Method and device for classifying and characterizing reservoirs, electronic device and storage medium

Abstract

The application provides a reservoir classification and characterization method and device, electronic equipment and a storage medium. A waveform component selection model is established in advance, the waveform component selection model is obtained based on sample data sets, each sample data in the sample data set includes second waveform component data and a reservoir type corresponding to the second waveform component data, the reservoir type is determined based on logging data and production geological information, after obtaining seismic data of a target area, waveform analysis is performed on the seismic data to obtain a plurality of first waveform component data, then the waveform component data is input into the waveform component selection model to determine the first waveform component corresponding to each reservoir type, and then the dominant waveform component body is determined based on the first waveform component, each reservoir is characterized by the dominant waveform component body, the classification and characterization of the reservoir is realized, and the accuracy of the characterization of the reservoir can be improved.

Description

Technical Field

[0001] This application relates to the field of oil and gas geophysical exploration, and in particular to a method, apparatus, electronic device and storage medium for classifying and characterizing reservoirs. Background Technology

[0002] Carbonate reservoirs are controlled by geological factors such as sedimentary environment, fluid properties, tectonic conditions, and karst landforms. The dominant controlling factors vary significantly under different geological backgrounds. Based on the development characteristics and dominant controlling factors of fracture-vuggy carbonate reservoirs, these reservoirs can be classified into different karst models, evolutionary stages, and fracture-vuggy reservoir systems. Currently, fracture-vuggy reservoirs are mainly classified into fracture-type, cavern-type, and fracture-void-type. These types of reservoirs differ greatly in morphology, scale, size, and spatial distribution, making it difficult to characterize the spatial distribution, internal structure, and reservoir heterogeneity of oil reservoirs. Classifying and identifying different types of reservoirs can enable targeted development of oilfield reservoir units and efficient reserve assessment. Currently, both domestically and internationally, the identification of different types of reservoirs mainly employs a method combining geological logging and 3D seismic data, including pre-stack and post-stack inversion and multi-attribute seismic joint analysis.

[0003] Most existing technologies summarize the seismic characteristics of different types of reservoirs, explore identification patterns, and then differentiate them through frequency division or different seismic attributes. The differences in scale among different types of reservoirs will produce different seismic tuning frequencies, which can be classified and characterized through frequency division. However, geologically, reservoirs of the same scale with varying degrees of mud filling will also produce different tuning frequency responses. Traditional methods will therefore result in inaccurate reservoir identification and cannot meet the needs of oilfield exploration and development. Summary of the Invention

[0004] To address the aforementioned problems, this application provides a method and related equipment for classifying and characterizing storage groups.

[0005] This application provides a method for classifying and characterizing storage groups, including:

[0006] Acquire seismic data for the target area;

[0007] The seismic data is decomposed into waveforms to obtain multiple first waveform component data volumes;

[0008] The first waveform component data volume is input into a pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information.

[0009] The dominant waveform component corresponding to each reservoir type is determined based on the first waveform component corresponding to each reservoir type.

[0010] Each reservoir within the target area is characterized based on the dominant waveform component corresponding to each reservoir type.

[0011] In some embodiments, the waveform decomposition of the seismic data to obtain multiple waveform component data volumes includes:

[0012] The reflection coefficient of each reservoir is determined based on the sonic logging curves of wells drilled within the target area;

[0013] Determine the combination of reflection coefficients based on the aforementioned reflection coefficients;

[0014] Based on the combination of reflection coefficients and the seismic data, multiple waveform component data volumes are determined.

[0015] In some embodiments, determining multiple waveform component data volumes based on the reflection coefficient combination and the seismic data includes:

[0016] Multiple waveform component data volumes are determined based on the following formula:

[0017]

[0018] Among them, W well(i) (i = 1, 2, ..., M) represents a wavelet sequence, R well(i) (i = 1, 2, ..., M) is a sequence function of a single reflection coefficient, satisfying... N(t) is noise.

[0019] In some embodiments, the method further includes:

[0020] Acquire well logging data, production geological information, and second waveform component data volume for the target area;

[0021] Based on the well logging data and production geological information, the reservoirs in the target area are divided into various reservoir types, wherein the similarity between the waveform component data volumes corresponding to any two reservoir types is less than the similarity threshold.

