A method and apparatus for reservoir classification based on well logs
Patent Information
- Application Number
- CN202510342864.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-21
- Publication Date
- 2026-09-22
AI Technical Summary
[0039]第五方面，本发明实施例提供了一种包含指令的计算机程序产品，当计算机程序产品在计算机设备上运行时，使得计算机设备执行如第一方面中所述的基于测井曲线的储层分类的方法。
Smart Images
Figure CN122796576A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of oil well logging processing and interpretation, and in particular to a method and apparatus for reservoir classification based on well logging curves. Background Technology
[0002] In well logging interpretation, reservoirs are identified based on well logging curves, and these reservoirs are further classified in detail. Since reservoirs of the same type often share similarities in geological characteristics and fluid content, this similarity manifests as morphological convergence in the well logging curves. Therefore, clustering methods are applied, which group reservoirs with similar logging responses into the same category based on the characteristics of the well logging curves.
[0003] Specifically, the clustering method first converts a series of well logging data from a formation sequence into feature vectors without strict sequence properties. These feature vectors can comprehensively and accurately reflect various reservoir attributes. Then, advanced clustering algorithms are used to classify these feature vectors, grouping reservoirs with similar well logging response characteristics into the same category. This approach not only improves the accuracy and efficiency of reservoir evaluation but also provides strong technical support for subsequent oil and gas exploration and development. Summary of the Invention
[0004] To improve the accuracy and reliability of reservoir classification, this invention provides a method for reservoir classification based on well logging curves.
[0005] In a first aspect, embodiments of the present invention provide a method for reservoir classification based on well logging curves, including:
[0006] Based on the multiple types of logging curves obtained for each well, multiple pairs of logging curves of common types are obtained for any two wells.
[0007] For any two wells, the corresponding mode distance is calculated based on each pair of logging curves of the common type.
[0008] For any two wells, the corresponding morphological distance is calculated using a dynamic time warping algorithm based on each pair of logging curves of the common type.
[0009] For any two wells, the corresponding structural distance is calculated based on the logging curves of each pair of common types.
[0010] For any two wells, the total similarity is calculated based on the pattern distance, morphological distance, and structural distance corresponding to each pair of logging curves of the common type. Based on the total similarity corresponding to all pairs of logging curves of the common type, the total similarity matrix between the two wells is obtained.
[0011] Clustering is performed on all the total similarity matrices obtained from pairwise intersections of all wells to obtain reservoir classification results.
[0012] Optionally, the step of calculating the corresponding structural distance for any two wells based on each pair of logging curves of the common type includes:
[0013] For each pair of logging curves of the common type for any two wells, the graph transformation method is used to transform each pair of logging curves of the common type to obtain a corresponding pair of graph structures;
[0014] Based on each pair of graph structures, a corresponding pair of statistical information is extracted;
[0015] Each pair of statistical information is arranged in a preset order to obtain a pair of multidimensional vectors;
[0016] Based on each pair of the multidimensional vectors, the corresponding structural distance is calculated using Euclidean distance.
[0017] Optionally, for any two wells, based on the pattern distance, morphological distance, and structural distance corresponding to each pair of logging curves of the common type, a total similarity is calculated. Based on the total similarity corresponding to all pairs of logging curves of the common type, a total similarity matrix between the two wells is obtained, including:
[0018] For each pair of logging curves of the common type for any two wells, the corresponding pattern distance, morphological distance, and structural distance are summed to obtain the corresponding total distance;
[0019] The total distance corresponding to each pair of logging curves of the common type is negativeed to obtain the total similarity corresponding to each pair of logging curves of the common type.
[0020] A total similarity matrix is constructed based on the total similarity corresponding to all logging curves of the common type.
[0021] Optionally, the step of obtaining multiple pairs of common-type logging curves for any two wells based on the acquired logging curves of multiple types for each well includes:
[0022] For any two wells, obtain multiple types of logging curves for each well, where each type of logging curve includes a type label and logging data;
[0023] By comparing the type labels corresponding to the two wells, multiple common type labels are identified;
[0024] Based on the multiple shared type labels, logging data matching the shared type labels are extracted from multiple logging data corresponding to the two wells to obtain multiple pairs of logging curves of the shared type.
