Interlayer identification method, system and equipment based on pattern identification and medium
Through pattern recognition technology, the Fisher dimensionality reduction method and the minimum square error criterion function are used to achieve accurate identification of interlayers, solving the problems of low recognition accuracy and cumbersome steps in the existing technology, and improving the accuracy and applicability of interlayer recognition .
Patent Information
- Application Number
- CN202310832184.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-06
- Publication Date
- 2025-01-07
AI Technical Summary
The prior art has low recognition accuracy and cumbersome steps in the identification of interlayers, which cannot be effectively promoted, and it is difficult to accurately distinguish different interlayers, which affects oil field development and residual oil evaluation.
Using a pattern recognition-based method, select different types of core sample data, perform response feature analysis, combine inclination logging data, use Fisher dimensionality reduction method to obtain the normal function, and use the gradient descent algorithm of the minimum square error criterion function to obtain it. Discriminant function to achieve accurate identification of interlayers.
The accuracy of interlayer recognition is improved and the recognition steps are simplified, achieving higher recognition accuracy and reliability, wide applicability, and easy and concise calculations.
Smart Images
Figure CN119270363A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of well logging and relates to an interlayer identification method, system, equipment and medium based on pattern recognition. Background Art
[0002] Interlayers are ubiquitous in continental sediments. Their size, distribution and location in the reservoir have a significant impact on the flow of fluids. Interlayer data is an important part of reservoir description in the middle and late stages of oilfield development. The identification of interlayers in a single well and the comparison of interlayers between wells are of great significance for the potential of remaining oil.
[0003] With the deepening of oilfield development, oil well production has entered a high water-cut period. The core problem faced by development is bottom water cone advance, and interlayers have a good effect on inhibiting bottom water cone advance. Therefore, the identification of reservoir interlayers plays an important role in the fine characterization of reservoirs and the evaluation of remaining oil. At present, the main methods for reservoir interlayer identification by researchers at home and abroad include: using field observation results, conceptualizing with modern sedimentary theories and sedimentary models, and conducting qualitative and semi-quantitative research on the distribution of interlayers based on mathematical fitting; using core, logging, and well logging analysis as means, and individually supplemented with fine seismic interpretation, interwell seismic, etc. to identify interlayers; combining the logging response characteristics of different types of interlayers to establish an interlayer identification chart; with the advancement of computer technology, artificial intelligence technology has gradually been applied to interlayer identification, and the above methods have achieved certain applicable effects. However, the above methods either have low recognition accuracy or cumbersome recognition steps, and cannot be effectively promoted. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method, system, device and medium for interlayer identification based on pattern recognition, which can accurately distinguish different interlayers, improve the interlayer identification accuracy and simplify the cumbersome identification steps.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] A method for identifying interlayers based on pattern recognition comprises the following steps:
[0007] S1, select data of multiple core samples with different interlayer types;
[0008] S2, analyze the response characteristics of different interlayer types on conventional logging curves, and determine the sensitive logging curves of the selected core samples reflecting the interlayer types in combination with the dip logging data;
[0009] S3, using the interlayer sensitive logging curve, the canonical function is obtained by Fisher dimensionality reduction method;
[0010] S4, using canonical functions, reducing the dimension of multi-parameter high-dimensional logging parameter data, and obtaining the discriminant functions corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function;
[0011] S5, substituting the logging value of the depth section to be judged into the above discriminant functions for calculation, comparing the calculated discriminant function values, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
[0012] Preferably, in S1, the interlayer type of the core sample is obtained by a core analysis experimental method.
[0013] Preferably, in S2, the correlation between the interlayer and conventional logging and dip logging data is established through core and logging data, and then the sensitive logging curve reflecting the interlayer type of the selected core sample is determined.
[0014] Preferably, in S3, using the determined sensitive logging curve, a k-type logging parameter vector matrix X is constructed. k =(x k1 ,x k2 ,…,x kn ) T , k is the number of interlayer types, n is the number of samples in each type, x ki is the logging parameter vector corresponding to the i-th sample point of the k-th category; the principle of Fisher discriminant is used to find the projection direction a that can separate the population as much as possible, and the coordinate transformation of the original data system is performed to extract the discriminant features of interlayers.
