METHOD AND APPARATUS FOR ESTABLISHING AN INTERPRETATION MODEL FOR STRIKE FAILURES, AND INTERPRETATION METHOD AND APPARATUS FOR STRIKE FAILURES

Abstract

The present invention relates to the technical field of seismic fault detection, and implements a method and apparatus for establishing an interpretation model for strike-slip faults, an interpretation method and apparatus for strike-slip faults. The method for establishing an interpretation model for strike-slip faults comprises the steps of: receiving seismic data and logging information; determining initial strike-slip fault characterization images based on the seismic data; adjusting the initial strike-slip fault characterization images based on the logging information and geological evolution laws to obtain spatial topological strike-slip fault images;considering seismic profile data in the seismic data as an input, considering label information of strike-slip fault spatial topological images as an output, constructing training samples according to the seismic profile data and the corresponding strike-slip fault spatial topological images, and constructing a neural network model based on the input and the output; and performing training of the neural network model by means of the training samples, and considering a neural network model trained by learning as an interpretation model for strike-slip faults. According to the embodiments of the present invention, the problems of low detection and recognition accuracy of small-sized faults and heavy workload of seismic interpreters are solved. Fig. 1;

Description

Title of the invention: METHOD AND APPARATUS FOR ESTABLISHING AN INTERPRETATION MODEL FOR STRIKE FAILURES, AND INTERPRETATION METHOD AND APPARATUS FOR STRIKE FAILURES Technical field

[0001] The present invention relates to the technical field of seismic fault detection, and in particular to a method and apparatus for establishing an interpretation model for strike-slip faults, and to an interpretation method and apparatus for strike-slip faults. STATE OF PRIOR ART

[0002] The purpose of this section is to provide background information or context for embodiments of the present invention recited in the claims. The description herein should not be construed as prior art simply because it is included in this section.

[0003] As the most important stratum for exploration and development purposes, the carbonate formation is affected by regional compressive and torsional stresses, and a large number of strike-slip faults are developed therein. Different segments of strike-slip faults with different characteristics or the same characteristics have obvious differences in terms of reservoir control, and they have a positive effect on the reconstruction of the carbonate reservoir and the filling of oil and gas. But at the same time, there are also negative impacts of destroying the condensation preservation conditions, and communicating with the bottom water of the oil field and then reducing the development efficiency of the oil field.Due to the fact that the strike-slip fault has the characteristics that the fault displacement in the vertical direction is small or even not recognizable, the formation structures on both sides of the fault are greatly changed, and the associated faults are complex, there is a great uncertainty in the interpretation of the strike-slip fault in carbonate rocks.

[0004] Currently, the interpretation method for strike-slip faults mainly includes two types, manually and artificial intelligence strike-slip fault analysis and interpretation, respectively. Based on seismic data processing or manual analysis of seismic attribute imaging, this This process is very costly in terms of time and labor, and there are inevitable personal errors. Existing AI fault interpretation processes only characterize large faults through seismic attributes, and cannot achieve precision detection and recognition for small faults, thus resulting in high ambiguity and uncertainty in interpretation, and tending to miss or misdetect faults. In addition, many manual interpretations are required for intervention purposes to supplement corrections and filtrations, and the manual workload remains high.Therefore, there is an urgent need to implement an intelligent interpretation method for strike-slip faults to solve the problems of low detection and recognition accuracy of small-sized faults and heavy characterization workload of seismic interpreters. Statement of the invention

[0005] Due to the low accuracy of detection and recognition of small-sized faults and the heavy characterization workload of seismic interpreters, this solution is proposed to overcome or at least partially solve the above problems.

[0006] According to one aspect, embodiments of the present invention are directed to implementing a method for establishing an interpretation model for strike-slip faults, comprising the steps of:

[0007] receive seismic data and logging information;

[0008] determining initial strike-slip fault characterization images based on seismic data;

[0009] adjusting the initial strike-slip fault characterization images based on the logging information and geological evolution laws in order to obtain spatial topological strike-slip fault images;

[0010] considering the seismic profile data in the seismic data as an input, considering the label information of the strike-slip fault spatial topological images as an output, constructing training samples according to the seismic profile data and the corresponding strike-slip fault spatial topological images, and constructing a neural network model based on the input and the output; and

[0011] performing training of the neural network model using the training samples, and considering a trained neural network model as an interpretation model for strike-slip faults.

[0012] Furthermore, the step of determining initial strike-slip fault characterization images based on the seismic data comprises the steps of:

[0013] determining seismic attributes and seismic reflection features based on the seismic data;

[0014] determining sensitive seismic attributes based on seismic reflection features in conjunction with two modes of seismic attribute analysis in terms of profile attributes and slice attributes; and

[0015] determining initial strike-slip fault characterization images based on sensitive seismic attributes and seismic reflection features.

[0016] Furthermore, the step of adjusting the initial strike-slip fault characterization images based on the logging information and the geological evolution laws in order to obtain spatial topological strike-slip fault images comprises the steps of:

[0017] obtaining corresponding petrophysical information based on the logging information;

[0018] obtaining results of analysis of geological structures which correspond by the corresponding petro-physical information based on the laws of geological evolution; and

[0019] adjusting the initial strike-slip fault characterization images based on the analysis results of corresponding geological structures to obtain the spatial topological strike-slip fault images.

[0020] Furthermore, the neural network model is constructed by a UNet++ neural network; and parameters in the neural network model are trained by learning and optimized by means of an adaptive type moment estimation optimizer.