[0025] Optionally, after obtaining multiple types of logging curves for each of any two wells, the method further includes:
[0026] All logging data from each well are preprocessed to obtain multiple processed logging data sets.
[0027] Optionally, the clustering based on the total similarity matrix obtained from pairwise intersections of all wells to obtain the reservoir classification result includes:
[0028] Based on the total similarity matrix and preset preference values obtained from pairwise intersection of all wells, the affinity propagation algorithm is executed to obtain the calculation results. The algorithm is iterated until the change in the calculation results is less than the preset threshold or the number of iterations reaches the maximum number of iterations, and the clustering label corresponding to each well is obtained.
[0029] Based on the clustering labels corresponding to each well, the reservoir classification results are obtained.
[0030] Secondly, embodiments of the present invention provide an apparatus for reservoir classification based on well logging curves, comprising:
[0031] The data acquisition module is used to obtain multiple pairs of common logging curves between any two wells based on the multiple types of logging curves acquired for each well.
[0032] The first calculation module is used to calculate the corresponding mode distance for any two wells based on each pair of logging curves of the common type.
[0033] The second calculation module is used to calculate the corresponding morphological distance for any two wells based on the logging curves of each pair of common types using a dynamic time warping algorithm.
[0034] The third calculation module is used to calculate the corresponding structural distance for any two wells based on the logging curves of each pair of common types.
[0035] The fourth calculation module is used to calculate the total similarity between any two wells based on the pattern distance, morphological distance and structural distance corresponding to each pair of logging curves of the common type, and to obtain the total similarity matrix between the two wells based on the total similarity corresponding to all pairs of logging curves of the common type.
[0036] The clustering analysis module is used to perform clustering based on the total similarity matrix obtained from pairwise intersections of all wells, and to obtain reservoir classification results.
[0037] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, characterized in that, when executed by a processor, the program implements the reservoir classification method based on logging curves as described in the first aspect.
[0038] Fourthly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the reservoir classification method based on well logging curves as described in the first aspect.
[0039] Fifthly, embodiments of the present invention provide a computer program product containing instructions that, when run on a computer device, cause the computer device to perform the reservoir classification method based on well logging curves as described in the first aspect.
[0040] The beneficial effects of the above-mentioned technical solutions provided in the embodiments of the present invention include at least the following:
[0041] This invention provides a method for reservoir classification based on well logging curves. For each pair of well logging curves of common type between any two wells, the mode distance is calculated, the corresponding morphological distance is calculated using a dynamic time warping algorithm, and the structural distance is also calculated. A total similarity is obtained based on the mode distance, morphological distance, and structural distance. A total similarity matrix between the two wells is obtained based on the total similarity of all pairs of well logging curves of common type. Clustering is performed on all total similarity matrices obtained from pairwise intersections of all wells to obtain the reservoir classification result. This method can determine the reservoir similarity between two wells from the similarity of the overall and local morphology of well logging curves, eliminating the need for physical property parameter calculations and avoiding the influence of overall well logging curve drift. Moreover, since the total similarity is calculated for any two wells, it effectively addresses the problem of missing well logging curves; only the common well logging curves of the two wells are used in the calculation, without the need for unified parameters. By comprehensively considering multiple distance dimensions, it can more effectively identify the similarity between well logging curves, thereby improving the accuracy of reservoir classification.
[0042] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.
