Knowledge reasoning and feature constraint-based wrinkle structure identification method and equipment

Abstract

The invention discloses a wrinkle structure recognition method and device based on knowledge reasoning and feature constraint. The method comprises the steps that a fault set F and a stratum data set S are acquired; taking each stratum in the S as a stratum node in the stratum knowledge graph, generating a relationship between the stratum nodes, and endowing each stratum node with a unique identifier to obtain the stratum knowledge graph; obtaining an adjacent path unique identifier character string of each node in the stratum knowledge graph, and storing the adjacent path unique identifier character string into a stratum sequence set R; nodes with symmetrical adjacent paths in the R are searched by adopting a regular expression to serve as candidate wrinkle core stratum nodes, and a wrinkle construction candidate region subgraph set SUB is constructed; constructing a sub-graph stratum sequence set for each sub-graph in the SUB, identifying a wrinkle kernel stratum and a wing stratum, and updating a stratum knowledge graph; and based on a preset knowledge inference engine, performing wrinkle structure mode matching on the stratum knowledge graph, and determining the type and range of the wrinkle structure. The method is high in efficiency and precision.

Description

Technical Field

[0001] The present invention relates to geographic information technology, and in particular to a method and device for identifying fold structures based on knowledge reasoning and feature constraints. Background Art

[0002] Folds refer to the phenomenon in which rock formations change, bend, and deform due to tectonic movement. Accurate detection and identification of fold structures has important research and application value in areas such as paleogeographic restoration, geological structure identification, land management, and geological disaster prevention.

[0003] Traditionally, the identification and interpretation of fold structures relies primarily on manual analysis by experts or hard-coded automated interpretation of geological maps or remote sensing images. Manual interpretation is inefficient, costly, and subject to subjective influence by the interpreters. Hard-coded automated interpretation is still in its early stages of exploration, with various algorithms being highly targeted, limited in reusability, and still primarily focused on coarse spatial and temporal scales.

[0004] In 2021, Liu Xianyu proposed an automatic analysis method for fold structures based on knowledge graphs. Through the instance knowledge graph constructed by geological maps and cross-section maps, according to the specific geological knowledge rules and with the help of the spatiotemporal reasoning engine, the automatic analysis and identification of fold structures are realized. However, this method has fewer constraints on the identification of fold cores, and the analysis method for identifying fold wings is relatively complex. The recognition effect and efficiency need to be further improved. In 2024, David et al. proposed an automatic identification method for folds in planar geological maps based on machine learning. Through machine learning, a large number of rays passing through the planar geological map and the midpoints of the intersection lines of the geological map strata are clustered and fitted to find the fold axis. The fold structure is identified based on the fold axis, thereby realizing the automatic identification of the fold structure. However, this method requires the construction of a large number of rays intersecting with the strata, which is labor-intensive and inefficient, and the effect of distinguishing the fold wings needs to be improved. Summary of the Invention

[0005] In view of the problems existing in the prior art, the purpose of the present invention is to provide a method and device for recognizing wrinkle structures based on knowledge reasoning and feature constraints with high efficiency and high recognition accuracy.

[0006] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0007] A fold structure recognition method based on knowledge reasoning and feature constraints includes the following steps:

[0008] (1) Read the fault line vector data and stratigraphic data of the area to be identified into the fault set F and stratigraphic data set S respectively;

[0009] (2) Each stratum in the stratigraphic data set S is treated as a stratigraphic node in the stratigraphic knowledge graph. Based on the fault set F and the spatial data model, the relationship between stratigraphic nodes is generated, and a unique identifier is assigned to each stratigraphic node to obtain the stratigraphic knowledge graph.