[0022] Determine the correspondence between each second sample waveform component data volume and each storage volume type;

[0023] Based on the correspondence, sample data is determined to obtain the sample dataset;

[0024] The waveform component selection model is obtained by performing neural network learning based on the sample dataset.

[0025] In some embodiments, the sample dataset includes: first sample data; based on the sample dataset, neural network learning is performed to obtain the waveform component selection model, including:

[0026] The second waveform component data volume of the first sample data is input into the neural network model to determine the predicted storage type;

[0027] The error between the storage type corresponding to the second waveform component data volume based on the first sample data and the predicted storage type is calculated using an error function.

[0028] The weight coefficients of the neural network model are adjusted based on the error to obtain the waveform component selection model.

[0029] In some embodiments, determining the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type includes:

[0030] The first waveform component corresponding to each reservoir type is superimposed to obtain the dominant waveform component corresponding to each reservoir type.

[0031] In some embodiments, characterizing each reservoir within the target region based on the dominant waveform component corresponding to each reservoir type includes:

[0032] The attribute body of each reservoir type is obtained by extracting the target attributes of the dominant waveform component volume corresponding to each reservoir type.

[0033] Each storage group is characterized based on its attribute data of each storage group type; or,

[0034] The target attributes are inverted for the dominant waveform component volume corresponding to each reservoir type to obtain the inversion volume of each reservoir type;

[0035] Each reservoir type is characterized based on its inversion model.

[0036] This application provides a classification and characterization device for a storage collection, including:

[0037] The acquisition module is used to acquire seismic data for the target area.

[0038] The decomposition module is used to perform waveform decomposition on the seismic data to obtain multiple first waveform component data volumes;

[0039] The first determining module is used to input the first waveform component data volume into a pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information.

[0040] The second determining module is used to determine the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type.

[0041] The third determining module is used to characterize each reservoir within the target area based on the dominant waveform component corresponding to each reservoir type.

[0042] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, performs the classification and characterization method of the memory as described above.

[0043] This application provides a storage medium storing a computer program that can be executed by one or more processors and can be used to implement the classification and characterization method of the storage group described in any of the above claims.

[0044] This application provides a method, apparatus, electronic device, and storage medium for classifying and characterizing reservoirs. It pre-establishes a waveform component selection model, trained on a sample dataset. Each sample data point in the dataset includes a second waveform component data volume and the corresponding reservoir type. The reservoir type is determined based on well logging data and production geological information. After acquiring seismic data for the target area, waveform analysis is performed to obtain multiple first waveform component data volumes. These data volumes are then input into the waveform component selection model to determine the first waveform component corresponding to each reservoir type. Based on these first waveform components, a dominant waveform component volume is determined. By characterizing each reservoir using the dominant waveform component volume, the classification and characterization of reservoirs are achieved, and the accuracy of reservoir characterization is improved. Attached Figure Description

[0045] The present application will be described in more detail below based on embodiments and with reference to the accompanying drawings.

[0046] Figure 1 A schematic diagram illustrating the implementation process of a method for classifying and characterizing a storage group, provided in an embodiment of this application;

[0047] Figure 2 A schematic diagram illustrating the implementation process of another method for classifying and characterizing storage groups provided in this application embodiment;

[0048] Figure 3 A schematic diagram illustrating the implementation process of another method for classifying and characterizing storage groups provided in this application embodiment;

[0049] Figure 4 A schematic diagram of the structure of a classification and characterization device for a storage collection provided in an embodiment of this application;

[0050] Figure 5 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application.

[0051] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0054] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0056] To address the problems existing in related technologies, this application provides a method for classifying and characterizing storage collections. This method is applied to electronic devices, such as computers and mobile terminals. The functions implemented by the storage collection classification and characterization method provided in this application can be achieved by the processor of the electronic device calling program code, wherein the program code can be stored in a computer storage medium.

[0057] Example 1

[0058] This application provides a method for classifying and characterizing storage groups. Figure 1 This is a schematic diagram illustrating the implementation process of a method for classifying and characterizing storage groups provided in an embodiment of this application. Figure 1 As shown, it includes:

[0059] Step S101: Obtain seismic data for the target area.

[0060] In this embodiment, the electronic device can communicate with the seismic data measurement equipment to acquire seismic data. The target area can be the area of ​​interest for the storage group's research. In this embodiment, the basic model of the seismic data is a convolution model, meaning that each seismic trace is a convolution of different seismic wavelets and reflection coefficients, plus a certain amount of noise. Each seismic trace can be represented by the following formula:

[0061] S(t) = W(t) * R(t) + N(t);

[0062] Where R(t) is the reflection coefficient sequence function, W(t) is the seismic wavelet, N(t) is the noise term, and S(t) is the seismic trace.