[0015] Furthermore, the process of extracting the discriminant features of the interlayer is as follows: Sample point x ki The projection in the direction a is a T x ki , T represents the transpose of the matrix, and the projection of various logging data is: G k :a T x k1 …a T x kn k=1,2…5G k The mean of the data projection in the group is recorded as In order to better separate the categories, the ratio of the deviation between groups to the deviation within the group is maximized, that is: Solved using Lagrange multiplier algorithm The maximum value of is a, where B is the inter-group scatter matrix, E is the intra-group scatter matrix, and m is the counting symbol; the function constituting the eigenvalue of the projection direction is: F(X) = a T X k Where X kis the k-th type lithology sample vector matrix, and F(X) is the projection dimension reduction function, that is, the canonical function.
[0016] Preferably, in S4, a multi-parameter high-dimensional logging parameter data body is obtained from the existing logging data, and the multi-parameter high-dimensional logging parameter data body is reduced in dimension using a canonical function, and F(X)=a T X transformation can get a new logging parameter sample vector y. According to the gradient descent algorithm of the least square error criterion function, w is found so that for any given y∈X k make Established, w is the discriminant weight matrix, thus obtaining the discriminant function f i (y).
[0017] Preferably, in S1, data of core samples of at least three different interlayer types are selected.
[0018] A method for identifying interlayers based on pattern recognition, comprising:
[0019] The core sample selection module is used to select data of core samples with different interlayer types;
[0020] The sensitive logging curve determination module is used to analyze the response characteristics of different interlayer types on conventional logging curves, and determine the sensitive logging curves of the selected core samples reflecting the interlayer types in combination with the dip logging data;
[0021] The canonical function obtaining module is used to obtain the canonical function by using the interlayer sensitive logging curve through the Fisher dimensionality reduction method;
[0022] The discriminant function obtaining module is used to reduce the dimension of the multi-parameter high-dimensional logging parameter data body by using the canonical function, and obtain the discriminant function corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function;
[0023] The interlayer identification module is used to substitute the logging value of the depth section to be judged into the above discriminant functions for calculation, compare the discriminant function values after calculation, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
[0024] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the interlayer identification method based on pattern recognition when executing the computer program.
[0025] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the interlayer identification method based on pattern recognition.
[0026] Compared with the prior art, the present invention has the following beneficial effects:
[0027] The present invention first selects core samples with different types of interlayers to determine the interlayer types of the selected core samples; then, response characteristic analysis is performed on conventional logging curves of different interlayer types, and combined with inclination logging data, a sensitive logging curve reflecting the interlayer type of the selected core sample is determined; using the interlayer sensitive logging curve, a canonical function is obtained by Fisher dimensionality reduction method, and an interlayer analysis chart is established; using the canonical function, a multi-parameter high-dimensional logging parameter data body is dimensionally reduced, and a discriminant function of different interlayer types is obtained according to the gradient descent algorithm of the least square error criterion function; finally, the logging value of the depth section to be judged is substituted into the above discriminant functions for calculation, and the calculated f values are compared, and the interlayer category corresponding to the maximum function value is the interlayer discrimination result of the depth section, thereby predicting the unknown interlayer. The problem that the existing interlayer identification methods are not convenient, applicable and scalable is solved. By introducing the Fisher discriminant analysis method of pattern recognition, a canonical function is obtained, and finally discriminant functions of different types of interlayers are formed. It has higher interlayer identification accuracy and reliable theoretical basis, is convenient and concise in calculation, has strong versatility and wide applicability. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 It is a flow chart of interlayer identification of the present invention;
[0029] Figure 2 This is a core image of the calcareous interlayer of the present invention;
[0030] Figure 3 This is the core image of the muddy interlayer of the present invention;
[0031] Figure 4 This is the physical property interlayer core diagram of the present invention;
[0032] Figure 5 The radar diagram of the logging response characteristics of different interlayers of the present invention;
[0033] Figure 6 It is the Fisher discriminant analysis diagram of the interlayer of the present invention. DETAILED DESCRIPTION
[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0035] It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to directions in the drawings, and the words "inside" and "outside" refer to directions toward or away from the geometric center of a specific component, respectively.
[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.
[0037] like Figure 1 As shown, the interlayer identification method based on pattern recognition according to the present invention comprises the following steps:
[0038] 1) Select data of at least three different types of interlayer core samples to determine the interlayer type of the selected core samples.
[0039] The interlayer types of the selected core samples were obtained through core analysis experimental methods.
[0040] 2) Classify the different types of interlayers obtained in step 1), analyze the response characteristics of different interlayer types on conventional logging curves, and determine the sensitive logging curves of the selected core samples reflecting the interlayer types in combination with the dip logging data. Different curves respond differently to different samples.
[0041] Using the experimental data of core samples and the response relationship between conventional logging curves and dip logging, there are certain limitations in the identification of a single logging curve, so it is necessary to determine the response characteristics of various logging curves of interlayers. Natural gamma, density, acoustic wave, resistivity and dip logging conductivity curves have different response characteristics for different types of interlayers. Through core and logging data, the correlation between interlayers and conventional logging and dip logging data is established.
[0042] 3) Using the interlayer sensitive logging curve obtained in step 2), the canonical function is obtained through the Fisher dimensionality reduction method, and an interlayer analysis chart is established.
[0043] Using the determined sensitive logging curve, construct the k-type logging parameter vector matrix X k =(x k1 ,x k2 ,…,x kn ) T , k is the number of interlayer types, n is the number of samples in each type, x kiis the logging parameter vector corresponding to the i-th sample point of the k-th class. Using the principle of Fisher discriminant, we find the projection direction a that can separate the population as much as possible, transform the coordinates of the original data system, and then extract the discriminant features of the interlayer. Sample point x ki The projection in the direction a is a T x ki , T represents the transpose of the matrix, and the projection of various logging data is: G k :a T x k1 …a T x kn k=1,2…5G k The mean of the data projection in the group is recorded as In order to better separate the categories, the ratio of the deviation between groups to the deviation within the group is maximized, that is: Solved using Lagrange multiplier algorithm The maximum value of is a, where B is the inter-group scatter matrix, E is the intra-group scatter matrix, and m is the counting symbol. The function that constitutes the eigenvalue of the projection direction is: F(X) = a T X k Where X k is the vector matrix of the k-th type of lithology samples, F(X) is the projection dimension reduction function, that is, the canonical function, and an interlayer identification plate is established, and the identification plate is determined by the canonical function.
[0044] 4) directly obtain multi-parameter high-dimensional logging parameter data from existing logging data, and use the canonical function F(X) obtained in step 3) to reduce the dimension of the multi-parameter high-dimensional logging parameter data, and then use F(X)=a T X transformation can get a new logging parameter sample vector y. According to the gradient descent algorithm of the least square error criterion function, w is found so that for any given y∈X k make Established, w is the discriminant weight matrix, thus obtaining the discriminant function f i (y).
[0045] 5) According to the discriminant function obtained in step 4), the logging value of the depth section to be judged is substituted into the above discriminant functions for calculation, and the calculated f values are compared. f is the discriminant function. The interlayer category corresponding to the largest function value is the interlayer discrimination result of the depth section, thereby identifying the unknown interlayer.
[0046] Combined with Fisher discriminant function, the logging curve is brought into the calculation, and the calculated f value is compared. The interlayer category corresponding to the largest function value is the interlayer discrimination result of the depth section, so as to identify the unknown interlayer. The quantitative division of interlayer can be realized.
[0047] Below, the specific embodiments of this embodiment are further described in detail to support the technical problem to be solved by the present invention.
[0048] 1. Based on the core data, select volcanic rock core samples with different lithologies, conduct rock thin section identification and analysis experiments, and determine their lithology types.
[0049] In this example, the Carboniferous stratum in an oil field block was selected as the target layer for research, the core data of the target layer section were collected, and 44 representative core samples were selected based on the data. Figure 2-4 As shown, in this embodiment, the so-called representative core refers to the interlayer type with obvious characteristics.
[0050] 2. Well logging response is a comprehensive reflection of lithology, porosity, pore structure and fluid properties. The logging curve responses of different types of interlayers are often different. Since interlayers in the study area are widely developed and thin, there are certain limitations in the identification of a single logging curve. Therefore, it is necessary to judge the response characteristics of various logging curves of interlayers. Natural gamma, density, sonic wave, resistivity and dip logging conductivity curves have different response characteristics for different types of interlayers. Mud interlayers have the characteristics of large natural gamma and sonic wave values, calcium interlayers have the characteristics of small natural gamma and sonic wave logging values, dip conductivity curves have the characteristics of "swinging tip", and physical interlayers have medium natural gamma and sonic wave curve values. For example Figure 5 As shown, the correlation between different types of interlayers and logging curves is finally determined.
[0051] 3. Using the determined sensitive logging curve, the canonical function is obtained by Fisher dimensionality reduction method: F1 = 7.15*ΔGR + 0.029*RXO - 0.152*RFC + 0.211*AC + 0.955*DEN - 19.833F2 = 5.17*ΔGR + 0.008*RXO + 0.265*RFC - 0.095*AC + 7.73*
[0052] DEN-10.833, where like Figure 6 As shown, a sandwich identification plate is established.
[0053] 4. Use the canonical function F(X) to reduce the dimension of the multi-parameter high-dimensional logging parameter data body, and then use F(X)=a T X k The transformation can obtain a new logging parameter sample vector y. According to the gradient descent algorithm of the least square error criterion function, w is found so that for any given y∈X k Make k (y) = max(w iT y,i=1,2,…,5) holds true, thus obtaining the discriminant function f i (y).
[0054] Among them, calcareous interlayer f1 = 15.236△GR + 17.268RXO + 29.357RFC + 11.306AC + 15.231DEN-1151.03, physical interlayer f2 = 16.875△GR + 16.851RXO + 26.407RFC + 10.926AC + 14.387DEN-1107.625, muddy interlayer f3 = 18.347△GR + 18.697RXO + 25.389RFC + 28.359AC + 16.573DEN-1095.312.
[0055] 5. Substitute the logging value of the depth section to be judged into the above discriminant functions for calculation, compare the calculated f values, and the interlayer category corresponding to the largest function value is the interlayer discrimination result of the depth section, so as to identify the unknown interlayer.
[0056] The following are device embodiments of the present invention, which can be used to perform method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.
[0057] In yet another embodiment of the present invention, a pattern recognition-based interlayer identification system is provided. The pattern recognition-based interlayer identification system can be used to implement the above-mentioned pattern recognition-based interlayer identification method. Specifically, the pattern recognition-based interlayer identification system includes a core sample selection module, a sensitive logging curve determination module, a canonical function calculation module, a discriminant function calculation module and an interlayer identification module.
[0058] Among them, the core sample selection module is used to select data of core samples of multiple different interlayer types.
[0059] The sensitive logging curve determination module is used to analyze the response characteristics of different interlayer types on conventional logging curves, and combine the dip logging data to determine the sensitive logging curve of the selected core sample that reflects the interlayer type.
[0060] The canonical function obtaining module is used to obtain the canonical function by using the interlayer sensitive logging curve through the Fisher dimensionality reduction method.
[0061] The discriminant function obtaining module is used to reduce the dimension of multi-parameter high-dimensional logging parameter data using canonical functions, and obtain the discriminant functions corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function.
[0062] The interlayer identification module is used to substitute the logging value of the depth section to be judged into the above discriminant functions for calculation, and compare the discriminant function values after calculation. The interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
[0063] In another embodiment of the present invention, a terminal device is provided, the terminal device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the interlayer identification method based on pattern recognition, including: S1, selecting data of multiple core samples of different interlayer types; S2, performing response characteristic analysis on different interlayer types on conventional logging curves, Combined with the inclination logging data, determine the sensitive logging curve of the selected core sample reflecting the interlayer type; S3, use the interlayer sensitive logging curve to obtain the canonical function through the Fisher dimensionality reduction method; S4, use the canonical function to reduce the dimension of the multi-parameter high-dimensional logging parameter data body, and obtain the discriminant function corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function; S5, substitute the logging value of the depth section to be judged into the above discriminant functions for calculation, compare the calculated discriminant function values, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
[0064] In another embodiment, the present invention further provides a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.
[0065] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the interlayer identification method based on pattern recognition in the above embodiment; one or more instructions in the computer-readable storage medium are loaded by the processor and the following steps are executed: S1, selecting data of multiple core samples of different interlayer types; S2, performing response characteristic analysis on conventional well logging curves for different interlayer types, and determining the sensitive well logging curves reflecting the interlayer types of the selected core samples in combination with the dip logging data; S3, using the interlayer sensitive well logging curves to obtain a canonical function by the Fisher dimensionality reduction method; S4, using the canonical function to reduce the dimension of the multi-parameter high-dimensional well logging parameter data body, and obtaining the discriminant function corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function; S5, substituting the logging value of the depth section to be judged into the above discriminant functions for calculation, comparing the calculated discriminant function values, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
[0066] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0067] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0068] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0069] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0070] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0071] It should be understood that the above description is for illustration and not for limitation. Many embodiments and many applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of the present teachings should not be determined with reference to the above description, but rather with reference to the foregoing claims and the full scope of equivalents to which such claims are entitled. For the purpose of comprehensiveness, all articles and references, including disclosures of patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the foregoing claims is not intended to be a waiver of such subject matter, nor should it be considered that the applicant has not considered such subject matter to be part of the disclosed inventive subject matter.
