Method and device for constructing operation design domain ontology of high-speed railway train control system
By constructing a safety risk database and using neural network analysis for the high-speed railway train control system, an ontology model was generated, which solved the problem of decentralized information management, improved the system's reliability and operational efficiency, and reduced safety risks and operating costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-23
- Publication Date
- 2026-04-14
AI Technical Summary
The existing design methods for high-speed railway train control systems rely on manual information integration, resulting in fragmented information management, incomplete understanding of safe operating conditions by operating units, and problems such as safety risks and high operating costs.
A safety risk database is constructed, and the actual operating data of the train control system is analyzed using knowledge extraction models and neural networks to generate an ontology model containing concepts, attributes, and logical relationships. Through topology optimization and reconstruction, the final ontology model is formed to support system design, operation, and maintenance.
The system integrates information from the train control system, improving system reliability and operational efficiency while reducing safety risks and operating costs.
Smart Images
Figure CN121858533A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of high-speed railway technology, and in particular to a method and apparatus for constructing the ontology of the operation design domain of a high-speed railway train control system. Background Technology
[0002] With the rapid development of high-speed railways, the train control system, as the core system ensuring the safe and efficient operation of trains, is becoming increasingly complex. During the design phase, the train control system defines the constraints and assumptions for safe train operation. The content involved in this process is scattered across design documents (system specifications, subsystem specifications, etc.), hazard and risk analysis documents (phase verification reports, hazard logs, etc.), and operation handover documents (safe operating conditions, operation manuals, etc.). Existing methods rely on manual integration, making it difficult to effectively integrate and manage this information. However, after the high-speed railway train control system is put into use, the risk responsibility also transfers to the operating unit, which may lead to incomplete or unsystematic understanding of the safe operating conditions of the train control system by operation and maintenance personnel, resulting in problems in system analysis, decision-making, and optimization. Therefore, to ensure the safe operation of high-speed railways, it is necessary to develop a new method to construct the operational design domain ontology of its train control system, thereby providing a theoretical basis for building a decision support system for high-speed railway operation risk monitoring and emergency response. Summary of the Invention
[0003] The purpose of this application is to provide a method and apparatus for constructing an ontology of the operation design domain of a high-speed railway train control system. This method and apparatus realize the construction of an ontology model, can systematically integrate relevant information of the train control system, provide effective support for the design, operation and maintenance of the train control system, effectively improve the reliability and operating efficiency of the system, and significantly reduce safety risks and high-speed railway operating costs.
[0004] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for constructing an ontology of the operation design domain of a high-speed railway train control system, including: Construct a safety risk database; the safety risk database includes actual operational data from several high-speed railway train control systems. Using a knowledge extraction model, knowledge is extracted from the security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships. The extracted concepts are used as nodes, and the attributes are used as node feature descriptions. A topology graph is constructed using the logical relationships between the concepts. Using neural networks, the semantic similarity of concept pairs is analyzed based on actual operational data to optimize and reconstruct the topology graph, resulting in the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk monitoring, and emergency decision-making of the high-speed railway train control system. A concept pair consists of two nodes in the topology graph that have a logical relationship, corresponding to the concepts.
[0005] Optionally, a knowledge extraction model is used to extract knowledge from the security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships, specifically including: The terminology of the safety risk database is standardized using a thesaurus for the rail transit field to obtain a standardized safety risk database. Using a knowledge extraction model, knowledge is extracted from the standardized security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships.
[0006] Optionally, a neural network is used to analyze the semantic similarity of concept pairs based on actual operational data to optimize and reconstruct the topology graph, resulting in the final ontology model of the high-speed railway train control system, specifically including: The semantic similarity components of concept pairs are calculated based on actual operational data; the semantic similarity components include concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity. Using the semantic similarity components of concept pairs as input to the neural network, the network parameters of the neural network are optimized and updated based on the loss function and backpropagation algorithm to obtain the updated network parameters; Calculate the overall node similarity score based on the updated network parameters; The topology graph is optimized and reconstructed based on the similarity score between nodes and the preset threshold to obtain the final ontology model of the high-speed railway train control system.
[0007] Optionally, the formulas for calculating concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity are expressed as follows: ; ; ; ; in, For concept name similarity, For concept instance similarity, Define similarity for concepts. For conceptual structural similarity; For the concept With concept Minimum edit distance of the name string; Concepts With concept The length of the name string; and Concepts With concept The probability of the number of intersections and unions of instances; and Representing concepts respectively and A set of descriptions; Set and The number of elements in the intersection of the two sets. Indicates belonging to a set Not belonging to a set The number of elements; This is a scaling factor.
[0008] Optionally, the neural network is a backpropagation (BP) neural network, including an input layer, hidden layers, and an output layer; the formula for calculating the similarity score between nodes is as follows: ; in, To score the similarity between nodes, It is the hidden layer. The weights from the node to the output layer It is the threshold of the output layer. It is the hidden layer. The activation value of a node.
[0009] Optionally, the topology graph is optimized and reconstructed based on the comprehensive node similarity score and a preset threshold to obtain the final ontology model of the high-speed railway train control system, specifically including: If the overall similarity score between nodes exceeds a preset threshold, then: Establish bidirectional relationships between nodes in the topology graph whose overall similarity score is greater than a similarity threshold; Merge nodes with overlapping attributes into a single attribute; Based on expert experience, check whether there are logical conflicts or whether the weight values of each part are reasonable. If there are logical conflicts or unreasonable weight values, modify the connection weights or remove abnormal relationships.
[0010] Optionally, the actual operating data of the high-speed railway train control system includes historical fault records, safety specification documents, and system design documents.
[0011] Secondly, this application provides a device for constructing the ontology of the operation design domain of a high-speed railway train control system, comprising: A safety risk database construction module is used to construct a safety risk database; the safety risk database includes actual operating data from several high-speed railway train control systems. The knowledge extraction module is used to extract knowledge from the security risk database using a knowledge extraction model to obtain a primary ontology model containing concepts, attributes, and logical relationships. The topology graph construction module is used to construct a topology graph by taking extracted concepts as nodes, attributes as node feature descriptions, and utilizing the logical relationships between concepts. The topology graph optimization and reconstruction module is used to optimize and reconstruct the topology graph by using neural networks and analyzing the semantic similarity of concept pairs based on actual operating data, so as to obtain the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk supervision and emergency decision-making of the high-speed railway train control system. A concept pair is composed of two concepts corresponding to two nodes with logical relationships in the topology graph.
[0012] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the above-described method for constructing the ontology of the high-speed railway train control system operation design domain.
[0013] Fourthly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for constructing the ontology of the high-speed railway train control system operation design domain.
[0014] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and apparatus for constructing an ontology of the operation design domain of a high-speed railway train control system. The method involves building a safety risk database; using a knowledge extraction model to extract knowledge from the database to obtain a primary ontology model containing concepts, attributes, and logical relationships; using extracted concepts as nodes and attributes as node feature descriptions; and constructing a topology graph using the logical relationships between concepts. Finally, using a neural network, the semantic similarity of concept pairs is analyzed based on actual operational data to optimize and reconstruct the topology graph, resulting in the final ontology model of the high-speed railway train control system. This method achieves ontology model construction that closely reflects the actual operation of the high-speed railway train control system, enabling the system to integrate relevant information and providing effective support for the design, operation, and maintenance of the system. This significantly improves system reliability and operational efficiency, and substantially reduces safety risks and high-speed railway operating costs. Attached Figure Description
[0015] 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.
[0016] Figure 1 This is an application environment diagram of a method for constructing an operation design domain ontology of a high-speed railway train control system according to an embodiment of this application.
[0017] Figure 2 This is a flowchart illustrating a method for constructing the ontology of a high-speed railway train control system operation design domain, provided as an embodiment of this application.
[0018] Figure 3 This is a schematic diagram illustrating the specific process of constructing the ontology of the operation design domain of a high-speed railway train control system, as provided in an embodiment of this application.
[0019] Figure 4 This is a schematic diagram of the structure of a quantized node and relational topology graph provided in an embodiment of this application.
[0020] Figure 5 This is a schematic diagram of a neural network structure provided in an embodiment of this application.
[0021] Figure 6 This is a schematic diagram of the functional modules of a high-speed railway train control system operation design domain body construction device provided in an embodiment of this application.
[0022] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0023] 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.
[0024] 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.
[0025] The method for constructing the ontology of the operation design domain of a high-speed railway train control system provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send requests to be processed to server 104. After receiving the requests, server 104 constructs a security risk database, extracts knowledge from the database using a knowledge extraction model, and obtains a primary ontology model containing concepts, attributes, and logical relationships. The extracted concepts are used as nodes, and attributes are used as node feature descriptions. A topology graph is constructed using the logical relationships between concepts. A neural network is used to analyze the semantic similarity of concept pairs based on actual operational data to optimize and reconstruct the topology graph, resulting in the final ontology model of the high-speed railway train control system. Server 104 can then feed back the obtained final ontology model for the high-speed railway train control system to terminal 102. In addition, in some embodiments, the method for constructing the ontology of the high-speed railway train control system operation design domain can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly construct the ontology of the high-speed railway train control system operation design domain for the request to be processed, or the server 104 can obtain the request to be processed from the data storage system and construct the ontology of the high-speed railway train control system operation design domain for the request to be processed.
[0026] The terminal 102 can be, but is not limited to, various desktop computers and laptops. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.
[0027] In one exemplary embodiment, such as Figure 2 and Figure 3 As shown, a method for constructing the ontology of the operation design domain of a high-speed railway train control system is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 204.
[0028] Step 201: Construct a safety risk database; the safety risk database includes actual operating data of several high-speed railway train control systems.
[0029] Step 202: Using a knowledge extraction model, knowledge is extracted from the security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships.
[0030] Step 203: Use the extracted concepts as nodes and the attributes as node feature descriptions, and construct a topology graph using the logical relationships between the concepts.
[0031] Step 204: Using a neural network, the semantic similarity of concept pairs is analyzed based on actual operating data to optimize and reconstruct the topology graph, thereby obtaining the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk monitoring, and emergency decision-making of the high-speed railway train control system. A concept pair consists of two nodes in the topology graph that have a logical relationship, corresponding to the concepts.
[0032] By implementing steps 201 to 204 above, an ontology model that closely reflects the actual operation of the high-speed railway train control system is constructed. This model can systematically integrate relevant information of the train control system, providing effective support for the design, operation, and maintenance of the train control system. It effectively improves the reliability and operational efficiency of the system and significantly reduces safety risks and high-speed railway operating costs.
[0033] The actual operational data of the high-speed railway train control system includes historical fault records, safety specification documents, and system design documents. Input information is obtained through multi-source data integration; that is, by systematically sorting through information such as historical fault records, safety specification documents, and system design documents of the high-speed railway train control system, a safety risk database is generated.
[0034] A systematic review of historical fault records, safety specifications, and system design documents for high-speed railway train control systems is conducted. This includes compiling a hazard list from historical fault records; extracting the conditions and requirements for safe system operation from safety specifications; and summarizing system design assumptions from system design documents. Based on this data, including the hazard list, system safe operation conditions, and design assumptions, a safety risk database is obtained. This database serves as the foundational data for constructing the ontology model of the high-speed railway train control system's operation design domain.
[0035] Hazardous situations refer to malfunctions or emergencies that may lead to accidents, such as: 1) Damaged or aged insulation in the track circuit, causing the train control system to still display an "idle" status after the train passes through the track section (when in fact a car should be occupying it). 2) Excessive wear of braking components (such as brake shoes) without timely replacement leads to partial failure of the train's air braking system, increasing the train's braking distance. 3) Short-term heavy rainfall (≥50mm / h) reduces the wheel-rail adhesion coefficient, extending the train's braking distance and potentially causing it to run out of the safety protection zone.
[0036] Step 202 above specifically includes: using a thesaurus for the rail transit field to standardize the terminology of the safety risk database to obtain a standardized safety risk database; terminology standardization specifically includes unifying names, parameters, and status descriptions; using a knowledge extraction model to extract knowledge from the standardized safety risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships.
[0037] The knowledge extraction model can be a deep learning model. The construction process is based on a security risk dataset, where a deep learning model is trained to obtain the knowledge extraction model.
[0038] For example, the knowledge extraction model includes a pre-trained language unit, a bidirectional long short-term memory network unit, and a conditional random field unit. Security risk text data from a security risk database is input into the knowledge extraction model. The pre-trained language unit processes the security risk text data to obtain word vector results; the bidirectional long short-term memory network unit processes the word vector results in both forward and backward directions to obtain a predicted label sequence incorporating contextual information; the conditional random field unit decodes the predicted label sequence incorporating contextual information to obtain an optimal label sequence; the optimal label sequence includes concepts, attributes, and logical relationships.
[0039] The extracted core concepts are used as nodes, and the attributes are used as node feature descriptions. A topology graph is constructed using the logical relationships between the concepts. At the same time, based on historical data and expert experience, initial weight values are set for the nodes and connections. The node weight represents the impact of the concept on the safe operation of the system, while the connection weight represents the degree of association or the intensity of the impact.
[0040] The constructed topology graph is as follows Figure 4 As shown, the node weight of rainy weather (0.6) represents its impact on the safe operation of the system. Rainfall ≥ 50 mm / h is considered an attribute of this node and is applied to the track surface through a connection weight of 0.82, quantifying the degree of interference of external conditions on the track. The track may trigger a temporary speed limit through a connection weight of 0.82. The temporary speed limit changes the block occupancy status through a connection weight of 0.82. The block status is transmitted to the interlocking equipment through a connection weight of 0.82. The interlocking equipment, as the core control device, has a node weight of 0.85 representing its impact on the safe operation of the system. Through a connection weight of 0.82, it coordinates and controls the track circuits and signals to complete the safety assurance from the external environment to the system operation.
[0041] The system utilizes neural network algorithms to analyze actual operational data, including historical fault records, safety specifications, and system design documents, to dynamically adjust the weights of nodes and connections, thereby optimizing the initial topology. Semantic similarity between concepts is calculated using neural network algorithms. If the similarity between two concepts exceeds a preset threshold (i.e., the overall node similarity score exceeds the preset threshold), a bidirectional relationship is established in the topology, or related attributes are normalized and merged. The topology is reconstructed by completing relationships and merging redundant attributes. Expert experience is used to check for logical conflicts or reasonable weights in the relationships between different parts. The ontology model is continuously adjusted and optimized based on actual case test results. After multiple rounds of optimization and verification, a final ontology model that accurately reflects the actual operation of the high-speed railway train control system is obtained.
[0042] A schematic diagram of the neural network structure is shown below. Figure 5 As shown, the neural network can be a BP neural network, including an input layer, a hidden layer, and an output layer.
[0043] In step 204 above, a neural network is used to analyze the semantic similarity of concept pairs based on actual operational data, in order to optimize and reconstruct the topology graph and obtain the final ontology model of the high-speed railway train control system, specifically including: The semantic similarity components of concept pairs are calculated based on actual operational data; the semantic similarity components include concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity. Using the semantic similarity components of concept pairs as input to the neural network, the network parameters of the neural network are optimized and updated based on the loss function and backpropagation algorithm to obtain the updated network parameters; Calculate the overall node similarity score based on the updated network parameters; The topology graph is optimized and reconstructed based on the similarity score between nodes and the preset threshold to obtain the final ontology model of the high-speed railway train control system.
[0044] This embodiment also continuously adjusts and optimizes the final ontology model obtained in step 204 based on the test results of actual cases, resulting in a final ontology model that accurately reflects the actual operation of the high-speed railway train control system. It should be noted that the final ontology model is tested (virtual testing / actual road testing) to obtain actual case test results. The results are then checked to see if they match the actual situation. If not, the final ontology model is adjusted and optimized. If it matches, the final ontology model obtained in step 204 is used as the final output, thus obtaining a final ontology model that accurately reflects the actual operation of the high-speed railway train control system.
[0045] The input layer of the BP neural network has four nodes, each corresponding to one of the four semantic similarity components: concept name similarity. Concept-Instance Similarity Concept definition similarity Similarity to concept structure The formulas for calculating concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity are as follows: (1); (2); (3); (4); Here, a semantic radius r is chosen, and all semantic neighbors of the concept are found within the path distance p ≤ r, resulting in a set. Concepts from different ontologies thus yield two related sets, denoted as the description set. The concept structure similarity is then calculated using the concept definition similarity formula. . For the concept With concept The minimum edit distance of a name string is the minimum number of single-character edit operations required to convert one string into another. Concepts With concept The length of the name string; and Concepts With concept The probability of the number of intersections and unions of instances; and Representing concepts respectively and The description set (synonym set, feature set), assuming and Two concepts, obtained by finding semantic neighbors, result in two sets of descriptions. and ; Set and The number of elements in the intersection of the two sets. Indicates belonging to a set Not belonging to a set The number of elements; This is the scaling factor. It is expressed as follows: (5); in, Indicates from the concept The shortest path distance to the root. Indicates from the concept The shortest path distance to the root.
[0046] Number of hidden layer nodes The empirical formula yields 6 layers. (6); in, This represents the number of nodes in the output layer. The number of nodes in the input layer. This is a constant. The output layer nodes represent the overall similarity score between nodes, i.e., set to layer 1. Each hidden node is configured with... Type activation function, the activation function is expressed as .
[0047] Hidden layer The input of each node is Hidden layer The output of each node is The formula is as follows: (7); (8); in, It is the input layer. Node to hidden layer The weight of a node; = , Each corresponds to one of the four semantic similarity components. , It is the threshold of the intermediate layer (hidden layer).
[0048] The formula for calculating the overall similarity score between nodes is as follows: (9); in, For the concept With concept The overall node similarity score between the corresponding two nodes. It is the hidden layer. The weights from the node to the output layer It is the threshold of the output layer. It is the hidden layer. The activation value of a node (via Type functions are compressed to between 0 and 1.
[0049] When the four semantic similarity components of the high-speed railway train control system are input into the neural network, the four semantic similarity components are normalized to form the input vector. Based on the loss function, the initial weights of nodes and connections are dynamically adjusted using the backpropagation algorithm. The formula for the loss function is as follows: (10); in, Represents the loss function; This serves as the input to the neural network; The first derivative of the activation function is obtained by taking the derivative of the activation function. ,in .
[0050] The weight update formula is as follows: (11); (12); in, Represents the learning coefficient, which is used to test different values on the validation set. Choose the value that decreases the loss the fastest and is the most stable. This represents the momentum factor, used to accelerate convergence and reduce oscillations, and is generally in the range of [0.5-0.9]. express Weight increment at time step express The weight increment at any given time; Let these represent the partial derivatives of the loss function and the first derivative of the previous layer, respectively. The first neuron is moved to the next layer. The partial derivatives of the weights of each neuron.
[0051] The topology graph is optimized and reconstructed based on the comprehensive node similarity score and a preset threshold to obtain the final ontology model of the high-speed railway train control system. Specifically, this includes: if the comprehensive node similarity score exceeds the preset threshold, then: establish bidirectional relationships for nodes in the topology graph whose comprehensive node similarity score is greater than the similarity threshold; merge nodes with overlapping attributes into a unified attribute; and check whether there are logical conflicts or whether the weight values of each part of the relationship are reasonable based on expert experience, and modify the connection relationship weights or remove abnormal relationships.
[0052] Based on the topological graph of nodes and connections optimized by weights, the input layer After nonlinear mapping in the hidden layer, the output layer generates a comprehensive node similarity score. If the comprehensive node similarity score exceeds a preset threshold, the topology graph is reconstructed. This includes establishing bidirectional relationships between nodes with strong semantic similarity, i.e., when the comprehensive node similarity score between two nodes in the topology graph exceeds a certain threshold... When the similarity exceeds the threshold, a bidirectional relationship is established between the two nodes; nodes with overlapping attributes are merged into a unified attribute; based on expert experience, it is checked whether there are logical conflicts or whether the weight values are reasonable in each part of the relationship. If there are logical conflicts or unreasonable weight values, the connection weights are modified or abnormal relationships are removed.
[0053] This application integrates historical fault records, safety specification documents, and system design documents to construct a safety risk database; it standardizes terminology based on a thesaurus for the rail transit field and extracts knowledge using a knowledge extraction model to obtain a primary ontology model containing concepts, attributes, and relationships; it uses the extracted core concepts as nodes and sets initial weight values for nodes and connections based on historical data and expert experience; it analyzes actual operating data through neural networks to dynamically adjust the weight values of nodes and connections, calculates semantic similarity, and optimizes and reconstructs the topology graph through threshold judgment; and it adjusts the ontology through case testing to obtain an ontology model that closely reflects the actual operation of high-speed railway train control systems.
[0054] This application constructs an ontology model for the operation design domain of a high-speed railway train control system, integrating information from historical fault records, safety specifications, and system design documents to form a unified knowledge base—a safety risk database. This eliminates information fragmentation and improves the efficiency and comprehension of complex information. Simultaneously, it uses a thesaurus of the rail transit domain to parse semantics and employs a knowledge extraction model to extract core concepts and attribute associations, ensuring the standardization and professionalism of the ontology construction. Furthermore, this application uses a neural network algorithm to dynamically adjust weight values and calculate semantic similarity, optimizing the topology structure and providing scientific support for the iterative updates of the ontology model, enhancing its rationality and practicality. In addition, the constructed ontology model supports the design optimization, risk monitoring, and emergency decision-making of the high-speed railway train control system, effectively improving system reliability and operational efficiency, and significantly reducing safety risks and high-speed railway operating costs.
[0055] This application also provides an application scenario in which the above-mentioned ontology construction method for the operation design domain of a high-speed railway train control system is applied. Specifically, the ontology construction method for the operation design domain of a high-speed railway train control system provided in this embodiment can be applied to the risk supervision scenario of a high-speed railway train control system. The risk supervision scenario of a high-speed railway train control system includes a request issuance stage, an ontology construction link, and a risk supervision stage. The request to be processed enters the ontology construction link from the request issuance stage to obtain the final ontology model of the high-speed railway train control system, and then enters the downstream risk supervision stage. The ontology construction method for the operation design domain of a high-speed railway train control system provided in this embodiment belongs to the ontology construction link. Specifically, in the ontology construction link process for the request to be processed, a safety risk database can be constructed, and a knowledge extraction model can be used to extract knowledge from the safety risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships. The extracted concepts are used as nodes, and the attributes are used as node feature descriptions. A topology graph is constructed using the logical relationships between concepts. A neural network is used to analyze the semantic similarity of concept pairs based on actual operation data to optimize and reconstruct the topology graph, thereby obtaining the final ontology model of the high-speed railway train control system.
[0056] Based on the same inventive concept, this application also provides a high-speed railway train control system operation design domain ontology construction device for implementing the above-mentioned high-speed railway train control system operation design domain ontology construction method. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the high-speed railway train control system operation design domain ontology construction device provided below can be found in the limitations of the high-speed railway train control system operation design domain ontology construction method above, and will not be repeated here.
[0057] In one exemplary embodiment, such as Figure 6 As shown, a high-speed railway train control system operation design domain ontology construction device is provided, which includes the following modules.
[0058] The safety risk database construction module T1 is used to construct a safety risk database; the safety risk database includes actual operating data of several high-speed railway train control systems.
[0059] The knowledge extraction module T2 is used to extract knowledge from the security risk database using a knowledge extraction model to obtain a primary ontology model containing concepts, attributes, and logical relationships.
[0060] The topology graph construction module T3 is used to construct a topology graph by taking extracted concepts as nodes, attributes as node feature descriptions, and utilizing the logical relationships between concepts.
[0061] The topology graph optimization and reconstruction module T4 is used to optimize and reconstruct the topology graph by using a neural network to analyze the semantic similarity of concept pairs based on actual operating data, thereby obtaining the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk monitoring and emergency decision-making of the high-speed railway train control system. A concept pair is composed of the concepts corresponding to two nodes with logical relationships in the topology graph.
[0062] 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 7 As shown, the 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 operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data for constructing the ontology of the high-speed railway train control system's operation design domain. 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 constructing the ontology of the high-speed railway train control system's operation design domain.
[0063] Those skilled in the art will understand that Figure 7 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.
[0064] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0065] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0066] 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.
[0067] Those skilled in the art will understand that all or part of the processes in the methods of 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, and when executed, it can include the processes of the embodiments of the above methods. 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).
[0068] 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.
[0069] 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.
[0070] 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 constructing the ontology of the operation design domain of a high-speed railway train control system, characterized in that, The method for constructing the ontology of the high-speed railway train control system operation design domain includes: Construct a safety risk database; the safety risk database includes actual operational data from several high-speed railway train control systems. Using a knowledge extraction model, knowledge is extracted from the security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships. The extracted concepts are used as nodes, and the attributes are used as node feature descriptions. A topology graph is constructed using the logical relationships between the concepts. Using neural networks, the semantic similarity of concept pairs is analyzed based on actual operational data to optimize and reconstruct the topology graph, resulting in the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk monitoring, and emergency decision-making of the high-speed railway train control system. A concept pair consists of two nodes in the topology graph that have a logical relationship, corresponding to the concepts.
2. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 1, characterized in that, Using a knowledge extraction model, knowledge is extracted from the security risk database to obtain a preliminary ontology model containing concepts, attributes, and logical relationships, specifically including: The terminology of the safety risk database is standardized using a thesaurus for the rail transit field to obtain a standardized safety risk database. Using a knowledge extraction model, knowledge is extracted from the standardized security risk database to obtain a primary ontology model containing concepts, attributes, and logical relationships.
3. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 1, characterized in that, By utilizing neural networks and analyzing the semantic similarity of concept pairs based on actual operational data, the topology graph is optimized and reconstructed to obtain the final ontology model of the high-speed railway train control system, specifically including: The semantic similarity components of concept pairs are calculated based on actual operational data; the semantic similarity components include concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity. Using the semantic similarity components of concept pairs as input to the neural network, the network parameters of the neural network are optimized and updated based on the loss function and backpropagation algorithm to obtain the updated network parameters; Calculate the overall node similarity score based on the updated network parameters; The topology graph is optimized and reconstructed based on the similarity score between nodes and the preset threshold to obtain the final ontology model of the high-speed railway train control system.
4. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 3, characterized in that, The formulas for calculating concept name similarity, concept instance similarity, concept definition similarity, and concept structure similarity are as follows: ； ； ； ； in, For concept name similarity, For concept instance similarity, Define similarity for concepts. For conceptual structural similarity; For the concept With concept Minimum edit distance of the name string; Concepts With concept The length of the name string; and Concepts With concept The probability of the number of intersections and unions of instances; and Representing concepts respectively and A set of descriptions; Set and The number of elements in the intersection of the two sets. Indicates belonging to a set Not belonging to a set The number of elements; This is a scaling factor.
5. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 3, characterized in that, The neural network is a backpropagation (BP) neural network, comprising an input layer, hidden layers, and an output layer; the formula for calculating the similarity score between nodes is as follows: ； in, To score the similarity between nodes, It is the hidden layer. The weights from the node to the output layer It is the threshold of the output layer. It is the hidden layer. The activation value of a node.
6. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 3, characterized in that, The topology graph is optimized and reconstructed based on the similarity score between nodes and a preset threshold to obtain the final ontology model of the high-speed railway train control system, which specifically includes: If the overall similarity score between nodes exceeds a preset threshold, then: Establish bidirectional relationships between nodes in the topology graph whose overall similarity score is greater than a similarity threshold; Merge nodes with overlapping attributes into a single attribute; Based on expert experience, check whether there are logical conflicts or whether the weight values of each part are reasonable. If there are logical conflicts or unreasonable weight values, modify the connection weights or remove abnormal relationships.
7. The method for constructing the ontology of the operation design domain of a high-speed railway train control system according to claim 1, characterized in that, The actual operating data of the high-speed railway train control system includes historical fault records, safety specification documents, and system design documents.
8. A device for constructing the operational design domain of a high-speed railway train control system, characterized in that, The high-speed railway train control system operation design domain ontology construction device includes: A safety risk database construction module is used to construct a safety risk database; the safety risk database includes actual operating data from several high-speed railway train control systems. The knowledge extraction module is used to extract knowledge from the security risk database using a knowledge extraction model to obtain a primary ontology model containing concepts, attributes, and logical relationships. The topology graph construction module is used to construct a topology graph by taking extracted concepts as nodes, attributes as node feature descriptions, and utilizing the logical relationships between concepts. The topology graph optimization and reconstruction module is used to optimize and reconstruct the topology graph by using neural networks and analyzing the semantic similarity of concept pairs based on actual operating data, so as to obtain the final ontology model of the high-speed railway train control system. The final ontology model is used for the design optimization, risk supervision and emergency decision-making of the high-speed railway train control system. A concept pair is composed of two concepts corresponding to two nodes with logical relationships in the topology graph.
9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the method for constructing the operational design domain ontology of a high-speed railway train control system according to any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for constructing the ontology of the high-speed railway train control system operation design domain as described in any one of claims 1-7.