Methods, systems, equipment and media for multi-source heterogeneous geological data integration and feature extraction

Abstract

This application discloses a method, system, device, and medium for integrating and extracting features from multi-source heterogeneous geological data, relating to the field of geological data processing technology. The method includes: acquiring raw geological survey data of multiple geological objects and constructing a standardized geological database based on the raw geological survey data; extracting the geological objects to be represented from the standardized geological database and determining the multi-dimensional data features of the geological objects to be represented; numerically encoding and normalizing the qualitative descriptive data in the multi-dimensional data features, normalizing the quantitative descriptive data, and determining the target feature data from the normalized quantitative and qualitative descriptive data in the multi-dimensional data features; generating a comprehensive feature vector based on the target feature data, thus completing the integration and feature extraction of the geological data. This application achieves deep fusion, high-quality standardization, and comprehensive, computable quantitative characterization of multi-source heterogeneous geological data.

Description

Technical Field

[0001] This application relates to the field of geological data processing technology, and in particular to a method, system, equipment and medium for integrating and extracting features from multi-source heterogeneous geological data. Background Technology

[0002] Existing technologies still face significant challenges in integrating and applying multi-source heterogeneous survey data, mainly in the following three aspects: 1) Geological survey data comes from diverse sources and has a complex structure, covering various types such as borehole, well logging, remote sensing, and geochemistry. Traditional data integration methods usually rely on manual intervention and rule mapping, which makes it difficult to systematically solve problems such as inconsistencies in coordinate systems, semantic conflicts, and format redundancy (such as the difference between "sandstone" in engineering geology and sedimentology), resulting in low data fusion efficiency and difficulty in ensuring consistency.

[0003] 2) Existing feature extraction techniques are mostly limited to single data sources or low-dimensional feature analysis (such as analyzing only mineral composition or lithology), lacking the ability to comprehensively represent the spatial distribution, physical properties, and chemical composition of geological structures in a high-dimensional way. This results in insufficient abstraction of geological features, making it difficult to support the construction and analysis of complex geological models. Consequently, the feature vector dimensions are fragmented, making it difficult to support the comprehensive identification of complex geological bodies.

[0004] To address these issues, the geological survey industry currently relies primarily on manually defined ETL rules for data cleaning. However, when faced with massive amounts of heterogeneous data, the maintenance costs increase exponentially. Feature extraction often employs linear dimensionality reduction methods such as principal component analysis, which struggle to capture the nonlinear relationships between geological phenomena. Knowledge processing largely remains at the keyword matching level, and a theme evolution model consistent with the characteristics of the geological discipline has not yet been established. These technical shortcomings directly restrict the depth and breadth of application of geological survey big data in areas such as resource exploration and disaster early warning.

[0005] Therefore, it is necessary to provide a method for integrating and extracting features from multi-source heterogeneous geological data to solve the above problems. Summary of the Invention

[0006] The purpose of this application is to provide a method, system, device and medium for integrating and extracting features from multi-source heterogeneous geological data, so as to achieve deep fusion, high-quality standardization and comprehensive, computable quantitative characterization of multi-source heterogeneous geological data.

[0007] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for integrating and extracting features from multi-source heterogeneous geological data, the method comprising: The process involves acquiring raw geological survey data for multiple geological objects and constructing a standardized geological database based on this data. The raw geological survey data includes geological survey data from different sources and in different formats. The standardized geological database stores standardized geological data for multiple geological objects. Geological objects to be characterized are extracted from standardized geological databases, and multi-dimensional data features of the geological objects to be characterized are determined based on the standardized geological databases. The multi-dimensional data features include at least geometric spatial features, physical property features, and chemical composition features. Each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data. The qualitative descriptive data in the multi-dimensional data features are numerically encoded and normalized to obtain the normalized qualitative descriptive data of the geological object to be represented. The quantitative descriptive data in the multidimensional data features are normalized, and the normalized quantitative descriptive data and the normalized qualitative descriptive data in the multidimensional data features are determined as the target feature data of the geological object to be characterized. Based on the target feature data of the geological object to be represented, a comprehensive feature vector of the geological object to be represented is generated, thus completing the integration and feature extraction of geological data.

