CN117574269B - Intelligent identification method and system for natural cracks of land shale reservoir - Google Patents
Intelligent identification method and system for natural cracks of land shale reservoir Download PDFInfo
- Publication number
- CN117574269B CN117574269B CN202410064155.5A CN202410064155A CN117574269B CN 117574269 B CN117574269 B CN 117574269B CN 202410064155 A CN202410064155 A CN 202410064155A CN 117574269 B CN117574269 B CN 117574269B
- Authority
- CN
- China
- Prior art keywords
- natural
- data
- vertexes
- graph
- sampling points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 110
- 238000005070 sampling Methods 0.000 claims abstract description 137
- 238000013528 artificial neural network Methods 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 22
- 239000013598 vector Substances 0.000 claims description 64
- 230000004044 response Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 230000002776 aggregation Effects 0.000 claims description 12
- 238000004220 aggregation Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 12
- 238000003384 imaging method Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 abstract description 7
- 238000002372 labelling Methods 0.000 abstract description 3
- 206010017076 Fracture Diseases 0.000 description 66
- 208000010392 Bone Fractures Diseases 0.000 description 62
- 239000011159 matrix material Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000011161 development Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 239000011435 rock Substances 0.000 description 6
- 238000005096 rolling process Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 241000634997 Leucon Species 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- UQVDQSWZQXDUJB-UHFFFAOYSA-N hydron;7h-purin-6-amine;chloride Chemical compound Cl.NC1=NC=NC2=C1NC=N2 UQVDQSWZQXDUJB-UHFFFAOYSA-N 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 239000004215 Carbon black (E152) Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011010 flushing procedure Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013106 supervised machine learning method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
- E21B49/08—Obtaining fluid samples or testing fluids, in boreholes or wells
- E21B49/087—Well testing, e.g. testing for reservoir productivity or formation parameters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V11/00—Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/0985—Hyperparameter optimisation; Meta-learning; Learning-to-learn
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Probability & Statistics with Applications (AREA)
- Geophysics (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- Geochemistry & Mineralogy (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a land shale reservoir natural fracture intelligent identification method and a system, which relate to the technical field of machine learning, and the method comprises the following steps: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using corresponding natural crack labeling data of partial sampling points. The intelligent identification method and the intelligent identification system for the natural cracks of the land shale reservoir provided by the embodiment of the invention have the advantage that the accuracy of the identification of the natural cracks of the oil and gas reservoir is improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of machine learning, in particular to an intelligent identification method and system for natural cracks of a land shale reservoir.
Background
Natural fracture identification is the basis and key of research on natural fracture distribution rules of a land shale reservoir, and can provide reliable geological basis for efficient exploration and development of unconventional reservoirs.
Natural fracture identification is a key issue for unconventional reservoir hydrocarbon exploration and development. The rock core and imaging logging data have high interpretation precision of natural cracks, but have small quantity and high cost, and are not beneficial to understanding the development rule of the natural cracks of the reservoir in the whole area, so that the conventional logging curve is often adopted to interpret the development condition of the natural cracks of a single well. Because the logging response characteristics of the natural fracture are weak and extremely complex, the interpretation accuracy of the conventional logging natural fracture is low. The machine learning provides a good opportunity for solving the difficult problem, and the nonlinear logging response characteristics of the natural cracks can be better extracted by a machine learning method, so that the accuracy of identifying the single-well natural cracks is improved.
However, the existing machine learning method is used for recognizing natural cracks of the land shale reservoir, and the recognition accuracy is still low.
Disclosure of Invention
Aiming at the defects existing in the prior art, the embodiment of the invention provides an intelligent identification method and system for natural cracks of a land shale reservoir.
The embodiment of the invention provides an intelligent identification method for natural cracks of a land shale reservoir, which comprises the following steps: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
According to the method for intelligently identifying natural cracks of a land shale reservoir provided by the embodiment of the invention, the graph data is constructed according to the depth data, the logging curve value and the lithology identification result of the sampling points, and the method comprises the following steps: taking the sampling points as vertexes of a graph, and constructing feature vectors of the vertexes according to the log values of the log curves corresponding to the sampling points; constructing a first type edge between the vertexes with adjacent depths according to the depth data; for vertexes corresponding to the sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the characteristic vectors of the vertexes.
According to the intelligent identification method for natural cracks of the land shale reservoir, provided by the embodiment of the invention, a second type edge is constructed among the vertexes through similarity calculation, and the intelligent identification method comprises the following steps: calculating Euclidean distance between every two vertexes; in response to the euclidean distance being greater than a preset fractional number, not constructing the second type edge between the corresponding two vertices; and constructing the second type edge between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset fraction.