[0043] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0044] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:
[0045] Figure 1 This is a flowchart of a reservoir classification method based on well logging curves provided in an embodiment of the present invention;
[0046] Figure 2 This is a schematic diagram illustrating the results of preprocessing well logging data as provided in an embodiment of the present invention;
[0047] Figure 3 This is a schematic diagram showing the result of processing a pair of well logging curves using the DTW algorithm in an embodiment of the present invention;
[0048] Figure 4 This is a schematic diagram of the graph structure extracted using the Visibility Graph provided in an embodiment of the present invention;
[0049] Figure 5 This is a schematic diagram of the total similarity matrix provided in an embodiment of the present invention;
[0050] Figure 6 This is a schematic diagram of a reservoir classification device based on well logging curves provided in an embodiment of the present invention. Detailed Implementation
[0051] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0052] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," "far," "near," "front," and "rear," etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0053] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0054] The inventors discovered that commonly used reservoir clustering methods first calculate parameters reflecting reservoir physical properties, such as porosity and permeability, based on well logging curves, and then use the K-means clustering algorithm to perform cluster analysis on these parameters. This method has the following drawbacks:
[0055] First, the K-means clustering algorithm typically treats reservoir data from a single well as an independent processing unit. When it's necessary to assess which reservoirs within the same formation are similar across multiple wells in an entire region, it's crucial to ensure that all wells participating in the clustering use the same physical property parameters as the metric. Because the quality of logging data varies from well to well, it's impossible to establish a unified feature vector for physical property parameters. For example, some wells may have poor-quality logging curves related to permeability, making permeability calculation impossible. Therefore, either these wells should not participate in the clustering, or their feature vectors should not include permeability, thus reducing the representativeness and comprehensiveness of the clustering results.
[0056] Secondly, the K-means clustering method uses distance metrics based on sampling points, focusing on utilizing local information to classify reservoirs. While this method may perform well in handling local features of reservoirs within a single well, when the stratigraphic distribution is large and it's necessary to determine the similarity of reservoirs in two different formations or wells, using the same physical property parameters as the clustering feature vector for different wells can lead to significant errors due to overall parameter drift. Therefore, the K-means clustering method based on existing physical property parameters has limitations in macroscopic stratigraphic correlation and reservoir similarity assessment.
[0057] To address the aforementioned issues, the inventors developed a method and apparatus for reservoir classification based on well logging curves, aiming to improve the accuracy and reliability of reservoir classification.
[0058] Example 1
[0059] See Figure 1 This embodiment proposes a reservoir classification method based on well logging curves, the specific process of which may include the following steps:
[0060] Step S1: Based on the multiple types of logging curves obtained for each well, obtain multiple pairs of logging curves of common types for any two wells;
[0061] Step S2: For any two wells, calculate the corresponding mode distance based on the logging curves of each pair of common types;
[0062] Step S3: For any two wells, based on the logging curves of each pair of common types, calculate the corresponding morphological distance using the dynamic time warping algorithm;
[0063] Step S4: For any two wells, calculate the corresponding structural distance based on the logging curves of each pair of common types;
[0064] Step S5: For any two wells, calculate the total similarity based on the pattern distance, morphological distance and structural distance corresponding to each pair of logging curves of common type. Based on the total similarity corresponding to all pairs of logging curves of common type, obtain the total similarity matrix between the two wells.
[0065] Step S6: Cluster the total similarity matrices obtained from pairwise intersections of all wells to obtain the reservoir classification results.
[0066] To provide a clearer explanation of the above method for reservoir classification based on well logging curves, each step will be described in detail below.
[0067] In step S1 above, the specific process of obtaining multiple pairs of common-type logging curves for any two wells based on the acquired logging curves of multiple types for each well may include the following steps:
[0068] Step S11: For any two wells, obtain multiple types of logging curves for each well, where each type of logging curve includes a type label and logging data.
[0069] In step S11 above, the type labels can include natural gamma, resistivity, sonic transit time, etc. After acquiring multiple types of logging curves for each well, all logging data for each well can be preprocessed to obtain multiple processed logging data. The preprocessing process can include data cleaning, handling missing values, smoothing and denoising, and standardization or normalization. Logging data may contain noise or outliers, therefore smoothing and denoising are necessary, such as using moving averages, Gaussian filtering, or wavelet denoising to remove outliers and perform data normalization to ensure data quality and improve the accuracy of subsequent analysis. Furthermore, the dimensions of logging data of the same type from different wells may differ, therefore standardization (e.g., Z-score) or normalization (scaling to between 0 and 1) is required to eliminate the influence of dimensions and facilitate subsequent calculation and analysis. See also... Figure 2 These are a pair of logging curves of the same type from two wells, with the initial logging curves being as follows: Figure 2The blue and green curves in the image, after the above preprocessing operations, are respectively obtained as follows: Figure 2 The orange and red curves are shown in the image.
[0070] Step S12: Compare the type labels corresponding to the two wells to determine the multiple common type labels.
[0071] In step S12 above, for example, given two wells, well A and well B, all type tags for well A and well B are compared to find the common type tags. These common type tags represent logging projects that have been performed on both wells. When comparing type tags, efficient algorithms or data structures (such as hash tables) can be used to accelerate the comparison process and improve efficiency.
[0072] Step S13: Based on multiple common type labels, extract logging data that matches the common type labels from multiple logging data corresponding to the two wells respectively, and obtain multiple pairs of logging curves with common types.
[0073] In step S2 above, the specific process of calculating the mode distance for each pair of logging curves of common type for any two wells may include the following steps:
[0074] Step S21: Discretize each pair of logging data to obtain a corresponding pair of strings.
[0075] In step S21 above, before discretizing the logging data, the logging data can be standardized to ensure that the data format, units, resolution, etc. are consistent.
[0076] For each logging data point, the discretization process can include the following steps: The logging data is segmented using the Piecewise Aggregation Approximation (PAA) method, with each segment represented by a representative value (such as the mean or extreme value), thus simplifying the data and reducing computational load; then, the segmented results are mapped to discrete symbols using the Symbolic Aggregation Approximation (SAX) method to obtain a string. For example, logging data: [0.2, 0.5, 0.8, 0.3, 0.6, 0.9], segmented using PAA into three segments, represented by the mean: [0.35, 0.55, 0.75], and symbolized using SAX: ['a', 'b', 'c'].
[0077] Step S22: Based on the preset editing operation cost, traverse each character of a pair of strings and calculate the minimum editing distance, which is used as the pattern distance of each pair of logging curves of common type.
[0078] In step S22 above, the smaller the model distance, the higher the similarity and the smaller the difference between the two wells on this type of logging curve.
[0079] In step S3 above, the specific process of calculating the corresponding morphological distance for any two wells based on each pair of logging curves of common type using the dynamic time warping algorithm may include the following steps:
[0080] Step S31: For each pair of logging curves of common type, calculate the Euclidean distance between any two points between them and construct a distance matrix.
[0081] Step S32: Construct a cumulative distance matrix based on the distance matrix and each pair of logging curves of common type.
[0082] In step S32 above, the specific process of calculating the cumulative distance matrix may include: creating a cumulative distance matrix of the same size as the distance matrix to store the minimum cumulative distance from the starting point to each point; starting from the top left corner of the cumulative distance matrix, using the Dynamic Time Warping (DTW) algorithm to calculate the minimum distance to each point step by step along the shortest path.
[0083] Step S33: Calculate the morphological distance based on the cumulative distance matrix.
[0084] In step S33 above, the specific process of calculating the morphological distance may include: using a dynamic time warping algorithm, starting from the lower right corner of the cumulative distance matrix, backtracking to find an optimal path, i.e., the path with the minimum distance. (See also...) Figure 3 For a pair of preprocessed logging curves of the same type from two wells, the line connecting any two points represents the optimal path. This path indicates the optimal alignment between logging curves of the common type in the two wells. For example, Figure 3 The left image shows the logging curve of well A, and the right image shows the logging curve of well B. The logging value of well A at a depth of 260 meters is aligned with the logging value of well B at a depth of 190 meters. This alignment is the best-matching correspondence found by the DTW algorithm described above. The cumulative distance along the optimal path is the morphological distance between logging curves of the common type in the two wells. Calculating the morphological distance between each pair of logging curves of the common type using the DTW algorithm reflects both local morphological differences and reveals overall similarity, providing a quantitative basis for formation correlation and reservoir analysis.
[0085] In step S4 above, the specific process of calculating the structural distance of any pair of logging curves of common type for any two wells may include the following steps:
[0086] Step S41: For each pair of logging curves of common type between any two wells, use the graph transformation method to transform each pair of logging curves of common type to obtain the corresponding graph structure.
[0087] In step S41 above, the graph transformation method may include Recurrence Graph, Transition Network, and Visibility Graph, extracting statistics reflecting the overall topological characteristics from the graph structure. See also... Figure 4 To construct the graph structure corresponding to the well logging curves extracted using the Visibility Graph, each data point (or segmented window) of the well logging curve is used as a node in the graph structure, and edges are established between similar nodes.
