertainty parameter of each layer is determined, including:

[0090] Sampling the layer tracking results to obtain layer samples of each layer;

[0091] Based on the layer samples of each layer, the Bayesian theory is used to obtain the first-order moment estimation and second-order moment estimation of each layer sample;

[0092] The first-order moment estimate of the layer samples of the layer is determined as the mean of the layer, and the second-order moment estimate of the layer samples of the layer is determined as the variance of the layer.

[0093] In a possible implementation, based on the layer samples of each layer, the Bayesian theory is used to obtain the first-order moment estimation and the second-order moment estimation of each layer sample, including:

[0094] Acquire first relationship data corresponding to the Monte Carlo integral, where the first relationship data is used to represent the relationship between the layer samples of each layer, the probability density function of each layer and the mean value of each layer;

[0095] Based on the Bayesian estimation theory and the first assumption, the first relational data is converted into the second relational data, the first assumption being that the layers and the bombardment recording points in the first seismic profile image are independent and identically distributed, and the second relational data is used to represent the relationship between the layer sample, the conditional probability density function, the layer function and the mean value of each layer;

[0096] Based on the second assumption, the second relationship data is simplified to obtain the third relationship data, and the second assumption is that the conditional probability density function of the layer is a function;

[0097] Based on the third relation data, the Bayesian statistical theory is used to determine the first-order moment estimate and the second-order moment estimate of the layer samples of each layer.

[0098] In a possible implementation, the horizon tracking result is processed based on the uncertainty parameter of each horizon in the first seismic profile image, including:

[0099] When the uncertainty parameter of any layer in the first seismic profile is greater than a first threshold, the layer is deleted from the layer tracing result, or the layer tracing result is discarded.

[0100] In a possible implementation, any layer tracing method is used to process the first seismic profile image to obtain a layer tracing result of the first seismic profile image, including:

[0101] Using a layer tracing model, performing layer tracing on the first seismic section image to obtain a layer tracing result of the first seismic section image; or,

[0102] Based on the first seismic profile image, a conventional dynamic time warping algorithm is used to process the first seismic profile image to obtain a horizon tracing result of the first seismic profile image; or,

[0103] The layer tracing model is used to extract features of the first seismic profile image. The feature map of the first seismic profile image is processed based on the feature map using a traditional dynamic time warping algorithm to obtain the layer tracing result of the first seismic profile image.

[0104] In a possible implementation, a layer tracking model is used to extract features from the first seismic profile image, and the feature map of the first seismic profile image includes:

[0105] Using a layer tracking model, extracting features from the first seismic profile image to obtain a first feature map;

[0106] Sampling the first feature map to obtain a second feature map;

[0107] Repeat the process of inputting the currently obtained feature map into the layer tracking model, extracting features from the input feature map using the layer tracking model, outputting the feature map obtained by this feature extraction, and sampling the feature map obtained by this feature extraction until the termination condition is met.

[0108] In a possible implementation, the method further includes:

[0109] After obtaining the layer tracing results of the plurality of continuous seismic section images, determining the layer position distribution function based on the layer tracing results of the plurality of continuous seismic section images;

[0110] Based on the layer position distribution function, the layer position in the next seismic section image of a plurality of continuous seismic section images is predicted.

[0111] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0112] Figure 4 is a flowchart of a horizon interpretation method based on uncertainty quantification provided in an embodiment of the present application. The embodiment of the present application is illustrated by taking a computer device as an example. Figure 4 , the method comprising:

[0113] 401. A computer device acquires a first seismic profile image, where the first seismic profile image includes seismic waves of multiple seismic traces.

[0114] The first seismic profile image may be any seismic profile image, and the embodiment of the present application does not limit the first seismic profile image.

[0115] 402. The computer device processes the first seismic profile image using any layer tracking method to obtain a layer tracking result of the first seismic profile image, wherein the layer tracking result indicates a position of at least one layer in the first seismic profile image.