According to the method for intelligently identifying natural cracks of the land shale reservoir, which is provided by the embodiment of the invention, the graph data is input into a pre-trained natural crack identification model to obtain a natural crack identification result corresponding to the sampling point, and the method comprises the following steps: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex; and endowing the natural crack identification result of the vertex with the corresponding sampling point to obtain the natural crack identification result corresponding to the sampling point.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the natural crack identification model comprises a multi-layer graph convolutional neural network and an output layer; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex, wherein the method comprises the following steps of: inputting the graph data into the natural crack recognition model, and updating the feature vector of the vertex through each layer of graph convolution neural network; the method comprises the steps that each layer of graph convolution neural network is used for carrying out feature vector aggregation on neighbor vertexes of the vertexes, combining the aggregated feature vectors with the feature vectors of the vertexes and updating the feature vectors of the vertexes; and obtaining the natural crack identification result of each vertex according to the output result of the last layer of the graph convolution neural network after the output is processed by the activation function of the output layer.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the method further comprises the following steps: and acquiring the natural fracture marking data according to the core description data and the imaging logging data.
According to the intelligent identification method for natural cracks of the land shale reservoir, provided by the embodiment of the invention, before the map data is constructed according to the depth data, the logging curve value and the lithology identification result of the sampling points, the method further comprises the following steps: and carrying out preset data preprocessing on the conventional logging data.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the method further comprises the following steps: and visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
The embodiment of the invention also provides an intelligent identification system for natural cracks of the land shale reservoir, which comprises the following steps: an acquisition module for: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; a construction module for: constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; an identification module for: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
The embodiment of the invention also provides electronic equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the steps of the intelligent identification method for the natural cracks of the land shale reservoir are realized when the processor executes the program.
The embodiment of the invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the land shale reservoir natural fracture intelligent identification method as described in any one of the above.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the steps of the intelligent identification method for natural cracks of the land shale reservoir when being executed by a processor.
According to the intelligent identification method and system for natural cracks of the land shale reservoir, conventional logging data are obtained; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values, map data are built according to the depth data of the sampling points, the logging curve values and lithology recognition results, the map data are input into a pre-trained natural fracture recognition model, natural fracture recognition results corresponding to the sampling points are obtained, and accuracy of natural fracture recognition of an oil and gas reservoir is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is one of the flow diagrams of the intelligent identification method for natural cracks of a land shale reservoir provided by the embodiment of the invention;
Fig. 2 is a schematic diagram of a flow chart of generating graph data in the intelligent identification method of natural cracks of a land shale reservoir provided by the embodiment of the invention;
FIG. 3 is a second flow chart of the intelligent identification method for natural cracks of a land shale reservoir provided by the embodiment of the invention;
FIG. 4 is a third flow chart of the intelligent identification method for natural fractures of a land shale reservoir provided by the embodiment of the invention;
FIG. 5 is a schematic structural diagram of a land shale reservoir natural fracture intelligent identification system provided by an embodiment of the invention;
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Machine learning is a discipline that studies how to simulate human learning activities using computers, i.e., analyzing and processing past data by machines to obtain new knowledge and new skills. At present, the single-well natural fracture recognition method based on machine learning can be divided into an unsupervised learning method, a supervised learning method and a semi-supervised learning method. The non-supervision method does not need a label sample, and the identification accuracy is low due to lack of guidance of label information, and algorithms in single well natural crack identification comprise k-means, SOM neural network, gaussian Mixture Model (GMM) and the like; the supervised learning method needs a large number of labeled samples for learning, the recognition accuracy is relatively high, and algorithms in single-well natural fracture recognition include Fuzzy Inference Systems (FIS), tree models (decision trees, random forests and GBDT, XGBOOST, lightGBM), naive Bayes, bayesian networks, neural networks (such as BP, CNN, RNN, CGAN), kernel methods (such as SVM and KFD) and the like; the semi-supervised learning method only needs a small amount of samples with labels, is an effective method for solving the problem of small data size of the labels, and algorithms in single-well natural fracture recognition comprise a semi-supervised support vector machine, a Laplacian support vector machine and the like.
In general, the supervised learning method has better natural crack recognition effect than the unsupervised learning method, and is preferred when the number of labeled sample data is sufficient. The semi-supervised learning method is more suitable for model construction of limited labels. According to the embodiment of the invention, the marking data of the natural cracks are obtained by using the core observation and imaging logging interpretation results, and the core observation and imaging logging interpretation results are limited, so that the embodiment of the invention adopts a semi-supervised machine learning method to construct a natural crack identification model.
Fig. 1 is a schematic flow chart of a method for intelligently identifying natural cracks of a land shale reservoir according to an embodiment of the invention. As shown in fig. 1, the method includes:
S1, acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values.
Conventional logging data is data obtained by conventional logging methods. The conventional logging method mainly refers to a logging method which is required to be measured in the oil and gas exploration and development, well logging evaluation and well logging engineering. The conventional log data includes a plurality of log curves including depth data for sampling points and corresponding log curve values. Wherein the log values of different logs have different meanings.