[0010] (3) Based on the depth-first traversal algorithm and the preset fold structure wing path constraint conditions, the unique identifier string of the adjacent path of each stratigraphic node in the stratigraphic knowledge graph is obtained and stored in the stratigraphic sequence set R;

[0011] (4) Based on the stratigraphic sequence set R, a regular expression is used to search for stratigraphic nodes with symmetrical adjacent paths as candidate fold core stratigraphic nodes, and a subgraph set SUB of candidate fold structure regions is constructed based on the candidate fold core stratigraphic nodes;

[0012] (5) Traverse the set SUB, and for each subgraph, build a subgraph stratigraphic sequence set based on the depth-first traversal algorithm and the preset fold structure wing path constraints, identify the fold core strata and wing strata, and update them in the stratigraphic knowledge graph;

[0013] (6) Based on the preset knowledge reasoning engine, the fold structure pattern is matched on the stratigraphic knowledge map to determine the type and range of the fold structure and generate the recognition results.

[0014] Furthermore, step (1) specifically includes:

[0015] (1-1) Read fault line vector data to the fault set F = {f i |i=1,2,…,fn}; where f i represents the i-th fault segment, and fn represents the number of faults;

[0016] (1-2) Read the stratum data to the stratum set S = {S j |j=1,2,…,Sn}; where S j represents the jth stratum, and Sn represents the number of strata.

[0017] Furthermore, step (2) specifically includes:

[0018] (2-1) Sequentially read any two strata S in the stratum set S j 、S j+1 ;

[0019] (2-2) According to the mapping relationship table between the preset geological age and the unique identifier, find the stratum S j 、S j+1 The mapping relationship table sets a unique identifier for each stratigraphic geological age in the order of geological ages from earliest to latest;

[0020] (2-4) The formation S j 、S j+1 As two underlying nodes p in the stratum knowledge graph j 、p j+1 , add the unique identifier of the stratum to the attribute of the corresponding stratum node, and judge the stratum S based on the JTS spatial data model j 、S j+1 The spatial relationship between the two layers and whether there is a fault are added to the stratigraphic knowledge map. j 、p j+1 The spatial relationship between

[0021] (2-4) Loop through steps (2-1) to (2-3) until all strata in the stratum set S are traversed and a stratum knowledge graph is obtained.

[0022] Furthermore, steps (2-3) specifically include:

[0023] (2-3-1) The formation S j 、S j+1 As two nodes p in the stratum knowledge graph j 、p j+1 , added to the stratigraphic knowledge graph;

[0024] (2-3-2) Determine the stratigraphic S based on the JTS spatial data model j 、S j+1 Is there any adjacent boundary? If so, then add the stratum S j 、S j+1 The spatial relationship between them is an adjacency relationship, and the p j 、p j+1 The relationship between them is adjacency;

[0025] (2-3-3) Sequentially read each fault in the fault set F, and determine whether the fault is related to the formation S based on the JTS spatial data model. j 、S j+1 Adjacent, if adjacent, then determine S j 、S j+1 The spatial relationship between them is a fault relationship, and the p in the stratum knowledge map is deleted. j 、p j+1 Adjacency relationship between them, update p j 、p j+1 The relationship between them is a fault relationship.

[0026] Furthermore, step (3) specifically includes:

[0027] (3-1) Based on the depth-first algorithm, read any node p in the stratum knowledge graph j ;

[0028] (3-2) Taking node p j Taking as the starting point, search for all its adjacent paths according to the following conditions, and record the unique identifiers of all stratum nodes on the adjacent paths that meet the conditions. Concatenate them into the unique identifier string of the adjacent path in the order of retrieval:

[0029] A. The adjacent path of the wing strata does not exceed the boundary of the geological map sheet;

[0030] B. Adjacent path retrieval of wing strata does not allow crossing faults;

[0031] C. No loop path structure is allowed in the adjacent paths of the wing strata;

[0032] D. The geological ages of nodes in the adjacent paths of the wing strata change monotonically;

[0033] E. The strike of nodes in the adjacent paths of the wing strata is basically consistent;

[0034] (3-3) Loop through steps (3-1) and (3-2) until all nodes in the stratigraphic knowledge graph are retrieved in depth first, and the unique identifier strings of the adjacent paths of all nodes are stored in the stratigraphic sequence set R.