[0008] In one embodiment, the raw geological survey data includes at least two of the following: borehole data, remote sensing images, or geological survey data; Based on the aforementioned original geological survey data, a standardized geological database is constructed, specifically including: The original geological survey data is parsed and converted to obtain geological data after format conversion, with a set intermediate data type. The Bursa-Wolf projection transformation formula is used to uniformly transform the coordinate system of the geological data after the format conversion to the preset standard coordinate system, so as to obtain the geological data in the standard coordinate system. Based on a geological ontology library, semantic mapping and standardization are performed on geological data in the standard coordinate system to obtain standardized geological data. Standardized geological data is packaged and stored to obtain a standardized geological database.

[0009] In one implementation, based on a geological ontology library, geological data in a standard coordinate system is semantically mapped and standardized to obtain standardized geological data, specifically including: By mapping the inconsistent synonyms in geological data under the standard coordinate system to the unified standard terms in the geological ontology library, we obtain the semantically mapped geological data. The associated attribute values ​​in the semantically mapped geological data are converted to units and normalized to obtain standardized geological data.

[0010] In one embodiment, the geometric spatial features are geometric feature sub-vectors generated based on geometric spatial data; the geometric feature sub-vectors include at least one of: centroid coordinates, area, perimeter, shape index, orientation angle, and dip angle; The physical property features are physical feature sub-vectors generated based on physical property data; the physical property data includes measured density data, porosity detection data, resistivity geophysical data, and radioactivity intensity detection data; The chemical composition features are chemical composition sub-vectors generated based on the chemical composition data; the chemical composition data are the main element content parameters.

[0011] In one embodiment, the geological object to be characterized is extracted from a standardized geological database, and multi-dimensional data features of the geological object to be characterized are determined based on the standardized geological database, specifically including: Based on spatial scope and attribute rules, GIS spatial analysis methods are used to identify and extract geological objects to be represented from standardized geological databases; Extract geometric spatial data, physical property data, and chemical composition data of the geological objects to be characterized from standardized geological databases; Based on geometric spatial data, geometric feature sub-vectors of the geological objects to be represented are generated; the geometric spatial data includes spatial coordinate data and attitude record data. Based on physical property data, physical feature sub-vectors of the geological objects to be characterized are generated; Based on chemical composition data, chemical feature sub-vectors of the geological objects to be characterized are generated; The geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors of the geological object to be characterized are determined as the multi-dimensional data features of the geological object to be characterized.

[0012] In one embodiment, based on the target feature data of the geological object to be characterized, a comprehensive feature vector of the geological object to be characterized is determined, specifically including: The geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors in the target feature data of the geological object to be represented are concatenated and weighted to obtain the comprehensive feature vector of the geological object to be represented. Among them, the weighted fusion adopts the formula accomplish; in, This represents the comprehensive feature vector of the geological object to be characterized. , and These are the weighting coefficients; The geometric spatial features in the target feature data of the geological object to be characterized; The physical properties of the target feature data of the geological object to be characterized; Chemical composition features in the target feature data of the geological object to be characterized.

[0013] In one embodiment, the multi-source heterogeneous geological data integration and feature extraction method further includes: Natural language processing is used to extract quantifiable topic tags from unstructured geological survey documents; By associating topic tags with the comprehensive feature vectors of the geological objects to be represented, a semantic knowledge model of the geological objects to be represented is constructed.

[0014] Secondly, this application provides a multi-source heterogeneous geological data integration and feature extraction system. This system is used to implement the aforementioned multi-source heterogeneous geological data integration and feature extraction method. The multi-source heterogeneous geological data integration and feature extraction system includes: A standardized geological database construction unit is used to acquire raw geological survey data of multiple geological objects and construct a standardized geological database based on the raw geological survey data; the raw geological survey data includes geological survey data from different sources and in different formats; the standardized geological database stores standardized geological data of multiple geological objects; A multi-dimensional data feature determination unit is used to extract the geological object to be characterized from a standardized geological database and determine the multi-dimensional data features of the geological object to be characterized based on the standardized geological database. The multi-dimensional data features include at least geometric spatial features, physical property features, and chemical composition features. Each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data. The normalized qualitative description data determination unit is used to perform numerical encoding and normalization processing on the qualitative description data in the multi-dimensional data features to obtain the normalized qualitative description data of the geological object to be represented. The target feature data determination unit is used to normalize the quantitative descriptive data in the multi-dimensional data features, and to determine the normalized quantitative descriptive data and normalized qualitative descriptive data in the multi-dimensional data features as the target feature data of the geological object to be characterized. The comprehensive feature vector determination unit is used to generate a comprehensive feature vector of the geological object to be represented based on the target feature data of the geological object to be represented, thereby completing the integration and feature extraction of geological data.