The acoustic wave time difference is increased, the resistivity is reduced, the density is reduced, the compensated neutrons are increased, the gamma is middle and high, and the like in the natural fracture development section, but the natural fracture has weak response in a logging curve and the characteristic response does not uniquely correspond to the fracture result. Based on the conventional logging curve, various characteristic curves which are more sensitive to the natural crack development condition can be optimized, and the sensitivity of the characteristic curves can be analyzed.
Based on correlation analysis of logging curves and natural fractures, in the embodiment of the invention, 9 logging curves including acoustic time difference (AC), density logging (DEN), neutron porosity (CNL), natural potential (SP), natural Gamma (GR), undisturbed formation Resistivity (RT), invaded zone Resistivity (RI), flushing zone Resistivity (RXO) and borehole diameter (CAL) are selected for natural fracture identification.
And S2, constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results.
The graph data includes a graph structure and feature vectors. Vertices in the graph structure correspond to sampling points in conventional log data. And corresponding to each sampling point, the depth data and the logging curve values of a plurality of preset logging curves. And constructing the characteristic vector of the corresponding vertex according to the log values of the plurality of log corresponding to the sampling points. And constructing edges between the sampling points according to the association relation between the sampling points. The graph data is thus obtained based on the vertices, edges, and feature vectors of the vertices of the graph.
S3, inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
The natural crack identification model is obtained by training a model based on a graph convolution neural network by corresponding natural crack labeling data of partial sampling points.
The training process of the natural fracture recognition model comprises the following steps:
constructing a training sample based on conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and logging curve values;
Constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results;
Inputting the graph data into a neural network model constructed by a multi-layer graph convolutional neural network and an output layer, training the neural network model by taking natural crack labeling data corresponding to part of sampling points as output labels of corresponding vertexes, and obtaining a natural crack identification model after training is finished.
In the training process, the error function can be iteratively optimized through SGD or Adam according to the principle of total error minimization, so that the parameter optimization of the network is realized. The grid search method can be adopted to find the optimal value of the super parameter, and the basic idea is to divide the super parameter to be optimized into grids in a certain space range, and find the optimal solution of the super parameter by traversing all the intersection points in the grids.
Rock with different lithology has larger brittleness difference, and natural cracks are easier to develop in areas with large rock brittleness. Therefore, the embodiment of the invention blends lithology factors (lithology recognition results of sampling points) into the construction process of the graph data of the natural fracture recognition model, and can improve the recognition capability of the model on the natural fracture. The figure provides a clear representation of any relationship between entities as a data structure. Topology information among different fracture segments can be captured by calculating graph data through a graph neural network algorithm, and geological knowledge is integrated into a model. Meanwhile, compared with the calculation of the original sequence data, the calculation of the graph data has higher induction bias, the training efficiency is higher, the training time is shorter, and the generalization capability is stronger.
According to the intelligent identification method for natural cracks of the land shale reservoir, conventional logging data are obtained; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values, map data are built according to the depth data of the sampling points, the logging curve values and lithology recognition results, the map data are input into a pre-trained natural fracture recognition model, natural fracture recognition results corresponding to the sampling points are obtained, and accuracy of natural fracture recognition of an oil and gas reservoir is improved.
According to the method for intelligently identifying natural cracks of a land shale reservoir provided by the embodiment of the invention, the graph data is constructed according to the depth data, the logging curve value and the lithology identification result of the sampling points, and the method comprises the following steps: taking the sampling points as vertexes of a graph structure, and constructing feature vectors of the vertexes according to the log values of the log curves corresponding to the sampling points; constructing a first type edge between the vertexes with adjacent depths according to the depth data; for vertexes corresponding to the sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the characteristic vectors of the vertexes.
The construction of the graph structure requires two conditions to be satisfied: (1) The constructed graph data set needs to have a certain geological meaning; (2) The method is in line with the bottom logic of the graph convolution operation, and the natural crack recognition accuracy is improved as much as possible.
Fig. 2 is a schematic diagram of a flow chart of generating graph data in the intelligent identification method of natural cracks of a land shale reservoir. Based on the above two points, in connection with fig. 2, the construction of the graph structure includes: (1) And generating vertexes, namely taking a conventional logging curve sampling point as the vertexes, taking a logging curve value corresponding to each vertex as a characteristic vector of the vertex, and taking the number of logging curves as the dimension of the vector, wherein the characteristic vectors of the vertexes form a characteristic matrix. (2) Edge generation, wherein vertices with adjacent depths are connected by edges to obtain a first type of edge (a sequence edge in fig. 2); next, for vertices corresponding to sampling points corresponding to the same lithology recognition result, a second type edge (distance edge in fig. 2) is constructed between vertices by similarity calculation. The generated edges belong to undirected edges, and all vertexes and corresponding feature vectors form graph data together.