[0035] Furthermore, step (4) specifically includes:

[0036] (4-1) Construct a regular expression for fold structure symmetry detection, detect the stratigraphic sequence set R, find all stratigraphic nodes with symmetrical adjacent paths as candidate fold core stratigraphic nodes, store them in the candidate fold core stratigraphic node set CS, and store the adjacent path stratigraphic nodes of all candidate core stratigraphic nodes in the core stratigraphic adjacent path stratigraphic node set PR;

[0037] (4-2) Each candidate fold core stratigraphic node in CS and the corresponding adjacent path stratigraphic node in PR form a fold structure candidate region subgraph, and store it in the fold structure candidate region subgraph set SUB;

[0038] (4-3) Merge the same adjacent path stratigraphic nodes in the fold structure candidate area subgraph to simplify the knowledge graph.

[0039] Furthermore, step (4-3) specifically includes:

[0040] (4-3-1) Sequentially read any subgraph sub in the fold structure candidate region subgraph set SUB k , merge the layer nodes in the subgraph based on the following rules:

[0041] A. Merge adjacent stratigraphic nodes with the same geological age;

[0042] B. It is not allowed to merge the stratigraphic nodes on both sides of the fault;

[0043] C. The relationship between all nodes is preserved during the merging process;

[0044] (4-3-2) Repeat step (4-3-1) until all subgraphs in SUB are merged.

[0045] Furthermore, step (5) specifically includes:

[0046] (5-1) Sequentially read a subgraph sub in the fold structure candidate region subgraph set SUB k , and extract and construct subgraph sub k All stratigraphic nodes form the stratigraphic node dataset SUBP k ;

[0047] (5-2) Read SUBP k Any stratum node subp s ;

[0048] (5-3) Using the stratum node subp s Taking node as the starting point, retrieve all its adjacent paths and record the unique identifiers of all stratum nodes on the adjacent paths. Connect them into the unique identifier string of the adjacent paths in the order of retrieval.

[0049] (5-4) Loop through steps (5-2)-(5-3) until all SUBPs are traversed. k All the stratum nodes in the subgraph k The unique identifier string of the adjacent path of all stratum nodes in the subgraph sequence set SUBR k ;

[0050] (5-5) Construct a regular expression for symmetry detection of fold structures, and perform the following operations on the subgraph sequence set SUBR k Perform detection to find all stratum nodes with symmetrical adjacent paths as the fold core stratum. Extract the corresponding wing stratum nodes from all adjacent path nodes of each fold core stratum and update them into the stratum knowledge graph.

[0051] (5-6) Loop through steps (5-1) to (5-5) until all subgraphs in SUB are traversed.

[0052] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above method.

[0053] A computer program product comprises a computer program / instruction, wherein the computer program / instruction implements the above method when executed by a processor.

[0054] Compared with the existing technology, the present invention has the following beneficial effects: the present invention realizes automatic identification of fold structures in planar geological maps through a stratigraphic instance knowledge graph constructed from geological maps, based on knowledge reasoning and fold structure feature constraints, thereby improving the reusability, efficiency and accuracy of the identification method. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is the stratum and fault vector data used in the embodiment of the present invention;

[0056] Figure 2 1 is a flow chart of a wrinkle structure identification method based on knowledge reasoning and feature constraints provided by an embodiment of the present invention;

[0057] Figure 3 This is a partial stratigraphic knowledge graph provided by an embodiment of the present invention;

[0058] Figure 4 is a geological map resulting from identifying candidate fold structure areas in the test area according to an embodiment of the present invention;

[0059] Figure 5 It is a candidate fold structure region subgraph of a stratigraphic knowledge graph provided by an embodiment of the present invention;

[0060] Figure 6 It is another stratigraphic knowledge graph candidate fold structure region subgraph provided by an embodiment of the present invention;

[0061] Figure 7 It is the result of merging and simplifying a candidate fold structure region sub-graph according to an embodiment of the present invention;