[0015] Thirdly, this application provides a computer device, including: 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-described method for integrating and extracting features from multi-source heterogeneous geological data.

[0016] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for integrating and extracting features from multi-source heterogeneous geological data.

[0017] According to the specific embodiments provided in this application, this application has the following technical effects: This application discloses a method, system, equipment, and medium for integrating and extracting features from multi-source heterogeneous geological data. First, by constructing a standardized geological database, the structure and format of multi-source heterogeneous data are unified, solving the primary challenge of heterogeneous integration. Second, multi-dimensional data features are determined, and a representation system for geological objects is defined from a professional perspective, ensuring the comprehensiveness and structure of feature extraction. Third, the qualitative data is numerically encoded and all data is normalized, completing a key leap from heterogeneous information to homogeneous numerical values, achieving deep fusion and dimensional unification at the feature level. Finally, a comprehensive feature vector is generated as the output feature, integrating the processed multi-dimensional features into a structured and mathematical vector object, providing a standardized interface for intelligent analysis, marking the final realization of quantitative representation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of a method for integrating and extracting features from multi-source heterogeneous geological data according to an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] In one exemplary embodiment, such as Figure 1As shown, a method for integrating and extracting features from multi-source heterogeneous geological data is provided, including the following steps: Wherein: Step S1: Obtain raw geological survey data for multiple geological objects, and construct a standardized geological database based on the raw geological survey data; the raw geological survey data includes geological survey data from different sources and in different formats; the standardized geological database stores standardized geological data for multiple geological objects.

[0023] As an optional implementation, if the original geological survey data includes at least two of borehole data, remote sensing images, or geological survey data, then in step S1, a standardized geological database is constructed based on the original geological survey data, specifically including: Step S11: Perform data parsing and format conversion on the original geological survey data to obtain geological data after format conversion with the set intermediate data type.

[0024] Step S12: Using the Bursa-Wolf projection transformation formula, the coordinate system of the geological data after the format conversion is uniformly transformed to the preset standard coordinate system to obtain geological data in the standard coordinate system.

[0025] Step S13: Based on the geological ontology library, perform semantic mapping and standardization on the geological data in the standard coordinate system to obtain standardized geological data.

[0026] As an optional implementation, step S13 specifically includes: Step S131: Map the inconsistent synonyms in the geological data under the standard coordinate system to the unified standard terms in the geological ontology library to obtain the semantically mapped geological data. Step S132: Perform unit conversion and normalization on the associated attribute values ​​in the semantically mapped geological data to obtain standardized geological data.

[0027] Step S14: The standardized geological data is packaged and stored to obtain a standardized geological database.

[0028] Specifically, 1) The computer system accesses raw geological survey data from different sources, such as borehole data, remote sensing images, and geological survey data, through built-in multiple data interface adapters; 2) The system calls the corresponding format parsing engine to uniformly parse raw data in different formats, such as database files, spreadsheets, text files, and GIS files, into an intermediate data model defined within the system; 3) The computer system automatically identifies or matches the coordinate system information of each dataset (such as CGCS2000, WGS84, etc.); 4) The system calls the coordinate transformation engine to unify the coordinate system of all data based on the Bursa-Wolf projection transformation formula (Equations (1) to (3)). The purpose is to solve the fusion deviation caused by inconsistent spatial locations and transform the coordinate system of all datasets to the preset standard coordinate system; 5) The computer system loads a predefined geological domain ontology library containing standard terms; The system maps synonymous terms from different data sources with different expressions (such as "granite", "granite", "γ") to the unified standard terms in the geological domain ontology library through the semantic mapping module, and performs unit conversion and normalization processing on the associated attribute values; 6) Standardized data output. The computer system encapsulates and stores standardized geological data in a standardized geological database, providing a high-quality, unified data foundation for subsequent high-dimensional feature extraction and analysis. The coordinate transformation formula is as follows: (1) (2) (3) in, Source data coordinates (known quantities); a~ l All are preset conversion parameters (known quantities). The coordinates are in the target coordinate system (unknown quantities).

[0029] Step S2: Extract the geological object to be characterized from the standardized geological database, and determine the multi-dimensional data features of the geological object to be characterized based on the standardized geological database; the multi-dimensional data features include at least geometric spatial features, physical property features and chemical composition features; wherein, each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data.