It should be noted that the first type of edge and the second type of edge are merely different methods for constructing the representing edges.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the sampling points are used as the vertexes of the graph structure, and the feature vectors of the vertexes are constructed according to the logging curve values of the logging curves corresponding to the sampling points; constructing a first type edge between vertices adjacent in depth according to the depth data; for vertexes corresponding to sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the feature vectors of all vertexes, and improving the adaptability of the graph data for crack identification, so that the accuracy of natural crack identification of the oil and gas reservoir is further improved.
According to the intelligent identification method for natural cracks of the land shale reservoir, provided by the embodiment of the invention, a second type edge is constructed among the sampling points through similarity calculation, and the intelligent identification method comprises the following steps: calculating Euclidean distance between every two vertexes; in response to the euclidean distance being greater than a preset fractional number, not constructing the second type edge between the corresponding two vertices; and constructing the second type edge between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset fraction.
When a second type edge is built between vertexes through similarity calculation, euclidean distance between every two vertexes is calculated, a reasonable quantile is set, if the Euclidean distance is larger than the preset quantile, the second type edge is not built between the corresponding two vertexes, and if the Euclidean distance is smaller than or equal to the preset quantile, the second type edge is built between the corresponding two vertexes.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the Euclidean distance between every two vertexes is calculated, the second type edge is not constructed between the corresponding two vertexes in response to the Euclidean distance being larger than the preset quantile, and the second type edge is constructed between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset quantile, so that the relation network between the nodes is thinned, and the accuracy of the natural crack identification of the oil and gas reservoir is further improved.
According to the method for intelligently identifying natural cracks of the land shale reservoir, which is provided by the embodiment of the invention, the graph data is input into a pre-trained natural crack identification model to obtain a natural crack identification result corresponding to the sampling point, and the method comprises the following steps: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex; and endowing the natural crack identification result of the vertex with the corresponding sampling point to obtain the natural crack identification result corresponding to the sampling point.
When the graph data is input into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point, the graph data is input into the pre-trained natural crack recognition model to obtain the natural crack recognition result of each vertex in the graph data, and the natural crack recognition result of each vertex is endowed to the corresponding sampling point because the vertex corresponds to the sampling point, so that the natural crack recognition result corresponding to each sampling point is obtained.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the natural crack identification result of the vertex is obtained by inputting the graph data into a pre-trained natural crack identification model; and endowing the natural crack identification result of the vertex with a corresponding sampling point to obtain a natural crack identification result corresponding to the sampling point, thereby realizing quick acquisition of the natural crack identification result corresponding to the sampling point.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the natural crack identification model comprises a multi-layer graph convolutional neural network and an output layer; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex, wherein the method comprises the following steps of: inputting the graph data into the natural crack recognition model, and updating the feature vector of the vertex through each layer of graph convolution neural network; the method comprises the steps that each layer of graph convolution neural network is used for carrying out feature vector aggregation on neighbor vertexes of the vertexes, combining the aggregated feature vectors with the feature vectors of the vertexes and updating the feature vectors of the vertexes; and obtaining the natural crack identification result of each vertex according to the output result of the last layer of the graph convolution neural network after the output is processed by the activation function of the output layer.
The structure of the crack identification model comprises a multi-layer graph convolutional neural network and an output layer. And for each layer of graph convolution neural network, performing feature vector aggregation on neighbor vertexes of the vertexes to obtain aggregated feature vectors, combining the aggregated feature vectors with the feature vectors of the vertexes, and updating the feature vectors of the vertexes. The feature vector aggregation can be performed on the neighbor vertexes of the vertexes by using an aggregation function, and the feature vector of the vertexes is updated by a nonlinear activation function.
Therefore, when the graph data is input into the crack recognition model and the natural crack recognition result corresponding to the vertex is output, the graph data is input into the crack recognition model, the characteristic vector of the vertex is updated through each layer of graph convolution neural network, the output of the last layer of graph convolution neural network is processed by the activation function of the output layer and then outputs a probability value, and the natural crack recognition result is obtained according to the output probability value. For example, when training, a crack is set as a label 1, and no crack is set as a label 0, if a crack exists, the output probability value is close to 1; if no crack exists, the output probability value is close to 0.
The feature vectors of the vertices include measured curve values of corresponding depth data, and the feature vector of each vertex in the graph data may be represented as a feature matrix. When the characteristic vector of the vertex is updated by the graph convolution neural network of each layer, the characteristic vector of the vertex is updated according to the neighbor relation between the vertices (namely, the graph structure in the graph data), namely, the information fusion of the characteristic matrix and the graph structure is realized.