[0062] Figure 8 It is the result of merging and simplifying another candidate fold structure region sub-graph in the embodiment of the present invention;

[0063] Figure 9 This is a result diagram of distinguishing the core strata and wing strata of the candidate fold structure region sub-graph according to an embodiment of the present invention;

[0064] Figure 10 This is a diagram showing the results of identifying the type of fold structure in the test area according to an embodiment of the present invention;

[0065] Figure 11 This is a result diagram of the fold structure identification range of the test area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0067] The experimental data of this embodiment uses the geological vector data of Ningzhen Mountain Range and its surrounding areas ( Figure 1 ); and, in order to focus on the identification of fold structures, only the vector data of the bedrock strata were retained. The experimental data coordinates used the WGS84 coordinate system, and the projection coordinate system used in the data involving spatial operations was the Universal Mercator projection.

[0068] like Figure 2 As shown, an embodiment of the present invention provides a wrinkle structure recognition method based on knowledge reasoning and feature constraints, comprising the following steps:

[0069] (1) Read the fault line vector data and stratigraphic data of the area to be identified into the fault set F and stratigraphic data set S respectively.

[0070] This step specifically includes:

[0071] (1-1) Read fault line vector data to the fault set F = {f i |i=1,2,…,fn}; where f i represents the i-th fault segment, and fn represents the number of faults; in this embodiment, fn=553.

[0072] (1-2) Read the stratum data to the stratum set S = {S j |j=1,2,…,Sn}; where S j represents the jth stratum, and Sn represents the number of strata. In this embodiment, Sn=548.

[0073] (2) Each stratum in the stratigraphic data set S is regarded as a stratigraphic node in the stratigraphic knowledge graph. According to the fault set F and the spatial data model, the relationship between stratigraphic nodes is generated, and a unique identifier is assigned to each stratigraphic node to obtain the stratigraphic knowledge graph.

[0074] This step specifically includes:

[0075] (2-1) Sequentially read any two strata S in the stratum set S j 、S j+1 ; j value starts from 1;

[0076] (2-2) According to the mapping relationship table between the preset geological age and the unique identifier, find the stratum S j 、S j+1The mapping relationship table sets a unique identifier for each stratum geological age in the order of geological age from ancient to recent. The age can be determined by the unique identifier. For example, the unique identifiers are set in the order of A, B, C... from ancient to recent geological age. The stratum age with the unique identifier A is older than the stratum age with the unique identifier B; as shown in Table 1 below:

[0077] Table 1 Comparison table of unique identifiers for some geological eras

[0078]

[0079] (2-3) The formation S j 、S j+1 As two underlying nodes p in the stratum knowledge graph j 、p j+1 , add the unique identifier of the stratum to the attribute of the corresponding stratum node, and determine the stratum S based on the JTS (Java Topology Suite) spatial data model of the area to be identified j 、S j+1 The spatial relationship between the two layers and whether there is a fault are added to the stratigraphic knowledge map. j 、p j+1 The spatial relationship between them; the attributes of the stratum node also include stratum name, stratum trend, stratum polygon vertex set, etc.

[0080] Steps (2-3) specifically include:

[0081] (2-3-1) The formation S j 、S j+1 As two nodes p in the stratum knowledge graph j 、p j+1 , added to the stratigraphic knowledge graph;

[0082] (2-3-2) Determine the stratigraphic S based on the JTS spatial data model j 、S j+1 Is there any adjacent boundary? If so, then add the stratum S j 、S j+1 The spatial relationship between them is an adjacency relationship, and the p j 、p j+1 The relationship between them is adjacency;

[0083] (2-3-3) Sequentially read each fault in the fault set F, and determine whether the fault is related to the formation S based on the JTS spatial data model. j 、S j+1 Adjacent, if adjacent, then determine S j 、S j+1The spatial relationship between them is a fault relationship, and the p in the stratum knowledge map is deleted. j 、p j+1 Adjacency relationship between them, update p j 、p j+1 The relationship between them is a fault relationship.