[0030] As an optional implementation, in step S2, the geometric space features are geometric feature sub-vectors generated based on geometric space data; the geometric feature sub-vectors include at least one of the following: centroid coordinates, area, perimeter, shape index, orientation angle and dip angle.

[0031] The physical property features are physical feature sub-vectors generated based on physical property data; the physical property data includes measured density data, porosity detection data, resistivity geophysical data, and radioactivity intensity detection data.

[0032] The chemical composition features are chemical composition sub-vectors generated based on the chemical composition data; the chemical composition data are the main element content parameters.

[0033] As an optional implementation method, step S2 specifically includes: Step S21: Based on spatial scope and attribute rules, GIS spatial analysis methods are used to identify and extract the geological objects to be characterized from the standardized geological database. The geological objects to be characterized can be a rock mass, a fault, or a mineral deposit, etc.

[0034] Step S22: Extract the geometric spatial data, physical property data, and chemical composition data of the geological object to be characterized from the standardized geological database.

[0035] Step S23: Based on geometric spatial data, generate geometric feature sub-vectors of the geological object to be represented; the geometric spatial data includes spatial coordinate data (such as vertex coordinates and boundary coordinates) and occurrence record data (derived from remote sensing images and geological survey data).

[0036] Specifically, the centroid coordinates are obtained by averaging the coordinates of all spatial vertices of the geological object; the area is calculated based on a GIS spatial analysis algorithm, determining the area of ​​the two-dimensional / three-dimensional region enclosed by the object's boundaries; the perimeter is calculated by statistically analyzing the total length of the object's boundaries; and the shape index is calculated using the formula "4π × area / perimeter". 2 "Calculation; strike and dip angles are extracted from the occurrence records of geological survey data, or derived from the slope and azimuth of rock strata / rock mass boundaries through remote sensing image analysis. Finally, the geometric feature sub-vectors are constructed from the aforementioned centroid coordinates, area, perimeter, shape index, strike angle, and dip angle." .

[0037] Step S24: Based on the physical property data, generate physical feature sub-vectors of the geological object to be characterized.

[0038] Specifically, the measured density data, porosity detection data, resistivity geophysical data, and radioactivity intensity detection data are normalized to generate physical feature sub-vectors. .

[0039] Step S25: Based on the chemical composition data, generate chemical feature sub-vectors of the geological object to be characterized.

[0040] Specifically, the contents of major elements, such as the contents of major oxides (SiO2, Al2O3, etc.), trace element contents, and isotope ratios, are obtained based on spectral analysis or laboratory reports. These chemical properties are then compressed into low-dimensional vectors to obtain chemical feature sub-vectors. .

[0041] Step S26: Determine the geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors of the geological object to be characterized as the multi-dimensional data features of the geological object to be characterized.

[0042] In addition, other types of data features can be extracted to form multi-dimensional data features.

[0043] Step S3 involves numerically encoding and normalizing the qualitative descriptive data in the multi-dimensional data features to obtain normalized qualitative descriptive data of the geological object to be represented.

[0044] Specifically, the computer system uses a feature encoder to numerically encode the qualitative descriptive data in all the aforementioned feature sub-vectors and normalizes all numerical features to eliminate the influence of dimensions. The qualitative descriptive data requiring numerical encoding mainly includes: Qualitative descriptive data in geometric feature subvectors include occurrence descriptions (e.g., "steeply dipping") and morphological descriptions (e.g., "lenticular"). Qualitative descriptive data in physical feature subvectors include lithological descriptions (e.g., "granite"), color (e.g., "dark gray"), and weathering degree (e.g., "strongly weathered"). Qualitative descriptive data in chemical feature subvectors include compositional descriptions (e.g., "high silica") and alteration type (e.g., "sericitization").

[0045] Step S4: Normalize the quantitative descriptive data in the multi-dimensional data features, and determine the normalized quantitative descriptive data and normalized qualitative descriptive data in the multi-dimensional data features as the target feature data of the geological object to be characterized.

[0046] Step S5: Based on the target feature data of the geological object to be represented, generate a comprehensive feature vector of the geological object to be represented, and complete the integration and feature extraction of geological data.

[0047] As an optional implementation, step S5 specifically includes: The geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors in the target feature data of the geological object to be represented are concatenated and weighted to obtain the comprehensive feature vector of the geological object to be represented.