Fig. 3 is a second flow chart of the intelligent identification method for natural cracks of the land shale reservoir provided by the embodiment of the invention. As shown in fig. 3, after inputting the graph data into the crack recognition model, performing first information integration on the graph structure and the feature matrix by using a graph convolution layer (graph convolution neural network), namely, performing an embedding process to obtain an embedded matrix, and performing second information integration on the embedded matrix which is activated by an activation function and then used as input data of a next layer to obtain the next embedded matrix; and obtaining an embedded matrix which is finally required to be output after updating for n times, and calculating the embedded matrix through a softmax function of an output layer to obtain a classification result of the vertex, namely, whether the crack identification result corresponding to the vertex is a crack or no crack classification result. The above process is put on the microscopic node scale, and each embedding process, also called an information transfer process, is divided into aggregation of node features and updating of node features.
The neural network may be GRAPHSAGE, or other neural networks may be used.
According to the land shale reservoir natural fracture intelligent identification method provided by the embodiment of the invention, the image data is input into the natural fracture identification model, the characteristic vector of the vertex is updated through each layer of image rolling neural network, wherein each layer of image rolling neural network combines the characteristic vector after aggregation with the characteristic vector of the vertex by carrying out characteristic vector aggregation on the neighbor vertex of the vertex, the characteristic vector of the vertex is updated, and the natural fracture identification result of each vertex is obtained according to the output result of the last layer of image rolling neural network after the activation function of the output layer is processed, so that the accuracy of the natural fracture identification of the oil and gas reservoir is further improved.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the method further comprises the following steps: and acquiring the natural fracture marking data according to the core description data and the imaging logging data.
Natural fracture interpretation can be performed on the rock core description data and the imaging logging data, natural fracture conditions are calibrated, and natural fracture marking data corresponding to partial sampling points are obtained.
According to the embodiment of the invention, the natural fracture marking data is obtained according to the rock core description data and the imaging logging data, so that the accuracy of the marking data is improved.
According to the intelligent identification method for natural cracks of the land shale reservoir, provided by the embodiment of the invention, before the lithology identification result of the sampling point is obtained, the method further comprises the following steps: and carrying out preset data preprocessing on the conventional logging data.
Because the difference of numerical ranges of different conventional logging curves is obvious, before the lithology recognition result of the sampling point is obtained, at least one of data cleaning, unit standardization, depth correction, data alignment and interpolation, trending, outlier processing, standardization and normalization, and data storage and backup pre-treatment are carried out on conventional logging data, so that the logging data quality is ensured.
The same data preprocessing process may be performed for the training samples when training the fracture identification model.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the quality of conventional logging data is improved by carrying out preset data preprocessing on the conventional logging data, so that the accuracy of natural crack identification is improved.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the method further comprises the following steps: and visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
And after the natural crack identification result corresponding to the sampling point is obtained, visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
According to the intelligent identification method for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the visual display of the natural crack identification result corresponding to the sampling point is realized by visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
Fig. 4 is a third flow chart of the intelligent identification method for natural cracks of the land shale reservoir provided by the embodiment of the invention. An embodiment is given below in conjunction with fig. 4.
And selecting a set of concave wind city groups of the Leucon palace basin and the Leucon palace basin as an example to perform natural crack identification. 2 typical wells GR, CAL, SP, AC, CNL, DEN, RXO, rt, ri these 9 conventional logs were selected. And carrying out data standardization processing on the conventional logging curve, and calibrating the processed conventional logging data by using core observation and imaging logging interpretation results. And generating vertexes of the graph structure according to the sampling points, constructing a first type edge based on a depth data sequence corresponding to the sampling points, constructing a second type edge by calculating Euclidean distances among sampling points of the same lithology logging, and generating edges according to the first type edge and the second type edge to obtain a global graph (graph data). And (3) carrying out vertex classification on the constructed global graph by using a crack identification model (obtained based on GRAPHSAGE training), thereby realizing natural crack identification. Wherein, the process of vertex classification of the constructed global graph by using the crack recognition model comprises updating the embedded representation (feature vector) of the vertex and carrying out natural crack recognition based on the normalized result of the softmax function.
The results show that the intelligent identification method for natural cracks of the land shale reservoir provided by the embodiment of the invention can better identify the intervals of the development of the natural cracks, and can accurately identify the natural cracks in the oil and gas reservoir.
The preferred embodiments of the present embodiment may be freely combined on the premise that the logic or structure does not conflict with each other, and the present invention is not limited to this.
The land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention is described below, and the land shale reservoir natural fracture intelligent recognition system described below and the land shale reservoir natural fracture intelligent recognition method described above can be correspondingly referred to each other.
Fig. 5 is a schematic structural diagram of a land shale reservoir natural fracture intelligent recognition system provided by an embodiment of the invention. As shown in fig. 5, the apparatus includes an acquisition module 10, a construction module 20, and an identification module 30, where: the acquisition module 10 is configured to: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; the construction module 20 is for: constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; the identification module 30 is configured to: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
According to the land shale reservoir natural fracture intelligent identification system provided by the embodiment of the invention, conventional logging data are acquired; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values, map data are built according to the depth data of the sampling points, the logging curve values and lithology recognition results, the map data are input into a pre-trained natural fracture recognition model, natural fracture recognition results corresponding to the sampling points are obtained, and accuracy of natural fracture recognition of an oil and gas reservoir is improved.