[0116] In a possible implementation, the computer device uses a neural network model to perform layer tracking. The computer device uses any layer tracking method to process the first seismic profile image to obtain a layer tracking result of the first seismic profile image, including: using a layer tracking model to perform layer tracking on the first seismic profile image to obtain a layer tracking result of the first seismic profile image.

[0117] In some embodiments, the neural network model is a convolutional neural network model. In some embodiments, the layer tracking model is trained using a predecessor algorithm. In some embodiments, the layer tracking model is a priori model constructed based on a Bayesian priori method. The embodiments of the present application are merely exemplary illustrations of the layer tracking model and do not limit the layer tracking model.

[0118] In another possible implementation, the computer device uses a traditional dynamic time warping algorithm to perform horizon tracking. The computer device uses any horizon tracking method to process the first seismic profile image to obtain a horizon tracking result of the first seismic profile image, including: based on the first seismic profile image, using a traditional dynamic time warping algorithm to process, to obtain a horizon tracking result of the first seismic profile image.

[0119] In another possible implementation, the computer device uses a neural network model to extract features to obtain a feature map, and uses a traditional dynamic time warping algorithm to perform layer tracking on the feature map. The computer device uses any layer tracking method to process the first seismic profile image to obtain a layer tracking result of the first seismic profile image, including: using a layer tracking model to extract features from the first seismic profile image, the feature map of the first seismic profile image, and based on the feature map, using a traditional dynamic time warping algorithm to process the feature map to obtain a layer tracking result of the first seismic profile image.

[0120] In some embodiments, the neural network model is a convolutional neural network model. In some embodiments, the layer tracking model is trained using a predecessor algorithm. In some embodiments, the layer tracking model is a priori model constructed based on a Bayesian priori method. The embodiments of the present application are merely exemplary illustrations of the layer tracking model and do not limit the layer tracking model.

[0121] Among them, the first seismic profile image is feature extracted through the layer tracing model, and the obtained feature map is the prior distribution of the layer of the first seismic profile image; based on the feature map, the traditional dynamic time warping algorithm is used for processing, and the layer tracing result of the first seismic profile image is the posterior distribution of the layer.

[0122] Wherein, when determining the layer tracking result, feature extraction can be performed only once, or multiple times to obtain the depth prior distribution of the layer. Optionally, a layer tracking model is used to extract features from the first seismic profile image, and the feature map of the first seismic profile image includes: using the layer tracking model to extract features from the first seismic profile image to obtain a first feature map; sampling the first feature map to obtain a second feature map; repeatedly inputting the currently obtained feature map into the layer tracking model, using the layer tracking model to extract features from the input feature map, outputting the feature map obtained by this feature extraction, and sampling the feature map obtained by this feature extraction until the termination condition is met. Wherein, the termination condition can be that the number of repeated executions reaches a preset number of times.

[0123] In some embodiments, the layer tracking model is a convolutional neural network model. In some embodiments, the layer tracking model is trained using a predecessor algorithm. In some embodiments, the layer tracking model is a priori model constructed based on a Bayesian priori method. The embodiments of the present application are merely exemplary illustrations of the layer tracking model and do not limit the layer tracking model.

[0124] The computer device may sample the feature map in any manner. For example, the feature map may be sampled using stochastic gradient Langevin dynamics; or the feature map may be sampled using the Markov chain Monte Carlo method; or the feature map may be randomly sampled. Or, the feature map may be sampled at preset intervals. The present application does not limit the sampling method of the feature map.

[0125] For example, Figure 5 As shown in the figure, convolutional neural network and Markov chain Monte Carlo sampling are used to process the seismic profile image to obtain the horizon tracking result.

[0126] 403. The computer device determines the uncertainty parameter of each layer based on the distribution of each layer in the first seismic profile image in the layer tracking result, and the uncertainty parameter of the layer includes the mean and variance of the layer.