[0048] Among them, the weighted fusion adopts the formula accomplish.

[0049] in, This represents the comprehensive feature vector of the geological object to be characterized. , and These are the weighting coefficients; The geometric spatial features in the target feature data of the geological object to be characterized; is the physical property feature in the target feature data of the geological object to be characterized; is the chemical composition feature in the target feature data of the geological object to be characterized.

[0050] Specifically, the computer system splices and fuses the encoded and normalized geometric feature sub-vectors, physical feature sub-vectors, and chemical feature sub-vectors to form a comprehensive high-dimensional feature vector representing the geological object, that is, the comprehensive feature vector, realizing the cross-dimensional upgrade of geological attributes from qualitative description to quantitative characterization. The computer system associates the high-dimensional feature vector with the unique identifier of the geological object and stores it in the feature vector library for subsequent analysis, comparison, and machine learning model use.

[0051] As an optional implementation manner, the multi-source heterogeneous geological data integration and feature extraction method further includes: Step S6, perform natural language processing on the unstructured geological survey documents to extract quantifiable topic tags.

[0052] Step S7, associate the topic tags with the comprehensive feature vector of the geological object to be characterized, and construct a semantic knowledge model of the geological object to be characterized.

[0053] Specifically, the computer system executes an automated process, analyzes the geological knowledge documents, and generates quantifiable topic tags based on natural language processing technology. The specific process is as follows: 1) The computer receives geological survey documents (such as investigation reports, academic papers) and a preset geological object semantic framework.

[0054] 2) The natural language processing module carried by the computer preprocesses the document: performs word segmentation operations (such as splitting geological domain words), removes stop words (such as "of", "and"), purifies the text, and retains the core semantic information.

[0055] 3) The natural language processing module calls the LDA topic modeling algorithm, inputs the preprocessed set of words, and calculates the topic-word probability distribution P(w|t), where w is the word, t is the topic, the known quantity is the word frequency, and the unknown quantity is the probability value.

[0056] 4) The computer system calculates the topic weight through the topic weight formula where W is the weight of topic t (unknown quantity); P(w i |t) is the probability that the i th core word w i appears in topic t; is the i th core word under topic t; is the i th core word The inverse document frequency values ​​of the words (known quantity).

[0057] 5) The computer system filters high-weight core topics and generates quantifiable topic tags in the format of "topic name-weight value". The purpose is to transform tacit knowledge into explicit and structured information.

[0058] 6) The computer associates quantifiable topic tags with the semantic framework of geological objects to construct a knowledge model of geological objects and their semantics.

[0059] The above steps involve integrating multi-source heterogeneous geological data to solve the problem of heterogeneity in multi-source data and producing a standardized and interoperable geological database; constructing high-dimensional feature vectors for geological objects to upgrade geological objects from qualitative description to quantitative representation and to build machine-readable high-dimensional feature vectors; and quantifying document topic tags to extract explicit topic knowledge from unstructured documents and supplement semantic dimensions.

[0060] Beneficial effects: Through a progressively layered technical system, a closed loop of "data-feature-knowledge" has been formed, constructing a unified and continuously iteratively evolving semantic knowledge model for geological objects. This systematically solves the key issues of data standardization, feature vectors, and topic tags in geological information processing, mainly in the following three aspects: 1) This application solves the "data silo" problem in geological surveys, achieving deep integration and high-quality standardization of multi-source heterogeneous data. Related technologies require significant manual intervention for data processing, format conversion, and terminology alignment in geological survey data processing. This application employs a format parsing engine, coordinate transformation engine, semantic mapping module, and geological ontology library technology to automatically and efficiently process raw geological data from diverse sources, with different formats, coordinate systems, and inconsistent terminology into a high-quality, standardized dataset with unified format, spatial structure, and semantics.

[0061] 2) This application overcomes the limitations of qualitative description of geological objects, achieving comprehensive and computable quantitative characterization. Related technologies often describe geological objects in a scattered, qualitative, or incomplete manner, making precise comparison, classification, and modeling difficult. This application employs techniques such as parallel extraction of multi-dimensional features, feature encoding and normalization, and high-dimensional feature vector fusion to provide comparative and quantitative descriptions of geological objects (such as rock masses and faults). The descriptions are no longer vague qualitative terms like "large," "hard," or "rich in silica and aluminum," but are precisely represented by a high-dimensional feature vector that integrates geometric, physical, and chemical properties, allowing direct application to mathematical calculations and machine analysis.