According to the intelligent identification system for natural cracks of the land shale reservoir provided by the embodiment of the invention, the construction module 20 is specifically configured to, when being configured to construct graph data according to the depth data, the logging curve value and the lithology identification result of the sampling point: taking the sampling points as vertexes of a graph structure, and constructing feature vectors of the vertexes according to the log values of the log curves corresponding to the sampling points; constructing a first type edge between the vertexes with adjacent depths according to the depth data; for vertexes corresponding to the sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the characteristic vectors of the vertexes.
According to the land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention, the sampling points are used as the vertexes of the graph structure, and the feature vectors of the vertexes are constructed according to the logging curve values of the plurality of logging curves corresponding to the sampling points; constructing a first type edge between vertices adjacent in depth according to the depth data; for vertexes corresponding to sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the feature vectors of all vertexes, and improving the adaptability of the graph data for crack identification, so that the accuracy of natural crack identification of the oil and gas reservoir is further improved.
According to the intelligent identification system for natural cracks of the land shale reservoir provided by the embodiment of the invention, the construction module 20 is specifically used for constructing a second type edge between the vertexes through similarity calculation: calculating Euclidean distance between every two vertexes; in response to the euclidean distance being greater than a preset fractional number, not constructing the second type edge between the corresponding two vertices; and constructing the second type edge between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset fraction.
According to the land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention, the Euclidean distance between every two vertexes is calculated, the second type edge is not constructed between the corresponding two vertexes in response to the Euclidean distance being larger than the preset quantile, and the second type edge is constructed between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset quantile, so that the relation network between the nodes is refined, and the accuracy of the natural fracture recognition of the oil and gas reservoir is further improved.
According to the intelligent recognition system for natural cracks of the land shale reservoir provided by the embodiment of the invention, the recognition module 30 is specifically used for inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point when being used for: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex; and endowing the natural crack identification result of the vertex with the corresponding sampling point to obtain the natural crack identification result corresponding to the sampling point.
According to the land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention, the natural fracture recognition result of the vertex is obtained by inputting the graph data into the natural fracture recognition model trained in advance; and endowing the natural crack identification result of the vertex with a corresponding sampling point to obtain a natural crack identification result corresponding to the sampling point, thereby realizing quick acquisition of the natural crack identification result corresponding to the sampling point.
According to the intelligent recognition system for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the natural crack recognition model comprises a multi-layer graph convolutional neural network and an output layer; the recognition module 30 is specifically configured to, when inputting the graph data into a pre-trained natural fracture recognition model to obtain the natural fracture recognition result of the vertex: inputting the graph data into the natural crack recognition model, and updating the feature vector of the vertex through each layer of graph convolution neural network; the method comprises the steps that each layer of graph convolution neural network is used for carrying out feature vector aggregation on neighbor vertexes of the vertexes, combining the aggregated feature vectors with the feature vectors of the vertexes and updating the feature vectors of the vertexes; and obtaining the natural crack identification result of each vertex according to the output result of the last layer of the graph convolution neural network after the output is processed by the activation function of the output layer.
According to the land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention, the image data is input into the natural fracture recognition model, the characteristic vector of the vertex is updated through each layer of image rolling neural network, wherein each layer of image rolling neural network combines the characteristic vector after aggregation with the characteristic vector of the vertex by carrying out characteristic vector aggregation on the neighbor vertex of the vertex, the characteristic vector of the vertex is updated, and the natural fracture recognition result of each vertex is obtained according to the output result of the last layer of image rolling neural network after the activation function of the output layer is processed, so that the accuracy of the natural fracture recognition of the oil and gas reservoir is further improved.
According to the intelligent identification system for the natural cracks of the land shale reservoir, provided by the embodiment of the invention, the device further comprises a marking data acquisition module for: and acquiring the natural fracture marking data according to the core description data and the imaging logging data.
According to the embodiment of the invention, the natural fracture marking data is obtained according to the rock core description data and the imaging logging data, so that the accuracy of the marking data is improved.
According to the intelligent identification system for natural cracks of the land shale reservoir, the device further comprises a preprocessing module, and the preprocessing module is used for preprocessing preset data of conventional logging data before the construction module 20 constructs map data according to the depth data of the sampling points, the logging curve values and lithology identification results.
According to the land shale reservoir natural fracture intelligent identification system provided by the embodiment of the invention, the quality of conventional logging data is improved by carrying out preset data preprocessing on the conventional logging data, so that the accuracy of natural fracture identification is improved.