Claims
1. A method for identifying interlayers based on pattern recognition, characterized in that: The following steps are involved: S1, select data of multiple core samples with different interlayer types; S2, analyze the response characteristics of different interlayer types on conventional logging curves, and determine the sensitive logging curves of the selected core samples reflecting the interlayer types in combination with the dip logging data; S3, using the interlayer sensitive logging curve, the canonical function is obtained by Fisher dimensionality reduction method; S4, using canonical functions, reducing the dimension of multi-parameter high-dimensional logging parameter data, and obtaining the discriminant functions corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function; S5, substituting the logging value of the depth section to be judged into the above discriminant functions for calculation, comparing the calculated discriminant function values, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
2. The interlayer identification method based on pattern recognition according to claim 1, characterized in that: In S1, the interlayer type of the core sample was obtained by core analysis experimental method.
3. The interlayer identification method based on pattern recognition according to claim 1, characterized in that: In S2, the correlation between the interlayer and conventional logging and dip logging data is established through core and logging data, and then the sensitive logging curve reflecting the interlayer type of the selected core sample is determined.
4. The interlayer identification method based on pattern recognition according to claim 1, characterized in that: In S3, the k-type logging parameter vector matrix X is constructed using the determined sensitive logging curves. k =(x k1 ,x k2 ,…,x kn ) T , k is the number of interlayer types, n is the number of samples in each type, x ki is the logging parameter vector corresponding to the i-th sample point of the k-th category; the principle of Fisher discriminant is used to find the projection direction a that can separate the population as much as possible, and the coordinate transformation of the original data system is performed to extract the discriminant features of interlayers.
5. The interlayer identification method based on pattern recognition according to claim 4 is characterized in that: The process of extracting interlayer discriminant features is as follows: Sample point x ki The projection in the direction a is a T x ki , T represents the transpose of the matrix, and the projection of various logging data is: G k :a T x k1 …a T x kn k=1,2…5G k The mean of the data projection in the group is recorded as In order to better separate the categories, the ratio of the deviation between groups to the deviation within the group is maximized, that is: Solved using Lagrange multiplier algorithm The maximum value of is a, where B is the inter-group scatter matrix, E is the intra-group scatter matrix, and m is the counting symbol; the function constituting the eigenvalue of the projection direction is: F(X) = a T X k Where X k is the k-th type lithology sample vector matrix, and F(X) is the projection dimension reduction function, that is, the canonical function.
6. The interlayer identification method based on pattern recognition according to claim 1, characterized in that: In S4, a multi-parameter high-dimensional logging parameter data body is obtained from the existing logging data, and the multi-parameter high-dimensional logging parameter data body is reduced in dimension using a canonical function. T X transformation can get a new logging parameter sample vector y. According to the gradient descent algorithm of the least square error criterion function, w is found so that for any given y∈X k make Established, w is the discriminant weight matrix, thus obtaining the discriminant function f i (y).
7. The interlayer identification method based on pattern recognition according to claim 1, characterized in that: In S1, data of core samples with at least three different interlayer types are selected.
8. A method for identifying interlayers based on pattern recognition, characterized in that: include: The core sample selection module is used to select data of core samples with different interlayer types; The sensitive logging curve determination module is used to analyze the response characteristics of different interlayer types on conventional logging curves, and determine the sensitive logging curves of the selected core samples reflecting the interlayer types in combination with the dip logging data; The canonical function obtaining module is used to obtain the canonical function by using the interlayer sensitive logging curve through the Fisher dimensionality reduction method; The discriminant function obtaining module is used to reduce the dimension of the multi-parameter high-dimensional logging parameter data body by using the canonical function, and obtain the discriminant function corresponding to different interlayer types according to the gradient descent algorithm of the least square error criterion function; The interlayer identification module is used to substitute the logging value of the depth section to be judged into the above discriminant functions for calculation, compare the discriminant function values after calculation, and the interlayer type corresponding to the largest discriminant function value is the interlayer type of the depth section.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the interlayer identification method based on pattern recognition as claimed in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the interlayer identification method based on pattern recognition as claimed in any one of claims 1 to 7 are implemented.