[0062] 3) This application mines "tacit knowledge" from unstructured documents and transforms it into associative and quantifiable "explicit knowledge." Existing technologies struggle to effectively utilize knowledge hidden within unstructured geological survey documents, as this knowledge cannot be directly processed and applied by computers. This application employs LDA topic modeling and topic weighting algorithms to automatically extract key knowledge topics from massive amounts of geological reports, papers, and other documents, outputting them as quantitative "topic-weight" labels that can be associated with specific geological objects.

[0063] Based on the same inventive concept, this application also provides a system for integrating and extracting multi-source heterogeneous geological data to implement the aforementioned method for integrating and extracting multi-source heterogeneous geological data. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of the one or more embodiments of the system for integrating and extracting multi-source heterogeneous geological data provided below can be found in the limitations of the method for integrating and extracting multi-source heterogeneous geological data described above, and will not be repeated here.

[0064] In one exemplary embodiment, a multi-source heterogeneous geological data integration and feature extraction system is provided, comprising: A standardized geological database construction unit is used to acquire raw geological survey data of multiple geological objects and construct a standardized geological database based on the raw geological survey data; the raw geological survey data includes geological survey data from different sources and in different formats; the standardized geological database stores standardized geological data of multiple geological objects.

[0065] A multi-dimensional data feature determination unit is used to extract the geological object to be characterized from a standardized geological database and determine the multi-dimensional data features of the geological object to be characterized based on the standardized geological database. The multi-dimensional data features include at least geometric spatial features, physical property features, and chemical composition features. Each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data.

[0066] The normalized qualitative description data determination unit is used to perform numerical encoding and normalization processing on the qualitative description data in the multi-dimensional data features to obtain the normalized qualitative description data of the geological object to be characterized.

[0067] The target feature data determination unit is used to normalize the quantitative descriptive data in the multi-dimensional data features, and to determine the normalized quantitative descriptive data and normalized qualitative descriptive data in the multi-dimensional data features as the target feature data of the geological object to be characterized.

[0068] The comprehensive feature vector determination unit is used to generate a comprehensive feature vector of the geological object to be represented based on the target feature data of the geological object to be represented, thereby completing the integration and feature extraction of geological data.

[0069] Specifically, the multi-source heterogeneous geological data integration and feature extraction system of this application adopts a layered architecture design, which includes, from bottom to top: a data fusion and standardization layer, a quantitative characterization and modeling layer, a knowledge discovery and semantic layer, and a unified knowledge model at the top. The layers are interconnected through explicit data flows to collaboratively complete the entire process from raw data to structured knowledge.

[0070] (I) First layer: Data fusion and standardization layer 1) Multi-source data interface adapter: It is responsible for connecting to external data sources and serves as the system's "input port," handling the access of raw geological data in various formats and protocols.

[0071] 2) Format parsing engine: The heterogeneous data received is parsed from its native format into a unified intermediate data model within the system, thus solving the problem of heterogeneous format.

[0072] It receives raw data from the "interface adapter" and outputs intermediate data in a standard structure to the "coordinate transformation engine" and the "semantic mapping module".

[0073] 3) Coordinate transformation engine: The system identifies the spatial reference of the data and uses transformation formulas (such as the Burshall-Wolf formula) to transform all data to a unified coordinate system, thus resolving the problem of inconsistent spatial references. It receives intermediate data from the "format parsing engine" and outputs data with a unified spatial reference to the "semantic mapping module".

[0074] 4) Semantic mapping module: Based on a geological ontology database, terms and units from different sources are mapped to standard terms and units, resolving semantic ambiguity. It receives spatially unified data from a coordinate transformation engine and outputs semantically consistent standard data to a standardized geological database.

[0075] 5) Standardized geological database: The system's foundational data warehouse stores and manages high-quality data that has undergone comprehensive standardization. This data supports the second-layer "geological object identification module."

[0076] (ii) Second layer: Quantitative characterization and modeling layer 1) Geological object identification module: Based on rules, specific geological objects (such as rock masses and faults) are automatically identified and delineated from a standardized geological database. Data is read from the "standardized geological database," and the identified objects are passed to the "feature extraction and fusion engine."