[0084] (2-4) Loop through steps (2-1) to (2-3) until all strata in the stratum set S are traversed and a stratum knowledge graph is obtained.

[0085] All stratigraphic nodes in the stratigraphic knowledge graph form a stratigraphic node dataset P = {p j |j=1,2,…,Sn}, in this embodiment, some stratigraphic node data are shown in Table 2 below, and some stratigraphic knowledge graphs are shown in Table 2 below. Figure 3 shown.

[0086] Table 2 Example of stratigraphic node data point set

[0087]

[0088] (3) Based on the depth-first traversal algorithm and the preset fold structure wing path constraints, the unique identifier string of the adjacent path of each stratigraphic node in the stratigraphic knowledge graph is obtained and stored in the stratigraphic sequence set R.

[0089] This step specifically includes:

[0090] (3-1) Based on the depth-first algorithm, read any node p in the stratum knowledge graph j ;

[0091] (3-2) Taking node p j Taking as the starting point, search for all its adjacent paths according to the following conditions, and record the unique identifiers of all stratum nodes on the adjacent paths that meet the conditions. Concatenate them into the unique identifier string of the adjacent path in the order of retrieval:

[0092] A. The adjacent path of the wing strata does not exceed the boundary of the geological map sheet;

[0093] B. Adjacent path retrieval of wing strata does not allow crossing faults;

[0094] C. No loop path structure is allowed in the adjacent paths of the wing strata;

[0095] D. The geological age of the nodes in the adjacent paths of the wing strata changes monotonically (that is, the age change of the strata must strictly follow the change from old to young or from young to old)

[0096] E. The strike of nodes in the adjacent paths of the wing strata is basically consistent;

[0097] Among them, the wing strata refer to the rock layers on both sides of the fold bend, that is, the rock layer area between the fold core (central part) and the turning end (bending apex), located on both sides of the fold core and extending toward the turning end.

[0098] The adjacent path refers to the search for nodes along the adjacent edges starting from the current node. Once a path that does not meet the above conditions appears, the current adjacent path is abandoned and the next adjacent path is searched. Figure 6 As shown, assuming that the point indicated by Q is the starting point, QX, QZ, and QST are all adjacent paths starting from Q. However, assuming that QX does not meet any of the above conditions, it is not an eligible adjacent path and is abandoned and deleted.

[0099] (3-3) Loop through steps (3-1) and (3-2) until all nodes in the stratigraphic knowledge graph are retrieved in depth first, and the unique identifier strings of the adjacent paths of all nodes are stored in the stratigraphic sequence set R.

[0100] Stratigraphic sequence set R = {r j ||j=1,2,…,Sn}; where r j The stratigraphic sequence representing the j-th stratigraphic node, i.e., the unique identifier string of the adjacency path; in this embodiment, a partial stratigraphic sequence set is shown in Table 3 below.

[0101] Table 3 Example stratigraphic sequence collection

[0102]

[0103] (4) Based on the stratigraphic sequence set R, regular expressions are used to search for stratigraphic nodes with symmetrical adjacent paths as candidate fold core stratigraphic nodes, and a subgraph set SUB of candidate fold structure regions is constructed based on the candidate fold core stratigraphic nodes.

[0104] This step specifically includes:

[0105] (4-1) Construct a regular expression for fold structure symmetry detection, detect the stratigraphic sequence set R, find all stratigraphic nodes with symmetrical adjacent paths as candidate fold core stratigraphic nodes, and store them in the candidate fold core stratigraphic node set CS = {cs u |u=1,2,…,csn}; where cs u represents the uth candidate fold core stratum node, csn represents the number of candidate fold core stratum nodes, and the adjacent path stratum nodes of all candidate core stratum nodes are stored in the core stratum adjacent path stratum node set PR = {pr v |v=1,2,…,prn}, where pr vrepresents the adjacent path stratum node of the vth candidate fold core stratum node, and prn represents the number of adjacent path stratum nodes of the candidate fold core stratum node. The regular expression is as follows:

[0106] (\w)(\w)(\w)*? \1\2#(1)

[0107] (\w)(\w)\1.#(2)

[0108] The above regular expression is used to detect whether there is a symmetrical string in the stratigraphic sequence set R, and w represents a character in the range [0-9, az, AZ];

[0109] (4-2) Each candidate fold core stratum node cs in CS u and the corresponding adjacent path stratum node pr in PR v , forming a fold structure candidate region subgraph sub k , store the candidate fold structure region subgraph set SUB = {sub k |k=1,2,…,subn}; where sub k represents the kth fold structure candidate region sub-map, subn represents the number of fold structure candidate region sub-maps; in this embodiment, the geological map of the fold structure candidate region is as follows Figure 4 As shown in the figure, the candidate region subgraph of the stratigraphic knowledge graph fold structure is as follows Figure 5 、 Figure 6 As shown;

[0110] (4-3) Merge the same adjacent path stratigraphic nodes in the fold structure candidate area subgraph to simplify the knowledge graph.

[0111] Wherein, step (4-3) specifically includes:

[0112] (4-3-1) Sequentially read any subgraph subk in the fold structure candidate region subgraph set SUB, and merge the stratigraphic nodes in the subgraph based on the following rules. Merging is to merge two nodes into one node, and the merged node inherits all the relationships of the original two nodes;

[0113] A. Merge adjacent stratigraphic nodes with the same geological age;

[0114] B. It is not allowed to merge the stratigraphic nodes on both sides of the fault;

[0115] C. The relationship between all nodes is preserved during the merging process;

[0116] (4-3-2) Repeat step (4-3-1) until all subgraphs in SUB are merged. The result after subgraph simplification in this embodiment is as follows Figure 7 、 Figure 8 shown.

[0117] (5) Traverse the set SUB and construct a subgraph stratigraphic sequence set for each subgraph based on the depth-first traversal algorithm and the preset fold structure wing path constraints, identify the fold core strata and wing strata, and update them in the stratigraphic knowledge graph.

[0118] This step specifically includes:

[0119] (5-1) Sequentially read a subgraph sub in the fold structure candidate region subgraph set SUB k , and extract and construct subgraph sub k All stratigraphic nodes form the stratigraphic node dataset SUBP k ;

[0120] (5-2) Read SUBP k Any stratum node subp s ;

[0121] (5-3) Using the stratum node subp s Taking as the starting point, retrieve all its adjacent paths and record the unique identifiers of all stratum nodes on the adjacent paths. Connect them in the order of retrieval to form the unique identifier string of the adjacent paths. The conditions that need to be met for the adjacent path retrieval are the same as those in step (3-2) and will not be repeated here.

[0122] (5-4) Loop through steps (5-2)-(5-3) until all SUBPs are traversed. k All the stratum nodes in the subgraph k The unique identifier string of the adjacent path of all stratum nodes in the subgraph sequence set SUBR k ;

[0123] (5-5) Construct a regular expression for symmetry detection of fold structures, and perform the following operations on the subgraph sequence set SUBR k Perform a test to find all stratum nodes with symmetrical adjacent paths as the fold core stratum. Extract the corresponding wing stratum nodes from all adjacent path nodes of each fold core stratum and update them into the stratum knowledge graph. The regular expression construction is the same as step (4-1).

[0124] (5-6) Execute steps (5-1) to (5-5) repeatedly until all subgraphs in SUB are traversed. In this embodiment, the updated stratum knowledge graph fold structure subgraph is as follows Figure 9 shown.

[0125] (6) Based on the preset knowledge reasoning engine, the fold structure pattern is matched on the stratigraphic knowledge map to determine the type and range of the fold structure and generate the recognition results.