According to the intelligent identification system for natural cracks of the land shale reservoir, provided by the embodiment of the invention, the device further comprises a display module for: and visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
According to the land shale reservoir natural fracture intelligent recognition system provided by the embodiment of the invention, visual display of the natural fracture recognition result corresponding to the sampling point is realized by visually displaying the natural fracture recognition result corresponding to the sampling point according to the depth data corresponding to the sampling point.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 6, the electronic device may include: processor 410, communication interface (Communications Interface) 420, memory 430, and communication bus 440, wherein processor 410, communication interface 420, and memory 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform a method of intelligent identification of natural fractures of a land shale reservoir, the method comprising: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, where the computer program may be stored on a non-transitory computer readable storage medium, where the computer program, when executed by a processor, can perform a method for intelligently identifying natural cracks of a land-phase shale reservoir provided by the above methods, where the method includes: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
In yet another aspect, an embodiment of the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the method for intelligently identifying natural fractures of a land-phase shale reservoir provided by the above methods, the method comprising: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values; constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results; inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A land shale reservoir natural fracture intelligent identification method is characterized by comprising the following steps:
acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values;
Constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results;
Inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points;
the constructing graph data according to the depth data, the logging curve value and the lithology recognition result of the sampling point comprises the following steps:
Taking the sampling points as vertexes of a graph structure, and constructing feature vectors of the vertexes according to the log values of the log curves corresponding to the sampling points;
Constructing a first type edge between the vertexes with adjacent depths according to the depth data;
for vertexes corresponding to the sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation;
Obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges;
And obtaining graph data according to the graph structure and the characteristic vectors of the vertexes.
2. The intelligent identification method of natural fractures of a land shale reservoir according to claim 1, wherein said constructing a second type of edge between said vertices by similarity calculation comprises:
calculating Euclidean distance between every two vertexes;
In response to the euclidean distance being greater than a preset fractional number, not constructing the second type edge between the corresponding two vertices;
and constructing the second type edge between the corresponding two vertexes in response to the Euclidean distance being smaller than or equal to the preset fraction.
3. The intelligent identification method of natural fractures of a land shale reservoir according to claim 1, wherein the step of inputting the graph data into a pre-trained natural fracture identification model to obtain natural fracture identification results corresponding to the sampling points comprises the steps of:
Inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex;
and endowing the natural crack identification result of the vertex with the corresponding sampling point to obtain the natural crack identification result corresponding to the sampling point.
4. The intelligent identification method of natural fractures of a land shale reservoir according to claim 3, wherein the natural fracture identification model comprises a multi-layer graph convolutional neural network and an output layer;
inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result of the vertex, wherein the method comprises the following steps of:
Inputting the graph data into the natural crack recognition model, and updating the feature vector of the vertex through each layer of graph convolution neural network; the method comprises the steps that each layer of graph convolution neural network is used for carrying out feature vector aggregation on neighbor vertexes of the vertexes, combining the aggregated feature vectors with the feature vectors of the vertexes and updating the feature vectors of the vertexes;
And obtaining the natural crack identification result of each vertex according to the output result of the last layer of the graph convolution neural network after the output is processed by the activation function of the output layer.
5. The method for intelligently identifying natural fractures of a land shale reservoir according to claim 1, further comprising:
and acquiring the natural fracture marking data according to the core description data and the imaging logging data.
6. The method of claim 1, wherein prior to constructing map data from the depth data, the log values, and lithology recognition results for the sampling points, the method further comprises:
And carrying out preset data preprocessing on the conventional logging data.
7. The method for intelligently identifying natural fractures of a land shale reservoir according to claim 1, further comprising:
And visually displaying the natural crack identification result corresponding to the sampling point according to the depth data corresponding to the sampling point.