[0127] In one possible implementation, Figure 5 As shown, the computer device determines the mean and variance of the layer based on the Bayesian theory. The computer device determines the uncertainty parameter of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image, including: sampling the layer tracking result to obtain a layer sample of each layer; based on the layer sample of each layer, using the Bayesian theory to obtain the first-order moment estimate and the second-order moment estimate of each layer sample; determining the first-order moment estimate of the layer sample of the layer as the mean of the layer, and determining the second-order moment estimate of the layer sample of the layer as the variance of the layer.

[0128] Optionally, the computer device uses Bayesian theory based on the stratigraphic samples of each layer to obtain first-order moment estimates and second-order moment estimates for each stratigraphic sample, including: obtaining first relationship data corresponding to the Monte Carlo integral, the first relationship data being used to represent the relationship between the stratigraphic samples of each layer, the probability density function of each layer and the mean of each layer; based on the Bayesian estimation theory and the first assumption, converting the first relationship data into second relationship data, the first assumption being that the layers and the shelling record points in the first seismic profile image are independent and identically distributed, and the second relationship data being used to represent the relationship between the stratigraphic samples, the conditional probability density function, the stratigraphic function and the mean of each layer; based on the second assumption, simplifying the second relationship data to obtain third relationship data, the second assumption being that the conditional probability density function of the layer is a delta function; based on the third relationship data, using Bayesian statistical theory, determining the first-order moment estimates and second-order moment estimates of the stratigraphic samples of each layer.

[0129] In some embodiments, the first relationship data corresponding to the Monte Carlo integral is obtained. The first relationship data corresponding to the Monte Carlo integral is obtained as shown in the following formula (1):

[0130]

[0131] Among them, a and b are constants, is the definite integral, d is the differential, f is the layer function, h is the layer, f(h) is the layer function of the layer, also represents the layer sample of the layer, f(h)>0, pdf(h) is the probability density function of the layer, E is the expectation, also represents the mean of the layer.

[0132] Before performing Bayesian estimation, it is necessary to assume that in the seismic profile image, the layer h and the bombardment record point c are independent and identically distributed. Based on the Bayesian estimation theory and the independent and identical distribution of the layer and the bombardment record point, the above formula (1) is converted into the following formula (2).

[0133]

[0134] Among them, is the definite integral, d is the differential, h is the horizon, f(h) is the horizon function of the horizon, also represents the horizon sample of the horizon, represents the seismic profile image, represents the posterior distribution of the horizon h, that is, the horizon sample of the horizon. represents the posterior distribution of the seismic profile image. represents the conditional probability density function of the horizon, represents the expectation of layer h in the posterior distribution of layer h, represents the expectation of the seismic section image in the posterior distribution of the seismic section image, and represents the seismic section image under the probability distribution of the shot record point c.

[0135] When multi-layer position tracking is performed in a given seismic profile image, if the layer position of the seismic profile image is determined, the conditional probability density function of the layer position is the (Dirac) function, that is, Then the above formula (2) can be simplified to the following formula (3).

[0136]

[0137] Among them, is the number of samples of the conditional probability density function, and is the number of samples of the posterior distribution of the seismic profile image. is the kth sample obtained from the conditional probability density function, is the jth sample obtained from the posterior distribution of the seismic profile image, and is the summation function.

[0138] According to formula (2) and formula (3), combined with the principle of Bayesian statistics, we can obtain the first-order moment estimate and second-order moment estimate of the sample, that is, the mean and variance of the layer sample.

[0139] Among them, the mean of the layer samples is

[0140] The variance of the layer samples is

[0141] Therefore, the 99% confidence interval for the horizon is estimated to be (μ h -2.576σ h ,μ h +2.576σ h ).

[0142] 404. The computer device determines the confidence interval of each layer based on the mean and variance of each layer in the first seismic image, and for each layer in the layer tracking result, deletes the points that do not belong to the confidence interval of the layer.

[0143] 405. After obtaining the layer tracing results of a plurality of consecutive seismic profile images, the computer device determines a layer position distribution function based on the layer tracing results of the plurality of consecutive seismic profile images, and predicts the layer in the next seismic profile image of the plurality of consecutive seismic profile images based on the layer position distribution function.

[0144] In a possible implementation, the layer position distribution function is determined by using an inter-frame difference method. For example, the layer contour positions of continuous seismic profile images are obtained by using the inter-frame difference method, and the layer distribution function is determined based on the layer contour positions of continuous seismic profile images.

[0145] Among them, the inter-frame difference method obtains the position of the layer contour based on the difference calculation between the layer positions between two adjacent seismic profile images. Assuming that the multiple continuous seismic profile images that have obtained the layer tracking results are X, and the seismic profile image that needs to predict the layer is Y, then the dichotomy method is used to divide each seismic profile image X into two parts, the layer part and the non-layer part, where the layer part is set to 1 and the non-layer part is set to 0. Then the difference in the position between the layers of two adjacent seismic profile images can be calculated. In some embodiments, the first seismic trace, the middle seismic trace, the last seismic trace and the special seismic trace of the fault on the layer can be used as the basis for judgment to calculate the difference D. The difference formula is as follows:

[0146]

[0147] in, Indicates the layer position corresponding to the ith seismic trace of the nth seismic profile image, It represents the layer position corresponding to the ith seismic trace of the n-1th seismic profile image. It represents the difference in the layer corresponding to the i-th seismic trace of the n-th seismic profile image and the n-1-th seismic profile image, where n is any integer greater than 1 and m is any positive integer.

[0148] By analogy, we obtain the differences in seismic trace positions between multiple seismic profile images, fit the layer position distribution function, and thus predict the layer position of the seismic profile image Y.

[0149] It should be noted that the embodiment of the present application is only illustrative of the example of executing the above step 404 and then executing the above step 405. In fact, the computer device can adopt any layer tracking method to obtain the layer tracking result. After obtaining the layer tracking results of multiple continuous seismic profile images, the inter-frame difference method is adopted to obtain the layer position distribution function, and based on the layer position distribution function, the layer in the next seismic profile image is predicted.

[0150] For example, Figure 6 As shown, after acquiring the seismic data, the dynamic time warping algorithm is used to interpret the layer, and then the inter-frame difference method is used to obtain the layer position distribution function. Based on the layer position distribution function, the layer in the next seismic profile image is predicted.

[0151] The layer interpretation method based on uncertainty quantification provided in the embodiment of the present application will determine the uncertainty parameters of the layer tracing results after obtaining the layer tracing results, so as to more accurately understand the accuracy of the layer tracing results, and perform more accurate next step processing based on the layer tracing results, thereby improving the accuracy of the layer interpretation.

[0152] Furthermore, the embodiment of the present application can also predict the layer tracing results of subsequent seismic profile images based on the layer tracing results of multiple consecutive seismic profile images, thereby improving the layer tracing efficiency.

[0153] Figure 7 is a structural schematic diagram of a horizon interpretation device based on uncertainty quantification provided in an embodiment of the present application, such as Figure 7 As shown, the device comprises:

[0154] An acquisition module 701 is used to acquire a first seismic profile image, where the first seismic profile image includes seismic waves of multiple seismic traces;

[0155] A tracking module 702 is used to process the first seismic profile image by using any layer tracking method to obtain a layer tracking result of the first seismic profile image, wherein the layer tracking result indicates a position of at least one layer in the first seismic profile image;

[0156] A determination module 703 is used to determine an uncertainty parameter of each layer based on the distribution of each layer in the first seismic profile image in the layer tracing result, where the uncertainty parameter is used to indicate the uncertainty degree of the layer obtained by processing the first seismic profile image using the layer tracing method;

[0157] The processing module 704 is used to process the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image.

[0158] In a possible implementation, the determination module 703 is used to determine the mean and variance of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image.

[0159] In one possible implementation, the processing module 704 is used to determine the confidence interval of each layer based on the mean and variance of each layer in the first seismic image; for each layer in the layer tracking result, the points that do not belong to the confidence interval of the layer are deleted.

[0160] In one possible implementation, a determination module 703 is used to sample the layer tracking results to obtain layer samples for each layer; based on the layer samples for each layer, the Bayesian theory is used to obtain the first-order moment estimate and the second-order moment estimate of each layer sample; the first-order moment estimate of the layer samples of the layer is determined as the mean of the layer, and the second-order moment estimate of the layer samples of the layer is determined as the variance of the layer.

[0161] In one possible implementation, module 703 is determined to obtain first relationship data corresponding to the Monte Carlo integral, the first relationship data being used to represent the relationship between the stratigraphic samples of each layer, the probability density function of each layer and the mean of each layer; based on the Bayesian estimation theory and the first assumption, the first relationship data is converted into second relationship data, the first assumption being that the layers and the shelling record points in the first seismic profile image are independent and identically distributed, the second relationship data being used to represent the relationship between the stratigraphic samples of each layer, the conditional probability density function, the stratigraphic function and the mean; based on the second assumption, the second relationship data is simplified to obtain third relationship data, the second assumption being that the conditional probability density function of the layer is a delta function; based on the third relationship data, the Bayesian statistical theory is used to determine the first-order moment estimate and the second-order moment estimate of the stratigraphic samples of each layer.

[0162] In a possible implementation, the processing module 704 is used to delete the layer in the layer tracking result, or discard the layer tracking result, when the uncertainty parameter of any layer in the first seismic profile is greater than a first threshold.

[0163] In a possible implementation, the tracking module 702 is used to use a layer tracking model to perform layer tracking on the first seismic profile image to obtain a layer tracking result of the first seismic profile image; or,

[0164] A tracking module is used to obtain a layer tracking result of the first seismic profile image by processing the first seismic profile image using a traditional dynamic time warping algorithm; or,

[0165] The tracking module is used to use the layer tracking model to extract features of the first seismic profile image. The feature map of the first seismic profile image is processed based on the feature map using a traditional dynamic time warping algorithm to obtain the layer tracking result of the first seismic profile image.

[0166] In one possible implementation, the tracking module 702 is used to use a stratigraphic tracking model to perform feature extraction on a first seismic profile image to obtain a first feature map; sample the first feature map to obtain a second feature map; repeatedly input the currently obtained feature map into the stratigraphic tracking model, use the stratigraphic tracking model to perform feature extraction on the input feature map, output the feature map obtained by this feature extraction, sample the feature map obtained by this feature extraction, until a termination condition is met.

[0167] In a possible implementation, the device further includes:

[0168] The prediction module is used to determine the layer position distribution function based on the layer tracing results of multiple consecutive seismic section images after obtaining the layer tracing results of multiple consecutive seismic section images; based on the layer position distribution function, predict the layer in the next seismic section image of the multiple consecutive seismic section images.

[0169] It should be noted that: the stratum interpretation device based on uncertainty quantification provided in the above embodiment is only illustrated by the division of the above functional modules when performing stratum interpretation. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the stratum interpretation test device based on uncertainty quantification provided in the above embodiment belongs to the same concept as the stratum interpretation method embodiment based on uncertainty quantification. The specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0170] In some embodiments, the computer device is provided as a terminal. Figure 8 The terminal 800 includes a processor 801 and a memory 802 .

[0171] The processor 801 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 801 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 801 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 801 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 801 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0172] The memory 802 may include one or more computer-readable storage media, which may be non-transitory. The memory 802 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 802 is used to store at least one program code, which is used to be executed by the processor 801 to implement the horizon interpretation method based on uncertainty quantification provided by the method embodiment of the present application.

[0173] In some embodiments, the terminal 800 may further optionally include: a peripheral device interface 803 and at least one peripheral device. The processor 801, the memory 802 and the peripheral device interface 803 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 803 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 804, a display screen 805, a camera 806, an audio circuit 807, a positioning component 808 and a power supply 809.

[0174] The peripheral device interface 803 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 801 and the memory 802. In some embodiments, the processor 801, the memory 802, and the peripheral device interface 803 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 801, the memory 802, and the peripheral device interface 803 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0175] The display screen 805 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos and any combination thereof. When the display screen 805 is a touch display screen, the display screen 805 also has the ability to collect touch signals on the surface or above the surface of the display screen 805. The touch signal can be input to the processor 801 as a control signal for processing. At this time, the display screen 805 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 805 can be one, and the front panel of the terminal 800 is set; in other embodiments, the display screen 805 can be at least two, which are respectively set on different surfaces of the terminal 800 or are folded; in some other embodiments, the display screen 805 can be a flexible display screen, which is set on the curved surface or folded surface of the terminal 800. Even, the display screen 805 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 805 can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode) and the like.

[0176] The power supply 809 is used to power various components in the terminal 800. The power supply 809 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 809 includes a rechargeable battery, the rechargeable battery can support wired charging or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0177] Those skilled in the art will understand that Figure 8 The structure shown in the figure does not constitute a limitation on the terminal 800, and the terminal 800 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0178] In some embodiments, the computer device is provided as a server. Fig. 9 This is a schematic diagram of the structure of a server provided in an embodiment of the present application. The server 900 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 901 and one or more memories 902, wherein the memory 902 stores at least one program code, and the at least one program code is loaded and executed by the processor 901 to implement the methods provided in the above-mentioned various method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output, and the server may also include other components for implementing device functions, which will not be described in detail here.

[0179] The server 900 is used to execute the steps executed by the server in the above method embodiment.

[0180] An embodiment of the present application also provides a computer-readable storage medium, which stores at least one program code, and the at least one program code is loaded and executed by a processor to implement the layer interpretation method based on uncertainty quantification as described in any of the above implementation methods.

[0181] An embodiment of the present application also provides a computer program product, which includes at least one program code, and the at least one program code is loaded and executed by a processor to implement the layer interpretation method based on uncertainty quantification as described in any of the above implementation methods.

[0182] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected by a communication network. Multiple computer devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.

[0183] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A horizon interpretation method based on uncertainty quantification, characterized in that: The method comprises: Acquire a first seismic profile image, wherein the first seismic profile image includes seismic waves of a plurality of seismic traces; Using any layer tracing method to process the first seismic section image, to obtain a layer tracing result of the first seismic section image, wherein the layer tracing result indicates a position of at least one layer in the first seismic section image; Determine an uncertainty parameter of each layer based on the distribution of each layer in the layer tracing result in the first seismic profile image, wherein the uncertainty parameter is used to indicate the degree of uncertainty of the layer obtained by processing the first seismic profile image using the layer tracing method; The layer tracking result is processed based on the uncertainty parameter of each layer in the first seismic profile image.

2. The method according to claim 1, characterized in that The step of determining the uncertainty parameter of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image comprises: Based on the distribution of each layer in the layer tracking result in the first seismic profile image, the mean and variance of each layer are determined.

3. The method according to claim 2, characterized in that The step of processing the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image comprises: Determining a confidence interval for each horizon based on a mean and a variance of each horizon in the first seismic image; For each layer in the layer tracking result, the points that do not belong to the confidence interval of the layer are deleted.

4. The method according to claim 1 or 2, characterized in that: The step of determining the uncertainty parameter of each layer based on the distribution of each layer in the layer tracking result in the first seismic profile image comprises: Sampling the layer tracking results to obtain layer samples of each layer; Based on the layer samples of each layer, using Bayesian theory, a first-order moment estimation and a second-order moment estimation of each layer sample are obtained; The first-order moment estimate of the layer samples of the layer is determined as the mean of the layer, and the second-order moment estimate of the layer samples of the layer is determined as the variance of the layer.