[0063] In this embodiment, because the seismic data responses of strata or reservoirs with different physical properties, such as reservoirs and non-reservoirs, large caverns and small- to medium-scale fracture-vuggy groups, and oil- and gas-bearing reservoirs and non-oil- and gas-free reservoirs, are all different, the seismic wavelets are also modified differently when passing through these different geological targets, and their shapes will change differently. Therefore, the seismic data can be equivalent to a multi-wavelet model.

[0064] Step S102: Perform waveform decomposition on the seismic data to obtain multiple first waveform component data volumes.

[0065] In this embodiment, the first waveform component data volume can simultaneously consider frequency, amplitude, phase, etc. The waveform decomposition of the seismic data to obtain multiple first waveform component data volumes can be achieved in the following ways:

[0066] Step S1021: Determine the reflection coefficient of each reservoir based on the sonic logging curves of the wells drilled within the target area.

[0067] In this embodiment, the sonic logging curve of the well is obtained by actual measurement, and the reflection coefficient of the formation and each reservoir can be calculated based on the sonic logging curve of the well.

[0068] Step S1022: Determine the combination of reflection coefficients based on the reflection coefficients.

[0069] In this embodiment of the application, a series of reflection coefficient combinations R can be formed based on the reflection system. well (t), this combination contains information on stratigraphic lithology sequence and reservoir geological characteristics, and can guide waveform decomposition of subsequent seismic data through reflection coefficient combination.

[0070] Step S1023: Determine multiple waveform component data volumes based on the reflection coefficient combination and the seismic data.

[0071] In this embodiment of the application, multiple waveform component data volumes can be determined based on the following calculation formula:

[0072]

[0073] Among them, W well(i) (i = 1, 2, ..., M) represents a wavelet sequence, R well(i) (i = 1, 2, ..., M) is a sequence function of a single reflection coefficient, satisfying... N(t) is noise.

[0074] In this embodiment, the wavelets in each waveform component data volume have different shapes or spectral characteristics. Based on the above calculation formula, a reverse decomposition can be performed to obtain the wavelet sequence, thereby determining multiple waveform component data volumes based on the wavelet sequence.

[0075] Step S103: Input the first waveform component data volume into the pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information.

[0076] In this embodiment, well logging data and production geological information can be used to classify reservoir types. The production geological information may include well blowouts, drilling fluid losses, etc. In this embodiment, the characteristics of each reservoir type are distinct and similar. Each type of reservoir has the same or similar geological characteristics, and their influence on seismic wavelets is also relatively consistent.

[0077] In this embodiment, the reservoir type may include N types. For example, the reservoir type may include: fracture-vuggy reservoir and cavernous reservoir. The number of N can be determined according to the geological characteristics of the target area.

[0078] Step S104: Determine the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type.

[0079] In this embodiment, the first waveform components corresponding to each storage type can be superimposed to obtain the dominant waveform component corresponding to each storage type. Continuing the example above, when there are N storage types, there are also N types of dominant waveform components.

[0080] Step S105: Characterize each reservoir in the target area based on the dominant waveform component corresponding to each reservoir type.

[0081] In this embodiment, each reservoir can be characterized in the following ways: target attribute extraction is performed on the dominant waveform component volume corresponding to each reservoir type to obtain the attribute volume of each reservoir type; each reservoir is characterized based on the attribute volume of each reservoir type; or, the target attribute is inverted on the dominant waveform component volume corresponding to each reservoir type to obtain the inverted volume of each reservoir type; each reservoir is characterized based on the inverted volume of each reservoir type.

[0082] In this embodiment, each dominant waveform component represents a comprehensive representation of seismic waveform characteristics sensitive to a certain type of reservoir, including various information such as amplitude, frequency, phase, and waveform. Seismic attributes are extracted or inverted from these dominant waveform components to form different attribute volumes or inversion volumes. Each attribute volume or inversion volume can accurately characterize and analyze the reservoir type separately, realizing the classification, identification, and representation of the reservoir.