[0088] Step S42: Based on each pair of graph structures, extract the corresponding pair of statistical information.
[0089] In step S42 above, the statistical information may include the degree of the vertex, the number of edges, the path length, etc.
[0090] Step S43: Arrange each pair of statistical information in a preset order to obtain a pair of multidimensional vectors.
[0091] In step S43 above, the statistical information is standardized and combined into a comparable multidimensional vector. Each logging curve corresponds to a multidimensional vector. The preset order can be set as: multidimensional vector = {degree, number of edges, path length ...}.
[0092] Step S44: Based on each pair of multidimensional vectors, calculate the corresponding structural distance using Euclidean distance.
[0093] In step S44 above, for each pair of multidimensional vectors, since these multidimensional vectors have no sequentiality, the Euclidean distance can be used to measure the structural distance between them. Specifically, the structural distance is calculated using the following equation (1):
[0094]
[0095] In equation (1) above, x i and y i These are the i-th eigenvalues in the two multidimensional vectors, respectively.
[0096] Structural distance is calculated using Euclidean distance to quantify the difference between two multidimensional vectors, reflecting the structural similarity of this pair of logging curves. The smaller the structural distance, the more similar the logging curves; the larger the structural distance, the more significant the difference between the logging curves.
[0097] Step S5, for any two wells, the specific process of calculating the total similarity of any pair of logging curves of common type and constructing the total similarity matrix between the two wells may include the following steps:
[0098] Step S51: For each pair of logging curves of common type for any two wells, sum the corresponding mode distance, morphological distance and structural distance to obtain the corresponding total distance.
[0099] In step S51 above, for any two wells, the overall difference between the two wells on a certain common type of logging curve is quantified by combining the pattern distance, morphological distance and structural distance of each pair of common type logging curves.
[0100] Step S52: Negate the total distance for each pair of logging curves of the common type to obtain the total similarity for each pair of logging curves of the common type.
[0101] Step S53: Construct a total similarity matrix based on the total similarity of all pairs of logging curves of the common type.
[0102] In step S53 above, for any two wells, the total similarity of all common logging types is summarized to form a total similarity matrix that reflects the overall similarity between wells. Here, if any two wells have n types of logging curves, the total similarity matrix is n×n. For example, if the two wells have only two types of logging curves (e.g., GR and Rt), the total similarity matrix will be 2×2, containing four total similarity values: GR-GR, GR-Rt, Rt-GR, and Rt-Rt (note that GR-Rt and Rt-GR are equal, therefore the matrix is symmetric). See also... Figure 5 , which is the total similarity matrix between the two wells, and there are 20 common types of logging curves between the two wells.
[0103] Step S6, clustering based on the total similarity matrices obtained from pairwise intersections of all wells to obtain reservoir classification results, may include the following steps:
[0104] Step S61: Based on the total similarity matrix obtained from pairwise intersection of all wells and the preset preference value, execute the affinity propagation algorithm to obtain the calculation result. Iterate continuously until the change in the calculation result is less than the preset threshold or the number of iterations reaches the maximum number of iterations, and obtain the clustering label corresponding to each well.
[0105] In step S61 above, all well pairwise intersection means that all wells are analyzed in pairs. For example, if there are three wells, well A, well B and well C, the results of all well pairwise intersection are: well A-well B, well A-well C and well B-well C.
[0106] The specific process of executing the affinity propagation algorithm may include the following steps: Setting a responsibility matrix R(i,k) and an availability matrix A(i,k), initialized as all-zero matrices, where the dimension of the responsibility matrix R(i,k) is the same as the dimension of the total similarity matrix; setting a damping factor, typically between 0.5 and 0.9, to prevent numerical oscillations during iteration; performing iterative updates to obtain the calculated results, i.e., the updated responsibility matrix R(i,k) and the updated availability matrix A(i,k), where the updated responsibility matrix R(i,k) reflects the ability of any well i to select well k as a cluster center, and the updated availability matrix A(i,k) reflects the cumulative support of well k as a cluster center. Iterating continuously until the change in the calculated results is less than a preset threshold or the maximum number of iterations is reached; that is, stopping iteration when the update amount of the responsibility matrix R(i,k) and the availability matrix A(i,k) is less than the preset threshold, or stopping iteration when the maximum number of iterations is reached. For each well, select the k well with the largest R(i,k) + A(i,k) as its cluster center. When the iteration stops, output the cluster label corresponding to each well, which indicates which reservoir category each well is assigned to.