5. The method according to claim 4, characterized in that The first-order moment estimation and the second-order moment estimation of each layer sample are obtained by using the Bayesian theory based on the layer sample of each layer, including: Acquire first relationship data corresponding to the Monte Carlo integral, where the first relationship data is used to represent the relationship between the layer samples of each layer, the probability density function of each layer and the mean value of each layer; Based on the Bayesian estimation theory and a first assumption, the first relationship data is converted into second relationship data, the first assumption being that the layers and the bombardment recording points in the first seismic profile image are independent and identically distributed, and the second relationship data is used to represent the relationship between the layer sample, the conditional probability density function, the layer function and the mean of each layer; Based on a second assumption, the second relationship data is simplified to obtain third relationship data, wherein the second assumption is that the conditional probability density function of the horizon is a delta function; Based on the third relationship data, the Bayesian statistical theory is used to determine the first-order moment estimate and the second-order moment estimate of the layer samples of each layer.

6. The method according to claim 1, characterized in that The step of processing the horizon tracking result based on the uncertainty parameter of each horizon in the first seismic profile image comprises: When the uncertainty parameter of any layer in the first seismic profile is greater than a first threshold, the layer is deleted from the layer tracking result, or the layer tracking result is discarded.

7. The method according to claim 1, characterized in that The adopting any layer tracing method to process the first seismic profile image to obtain a layer tracing result of the first seismic profile image includes: Using a layer tracing model to perform layer tracing on the first seismic section image to obtain a layer tracing result of the first seismic section image; or, Based on the first seismic profile image, a conventional dynamic time warping algorithm is used to process the first seismic profile image to obtain a horizon tracing result of the first seismic profile image; or, A layer tracing model is used to extract features of the first seismic profile image. A feature map of the first seismic profile image is processed using a traditional dynamic time warping algorithm based on the feature map to obtain a layer tracing result of the first seismic profile image.

8. The method according to claim 7, characterized in that The layer tracking model is used to extract features from the first seismic profile image, and the feature map of the first seismic profile image includes: Using the layer tracking model, extracting features from the first seismic profile image to obtain a first feature map; Sampling the first feature map to obtain a second feature map; Repeat the process of inputting the currently obtained feature map into the layer tracking model, extracting features from the input feature map using the layer tracking model, outputting the feature map obtained by this feature extraction, and sampling the feature map obtained by this feature extraction until a termination condition is met.

9. The method according to claim 1, characterized in that: The method further comprises: After obtaining the layer tracing results of the plurality of continuous seismic section images, determining the layer position distribution function based on the layer tracing results of the plurality of continuous seismic section images; Based on the layer position distribution function, the layer in the next seismic section image of the plurality of continuous seismic section images is predicted.

10. A horizon interpretation device based on uncertainty quantification, characterized in that: The device comprises: An acquisition module, configured to acquire a first seismic profile image, wherein the first seismic profile image includes seismic waves of a plurality of seismic traces; A tracking module, configured to process the first seismic profile image by using any layer tracking method to obtain a layer tracking result of the first seismic profile image, wherein the layer tracking result indicates a position of at least one layer in the first seismic profile image; A determination module, configured to determine an uncertainty parameter of each layer based on the distribution of each layer in the layer tracing result in the first seismic profile image, wherein the uncertainty parameter is used to indicate the degree of uncertainty of the layer obtained by processing the first seismic profile image using the layer tracing method; A processing module is used to process the layer tracking result based on the uncertainty parameter of each layer in the first seismic profile image.

11. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the layer interpretation method based on uncertainty quantification as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: At least one program code is stored in the computer-readable storage medium, and the at least one program code is loaded and executed by the processor to implement the layer interpretation method based on uncertainty quantification as described in any one of claims 1 to 9.

Images