[0083] This application provides a method for classifying and characterizing reservoirs. It pre-establishes a waveform component selection model, trained on a sample dataset. Each sample data point in the dataset includes a second waveform component data volume and the corresponding reservoir type. The reservoir type is determined based on well logging data and production geological information. After acquiring seismic data for the target area, waveform analysis is performed to obtain multiple first waveform component data volumes. These data volumes are then input into the waveform component selection model to determine the first waveform component corresponding to each reservoir type. Based on these first waveform components, a dominant waveform component volume is determined. By characterizing each reservoir using the dominant waveform component volume, the method achieves reservoir classification and characterization, and improves the accuracy of reservoir characterization.

[0084] Example 2

[0085] Based on the foregoing embodiments, this application further provides a method for classifying and characterizing storage groups. Figure 2 A flowchart illustrating another method for classifying and characterizing storage groups provided in this application embodiment is shown below. Figure 2 As shown, the method includes:

[0086] Step S201: Obtain well logging data, production geological information, and second waveform component data of the target area.

[0087] In this embodiment of the application, the well logging data and production geological information can be obtained based on actual drilling data, and the second waveform component can be obtained by waveform decomposition of seismic data.

[0088] Step S202: Based on the well logging data and production geological information, the reservoirs in the target area are divided into various reservoir types, wherein the similarity between the waveform component data volumes corresponding to any two reservoir types is less than the similarity threshold.

[0089] In the embodiments of this application, each type of reservoir has the same or similar geological characteristics, and their influence on seismic wavelets is also relatively consistent.

[0090] Step S203: Determine the correspondence between each second sample waveform component data volume and each storage volume type.

[0091] Step S204: Based on the correspondence, determine the sample data to obtain the sample dataset.

[0092] For example, the sample dataset contains N samples, and the sample data (x k ,y k k = 1, 2, ..., N.

[0093] Step S205: Based on the sample dataset, perform neural network learning to obtain the waveform component selection model.

[0094] In this embodiment of the application, the neural network model may be a BP neural network.

[0095] Step S206: Obtain seismic data for the target area;

[0096] Step S207: Perform waveform decomposition on the seismic data to obtain multiple first waveform component data volumes;

[0097] Step S208: Input the first waveform component data volume into the pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information.

[0098] Step S209: Determine the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type.

[0099] Step S210: Characterize each reservoir in the target area based on the dominant waveform component corresponding to each reservoir type.

[0100] The reservoir classification and characterization method provided in this application embodiment pre-establishes a waveform component selection model, which is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information. After obtaining seismic data of the target area, waveform analysis is performed on the seismic data to obtain multiple first waveform component data volumes. Then, the waveform component data volumes are input into the waveform component selection model to determine the first waveform component corresponding to each reservoir type. Based on the first waveform component, the dominant waveform component volume is determined. By characterizing each reservoir through the dominant waveform component volume, the classification and characterization of reservoirs is achieved, and the accuracy of reservoir characterization can be improved.

[0101] Example 3

[0102] Based on the foregoing embodiments, the sample dataset includes first sample data, which can be any sample data in the sample dataset. Step S205, "performing neural network learning based on the sample dataset to obtain the waveform component selection model," can be implemented in the following way:

[0103] Step S2051: Input the second waveform component data volume of the first sample data into the neural network model to determine the predicted storage type.

[0104] In this embodiment of the application, the first sample data is represented by (x k ,y k ) represents the input x of the first sample data. k The network output is y k The output of node i is o ik The input to neural network node j is:

[0105] The type of the predicted reservoir can be obtained through calculation. This is the actual output of the neural network, i.e., the predicted storage type.

[0106] Step S2052: The error between the storage type corresponding to the second waveform component data volume of the first sample data and the predicted storage type is calculated using an error function.

[0107] In this embodiment, the error is calculated using the error function as follows:

[0108]

[0109] Where E represents the calculation error. For the actual output of the network, define the error for a single sample k. W ij The weights are the weights between network nodes i and j. When j is an output node... y k This is the storage type corresponding to the second waveform component data volume.

[0110] Step S2053: Adjust the weight coefficients of the neural network model based on the error to obtain the waveform component selection model.

[0111] During neural network training, initial weights are given, and each weight is assigned a small, random, non-zero value. A sample from the input dataset is given, and the actual output value of the neural network is calculated based on this input value, yielding the difference. The weights of the network model are adjusted, and the process is repeated with the input sample data until all samples in the dataset have been calculated. Finally, the error is recalculated using all the calculated weights against the sample data. If the accuracy meets the requirements, training ends; otherwise, sample data is re-inputted and the weights are adjusted again.

[0112] Example 4

[0113] Based on the foregoing embodiments, this application provides a method for classifying and characterizing reservoirs. This method is based on a superior waveform matching and identification method that leverages the differences in seismic wave amplitude energy and wavefield reflection structure among different types of reservoirs. Compared with related technologies, this method fundamentally decomposes the underground seismic wavefield into multiple wavefield components of different wavelet types. Wavefield components that show a significant response to the target reservoir type are matched and superimposed, while wavefields without a response are separated. This forms a superposition of different waveform components for identifying different types of reservoirs. These superpositions are then used for inversion or attribute calculation. Because the superpositions of different waveform components can simultaneously consider the response of the target reservoir type to multiple factors such as frequency, amplitude, and phase, different types of reservoirs can be characterized more accurately. Figure 3 A schematic diagram illustrating the implementation flow of another method for classifying and characterizing storage groups provided in this application embodiment is shown below. Figure 3 As shown, it includes:

[0114] Step S1: Waveform decomposition of seismic data.

[0115] The basic model for seismic data is the convolution model, where each seismic trace is a convolution of different seismic wavelets and reflection coefficients, plus a certain amount of noise, as shown in formula (a):

[0116] S(t)=W(t)*R(t)+N(t) (a);

[0117] Where R(t) is the reflection coefficient sequence function, W(t) is the seismic wavelet, and N(t) is the noise term.

[0118] Because strata or reservoirs with different physical properties, such as reservoirs and non-reservoirs, large caverns and small- to medium-scale fracture-vuggy groups, and oil- and gas-bearing reservoirs and non-oil- and gas-free reservoirs, have different seismic responses, seismic wavelets are modified differently when passing through these different geological targets, and their shapes change accordingly. Therefore, seismic data can be equivalent to a multi-wavelet model.

[0119] Based on the sonic logging curves from actual wells in the study area, the reflection coefficients of each formation and reservoir type are calculated, forming a series of reflection coefficient combinations R. well (t), this combination of reflection coefficients contains information on the stratigraphic lithology sequence and reservoir geological characteristics, which can guide subsequent seismic waveform decomposition.

[0120] The multi-wavelet model can be generally expressed as follows: given the reflection coefficient sequence function R well (t), whose corresponding seismic reflection signal S well (t) can represent formula (b):

[0121]

[0122] Among them, W well(i) (i = 1, 2, ..., M) represents a wavelet sequence, where the wavelets can have different shapes or spectral characteristics, R well(i) (i = 1, 2, ..., M) is a sequence function of a single reflection coefficient, satisfying... N(t) is noise.

[0123] According to formula (b) and the actual drilling reflection coefficient combination sequence, a seismic trace can be converted into a superposition of wavelets of different shapes. The reverse decomposition process of formula (b) can form different waveform component data volumes according to actual geological needs.

[0124] Step S2: Classification of storage group types.

[0125] Using drilling and logging data and production geological information, reservoirs in the target area are classified to clearly define the characteristics of each type of reservoir, enabling effective and precise classification of reservoirs based on the aforementioned data. Assume that N types of reservoirs can be identified based on the geological data. Each type of reservoir possesses similar or identical geological characteristics, and their influence on seismic wavelets is also relatively consistent.

[0126] Step S3: Selection of dominant waveform components.

[0127] The data volume of multiple waveform components decomposed by the dominant wavelet will be compared with N types of reservoirs encountered during drilling, and a supervised BP neural network will be trained. Labeled samples will be established based on drilling logging knowledge and seismic waveform components, with N samples (x... k ,y k k = 1, 2, ..., N. A given input x... k The network output is y k The output of node i is o ik The input for node j is given in formula (c):

[0128]

[0129] The error is calculated using a squared error function; see formula (d).

[0130]

[0131] in, For the actual output of the network, define the error for a single sample k. W ij The weights are the weights between network nodes i and j. When j is an output node...

[0132] During neural network training, weights are assigned to each of them, each with a small, random, non-zero value. A sample from the input label set is given, and the actual output value of the neural network is calculated based on this input value, yielding the difference. The connection weights are adjusted, and the process is repeated with the sample input until all samples have been calculated. Finally, the error is recalculated using all the calculated weights. If the accuracy meets the requirements, training ends; otherwise, samples are re-inputted and weights are adjusted again.

[0133] After training with a neural network, multiple waveform components that can distinguish between N types of reservoirs will be generated. The waveform components that are sensitive to the same type of reservoir will be superimposed to form a superior waveform component that is more sensitive and effective to that type of reservoir.

[0134] Step S4: Classification and characterization of the storage group.

[0135] For the N dominant waveform components obtained, each dominant waveform component represents a comprehensive embodiment of seismic waveform characteristics sensitive to a certain type of reservoir, including multiple information such as amplitude, frequency, phase, and waveform. Seismic attributes are extracted or inverted from these dominant waveform components to form different attribute volumes or inversion volumes. Each attribute volume or inversion volume can accurately characterize and analyze the reservoir type separately, realizing the classification, identification, and representation of reservoirs.

[0136] In this embodiment, by maximizing the use of drilling geological information, the characteristics of different types of reservoirs are comprehensively summarized, and waveform decomposition and dominant waveform matching are guided by drilling and logging data, rather than simply pure mathematical calculations of seismic data.

[0137] The superior waveform components selected through supervised neural network training are related to reservoir type in multiple seismic parameters, such as amplitude, frequency, and phase, making them more targeted to reservoir type, unlike traditional reservoir classification which focuses on seismic response based on single information such as scale or lithology.

[0138] The following example uses a carbonate reservoir in an oilfield in the Tarim Basin for illustration:

[0139] Step S11: For the seismic data of this oilfield, a time window of 150ms from the T74 marker layer downwards is selected for waveform decomposition calculation. This time window also coincides with the location of the target reservoir layer, which can meet the requirements for subsequent reservoir classification and characterization. The data volume is decomposed into waveform components of 35 different wavelet types.

[0140] In step S12, after analyzing all waveform components, it is determined that waveform components 1 and 2 have the strongest energy and mainly reflect the stratigraphic structure, thus interfering with reservoir identification. Therefore, components 3-35, excluding components 1 and 2, are selected for subsequent reservoir analysis.

[0141] Step S13: Based on drilling and other production data, the oilfield's reservoirs are divided into two categories: one is the karst-vuggy reservoir, characterized by well blowouts; the other is the fractured-vuggy reservoir, characterized by drilling fluid loss.

[0142] Step S14 involves supervised neural network training of multiple waveform components for these two types of reservoirs. The results show that components 3-10 are more sensitive to fracture-vuggy reservoirs, while components 11-35 are more sensitive to cavernous reservoirs. Components 3-10 and 11-35 are then superimposed to form two dominant waveform component volumes. Figure 5 The cross-sections and wavelet characteristics of these two dominant waveform components are shown.

[0143] Step S15: Invert the two dominant waveform components obtained above.

[0144] Example 5

[0145] Based on the foregoing embodiments, this application provides a classification and characterization device for a storage collection. The various modules and units included in the device can be implemented by a processor in a computer device; of course, they can also be implemented by specific logic circuits. In the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0146] This application provides a classification and characterization device for storage collections. Figure 4 This is a schematic diagram of the structure of a classification and characterization device for a storage collection provided in an embodiment of this application, as shown below. Figure 4 As shown, the classification and characterization device 400 for the storage group includes:

[0147] The first acquisition module 401 is used to acquire seismic data of the target area;

[0148] Decomposition module 402 is used to perform waveform decomposition on the seismic data to obtain multiple first waveform component data volumes;

[0149] The first determining module 403 is used to input the first waveform component data volume into a pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information.

[0150] The second determining module 404 is used to determine the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type.

[0151] The third determining module 405 is used to characterize each reservoir in the target area based on the dominant waveform component corresponding to each reservoir type.

[0152] This application provides a reservoir classification and characterization device. It pre-establishes a waveform component selection model, trained on a sample dataset. Each sample data point in the dataset includes a second waveform component data volume and the corresponding reservoir type. The reservoir type is determined based on well logging data and production geological information. After acquiring seismic data for the target area, waveform analysis is performed to obtain multiple first waveform component data volumes. These data volumes are then input into the waveform component selection model to determine the first waveform component corresponding to each reservoir type. Based on these first waveform components, a dominant waveform component volume is determined. The dominant waveform component volume is then used to characterize each reservoir, thus achieving reservoir classification and characterization and improving the accuracy of reservoir characterization.

[0153] In some embodiments, the decomposition module 402 includes:

[0154] The first determining unit is used to determine the reflection coefficient of each reservoir based on the sonic logging curves of wells drilled within the target area.

[0155] The second determining unit is used to determine the combination of reflection coefficients based on the reflection coefficients;

[0156] The third determining unit is used to determine multiple waveform component data volumes based on the reflection coefficient combination and the seismic data.

[0157] In some embodiments, the third determining unit includes:

[0158] A calculation subunit is used to determine multiple waveform component data volumes based on the following formula:

[0159]

[0160] Among them, W well(i) (i = 1, 2, ..., M) represents a wavelet sequence, R well(i) (i = 1, 2, ..., M) is a sequence function of a single reflection coefficient, satisfying... N(t) is noise.

[0161] In some embodiments, the classification and characterization device 400 for the storage collection further includes:

[0162] The second acquisition module is used to acquire well logging data, production geological information and second waveform component data volume of the target area;

[0163] The module is used to divide the reservoirs in the target area into various reservoir types based on the well logging data and production geological information, wherein the similarity between the waveform component data volumes corresponding to any two reservoir types is less than a similarity threshold.

[0164] The fourth determination module is used to determine the correspondence between each second sample waveform component data volume and each storage volume type;

[0165] The fifth determining module is used to determine sample data based on the correspondence to obtain the sample dataset;

[0166] The training module is used to perform neural network learning based on the sample dataset to obtain the waveform component selection model.

[0167] In some embodiments, the sample dataset includes: first sample data; and a training module, including:

[0168] The fourth determining unit is used to input the second waveform component data volume of the first sample data into the neural network model to determine the type of the predicted storage group;

[0169] An error calculation unit is used to calculate the error based on the storage type corresponding to the second waveform component data volume of the first sample data and the predicted storage type using an error function.

[0170] An adjustment unit is used to adjust the weight coefficients of the neural network model based on the error to obtain the waveform component selection model.

[0171] In some embodiments, the second determining module includes:

[0172] The superposition unit is used to superimpose the first waveform component of each reservoir type to obtain the dominant waveform component of each reservoir type.

[0173] In some embodiments, the third determining module includes:

[0174] The extraction unit is used to extract the target attributes of the dominant waveform component volume corresponding to each storage group type to obtain the attribute volume of each storage group type.

[0175] The first characterization unit is used to characterize each storage group based on the attribute bodies of each storage group type; or,

[0176] The inversion unit is used to invert the target attributes of the dominant waveform component volume corresponding to each reservoir type to obtain the inversion volume of each reservoir type.

[0177] The second characterization unit is used to characterize each reservoir based on the inversion of each reservoir type.

[0178] It should be noted that, in the embodiments of this application, if the above-mentioned method for determining development parameters is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0179] Accordingly, this application provides a storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps in the classification and characterization method of the storage collection provided in the above embodiments.

[0180] Example 6

[0181] This application provides an electronic device; Figure 5 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application, such as... Figure 5 As shown, the electronic device 500 includes: a processor 501, at least one communication bus 502, a user interface 503, at least one external communication interface 504, and a memory 505. The communication bus 502 is configured to enable communication between these components. The user interface 503 may include a display screen, and the external communication interface 504 may include standard wired and wireless interfaces. The processor 501 is configured to execute a program for determining development parameters stored in the memory, to implement the steps in the classification and characterization method of the memory set provided in the above embodiment.

[0182] The descriptions of the above embodiments of the electronic devices and storage media are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the embodiments of the computer devices and storage media of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0183] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0184] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0185] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0186] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0187] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0188] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0189] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0190] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for classifying and characterizing storage groups, characterized in that, The method includes: Acquire seismic data for the target area; The seismic data is decomposed into waveforms to obtain multiple first waveform component data volumes; The first waveform component data volume is input into a pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information. The dominant waveform component corresponding to each reservoir type is determined based on the first waveform component corresponding to each reservoir type. Each reservoir within the target area is characterized based on the dominant waveform components corresponding to each reservoir type. The step of determining the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type includes: The first waveform component corresponding to each reservoir type is superimposed to obtain the dominant waveform component corresponding to each reservoir type. The dominant waveform component is a comprehensive representation of the seismic waveform characteristics that are sensitive to a certain type of reservoir, including at least one of amplitude, frequency, phase and waveform information. The waveform decomposition of the seismic data yields multiple first waveform component data volumes, including: The reflection coefficient of each reservoir is determined based on the sonic logging curves of wells drilled within the target area; A combination of reflection coefficients is determined based on the reflection coefficients, wherein the combination of reflection coefficients includes information on stratigraphic lithology sequence and reservoir geological characteristics, so as to guide the waveform decomposition of subsequent seismic data through the combination of reflection coefficients; Based on the combination of reflection coefficients and the seismic data, multiple first waveform component data volumes are determined; The determination of multiple first waveform component data volumes based on the reflection coefficient combination and the seismic data includes: Multiple first waveform component data volumes are determined based on the following calculation formula: ； in, Represents a wavelet sequence, It is a sequence function of a single reflection coefficient, satisfying , It's noise.