[0077] 2) Feature extraction and fusion engine: This module comprises multiple parallel sub-modules (geometric, physical, and chemical) used to quantify the properties of geological objects from different dimensions and generate feature sub-vectors. It receives geological object information and outputs the extracted feature sub-vectors to the "feature encoding and normalization module".

[0078] 3) Feature encoding and normalization module: Qualitative descriptions are encoded into numerical values, and all numerical features are normalized to eliminate the influence of dimensions. The system receives the original feature sub-vectors and outputs normalized features that can be directly used for fusion to the "feature vector library".

[0079] 4) Feature vector library: Store high-dimensional feature vectors associated with unique identifiers of geological objects. Provide the stored quantitative features to the top-level "geological object semantic knowledge model".

[0080] (III) The third layer: knowledge discovery and semantic layer 1) Natural Language Processing Module: Unstructured geological documents undergo preprocessing such as word segmentation and stop word removal to transform the text into computer-analyzable data. The system receives the original document and outputs the purified word sequence to the "LDA Topic Modeling Engine".

[0081] 2) LDA Theme Modeling Engine: The system automatically mines potential semantic topics from a text collection to obtain the "topic-word" probability distribution. It receives preprocessed text and outputs the topic model results to the "topic quantization and label generation module".

[0082] 3) Topic Quantification and Tag Generation Module: Calculate the weight of each topic and generate quantifiable tags in the format of "topic name-weight value" to make tacit knowledge explicit. Receive topic models and output structured topic tags to the "knowledge topic base".

[0083] 4) Knowledge Topic Base: Store and manage quantified topic tags mined from documents. Provide the stored semantic topic information to the top-level "geological object semantic knowledge model".

[0084] (iv) Top level: Unified knowledge model 1) Semantic knowledge model of geological objects: This is the core output and final result of the entire system. It organically integrates quantitative features from the "feature vector library" and semantic topics from the "knowledge topic library" to form a digital and structured knowledge model that provides a comprehensive and in-depth description of geological objects.

[0085] In one exemplary embodiment, a computer device is provided, including: 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 a method for integrating and extracting features from multi-source heterogeneous geological data.

[0086] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements a method for integrating and extracting features from multi-source heterogeneous geological data.

[0087] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements a method for integrating and extracting features from multi-source heterogeneous geological data.

[0088] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 2 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a method for integrating and extracting features from multi-source heterogeneous geological data.

[0089] Those skilled in the art will understand that Figure 2 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0090] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0091] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0092] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for integrating and extracting features from multi-source heterogeneous geological data, characterized in that, The method for integrating and extracting features from multi-source heterogeneous geological data includes: The process involves acquiring raw geological survey data for multiple geological objects and constructing a standardized geological database based on this data. The raw geological survey data includes geological survey data from different sources and in different formats. The standardized geological database stores standardized geological data for multiple geological objects. Geological objects to be characterized are extracted from standardized geological databases, and multi-dimensional data features of the geological objects to be characterized are determined based on the standardized geological databases. The multi-dimensional data features include at least geometric spatial features, physical property features, and chemical composition features. Each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data. The qualitative descriptive data in the multi-dimensional data features are numerically encoded and normalized to obtain the normalized qualitative descriptive data of the geological object to be represented. The quantitative descriptive data in the multidimensional data features are normalized, and the normalized quantitative descriptive data and the normalized qualitative descriptive data in the multidimensional data features are determined as the target feature data of the geological object to be characterized. Based on the target feature data of the geological object to be represented, a comprehensive feature vector of the geological object to be represented is generated, thus completing the integration and feature extraction of geological data.

2. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 1, characterized in that, Raw geological survey data includes at least two of the following: borehole data, remote sensing images, or geological survey data; Based on the aforementioned original geological survey data, a standardized geological database is constructed, specifically including: The original geological survey data is parsed and converted to obtain geological data after format conversion, with a set intermediate data type. The Bursa-Wolf projection transformation formula is used to uniformly transform the coordinate system of the geological data after the format conversion to the preset standard coordinate system, so as to obtain the geological data in the standard coordinate system. Based on a geological ontology library, semantic mapping and standardization are performed on geological data in the standard coordinate system to obtain standardized geological data. Standardized geological data is packaged and stored to obtain a standardized geological database.

3. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 2, characterized in that, Based on a geological ontology library, semantic mapping and standardization are performed on geological data in a standard coordinate system to obtain standardized geological data, specifically including: By mapping the inconsistent synonyms in geological data under the standard coordinate system to the unified standard terms in the geological ontology library, we obtain the semantically mapped geological data. The associated attribute values ​​in the semantically mapped geological data are converted to units and normalized to obtain standardized geological data.

4. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 1, characterized in that, Geometric spatial features are geometric feature sub-vectors generated based on geometric spatial data; Geometric feature vectors include at least one of the following: centroid coordinates, area, perimeter, shape index, orientation angle, and dip angle; The physical property features are physical feature sub-vectors generated based on physical property data; the physical property data includes measured density data, porosity detection data, resistivity geophysical data, and radioactivity intensity detection data; The chemical composition features are chemical composition sub-vectors generated based on the chemical composition data; the chemical composition data are the main element content parameters.

5. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 4, characterized in that, Geological objects to be characterized are extracted from standardized geological databases, and multi-dimensional data features of these objects are determined based on these databases. Specifically, this includes: Based on spatial scope and attribute rules, GIS spatial analysis methods are used to identify and extract geological objects to be represented from standardized geological databases; Extract geometric spatial data, physical property data, and chemical composition data of the geological objects to be characterized from standardized geological databases; Based on geometric spatial data, geometric feature sub-vectors of the geological objects to be represented are generated; the geometric spatial data includes spatial coordinate data and attitude record data. Based on physical property data, physical feature sub-vectors of the geological objects to be characterized are generated; Based on chemical composition data, chemical feature sub-vectors of the geological objects to be characterized are generated; The geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors of the geological object to be characterized are determined as the multi-dimensional data features of the geological object to be characterized.

6. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 5, characterized in that, Based on the target feature data of the geological object to be represented, the comprehensive feature vector of the geological object to be represented is determined, specifically including: The geometric spatial features, physical feature sub-vectors, and chemical feature sub-vectors in the target feature data of the geological object to be represented are concatenated and weighted to obtain the comprehensive feature vector of the geological object to be represented. Among them, the weighted fusion adopts the formula accomplish; in, This represents the comprehensive feature vector of the geological object to be characterized. , and These are the weighting coefficients; The geometric spatial features in the target feature data of the geological object to be characterized; The physical properties of the target feature data of the geological object to be characterized; Chemical composition features in the target feature data of the geological object to be characterized.

7. The method for integrating and extracting features from multi-source heterogeneous geological data according to claim 1, characterized in that, The method for integrating and extracting features from multi-source heterogeneous geological data also includes: Natural language processing is used to extract quantifiable topic tags from unstructured geological survey documents; By associating topic tags with the comprehensive feature vectors of the geological objects to be represented, a semantic knowledge model of the geological objects to be represented is constructed.

8. A system for integrating and extracting features from multi-source heterogeneous geological data, characterized in that, The multi-source heterogeneous geological data integration and feature extraction system is used to implement the multi-source heterogeneous geological data integration and feature extraction method according to any one of claims 1-7, wherein the multi-source heterogeneous geological data integration and feature extraction system comprises: A standardized geological database construction unit is used to acquire raw geological survey data of multiple geological objects and construct a standardized geological database based on the raw geological survey data; the raw geological survey data includes geological survey data from different sources and in different formats; the standardized geological database stores standardized geological data of multiple geological objects; A multi-dimensional data feature determination unit is used to extract the geological object to be characterized from a standardized geological database and determine the multi-dimensional data features of the geological object to be characterized based on the standardized geological database. The multi-dimensional data features include at least geometric spatial features, physical property features, and chemical composition features. Each multi-dimensional data feature includes qualitative descriptive data and / or quantitative descriptive data. The normalized qualitative description data determination unit is used to perform numerical encoding and normalization processing on the qualitative description data in the multi-dimensional data features to obtain the normalized qualitative description data of the geological object to be represented. The target feature data determination unit is used to normalize the quantitative descriptive data in the multi-dimensional data features, and to determine the normalized quantitative descriptive data and normalized qualitative descriptive data in the multi-dimensional data features as the target feature data of the geological object to be characterized. The comprehensive feature vector determination unit is used to generate a comprehensive feature vector of the geological object to be represented based on the target feature data of the geological object to be represented, thereby completing the integration and feature extraction of geological data.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for integrating and extracting features from multi-source heterogeneous geological data as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for integrating and extracting features from multi-source heterogeneous geological data as described in any one of claims 1-7.

Images