[0107] S62. Based on the clustering labels corresponding to each well, the reservoir classification results are obtained.
[0108] In step S62 above, wells with the same cluster label belong to the same reservoir category. Based on the cluster label, the wells are assigned to different reservoir categories to obtain the reservoir classification results.
[0109] The method in this embodiment determines the reservoir similarity between two wells based on the similarity of their overall and local logging curves, eliminating the need for physical property parameter calculations and avoiding the impact of overall logging curve drift. Furthermore, since the overall similarity calculation is performed on any two wells, it effectively addresses the issue of missing logging curves; only common logging curves between the two wells are used in the calculation, eliminating the need for standardized parameters. By comprehensively considering multiple distance dimensions, the method in this embodiment can more effectively identify the similarity between logging curves, thereby improving the accuracy of reservoir classification.
[0110] Example 2
[0111] Based on the same inventive concept, see [reference] Figure 6 This embodiment proposes a device for reservoir classification based on well logging curves, comprising:
[0112] Data acquisition module 1 is used to obtain multiple pairs of common logging curves between any two wells based on the multiple types of logging curves acquired for each well.
[0113] The first calculation module 2 is used to calculate the corresponding mode distance for any two wells based on the logging curves of each pair of common types using the minimum edit distance.
[0114] The second calculation module 3 is used to calculate the corresponding morphological distance for any two wells based on each pair of logging curves of common type using a dynamic time warping algorithm.
[0115] The third calculation module 4 is used to calculate the corresponding structural distance for any two wells based on each pair of logging curves of common type.
[0116] The fourth calculation module 5 is used to calculate the total similarity between any two wells based on the pattern distance, morphological distance and structural distance corresponding to each pair of logging curves of common type. Based on the total similarity corresponding to all pairs of logging curves of common type, the total similarity matrix between the two wells is obtained.
[0117] Clustering analysis module 6 is used to perform clustering based on the total similarity matrix obtained from pairwise intersections of all wells, and to obtain reservoir classification results.
[0118] The reservoir classification device based on well logging curves provided in this embodiment of the invention has a similar implementation principle and technical effect to that of Embodiment 1, and will not be described again here.
[0119] Example 3
[0120] Based on the same inventive concept, this application also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the reservoir classification method based on well logging curves as described in Embodiment 1.
[0121] The computer-readable storage medium may be included in the device / apparatus described in the above embodiments; or it may exist independently and not assembled into the device / apparatus. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to Embodiment 1 of the present invention.
[0122] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
[0123] Example 4
[0124] Based on the same inventive concept, this application also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the reservoir classification method based on well logging curves as described in Embodiment 1.
[0125] Example 5
[0126] Based on the same inventive concept, this application proposes a computer program product containing instructions that, when run on a computer device, cause the computer device to execute the reservoir classification method based on well logging curves in Embodiment 1.
[0127] The principles by which the above-mentioned devices, clients, media, and related equipment in this embodiment of the invention solve the problem are similar to those of the aforementioned methods. Therefore, their implementation can refer to the implementation of the aforementioned methods, and repeated details will not be repeated.
[0128] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.
[0129] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0130] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0131] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0132] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. This disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.
Claims
1. A method for reservoir classification based on well logging curves, characterized in that, include: Based on the multiple types of logging curves obtained for each well, multiple pairs of logging curves of common types are obtained for any two wells. For any two wells, the corresponding mode distance is calculated based on each pair of logging curves of the common type. For any two wells, the corresponding morphological distance is calculated using a dynamic time warping algorithm based on each pair of logging curves of the common type. For any two wells, the corresponding structural distance is calculated based on the logging curves of each pair of common types. For any two wells, the total similarity is calculated based on the pattern distance, morphological distance, and structural distance corresponding to each pair of logging curves of the common type. Based on the total similarity corresponding to all pairs of logging curves of the common type, the total similarity matrix between the two wells is obtained. Clustering is performed on all the total similarity matrices obtained from pairwise intersections of all wells to obtain reservoir classification results.