[0021] Furthermore, the step of training the neural network model using the training samples comprises the steps of:

[0022] use the training samples and the neural network model to construct a loss function as in the following equation: B fl — j v- Loss = -^-) F 01og(M(xJ)) - ) F 01og (1 - M(x,)))

[0023] where, xt denotes the seismic profile data having been input into the model, M(xt) denotes a probability matrix obtained by processing through the neural network model, yt denotes label information of a strike-slip fault spatial topological image corresponding to the seismic profile data, F ( • ) denotes a summation of all elements in a matrix ' ■ \ i denotes a number of seismic profiles, Af denotes a total number of seismic profiles, and P is a balancing coefficient; and

[0024] training the neural network model based on the training samples and the loss function.

[0025] Furthermore, the step of training the neural network model using the training examples comprises the steps of:

[0026] dividing the training samples into a training set and a validation set;

[0027] training the neural network model using the training set;

[0028] respectively inputting seismic profile data in the validation set into the neural network model trained by learning;

[0029] calculating an accuracy based on model output results and strike-slip fault spatial topology images in the validation set; and

[0030] retrain the neural network model if the accuracy is below a set threshold, otherwise complete the training of the neural network model.

[0031] According to another aspect, embodiments of the present invention further provide an apparatus for establishing an interpretation model for strike-slip faults, comprising:

[0032] a receiving module configured to receive seismic data and logging information;

[0033] a strike-slip fault characterization initial image construction module configured to determine strike-slip fault characterization initial images based on the seismic data;

[0034] a strike-slip fault spatial topological image construction module configured to adjust the initial strike-slip fault characterization images based on the logging information and geological evolution laws to obtain strike-slip fault spatial topological images;

[0035] a neural network model building module configured to consider the seismic profile data in the seismic data as an input, consider the label information of the strike-slip fault spatial topological images as an output, construct training samples based on the seismic profile data and the corresponding strike-slip fault spatial topological images, and construct a neural network model based on the input and the output; and

[0036] a neural network model training module configured to train the neural network model using the training samples, and consider a trained neural network model as an interpretation model for strike-slip faults.

[0037] Based on the same concept of the invention, the embodiments of the present invention further implement an interpretation method for strike-slip faults, comprising the steps of:

[0038] receive seismic data from a well for analysis; and

[0039] enter seismic profile data into seismic data in a model interpretation for strike-slip faults which is trained by learning in the method according to any of the above embodiments, in order to obtain images of the strike-slip fault.

[0040] According to another aspect, the embodiments of the present invention further implement an interpretation apparatus for strike-slip faults, comprising:

[0041] a receiving module configured to receive seismic data from a well for analysis; and

[0042] an interpretation module configured to input the seismic profile data in the seismic data into an interpretation model for strike-slip faults that is trained in the method according to any of the above embodiments, to obtain images of the strike-slip fault.

[0043] According to another aspect, embodiments of the present invention further implement a computing device, comprising a memory, a processor, and a computer program stored in the memory, and when executed by the processor, the computer program implements instructions of the method described in any one of the aforementioned embodiments.

[0044] According to another aspect, embodiments of the present invention further implement a computer storage medium in which a computer program is stored, and when executed by a processor of a computing device, the computer program implements instructions of the aforementioned method.

[0045] One or more technical solutions according to the embodiments of the present invention have at least the following technical effects.

[0046] In embodiments of the present invention, initial strike-slip fault characterization images are obtained based on the seismic data, and then strike-slip fault spatial topological images are obtained by improving the initial strike-slip fault characterization images based on well-logging information and geological evolution laws, so as to obtain interpretation results for precision strike-slip faults, and the strike-slip fault spatial topology images correspond to the seismic profile data in the seismic data. Then, the seismic profile data in the seismic data is taken as an input, and the label information of the strike-slip fault spatial topology images is taken as an output; training samples are constructed based on the seismic profile data and the corresponding strike-slip fault spatial topology images; a neural network model is constructed based on the input and the output;Once the neural network model is trained by learning using the training samples, an interpretation model for strike-slip faults is finally obtained, which automatically performs precision interpretation of the strike-slip fault by a small number of training samples, thereby solving the problems of low accuracy in detecting and recognizing small-sized faults and heavy workload in characterizing seismic interpreters.

[0047] The above description is only an overview of the technical solutions of some embodiments of the present invention. In order to understand the technical means of some embodiments of the present invention more clearly, the implementation can be carried out according to the contents of the present invention, and in order to make the above and other objectives, features and advantages of some embodiments of the present invention more obvious and understandable, the specific implementations of some embodiments of the present invention are illustrated below. Brief description of the drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, a brief description of the drawings for the embodiments will be given below. Of course, the drawings in the following description merely illustrate certain embodiments of the present invention, and those skilled in the art may derive other drawings therefrom without any creative effort. In the drawings:

[0049] [Fig. 1] illustrates a flow diagram of a method for establishing an interpretation model for strike-slip faults according to certain embodiments of the present invention;

[0050] [Fig.2] illustrates a schematic diagram of partial 3D seismic data from a work area according to certain embodiments of the present invention;

[0051] [Fig.3] illustrates a flow diagram for determining initial images of strike-slip fault characterization according to certain embodiments of the present invention;

[0052] [Fig.4] illustrates a schematic diagram of original seismic profiles in line 75 according to certain embodiments of the present invention;

[0053] [Fig.5a] illustrates a schematic diagram of slices with consistent attributes of horizon level T74 according to certain embodiments of the present invention;

[0054] [Fig.5b] illustrates a schematic diagram of slices with consistent attributes of horizon level T76 according to certain embodiments of the present invention;

[0055] [Fig.5c] illustrates a schematic diagram of slices with consistent attributes of horizon level T80 according to certain embodiments of the present invention;

[0056] [Fig.6a] illustrates a schematic diagram of event particularities shown in a position marked by a frame in [Fig.4] according to certain embodiments of the present invention;

[0057] [Fig.6b] illustrates another schematic diagram of event particularities shown in a position marked by a frame in [Fig.4] according to certain embodiments of the present invention;

[0058] [Fig.7a] illustrates a schematic diagram of strike-slip fault structures and distributing online profile flaws according to certain embodiments of the present invention;

[0059] [Fig.7b] illustrates a schematic diagram of strike-slip fault structures and distribution of cross-line profile faults according to certain embodiments of the present invention;

[0060] [Fig.8] illustrates a schematic diagram of a process of determining seismic sensitive attributes based on seismic reflection features according to certain embodiments of the present invention;

[0061] [Fig.9] illustrates a flow diagram of a process of determining spatial topological images of strike-slip faults according to certain embodiments of the present invention;

[0062] [Fig. 10] illustrates a schematic diagram of strike-slip fault spatial topological images and original seismic amplitude profile data according to certain embodiments of the present invention;

[0063] [Fig. 11] illustrates a schematic diagram of the structure of a UNet++ neural network according to certain embodiments of the present invention;

[0064] [Fig. 12] illustrates a training flow diagram of a neural network model according to certain embodiments of the present invention;

[0065] [Fig. 13] illustrates another flow diagram of training a neural network model according to some embodiments of the present invention;

[0066] [Fig. 14] illustrates a flow diagram of an interpretation method for strike-slip faults according to some embodiments of the present invention;

[0067] [Fig. 15] illustrates a structure block diagram of an apparatus for establishing an interpretation model for strike-slip faults according to some embodiments of the present invention;

[0068] [Fig. 16] illustrates a structure block diagram of an interpretation apparatus for strike-slip faults according to some embodiments of the present invention; and

[0069] [Fig. 17] illustrates a structure block diagram of a computing device according to some embodiments of the present invention.DETAILED DESCRIPTION

[0070]

[0071]

[0072] Reference numbers 151: receiving module; 152: module for constructing initial images for strike-slip fault characterization;

[0073] 153: module for constructing spatial topological images of strike-slip fault;

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090] 154: neural network model construction module; 155: neural network model training module; 161: receiving module; 162: interpretation module; 1702: computing device; 1704: processor; 1706: memory; 1708: drive mechanism; 1710: input / output interface; 1712: input device; 1714: output device; 1716: presentation device; 1718: graphical user interface; 1720: network interface; 1722: communications link; 1724: communications bus.In order for those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in certain embodiments of the . The present invention will be clearly and fully described below with reference to the drawings in the embodiments of the present invention. Of course, those described are only a part, rather than the whole, of the embodiments of the present invention. Based on certain embodiments of the present invention, any other embodiments obtained by those of ordinary skill in the art without any creative effort should be within the scope of protection of the present invention.

[0091] It should be noted that the terms "first," "second," and similar terms in the description and claims of the present invention and the drawings above are used to distinguish like objects and not necessarily to describe a particular order or precedence. It should be understood that the data so used may be interchanged under appropriate circumstances so that the embodiments of the present invention described herein may be implemented in an order other than those illustrated or described herein. Furthermore, the terms "comprising," "comprise," "having," and any variation thereof are intended to cover all non-exclusive inclusions.For example, a process, method, apparatus, product or device, which comprises a series of steps or units, is not necessarily limited to those explicitly listed, but may include other steps or units which are not expressly listed or which are not inherent in the process, method, apparatus, product or device.

[0092] Existing interpretation methods for strike-slip faults only characterize large faults through seismic attributes, and cannot achieve precision detection and recognition for small faults, thereby resulting in high ambiguity and uncertainty in interpretation, and tending to miss or misdetect faults. In addition, many manual interpretations are required for intervention purposes to complete corrections and filtrations, and the manual workload remains high. Due to the low accuracy of detection and recognition of small faults and the heavy workload of seismic interpreters, this solution is proposed to overcome or at least partially solve the above problems.

[0093] With reference to [Fig.l], some embodiments of the present invention implement a method for establishing an interpretation model for strike-slip faults. The present invention provides the steps for performing the method as described in the embodiments or in the flow diagrams, but more or less steps for performing may be included depending on the routine or non-creative work. Any order of performing the steps of which The list set forth in the embodiments is only one of many orders of execution of the steps and does not represent a single order of execution. In the case of an actual system or an actual apparatus product, the steps may be performed in order or simultaneously depending on the method illustrated in the embodiments or the drawings. Specifically, as illustrated in [Fig.l], the method may comprise the steps of:

[0094] S100: receive seismic data and logging information;

[0095] S200: determine initial strike-slip fault characterization images in based on seismic data;

[0096] S300: adjust the initial strike-slip fault characterization images in based on logging information and geological evolution laws to obtain spatial topological images of strike-slip faults;

[0097] S400: consider seismic profile data in seismic data as being an input, considering the label information of the strike-slip fault spatial topological images as an output, constructing training samples according to the seismic profile data and the corresponding strike-slip fault spatial topological images, and constructing a neural network model based on the input and the output;

[0098] S500: Perform training of the neural network model by means of training samples, and consider a neural network model trained by learning as an interpretation model for strike-slip faults.

[0099] In the embodiments of the present invention, initial strike-slip fault characterization images are obtained based on the seismic data, and then spatial topological strike-slip fault images are obtained by improving the initial strike-slip fault characterization images based on logging information and geological evolution laws, so as to obtain interpretation results for precision strike-slip faults, and the spatial topological strike-slip fault images correspond to the seismic profile data in the seismic data.Then, the seismic profile data in the seismic data is taken as an input, and the label information of the strike-slip fault spatial topological images is taken as an output; training samples are constructed based on the seismic profile data and the corresponding strike-slip fault spatial topological images; a neural network model is constructed based on the input and the output; after the neural network model is trained by learning using the training samples, a model . interpretation model for strike-slip faults is finally obtained, a model that automatically performs a precision interpretation of the strike-slip fault by a small number of training samples, thereby solving the problem of low accuracy of detection and recognition of small faults and the heavy workload of seismic interpreters.

[0100] In some embodiments, the received seismic data may originate from platform institutions, such as seismic information centers in relevant regions or countries, and the logging information may be information, such as geological structures and hydrological information, which can be obtained and used for geological interpretation. In embodiments of the present invention, [Fig. 2] illustrates partial 3D seismic data of a working area having an original area of ​​about 200 square kilometers. Once 3D migration seismic data that corresponds to the strike-slip fault is intercepted, a corresponding size is 400 Inlines x 400 Crosslines x 448 time sampling points, a time interval is from 4502 ms to 5400 ms, and a time sampling interval is 2 ms. As illustrated in [Fig.2], a north high "X" type shear system is developed in the partial 3D seismic data area of ​​the working area, where the formation is generally flat, the development process is dominated by marine sediment, and strike-slip faults are mainly developed, as shown by the arrow in [Fig.2]. And, the strike-slip faults comprise faults of multiple levels and the system structure is complex. The embodiments of the present invention mainly focus on the fault precision interpretation of fault No. 5 (namely the fault in [Fig.2]) among the strike-slip faults, the label information of the strike-slip fault spatial topological images may, for example, comprise part or all of the strike-slip fault spatial topological images.

[0101] With reference to [Fig. 3], in some embodiments, the step of determining initial strike-slip fault characterization images based on the seismic data may comprise the steps of:

[0102] S210: determine seismic attributes and seismic reflection features based on seismic data;

[0103] S220: determine sensitive seismic attributes based on features seismic reflection in conjunction with two modes of seismic attribute analysis in terms of profile attributes and slice attributes;

[0104] S230: determine initial strike-slip fault characterization images in based on sensitive seismic attributes and seismic reflection characteristics.

[0105] Specifically, firstly, the received seismic data is pre-processed, and spectral decomposition is performed by a sustained wavelet transform and a time-frequency sustained wavelet transform to obtain the seismic data in different frequency domains, meanwhile, the seismic data in an angular domain is divided by angular decomposition. Various seismic attributes, such as coherence and curvature, are extracted based on the pre-processed seismic data. The seismic attribute is a measure of geometric, kinematic, dynamic, and statistical characteristics of the seismic data. The seismic attribute can also be understood as a characteristic for describing and quantifying the seismic data, and it is a subset of all the information in the original seismic data.

[0106] Currently, seismic attributes are mainly classified into amplitude attributes, frequency attributes, time attributes and geometry attributes. The amplitude attributes include root mean square amplitude, average absolute amplitude, maximum peak amplitude, average peak amplitude, maximum trough amplitude, average trough amplitude and average energy. The frequency attributes mainly include average instantaneous frequency, root mean square instantaneous frequency, reflection width, etc. The time attributes mainly include coherence, tilt angle, azimuth angle, curvature, etc.

[0107] Amplitude attributes are generally used for seismic lithology interpretation and reservoir prediction. Frequency attributes are mainly found in seismic wave propagation, the frequency spectrum changes due to attenuation, absorption and geometric scattering of the formation. Frequency attributes are important reference factors for reservoir thickness, and seismic wave scattering, absorption and attenuation, and time attributes are important reference factors for structural analysis purposes.

[0108] There are mature algorithms for extracting seismic attributes. For example, the third-generation coherent body algorithm can be used to extract coherent attributes. The third-generation coherent body algorithm is based on an eigenvalue structure algorithm that needs to calculate an eigenvalue of a covariance matrix, and has the advantages of high resolution and strong noise removal capability.

[0109] As illustrated in [Fig.4], the original seismic profile in line 75 is selected, it includes three horizons T74, T76 and T80 having different depths. The results illustrated in Figures 5a, 5b and 5c can be obtained once The coherent attribute slices are extracted from the three horizons T74, T76 and T80 with different depths. It can be seen that the coherent attribute slices of different horizons have rather different sharpness because the faults are influenced by the surrounding sedimentary environment. The T74 horizon is an intense event where distortion and dislocation are obvious, and the coherent attributes present very clear fault morphology, namely, the boundary characterization of large-scale faults is consistent and complete, there is no excessive background information interference around the faults, the associated secondary faults are relatively fractured, and the interior detection effect is slightly poor.The event reflections at the fault positions in the T76 and T80 horizons are disordered, the distortions are abnormal and incoherent, the background information in the coherent attributes is confused, the boundaries of large-sized faults are unclear, the interior information is interrupted and confused, and the detection is vague. Similarly, corresponding seismic attribute extraction results can be obtained based on other seismic attribute extraction algorithms.

[0110] In some embodiments, the seismic reflection features may show change characteristics of seismic attributes such as event, waveform, and seismic amplitude at different frequencies. [Fig. 6a] illustrates event features shown in a position marked by a box in [Fig. 4] at low frequency, and [Fig. 6b] illustrates event features shown in a position marked by a box at high frequency. Comparing [Fig. 6a] and [Fig. 6b], it can be seen that the event signal is stronger at low frequency. There are three obvious event dislocations in the lower part of [Fig. 6a], and the boundaries of large faults on both sides are clear. However, as illustrated in [Fig.6b], at high frequency, the number of events at the same position increases obviously, bifurcations and mergers appear, and the intermediate parts of large-sized faults on both sides can be determined as small-sized faults. The seismic reflection features of seismic profiles at various frequencies are varied, and the fault distribution morphologies are different. Therefore, the fault distribution situations should be obtained by comprehensive analysis of seismic reflection features at different frequencies.

[0111] Furthermore, since there are currently hundreds of seismic attributes, too many attributes will increase the computational load, and there is overlapping and redundancy of information among the seismic attributes, thus affecting the fault prediction result. Therefore, it is necessary to determine the sensitive seismic attributes. to strike-slip fault reflection to improve strike-slip fault prediction accuracy.

[0112] In some embodiments, the seismic attributes may be analyzed from two perspectives, namely, slices and profiles. For example, in Figures 5a, 5b, and 5c, coherent attribute slices are extracted from three horizons T74, T76, and T80 having different depths, respectively, so as to analyze the seismic attributes from the slice perspective. Figures 7a and 7b illustrate structure configurations and strike-slip fault distributions on line and cross-line profiles, respectively, in which breakpoints are corresponding salient positions on the fault profiles extracted from the seismic attribute slices, and marked box positions indicate fault distribution ranges.It can be seen that strike-slip fault structures on the line profile are obvious with rich characteristics and fixed and concentrated positions, and strike-slip fault structures on the line profile are distributed disorderly and vaguely. In order to reduce the error of the prediction result, the transverse line profile is selected as the main profile for prediction in the embodiment of the present invention, and meanwhile, the construction of the spatial topological structure of strike-slip faults on the line profile is subjected to breakpoint connection line constraints by using the transverse line breakpoints, so as to further ensure the characterization accuracy of strike-slip faults.

[0113] Furthermore, the seismic sensitive attributes can be determined by filtering the seismic attributes obtained based on the seismic reflection features using the two seismic attribute analysis modes in terms of profile attributes and slice attributes. As illustrated in [Fig.8], in the process of determining the seismic sensitive attributes based on the seismic reflection features, the seismic sensitive attributes of faults of different scales can be determined by analyzing the instantaneous amplitude, maximum positive curvature, and coherent attribute slices of the T74 horizon in the Inline 180 profile together with the seismic reflection features that correspond to the profile, so as to better recognize small-sized faults.

[0114] Once the sensitive seismic attributes are determined, it is possible to observe the imaging differences of large-scale and medium-scale strike-slip faults, especially small-scale at different frequencies based on the combination of seismic profile, slice and cube. Meanwhile, the distribution characteristics of the three-dimensional spatial geological structure are monitored to complete the characterizations of the main faults and the zones of marginal and internal secondary faults thereof, crush zones and other micro-geological structures, and further optimize the characterization results based on different amplitudes of strike-slip faults and the distortion and disturbance of the waveform with the change of fault morphology, etc. as indicated by the seismic reflection features, to thereby obtain the initial strike-slip fault characterization images.

[0115] With reference to [Fig.9], in some embodiments, the step of adjusting the initial strike-slip fault characterization images based on the logging information and geological evolution laws to obtain spatial topological strike-slip fault images may comprise the steps of:

[0116] S310:obtaining corresponding petro-physical information based on the logging information;

[0117] S320: Obtaining results of analysis of geological structures which correspond by the corresponding petrophysical information based on the laws of geological evolution;

[0118] S330: adjust the initial strike-slip fault characterization images by based on the results of analysis of corresponding geological structures in order to obtain the spatial topological images of strike-slip faults.

[0119] Specifically, the logging information may be information, such as geological structures and hydrological information, which can be obtained and used for geological interpretation. Petrophysical information relating to the strike-slip fault zone can be obtained from the logging information. Based on the laws of geological evolution (including, but not limited to, dynamic, geometric, kinematic and geological mechanisms formed in strike-slip faults), the corresponding results of the analysis of the geological structure can be obtained by the corresponding petrophysical information. The geological structure refers to the morphology left by the deformation or displacement of rock strata or rock masses under the effect of internal and external stresses of the earth.Using the corresponding results of geological structure analysis, the initial strike-slip fault characterization images obtained previously can be optimized and adjusted to obtain the spatial topological images of strike-slip fault. As shown in [Fig. 10], the spatial topological images of strike-slip fault with different branching types are finally obtained by corresponding interpretation processes of Figs. 3 and 9 for the original seismic amplitude profile data.

[0120] In some embodiments, the neural network model is constructed by a UNet++ neural network; and parameters in the neural network model are trained and optimized using an adaptive-type moment estimation optimizer.

[0121] Specifically, the structure diagram of the UNet++ neural network is shown in [Fig. 11], including an upsampling layer, a downsampling layer, and a skip connection layer for realizing channel dimension superposition. The adaptive type moment estimation optimizer is an optimization algorithm of a random objective type function based on a first-order gradient, and is used for the purpose of iteratively updating the weight of the neural network based on the training data. The adaptive type moment estimation optimizer has the advantages of simple implementation, high computational efficiency, and low storage capacity requirements, and is suitable for optimization purposes concerning large-scale parameters and data.

[0122] With reference to [Fig. 12], in some embodiments, the step of training the neural network model using the training samples comprises the steps of:

[0123] S510: Use training samples and neural network model to construct a loss function as in the following equation: R fl — Fl ',v Loss = - 77 > Olog(Mxj)) - ) Ffy^log (1 - Mf^))) NZ—ii=0 N ^—ij -0

[0124] where, xi denotes the seismic profile data having been inputted into the model, M(xi) denotes a probability matrix obtained by processing xi through the neural network model, yt denotes label information of a strike-slip fault spatial topological image corresponding to the seismic profile data, FL) denotes a summation of all elements in a matrix C ■ >, i denotes a number of seismic profiles, -V denotes a total number of seismic profiles, and F is a balancing coefficient.

[0125] S520: Perform training of the neural network model based on the training samples and the loss function.

[0126] Specifically, the training samples are the seismic profile data in the seismic data and the spatial topological images of strike-slip faults. Traditional fault detection often adopts a cross-entropy loss function, which has certain limitations due to the extremely small proportion of faults in the data. In order to alleviate the unbalanced proportions of positive and negative samples in the fault detection process, faults, the embodiments of the present invention adopt the balanced cross-entropy for the loss function of the model, where $ is a balancing coefficient to control the weights of positive and negative samples in the total loss. In the embodiments of the present invention, since the positive samples (faults) represent a small proportion in the whole samples, and in combination with the experimental tests, ends up being set to 0.75. The neural network model can be trained by learning based on the training samples and the loss function.

[0127] With reference to [Fig. 13], in some embodiments, the method for establishing an interpretation model for strike-slip faults may comprise the steps of:

[0128] S610: divide the training samples into a training set and a validation set;

[0129] S620: train the neural network model by means of the learning set;

[0130] S630: respectively input seismic profile data into the set of validation in a neural network model trained by learning;

[0131] S640: calculate an accuracy based on model output results and on spatial topological images of strike-slip faults in the validation set;

[0132] S650: Retrain the neural network model by learning if the accuracy is less than a set threshold, otherwise complete the training of the neural network model.

[0133] Specifically, the training samples include the seismic profile data in the seismic data and the strike-slip fault spatial topology images. The training samples are divided into a training set and a validation set. Given the time cost of manually labeling profiles and conventional deep learning ratios, a ratio of the training set to the verification set is 10:1 in embodiments of the present invention.After the neural network model is trained by learning with the training set, the seismic profile data in the validation set are respectively input into the trained neural network model, and the accuracy is calculated based on the output results of the model and the spatial topological images of strike-slip fault in the validation set.

[0134] Since the final prediction results, i.e., the results output from the model, constitute a probability graph, which shows that the probability value of each point in the final prediction results graph is between 0 and 1. As the probability value approaches 1, the possibility that the corresponding point is a fault increases, and as the probability value approaches 0, the possibility that the corresponding point is a fault decreases. The probability graph can be split with a threshold of 0.5 to obtain a binary graph containing only 0 and 1, so that the accuracy can be calculated with the strike-slip fault spatial topological images.

[0135] The formula for precision is: TP + TN yj — _________________________ TP + TN + FP + FN

[0136] where Ace denotes the accuracy, TP indicates that the strike-slip fault spatial topological image is a positive example, and the model output result is also a positive example; TN indicates that the strike-slip fault spatial topological image is a negative example, and the model output result is also a negative example; FP indicates that the strike-slip fault spatial topological image is a negative example, and the model output result is a positive example; FN indicates that the strike-slip fault spatial topological image is a positive example, and the model output result is a negative example; positive example indicates that the value is 1, and negative example indicates that the value is 0.

[0137] TP, TN, FP and FN in accuracy can be calculated from Table 1:

[0138] Table 1: Calculation table of TP, TN, FP and FN Predicted value = 1 (Model output) Predicted value = 0 (Model output) True value = 1 (Spatial topological image of strike-slip fault) TP FN True value = 0 (Spatial topological image of strike-slip fault) FP TN

[0139] The neural network model is retrained by learning if the accuracy is lower than a set threshold, otherwise the training of the neural network model is completed. In the embodiments of the present invention, the threshold is set in the range of 90% ±5% in consideration of preventing the possibility that the accuracy of the prediction result decreases to some extent due to the existence of individual profiles with large differences in the seismic data.

[0140] Based on the same concept of the invention, with reference to [Fig. 14], in certain embodiments, the interpretation method for strike-slip faults may comprise the steps of:

[0141] S710: receive seismic data from a well for analysis;

[0142] S720: Enter seismic profile data into seismic data in an interpretation model for strike-slip faults that is trained by learning in the method described in any of the embodiments mentioned above, in order to obtain images of the strike-slip fault.

[0143] In accordance with the above method for establishing an interpretation model for strike-slip faults, some embodiments of the present invention further provide an apparatus for establishing an interpretation model for strike-slip faults. Referring to [Fig. 15], in some embodiments, the apparatus for establishing an interpretation model for strike-slip faults may comprise:

[0144] a receiving module 151 configured to receive seismic data and logging information;

[0145] a strike-slip fault characterization initial image construction module 152 configured to determine strike-slip fault characterization initial images based on the seismic data;

[0146] a strike-slip fault spatial topological image construction module 153 configured to adjust the initial strike-slip fault characterization images based on the logging information and geological evolution laws to obtain strike-slip fault spatial topological images;

[0147] a neural network model building module 154 configured to consider the seismic profile data in the seismic data as an input, consider the label information of the strike-slip fault spatial topological images as an output, construct training samples based on the seismic profile data and the corresponding strike-slip fault spatial topological images, and construct a neural network model based on the input and the output; and

[0148] a neural network model training module 155 configured to train the neural network model using the training samples, and consider a trained neural network model as an interpretation model for strike-slip faults.

[0149] Based on the same concept of the invention, in correspondence with the above method for establishing an interpretation model for strike-slip faults, Some embodiments of the present invention further implement a method for interpreting strike-slip faults. Referring to [Fig. 16], in some embodiments, the strike-slip fault interpretation apparatus may comprise:

[0150] a receiving module 161 configured to receive seismic data from a well for analysis purposes;

[0151] an interpretation module 162 configured to input the seismic profile data in the seismic data into an interpretation model for strike-slip faults that is trained in the method described in any one of the embodiments mentioned above, to obtain images of the strike-slip fault.

[0152] For ease of description, the above apparatus is described as being divided into different units based on its functions and described respectively. Of course, the functions of the different units can be realized in the same one or more software and / or hardware when the present invention is implemented.

[0153] It should be noted that the user information (including, but not limited to, information relating to the user's device, the user's personal information, etc.) and data (including, but not limited to, data for analysis purposes, stored data, displayed data, etc.) involved in the embodiments of the present invention are all information and data permitted for use by the user and fully authorized by all parties.

[0154] Embodiments of the present invention further implement a computing device. As illustrated in [Fig. 17], in some embodiments of the present invention, the computing device 1702 may include one or more processors 1704, such as one or more central processing units (CPUs) or graphics processors (GPs), and each processing unit may implement one or more hardware threads. The computing device 1702 may further include any memory 1706 configured to store any type of information such as codes, settings, data, etc.In a specific embodiment, a computer program is stored in memory 1706 and is capable of being executed on processor 1704, and when executed by processor 1704, the computer program may implement instructions of a method according to any of the above embodiments. Without limitation, for example, memory 1706 may include any one or combinations of any type of RAM (random access memory), any type of ROM (read only memory), a flash memory device, a hard disk drive, an optical disk drive, etc. More generally, . Any memory may use any technology for storing information. Furthermore, any memory may provide volatile or non-volatile retention of information. Furthermore, any memory may represent a fixed or removable component of the computing device 1702. In a scenario where the processor 1704 executes associated instructions stored in any memory or combinations of memories, the computing device 1702 may perform any operation of the associated instructions. The computing device 1702 further includes one or more drive mechanisms 1708 (e.g., a hard disk drive mechanism, an optical disk drive mechanism, etc.) interacting with any memory.

[0155] The computing device 1702 may further include an input / output (I / O) interface 1710 configured to receive different inputs (via an input device 1712) and to provide different outputs (via an output device 1714). A specific output mechanism may include a presentation device 1716 and an associated graphical user interface (GUI) 1718. In other embodiments, the input / output (I / O) interface 1710, the input device 1712, and the output device 1714 may not be included, and the device is used solely as a computing device in the network. The computing device 1702 may further include one or more network interfaces 1720 configured to exchange data with other devices via one or more communication links 1722.One or more communication buses 1724 couple together the components described above.

[0156] The communication link 1722 may be implemented in any manner, for example, by a local area network, by a wide area network (e.g., the Internet), a point-to-point connection, etc., or by any combination thereof. The communication link 1722 may include any combination of a wired link, a wireless link, a router, a gateway function, a name server, etc. governed by any protocol or by a combination of protocols.

[0157] The present invention is described with reference to a flowchart and / or a block diagram of the method, apparatus (system), computer-readable storage medium and computer program product according to embodiments of the present invention. It should be appreciated that each flow and / or each block in the flowchart and / or the block diagram, and combinations of the flows and / or the blocks in the flowchart and / or the block diagram, may be implemented by means of computer program instructions. The computer program instructions may be implemented in a general-purpose computer, a special-purpose computer, an embedded processor or a processor of another device. programmable data processing, for forming a machine such that the instructions, which are executed by the computer or by the processor of another programmable data processing device, generate a means for carrying out the functions specified in one or more flows in the flow diagram and one or more blocks in the block diagram.

[0158] The computer program instructions may also be stored in a computer-readable memory capable of causing the computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory generate an article of manufacture comprising an instruction device that implements the designated function(s) in one or a plurality of flows in the flow diagram and / or in one or a plurality of blocks in the block diagram.

[0159] The instructions of this computer program may also be loaded onto a computer or other programmable data processing devices, such that a series of execution steps are executed on the computer or other programmable devices to generate the processing performed by the computer, whereby the instructions executed on the computer or other programmable devices provide the steps for implementing the designated function in one or a plurality of flows in the flow diagram and / or in one or a plurality of blocks in the block diagram.

[0160] In one possible configuration, the computing device comprises one or more processors (CPUs), an input / output interface, a network interface and memory.

[0161] The memory may comprise the form of volatile memory, random access memory (RAM) and / or non-volatile memory such as read only memory (ROM) or flash RAM, etc. among the computer readable media. The memory is an example of the computer readable media.

[0162] The computer-readable medium includes permanent and non-permanent, removable and non-removable media, which may perform the storage of information in any method or technique. The information may be computer-readable instructions, data structures, program modules, or other data. An example of the computer storage medium includes, but is not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other techniques of memory, a compact disc read-only memory (CD-ROM), a digital versatile disc (DVD) or other optical storage, magnetic tape cassettes, magnetic disks or other magnetic storage device or any other non-transmission medium, which can be used for the purpose of storing information accessible to a computing device. As defined herein, computer-readable medium does not include computer-readable media of the temporary type (transient media), such as modulated digital signal and carrier wave.

[0163] Those skilled in the art will appreciate that any embodiment of the present invention may be implemented as a method, system, or computer program product. Accordingly, the present invention may take the form of an all-hardware embodiment, an all-software embodiment, or a combination of software and hardware. Further, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, magnetic disk memory, CD-ROM, optical storage, etc.) containing computer-usable program codes therein.

[0164] Embodiments of the present invention may be described in the general context of computer-executable instructions that are executed by the computer, e.g., the program module. In general, the program module comprises a routine, program, object, component, data structure, etc. performing a particular task or realizing a particular abstract data type. Embodiments of the present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In distributed computing environments, the program modules may be located in the local and remote computer storage media including the storage device.

[0165] It should also be understood that, in embodiments of the present invention, the term "and / or" is only an association relationship that describes the associated objects, indicating that there can be three relationships. For example, A and / or B can mean that A exists alone, A and B both exist, and B exists alone. In addition, the character " / " in the present invention generally indicates that the contextually associated objects are in an "or" relationship.

[0166] The embodiments herein are all described in a step-by-step manner, and the same or similar parts of the embodiments may refer to each other. Each embodiment emphasizes its distinctions. compared to other embodiments. In particular, the system embodiment is described in a simple manner since it is substantially similar to the method embodiment, and it is sufficient to refer to the descriptions of the method embodiment for the relevant part.

[0167] In the description of the present invention, the description of the reference terms "an embodiment", "certain embodiments", "an example", "a specific example" or "examples" and the like means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment(s) or example(s) are included in at least one embodiment or example of the present invention. In the present invention, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples as appropriate.Additionally, those skilled in the art may combine different embodiments or examples described in the present invention and features thereof if there is no contradiction between them.

[0168] These modes described above are merely embodiments of the present invention, rather than limitations to the present invention. For those skilled in the art, the present invention is intended to cover any modification or variation. Any modifications, equivalent substitutions, improvements, etc. that are made in the spirit and principle of the present invention should fall within the scope of the claims of the present invention.

Claims

1. Claims A method for establishing an interpretation model for strike-slip faults, characterized in that the method for establishing an interpretation model for strike-slip faults comprises the steps of: receiving seismic data and well-logging information (S 100); determining initial strike-slip fault characterization images based on the seismic data (S200); adjusting the initial strike-slip fault characterization images based on the log information and geological evolution laws to obtain spatial topological strike-slip fault images (S300), wherein adjusting the initial strike-slip fault characterization images based on the log information and geological evolution laws to obtain spatial topological strike-slip fault images comprises: obtaining corresponding petrophysical information based on the logging information (S310); obtaining corresponding geological structure analysis results by the corresponding petrophysical information based on the geological evolution laws (S320); and adjusting the initial strike-slip fault characterization images based on the corresponding geological structure analysis results to obtain the strike-slip fault spatial topological images (S330); considering the seismic profile data in the seismic data as an input, considering the label information of the strike-slip fault spatial topological images as an output, constructing training samples according to the seismic profile data and the corresponding strike-slip fault spatial topological images, and constructing a neural network model based on the input and the output (S400); and perform the training of the neural network model using the training samples, and consider a network model of neurons trained by learning as an interpretation model for strike-slip faults (S500).

2. The method for establishing an interpretation model for strike-slip faults according to claim 1, wherein the step of determining initial strike-slip fault characterization images based on the seismic data comprises the steps of: determining seismic attributes and seismic reflection features based on the seismic data (S210), wherein the seismic reflection features indicate different strike-slip fault amplitudes and the distortion and disturbance of the waveform with the change of the fault morphology;determining sensitive seismic attributes based on the seismic reflection features in conjunction with two seismic attribute analysis modes in terms of profile attributes and slice attributes (S220) wherein the sensitive seismic attributes are seismic attributes that are sensitive to the reflection of the strike-slip fault; and determining initial strike-slip fault characterization images based on the sensitive seismic attributes and the seismic reflection features (S230).;

3. A method for establishing an interpretation model for strike-slip faults according to claim 1, wherein the neural network model is constructed by a UNet++ neural network; and parameters in the neural network model are trained by learning and optimized by means of an adaptive type moment estimation optimizer.

4. A method for establishing an interpretation model for strike-slip faults according to claim 1, wherein the step of training the neural network model using the training samples comprises the steps of: using the training samples and the neural network model to construct a loss function (S510) which is an equation of: / ? fl — Loss = > f(yiOlog(M(ril)) ? N Z_ij=0 ' AI J—>i=0 where, xi denotes the seismic profile data having been input into the model, denotes a probability matrix obtained by processing xt through the neural network model, Yi denotes label information of a strike-slip fault spatial topological image corresponding to the seismic profile data, F ' • denotes a summation of all elements in a matrix t \ i denotes a number of seismic profiles, N denotes a total number of seismic profiles, and is a balancing coefficient; and perform training of the neural network model based on the training samples and loss function.

5. A method for establishing an interpretation model for strike-slip faults according to claim 1, wherein the step of training the neural network model using the training examples comprises the steps of: dividing the training samples into a training set and a validation set (S610); training the neural network model using the training set (S620); respectively inputting seismic profile data in the validation set into a neural network model trained by learning (S630); calculating an accuracy based on model output results and strike-slip fault spatial topology images in the validation set (S640); and retraining the neural network model if the accuracy is below a set threshold, otherwise completing the training of the neural network model (S650).

6. An interpretation method for strike-slip faults, characterized in that the interpretation method for strike-slip faults comprises the steps of: receive seismic data from a well for analysis (S710); and inputting the seismic profile data into the seismic data into an interpretation model for strike-slip faults that is trained by learning in the process according to any

7. of claims 1 to 6, with a view to obtaining images of the strike-slip fault (S720). An apparatus for establishing an interpretation model for strike-slip faults, characterized in that the apparatus for establishing an interpretation model for strike-slip faults comprises: a receiving module (151) configured to receive seismic data and logging information; a strike-slip fault characterization initial image construction module (152) configured to determine strike-slip fault characterization initial images based on the seismic data; a strike-slip fault spatial topological image construction module (153) configured to adjust the initial strike-slip fault characterization images based on the logging information and geological evolution laws to obtain strike-slip fault spatial topological images, wherein the strike-slip fault spatial topological image construction module is further configured to obtain corresponding petrophysical information based on the logging information; obtain corresponding geological structure analysis results by the corresponding petrophysical information based on the geological evolution laws and adjust the initial strike-slip fault characterization images based on the corresponding geological structure analysis results to obtain the strike-slip fault spatial topological images; a neural network model building module (154) configured to consider the seismic profile data in the seismic data as an input, consider the label information of the strike-slip fault spatial topological images as an output, construct training samples based on the seismic profile data and the corresponding strike-slip fault spatial topological images, and construct a neural network model based on the input and the output; and a neural network model training module (155) configured to train by learning the neural network model neurons using the training samples, and consider a neural network model trained by learning as an interpretation model for strike-slip faults.

8. An interpretation apparatus for strike-slip faults, characterized in that the interpretation apparatus for strike-slip faults comprises: a receiving module (161) configured to receive seismic data from a well for analysis; and an interpretation module (162) configured to input the seismic profile data in the seismic data into an interpretation model for strike-slip faults which is trained by learning in the method according to any one of claims 1 to 5, in order to obtain images of the strike-slip fault.

9. A computing device, comprising a memory (1706), a processor (1704), and a computer program stored in the memory (1706), wherein, when executed by the processor (1704), the computer program implements instructions of the method according to any one of claims 1 to 6.

Images