[0126] The knowledge inference engine is a software system loaded from the outside. It takes the structured knowledge in the knowledge graph as input and automatically derives implicit new knowledge or conclusions through preset inference rules or algorithm models. In this embodiment, the knowledge inference engine has preset fold structure pattern recognition rules. The specific rules are existing technologies and will not be described in detail. In this embodiment, the fold structure recognition results are as follows: Figure 10 、 Figure 11 shown.

[0127] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above method.

[0128] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which implements the above method when executed by a processor.

[0129] It should be understood that the above embodiments and descriptions only describe the principles, main features and advantages of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, and these changes and improvements all fall within the scope of protection of the present invention.

Claims

1. A wrinkle structure recognition method based on knowledge reasoning and feature constraints, characterized by: The steps include: (1) Read the fault line vector data and stratigraphic data of the area to be identified into the fault set F and stratigraphic data set S respectively; (2) Each stratigraphic layer in the stratigraphic data set S is treated as a stratigraphic node in the stratigraphic knowledge graph. Based on the fault set F and the spatial data model, the relationship between stratigraphic nodes is generated, and each stratigraphic node is assigned a unique identifier that identifies the geological age of the stratigraphic layer, thereby obtaining a stratigraphic knowledge graph. (3) Based on the depth-first traversal algorithm and the preset fold structure wing path constraint conditions, the unique identifier string of the adjacent path of each stratigraphic node in the stratigraphic knowledge graph is obtained and stored in the stratigraphic sequence set R; (4) Using regular expressions to search for stratigraphic nodes with symmetrical adjacent paths in the stratigraphic sequence set R as candidate fold core stratigraphic nodes, and constructing a fold structure candidate region subgraph set SUB based on the candidate fold core stratigraphic nodes; (5) Traverse the set SUB, and for each subgraph, build a subgraph stratigraphic sequence set based on the depth-first traversal algorithm and the preset fold structure wing path constraints, identify the fold core strata and wing strata, and update them in the stratigraphic knowledge graph; (6) Based on the preset knowledge reasoning engine, the fold structure pattern is matched on the stratigraphic knowledge map to determine the type and range of the fold structure and generate the recognition results.

2. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 1 is characterized in that: Step (1) specifically includes: (1-1) Read fault line vector data to the fault set F = {f i |i=1,2,…,fn}; where f i represents the i-th fault segment, and fn represents the number of faults; (1-2) Read the stratum data to the stratum set S = {S j |j=1,2,…,Sn}; where S j represents the jth stratum, and Sn represents the number of strata.

3. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 1 is characterized in that: Step (2) specifically includes: (2-1) Sequentially read any two strata S in the stratum set S j 、S j+1 ; (2-2) According to the mapping relationship table between the preset geological age and the unique identifier, find the stratum S j 、S j+1 The mapping relationship table sets a unique identifier for each stratigraphic geological age in the order of geological ages from earliest to latest; (2-3) The formation S j 、S j+1 As two underlying nodes p in the stratum knowledge graph j 、p j+1 , add the unique identifier of the stratum to the attribute of the corresponding stratum node, and judge the stratum S based on the JTS spatial data model j 、S j+1 The spatial relationship between the two layers and whether there is a fault are added to the stratigraphic knowledge map. j 、p j+1 The spatial relationship between (2-4) Loop through steps (2-1) to (2-3) until all strata in the stratum set S are traversed and a stratum knowledge graph is obtained.

4. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 3 is characterized in that: Steps (2-3) specifically include: (2-3-1) The formation S j 、S j+1 As two nodes p in the stratum knowledge graph j 、p j+1 , added to the stratigraphic knowledge graph; (2-3-2) Determine the stratigraphic S based on the JTS spatial data model j 、S j+1 Is there any adjacent boundary? If so, then add the stratum S j 、S j+1 The spatial relationship between them is an adjacency relationship, and the p j 、p j+1 The relationship between them is adjacency; (2-3-3) Sequentially read each fault in the fault set F, and determine whether the fault is related to the formation S based on the JTS spatial data model. j 、S j+1 Adjacent, if adjacent, then determine S j 、S j+1 The spatial relationship between them is a fault relationship, and the p in the stratum knowledge map is deleted. j 、p j+1 Adjacency relationship between them, update p j 、p j+1 The relationship between them is a fault relationship.

5. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 1 is characterized in that: Step (3) specifically includes: (3-1) Based on the depth-first algorithm, read any node p in the stratum knowledge graph j ; (3-2) Taking node p j Taking as the starting point, search for all its adjacent paths according to the following conditions, and record the unique identifiers of all stratum nodes on the adjacent paths that meet the conditions. Concatenate them into the unique identifier string of the adjacent path in the order of retrieval: A. The adjacent path of the wing strata does not exceed the boundary of the geological map sheet; B. Adjacent path retrieval of wing strata does not allow crossing faults; C. No loop path structure is allowed in the adjacent paths of the wing strata; D. The geological ages of nodes in the adjacent paths of the wing strata change monotonically; E. The strike of nodes in the adjacent paths of the wing strata is basically consistent; (3-3) Loop through steps (3-1) and (3-2) until all nodes in the stratigraphic knowledge graph are retrieved in depth first, and the unique identifier strings of the adjacent paths of all nodes are stored in the stratigraphic sequence set R.

6. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 1 is characterized in that: Step (4) specifically includes: (4-1) Construct a regular expression for fold structure symmetry detection, detect the stratigraphic sequence set R, find all stratigraphic nodes with symmetrical adjacent paths as candidate fold core stratigraphic nodes, store them in the candidate fold core stratigraphic node set CS, and store the adjacent path stratigraphic nodes of all candidate core stratigraphic nodes in the core stratigraphic adjacent path stratigraphic node set PR; (4-2) Each candidate fold core stratigraphic node in CS and the corresponding adjacent path stratigraphic node in PR form a fold structure candidate region subgraph, and store it in the fold structure candidate region subgraph set SUB; (4-3) Merge the same adjacent path stratigraphic nodes in the fold structure candidate area subgraph to simplify the knowledge graph.

7. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 6 is characterized in that: Step (4-3) specifically includes: (4-3-1) Sequentially read any subgraph sub in the fold structure candidate region subgraph set SUB k , merge the layer nodes in the subgraph based on the following rules: A. Merge adjacent stratigraphic nodes with the same geological age; B. It is not allowed to merge the stratigraphic nodes on both sides of the fault; C. The relationship between all nodes is preserved during the merging process; (4-3-2) Repeat step (4-3-1) until all subgraphs in SUB are merged.

8. The wrinkle structure identification method based on knowledge reasoning and feature constraints according to claim 6 is characterized in that: Step (5) specifically includes: (5-1) Sequentially read a subgraph sub in the fold structure candidate region subgraph set SUB k , and extract and construct subgraph sub k All stratigraphic nodes form the stratigraphic node dataset SUBP k ; (5-2) Read SUBP k Any stratum node subp s ; (5-3) Using the stratum node subp s Taking node as the starting point, retrieve all its adjacent paths and record the unique identifiers of all stratum nodes on the adjacent paths. Connect them into the unique identifier string of the adjacent paths in the order of retrieval. (5-4) Loop through steps (5-2)-(5-3) until all SUBPs are traversed. k All the stratum nodes in the subgraph k The unique identifier string of the adjacent path of all stratum nodes in the subgraph sequence set SUBR k ; (5-5) Construct a regular expression for symmetry detection of fold structures, and perform the following operations on the subgraph sequence set SUBR k Perform detection to find all stratum nodes with symmetrical adjacent paths as the fold core stratum. Extract the corresponding wing stratum nodes from all adjacent path nodes of each fold core stratum and update them into the stratum knowledge graph. (5-6) Loop through steps (5-1) to (5-5) until all subgraphs in SUB are traversed.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the computer program to implement the method according to any one of claims 1 to 8.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images