2. The method for reservoir classification based on well logging curves according to claim 1, characterized in that, For any two wells, the corresponding structural distance is calculated based on each pair of logging curves of the common type, including: For each pair of logging curves of the common type for any two wells, the graph transformation method is used to transform each pair of logging curves of the common type to obtain a corresponding pair of graph structures; Based on each pair of graph structures, a corresponding pair of statistical information is extracted; Each pair of statistical information is arranged in a preset order to obtain a pair of multidimensional vectors; Based on each pair of the multidimensional vectors, the corresponding structural distance is calculated using Euclidean distance.
3. The method for reservoir classification based on well logging curves according to claim 1, characterized in that, For any two wells, based on the pattern distance, morphological distance, and structural distance corresponding to each pair of logging curves of the common type, a total similarity is calculated. Based on the total similarity corresponding to all pairs of logging curves of the common type, a total similarity matrix between the two wells is obtained, including: For each pair of logging curves of the common type for any two wells, the corresponding pattern distance, morphological distance, and structural distance are summed to obtain the corresponding total distance; The total distance corresponding to each pair of logging curves of the common type is negativeed to obtain the total similarity corresponding to each pair of logging curves of the common type. A total similarity matrix is constructed based on the total similarity corresponding to all logging curves of the common type.
4. The method for reservoir classification based on well logging curves according to claim 1, characterized in that, Based on the acquired logging curves of various types for each well, multiple pairs of logging curves of common types are obtained for any two wells, including: For any two wells, obtain multiple types of logging curves for each well, where each type of logging curve includes a type label and logging data; By comparing the type labels corresponding to the two wells, multiple common type labels are identified; Based on the multiple shared type labels, logging data matching the shared type labels are extracted from multiple logging data corresponding to the two wells to obtain multiple pairs of logging curves of the shared type.
5. The method for reservoir classification based on well logging curves according to claim 4, characterized in that, After obtaining multiple types of logging curves for each of any two wells, the process further includes: All logging data from each well are preprocessed to obtain multiple processed logging data sets.
6. The method for reservoir classification based on well logging curves according to claim 1, characterized in that, The reservoir classification results are obtained by clustering all the total similarity matrices obtained from pairwise intersections of all wells, including: Based on the total similarity matrix and preset preference values obtained from pairwise intersection of all wells, the affinity propagation algorithm is executed to obtain the calculation results. The algorithm is iterated until the change in the calculation results is less than the preset threshold or the number of iterations reaches the maximum number of iterations, and the clustering label corresponding to each well is obtained. Based on the clustering labels corresponding to each well, the reservoir classification results are obtained.
7. A device for reservoir classification based on well logging curves, characterized in that, include: The data acquisition module is used to obtain multiple pairs of common logging curves between any two wells based on the multiple types of logging curves acquired for each well. The first calculation module is used to calculate the corresponding mode distance for any two wells based on each pair of logging curves of the common type. The second calculation module is used to calculate the corresponding morphological distance for any two wells based on the logging curves of each pair of common types using a dynamic time warping algorithm. The third calculation module is used to calculate the corresponding structural distance for any two wells based on the logging curves of each pair of common types. The fourth calculation module is used to calculate the total similarity between any two wells based on the pattern distance, morphological distance and structural distance corresponding to each pair of logging curves of the common type, and to obtain the total similarity matrix between the two wells based on the total similarity corresponding to all pairs of logging curves of the common type. The clustering analysis module is used to perform clustering based on the total similarity matrix obtained from pairwise intersections of all wells, and to obtain reservoir classification results.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the reservoir classification method based on logging curves as described in any one of claims 1-6.
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 program, it implements the reservoir classification method based on logging curves as described in any one of claims 1-6.
10. A computer program product containing instructions that, when run on a computer device, cause the computer device to perform the reservoir classification method based on well logging curves as described in any one of claims 1-6.