2. The method according to claim 1, characterized in that, The method further includes: Acquire well logging data, production geological information, and second waveform component data volume for the target area; Based on the well logging data and production geological information, the reservoirs in the target area are divided into various reservoir types, wherein the similarity between the waveform component data volumes corresponding to any two reservoir types is less than the similarity threshold. Determine the correspondence between each second sample waveform component data volume and each storage volume type; Based on the correspondence, sample data is determined to obtain the sample dataset; The waveform component selection model is obtained by performing neural network learning based on the sample dataset.

3. The method according to claim 1, characterized in that, The sample dataset includes: first sample data; based on the sample dataset, a neural network is learned to obtain the waveform component selection model, including: The second waveform component data volume of the first sample data is input into the neural network model to determine the predicted storage type; The error between the storage type corresponding to the second waveform component data volume based on the first sample data and the predicted storage type is calculated using an error function. The weight coefficients of the neural network model are adjusted based on the error to obtain the waveform component selection model.

4. The method according to claim 1, characterized in that, The characterization of each reservoir within the target region based on the dominant waveform component corresponding to each reservoir type includes: The attribute body of each reservoir type is obtained by extracting the target attributes of the dominant waveform component volume corresponding to each reservoir type. Each storage group is characterized based on its attribute data of each storage group type; or, The target attributes are inverted for the dominant waveform component volume corresponding to each reservoir type to obtain the inversion volume of each reservoir type; Each reservoir type is characterized based on its inversion model.

5. A classification and characterization device for a storage group, characterized in that, include: The acquisition module is used to acquire seismic data for the target area. The decomposition module is used to perform waveform decomposition on the seismic data to obtain multiple first waveform component data volumes; The first determining module is used to input the first waveform component data volume into a pre-established waveform component selection model to determine the first waveform component volume corresponding to each reservoir type. The waveform component selection model is trained based on a sample dataset. Each sample data in the sample dataset includes a second waveform component data volume and the reservoir type corresponding to the second waveform component data volume. The reservoir type is determined based on well logging data and production geological information. The second determining module is used to determine the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type. The third determining module is used to characterize each reservoir in the target area based on the dominant waveform component corresponding to each reservoir type. The step of determining the dominant waveform component corresponding to each reservoir type based on the first waveform component corresponding to each reservoir type includes: The first waveform component corresponding to each reservoir type is superimposed to obtain the dominant waveform component corresponding to each reservoir type. The dominant waveform component is a comprehensive representation of the seismic waveform characteristics that are sensitive to a certain type of reservoir, including at least one of the following information: amplitude, frequency, phase and waveform. The waveform decomposition of the seismic data to obtain multiple waveform component data volumes includes: The reflection coefficient of each reservoir is determined based on the sonic logging curves of wells drilled within the target area; A combination of reflection coefficients is determined based on the reflection coefficients, wherein the combination of reflection coefficients includes information on stratigraphic lithology sequence and reservoir geological characteristics, so as to guide the waveform decomposition of subsequent seismic data through the combination of reflection coefficients; Based on the combination of reflection coefficients and the seismic data, multiple waveform component data volumes are determined. The determination of multiple waveform component data volumes based on the reflection coefficient combination and the seismic data includes: Multiple waveform component data volumes are determined based on the following formula: ； in, Represents a wavelet sequence, It is a sequence function of a single reflection coefficient, satisfying , It's noise.

6. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, performs the classification characterization method of the memory as described in any one of claims 1 to 4.

7. A storage medium, characterized in that, The computer program stored in the storage medium can be executed by one or more processors and can be used to implement the classification and characterization method of the storage group as described in any one of claims 1 to 4.

Images