8. A land shale reservoir natural crack intelligent recognition system is characterized by comprising:
an acquisition module for: acquiring conventional logging data; the conventional logging data comprise depth data of sampling points of a plurality of preset logging curves and corresponding logging curve values;
A construction module for: constructing graph data according to the depth data of the sampling points, the logging curve values and lithology recognition results;
An identification module for: inputting the graph data into a pre-trained natural crack recognition model to obtain a natural crack recognition result corresponding to the sampling point; the natural crack identification model is obtained by training a model based on a graph convolution neural network by using part of natural crack marking data corresponding to the sampling points;
The construction module is used for constructing graph data according to the depth data, the logging curve value and the lithology recognition result of the sampling points, and is specifically used for: taking the sampling points as vertexes of a graph structure, and constructing feature vectors of the vertexes according to the log values of the log curves corresponding to the sampling points; constructing a first type edge between the vertexes with adjacent depths according to the depth data; for vertexes corresponding to the sampling points corresponding to the same lithology recognition result, constructing a second type edge between the vertexes through similarity calculation; obtaining a graph structure according to the constructed vertexes, the first type edges and the second type edges; and obtaining graph data according to the graph structure and the characteristic vectors of the vertexes.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the land shale reservoir natural fracture intelligent identification method of any of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the land phase shale reservoir natural fracture intelligent identification method of any of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410064155.5A CN117574269B (en) | 2024-01-17 | 2024-01-17 | Intelligent identification method and system for natural cracks of land shale reservoir |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410064155.5A CN117574269B (en) | 2024-01-17 | 2024-01-17 | Intelligent identification method and system for natural cracks of land shale reservoir |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117574269A CN117574269A (en) | 2024-02-20 |
CN117574269B true CN117574269B (en) | 2024-04-19 |
Family
ID=89895914
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410064155.5A Active CN117574269B (en) | 2024-01-17 | 2024-01-17 | Intelligent identification method and system for natural cracks of land shale reservoir |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117574269B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106526693A (en) * | 2016-09-30 | 2017-03-22 | 中国石油天然气股份有限公司 | crack identification method and device |
CN116958014A (en) * | 2022-04-13 | 2023-10-27 | 中国石油大学(北京) | Target detection-based unconventional reservoir single well crack intelligent identification method and system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111880234B (en) * | 2020-06-24 | 2022-09-30 | 西南石油大学 | Fine identification method for tight reservoir fractures based on conventional well logging |
-
2024
- 2024-01-17 CN CN202410064155.5A patent/CN117574269B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106526693A (en) * | 2016-09-30 | 2017-03-22 | 中国石油天然气股份有限公司 | crack identification method and device |
CN116958014A (en) * | 2022-04-13 | 2023-10-27 | 中国石油大学(北京) | Target detection-based unconventional reservoir single well crack intelligent identification method and system |
Non-Patent Citations (4)
Title |
---|
Lithology identification using graph neural network in continental shale oil reservoirs: A case study in Mahu Sag, Junggar Basin, Western China;Guoqing Lu 等;Marine and Petroleum Geology;20230206;1-19 * |
准噶尔盆地玛湖凹陷风城组陆相页岩油储层测井裂缝智能识别;陆国青 等;地球科学;20230731;第48卷(第07期);2690-2702 * |
基于样本优化的神经网络方法在储层裂缝识别中的应用;蓝茜茜 等;科学技术与工程;20200728;20(21);8530-8536 * |
富有机质页岩天然裂缝研究进展;曾联波 等;地球科学;20230731;第48卷(第07期);2427-2442 * |
Also Published As
Publication number | Publication date |
---|---|
CN117574269A (en) | 2024-02-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220381948A1 (en) | System for improved reservoir exploration and production | |
Shi et al. | Development of subsurface geological cross-section from limited site-specific boreholes and prior geological knowledge using iterative convolution XGBoost | |
Ren et al. | A novel hybrid method of lithology identification based on k-means++ algorithm and fuzzy decision tree | |
US20170254910A1 (en) | Expert guided knowledge acquisition system for analyzing seismic data | |
Na et al. | Object‐based large‐scale terrain classification combined with segmentation optimization and terrain features: A case study in China | |
CN113902861A (en) | Three-dimensional geological modeling method based on machine learning | |
Zhu et al. | An automatic identification method of imbalanced lithology based on Deep Forest and K-means SMOTE | |
Juliani et al. | Deep learning of terrain morphology and pattern discovery via network-based representational similarity analysis for deep-sea mineral exploration | |
CN117593653B (en) | Geological disaster early warning method based on remote sensing monitoring | |
Abdelaziem et al. | Application of computer vision in machine learning-based diagnosis of water production mechanisms in oil wells | |
CN117473305A (en) | Method and system for predicting reservoir parameters enhanced by neighbor information | |
US20240028828A1 (en) | Machine learning model architecture and user interface to indicate impact of text ngrams | |
CN117574269B (en) | Intelligent identification method and system for natural cracks of land shale reservoir | |
CN117093922A (en) | Improved SVM-based complex fluid identification method for unbalanced sample oil reservoir | |
US20240127039A1 (en) | Geologic learning framework | |
CN117633658B (en) | Rock reservoir lithology identification method and system | |
Suyal et al. | An Agile Review of Machine Learning Technique | |
Qiao et al. | Gas production prediction using AM-BiLSTM model optimized by Whale Optimization Algorithm | |
Kor | Decision-Driven Data Analytics for Well Placement Optimization in Field Development Scenario-Powered by Machine Learning | |
CN115828072B (en) | Reconstruction method of missing section of unconventional logging curve and storage medium | |
Yu et al. | Engineering design of neural network architectures for estimation of inter-well connectivity and production performance | |
CN114580372B (en) | Text processing method, device, equipment and storage medium | |
US20240062119A1 (en) | Field equipment data system | |
US20240062124A1 (en) | Field equipment data system | |
de Magalhães Correia et al. | HOG-CNN based evaluation of cement integrity using 2D dispersion curves from an experimental through tubing logging setup |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |