A rapid modeling method and system for the service status of high-speed railway infrastructure based on semantic constraints

Abstract

The invention discloses a method and system for rapid modeling of the service status of high-speed railway infrastructure based on semantic constraints, which belongs to the field of high-speed railway science and technology. It solves the problem that the existing technology cannot obtain the semantic association relationship of the service status of in-service high-speed railway infrastructure, and thus cannot perform unified semantic understanding and knowledge mapping in terms of defects, resulting in inconsistent cognition of the high-speed railway track damage feature library and difficulty in organizing and retrieving massive defect samples. The present invention encodes the acquired defect data of in-service high-speed railway infrastructure, and at the same time builds a defect database based on the monitoring data of in-service high-speed railway infrastructure; constructs a knowledge graph of the service status of in-service high-speed railway infrastructure with unified knowledge mapping based on the structured data in the defect database; the method for rapid modeling of infrastructure service status based on semantic constraints, the knowledge graph of in-service high-speed railway infrastructure service status and the defect database semantically guide and constrain the defect data to rapidly model and update the service status of high-speed railway infrastructure. The present invention is used for rapid modeling of the service status of high-speed railway infrastructure.

Description

Technical Field

[0001] A semantic constraint-based rapid modeling method and system for high-speed railway infrastructure service status, which are used for rapid modeling of high-speed railway infrastructure service status, belong to the field of high-speed railway science and technology. Background Art

[0002] With the continuous increase in the service life of high-speed railways and the influence of complex environmental factors in different regions, high-speed railways have gradually entered the maintenance and repair stage from the maintenance stage. Various infrastructures of high-speed railways inevitably have different types and degrees of diseases. The disease manifestations caused by various types of diseases seriously affect their durability and endanger the safety of train operation. Therefore, it is a key issue to ensure the safety of train operation if the various diseases involved in the service status of high-speed railway infrastructure can be accurately described and expressed efficiently.

[0003] However, the existing modeling methods have the following technical problems:

[0004] 1. It is impossible to obtain the semantic association relationship of the service status of the in-service high-speed railway infrastructure, so it is impossible to carry out unified semantic understanding and knowledge mapping in terms of defects, which will cause problems such as inconsistent cognition of the high-speed railway track damage feature library and difficulty in organizing and retrieving massive defect samples;

[0005] 2. In the process of batch parameter modeling of the high-speed rail service status, a large amount of data needs to be processed, which makes the modeling process too time-consuming; since the existing high-speed rail infrastructure is dynamically changing, the existing modeling methods cannot solve the problem of real-time model updates. Summary of the invention

[0006] In response to the above-mentioned research problems, the purpose of the present invention is to provide a method and system for rapid modeling of the service status of high-speed railway infrastructure based on semantic constraints, so as to solve the problem that the prior art is unable to obtain the semantic association relationship of the service status of in-service high-speed railway infrastructure, and thus is unable to perform unified semantic understanding and knowledge mapping in terms of defects, resulting in inconsistent cognition of the high-speed railway track damage feature library and difficulty in organizing and retrieving massive defect samples.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A method for rapid modeling of the service status of high-speed railway infrastructure based on semantic constraints comprises the following steps:

[0009] Step 1: Encode the acquired disease data of the in-service high-speed railway infrastructure, and construct a disease database based on the monitoring data of the in-service high-speed railway infrastructure, wherein the disease database is a relational database;

[0010] Step 2: construct a service status knowledge graph of in-service high-speed railway infrastructure with unified knowledge mapping based on the structured data in the disease database;

[0011] Step 3: A rapid infrastructure service status modeling method based on semantic constraints, a knowledge graph of the service status of in-service high-speed railway infrastructure, and a defect database are used to semantically guide and constrain defect data to rapidly model and update the service status of high-speed railway infrastructure.

[0012] Further, the specific steps of step 1 are:

[0013] Step 1.1, classify the acquired disease data of in-service high-speed railway infrastructure, and use the Omniclass classification method for classification coding and hierarchical coding after classification. Among them, the disease types obtained after classification include three categories: crack disease type, rust disease type and wear disease type. The disease types include cracks, rust and wear. The crack disease types include rail cracks, track plate cracks and turnout microcracks. The rust disease types include rail rust and rail corrugation. The wear disease types include point rail wear. Classification coding is used for the disease types, and hierarchical coding is used for the disease names and corresponding disease damage degrees under each disease type. After coding, a 6-digit code number is obtained, the first two digits are the disease type code, the middle two digits are the disease name code, and the last two digits are the disease damage degree code. The connection symbol between the two codes is represented by ".";

[0014] Step 1.2: Based on the coded disease data and the monitoring data of the in-service high-speed railway infrastructure, a disease database is constructed. The disease database includes disease type, disease code, discovery time, processing time, monitoring time, damage degree and material structure.

[0015] Further, the specific steps of step 2 are:

[0016] On the basis of the damage database, based on the process of ontology element definition - data screening and cleaning - image acquisition and annotation - image storage and knowledge retrieval, a knowledge graph of the service status of the in-service high-speed railway infrastructure with unified knowledge mapping is formed with the in-service high-speed railway infrastructure as the driver. Among them, the ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain the coding file, graph storage refers to storing the graph in the graph database to form a knowledge graph, knowledge retrieval refers to the management and rapid retrieval of damage types, and the ontology elements refer to the in-service high-speed railway infrastructure and damage types.

[0017] Further, the specific steps of step 3 are:

[0018] Step 3.1, setting semantic constraint rules, including attribute semantic rule constraints and geometric parameter rule constraints, wherein the attribute semantic rule includes constraint time, category, material structure and object encoding parameter attribute semantic information, and the geometric parameter rule constraints include length, width, shape, texture, wavelength and peak parameter geometric parameter information;

[0019] Step 3.2: Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and semantic constraint rules, a rapid infrastructure service status modeling method based on semantic constraints is constructed. The specific steps are as follows:

[0020] Step 3.21, constructing a semantic description attribute table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the attribute semantic rule constraints, wherein the semantic description attribute table includes an attribute table and semantic description attribute information, the attribute table includes fields, data types, descriptions and value fields, and the semantic description attribute information includes the values ​​of each field in the attribute table corresponding to different types of defects;

[0021] Step 3.22, constructing a geometric parameter table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the geometric parameter rule constraints, wherein the geometric parameter table includes a geometric table and geometric parameter information, the geometric table includes the defect type, defect manifestation and specific geometric constraint parameter fields, and the geometric parameter information includes the values ​​of each field in the geometric table corresponding to different defect types;

[0022] Step 3.23, based on steps 3.21-3.22 and the parametric modeling method, a rapid modeling method of infrastructure service status based on semantic constraints is obtained;

[0023] Step 3.3: The infrastructure service status rapid modeling method based on semantic constraints is used to model the service status of high-speed railway infrastructure, obtain a damage model, and update it in real time.

[0024] Further, the specific steps of step 3.3 are:

[0025] Step 3.31, firstly, by checking the in-service high-speed railway infrastructure disease data, the disease data that needs to be parameterized modeling in the in-service high-speed railway infrastructure service status knowledge graph is collected, and various expression modes of the disease data are integrated based on the disease data, where the expression mode includes normal, under maintenance or fault;

[0026] Step 3.32, storing the data obtained in step 3.31 in the disease database, and extracting the attribute table and geometry table of each disease type from the disease data;

[0027] Step 3.33, abstract the knowledge in the service status knowledge graph of the in-service high-speed railway infrastructure into semantic description information, and extract the semantic description attribute information and the geometric parameter information according to the attribute table and the geometric table, respectively, to obtain the semantic description attribute table and the geometric parameter table;

[0028] Step 3.34, guiding and constraining the disease data that need parametric modeling based on the semantic description attribute table and the geometric parameter table, and using the parametric modeling method to construct the disease model corresponding to each disease in each expression mode;

[0029] Step 3.35: Compare the damage data obtained in real time with the damage data to see if they are consistent. If they are inconsistent, update the service status knowledge graph of the in-service high-speed railway infrastructure, analyze the updated service status knowledge graph of the in-service high-speed railway infrastructure, update the semantic description information and geometric parameter information, and update the parameters in the damage model to generate a new damage model.

[0030] A semantic constraint-based rapid modeling system for high-speed railway infrastructure service status, including:

[0031] Disease database construction module: encode the acquired disease data of in-service high-speed railway infrastructure, and construct a disease database based on the monitoring data of in-service high-speed railway infrastructure, where the disease database is a relational database;

[0032] Knowledge graph construction module: constructs a knowledge graph of the service status of in-service high-speed railway infrastructure with unified knowledge mapping based on the structured data in the disease database;

[0033] Defect modeling module: A rapid modeling method for infrastructure service status based on semantic constraints, a knowledge graph of the service status of in-service high-speed railway infrastructure, and a defect database to semantically guide and constrain defect data to rapidly model and update the service status of high-speed railway infrastructure.

[0034] Furthermore, the specific implementation steps of the disease database construction module are:

[0035] Step 1.1, classify the acquired disease data of in-service high-speed railway infrastructure, and use the Omniclass classification method for classification coding and hierarchical coding after classification. Among them, the disease types obtained after classification include three categories: crack disease type, rust disease type and wear disease type. The disease types include cracks, rust and wear. The crack disease types include rail cracks, track plate cracks and turnout microcracks. The rust disease types include rail rust and rail corrugation. The wear disease types include point rail wear. Classification coding is used for the disease types, and hierarchical coding is used for the disease names and corresponding disease damage degrees under each disease type. After coding, a 6-digit code number is obtained, the first two digits are the disease type code, the middle two digits are the disease name code, and the last two digits are the disease damage degree code. The connection symbol between the two codes is represented by ".";

[0036] Step 1.2: Based on the coded disease data and the monitoring data of the in-service high-speed railway infrastructure, a disease database is constructed. The disease database includes disease type, disease code, discovery time, processing time, monitoring time, damage degree and material structure.

[0037] Furthermore, the specific implementation steps of the knowledge graph construction module are:

[0038] On the basis of the damage database, based on the process of ontology element definition - data screening and cleaning - image acquisition and annotation - image storage and knowledge retrieval, a knowledge graph of the service status of the in-service high-speed railway infrastructure with unified knowledge mapping is formed with the in-service high-speed railway infrastructure as the driver. Among them, the ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain the coding file, graph storage refers to storing the graph in the graph database to form a knowledge graph, knowledge retrieval refers to the management and rapid retrieval of damage types, and the ontology elements refer to the in-service high-speed railway infrastructure and damage types.

[0039] Furthermore, the specific implementation steps of the disease model building module are:

[0040] Step 3.1, setting semantic constraint rules, including attribute semantic rule constraints and geometric parameter rule constraints, wherein the attribute semantic rule includes constraint time, category, material structure and object encoding parameter attribute semantic information, and the geometric parameter rule constraints include length, width, shape, texture, wavelength and peak parameter geometric parameter information;

[0041] Step 3.2: Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and semantic constraint rules, a rapid infrastructure service status modeling method based on semantic constraints is constructed. The specific steps are as follows:

[0042] Step 3.21, constructing a semantic description attribute table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the attribute semantic rule constraints, wherein the semantic description attribute table includes an attribute table and semantic description attribute information, the attribute table includes fields, data types, descriptions and value fields, and the semantic description attribute information includes the values ​​of each field in the attribute table corresponding to different types of defects;

[0043] Step 3.22, constructing a geometric parameter table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the geometric parameter rule constraints, wherein the geometric parameter table includes a geometric table and geometric parameter information, the geometric table includes the defect type, defect manifestation and specific geometric constraint parameter fields, and the geometric parameter information includes the values ​​of each field in the geometric table corresponding to different defect types;

[0044] Step 3.23, based on steps 3.21-3.22 and the parametric modeling method, a rapid modeling method of infrastructure service status based on semantic constraints is obtained;

[0045] Step 3.3: The infrastructure service status rapid modeling method based on semantic constraints is used to model the service status of the high-speed railway infrastructure, obtain the damage model, and update it in real time.

[0046] Further, the specific steps of step 3.3 are:

[0047] Step 3.31, firstly, by checking the in-service high-speed railway infrastructure disease data, the disease data that needs to be parameterized modeling in the in-service high-speed railway infrastructure service status knowledge graph is collected, and various expression modes of the disease data are integrated based on the disease data, where the expression mode includes normal, under maintenance or fault;

[0048] Step 3.32, storing the data obtained in step 3.31 in the disease database, and extracting the attribute table and geometry table of each disease type from the disease data;

[0049] Step 3.33, abstract the knowledge in the service status knowledge graph of the in-service high-speed railway infrastructure into semantic description information, and extract the semantic description attribute information and the geometric parameter information according to the attribute table and the geometric table, respectively, to obtain the semantic description attribute table and the geometric parameter table;

[0050] Step 3.34, guiding and constraining the disease data that need parametric modeling based on the semantic description attribute table and the geometric parameter table, and using the parametric modeling method to construct the disease model corresponding to each disease in each expression mode;

[0051] Step 3.35: Compare the damage data obtained in real time with the damage data to see if they are consistent. If they are inconsistent, update the service status knowledge graph of the in-service high-speed railway infrastructure, analyze the updated service status knowledge graph of the in-service high-speed railway infrastructure, update the semantic description information and geometric parameter information, and update the parameters in the damage model to generate a new damage model.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention establishes semantic constraint rules about defects, which can effectively guide and constrain the complex modeling process, ensure the accuracy of infrastructure service status (i.e., defects) modeling, and solve technical problems such as complex operation, poor standardization, and excessive time consumption of traditional modeling. At the same time, it can be used for batch parameter modeling of infrastructure service status based on real-time updates. Specifically, it is embodied as follows:

[0054] 1. The present invention obtains the semantic association relationship of the service status of the in-service high-speed railway infrastructure to perform unified semantic understanding and knowledge mapping, such as unified cognition of the high-speed railway track damage feature library, and simple organization and retrieval of massive disease samples;

[0055] 2. The present invention uses the constraints of semantic constraint rules to avoid processing a large amount of data during batch parameter modeling of the high-speed rail service status, thereby improving modeling efficiency; at the same time, real-time updating of the model can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is the overall research idea diagram of the present invention;

[0057] Figure 2 A schematic diagram for constructing a knowledge graph of the service status of in-service high-speed rail infrastructure in the present invention;

[0058] Figure 3 It is a schematic diagram of the service status knowledge graph of the in-service high-speed rail infrastructure in the present invention;

[0059] Figure 4 is a schematic diagram of the disease database management method of the present invention;

[0060] Figure 5 A schematic diagram of a method for rapid infrastructure service status modeling based on semantic constraints in the present invention;

[0061] Figure 6 A schematic diagram of the process of rapid infrastructure service status modeling based on semantic constraints in the present invention. DETAILED DESCRIPTION

[0062] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation methods.

[0063] First, the acquired disease data of in-service high-speed railway infrastructure is encoded, and a disease database is constructed based on the monitoring data of in-service high-speed railway infrastructure. The disease database is a relational data inventory. The disease data is encoded to facilitate data entry and docking with the disease model. The disease database is constructed based on the high-speed railway monitoring data, model semantic information, and model parameter information. At the same time, the disease database has excellent storage capacity and convenient query, addition, and modification capabilities.

[0064] Secondly, a knowledge graph of the service status of in-service high-speed railway infrastructure is constructed based on the data in the disease database to form a unified knowledge mapping.

[0065] Finally, the expression pattern of the service status of high-speed railway infrastructure is analyzed, and the defect data is guided and constrained based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and semantic constraint rules. The rapid modeling of the infrastructure service status is achieved by calling parameters, forming a common parameterized defect modeling method, that is, the rapid modeling method of infrastructure service status based on semantic constraints and the knowledge graph of the service status of in-service high-speed railway infrastructure are used to guide and constrain the defect model, and the defect model is obtained by calling the parameters in the attribute constraints and geometric constraints to rapidly model the service status of the high-speed railway infrastructure.

[0066] Coding disease data is an important method for coordinating disease data information. Coding diseases and forming disease coding references are of great significance for disease data entry and docking with disease models. This case is driven by high-speed rail monitoring data, combined with literature research and on-site investigation, and in-depth design and research of high-speed rail service status database, analysis and integration of multi-information resources of high-speed rail infrastructure disease data feature system, storage and rapid retrieval.

[0067] The defect database constructed on the basis of the service status of the in-service high-speed railway infrastructure (the service status refers to the defect data) requires the support of high-speed railway monitoring data, model semantic information, model parameter information, etc. to build the defect database, so as to meet the requirements of the informatization and modernization of the service status of high-speed railways. At the same time, the defect database should have excellent storage capacity and convenient query, addition, modification and other capabilities. Traditional databases can store and process data efficiently, such as SQL sever, My SQL and Oracle. This case uses the Access database to effectively connect to the defect, so that the defect data can be stored, queried, added and modified in large quantities in the Access database.

[0068] (1) Disease database storage

[0069] Databases can manage massive amounts of data according to certain data structures, which is conducive to data storage, sharing and control. Using databases to manage defects can solve problems such as efficient data retrieval, information sharing and dynamic scheduling. Implement the database from the aspects of demand analysis, logical design, physical design and implementation, and then effectively manage defects through the database to provide a data source for high-speed railway infrastructure service status modeling. Using databases to manage defects requires three parts, namely functional design, data organization, and database implementation, as shown in the figure.

[0070] (2) Disease database function design---connecting semantic parameter information in the database

[0071] Aiming at the database management and service needs of diseases, the functions of the designed database include:

[0072] ① Disease information integration. Perform parameterized modeling on the acquired disease data, redesign the table of the acquired disease data according to the query requirements, and integrate the attribute information such as disease category, disease code, discovery time, processing time, monitoring time, degree of damage, material structure, etc. in the table according to the requirements of the attribute table using data integration methods, integrate the acquired relevant disease information, and facilitate the rapid acquisition and search of data.

[0073] ②Damage compression, encryption, and decompression. There are a lot of disease data, and there will be texture data, which occupies a large amount of memory. You can choose to use the ASTC compression algorithm to compress the texture data to reduce its memory usage; then use the AES algorithm to encrypt it to protect data security.

[0074] ③ Operations such as adding, modifying, and deleting defect records. For defect records in the system, SQL statements can be used to modify them according to changes or needs to ensure the currentness of the data.

[0075] The specific steps are:

[0076] The acquired disease data of in-service high-speed railway infrastructure are classified, and the Omni cl ass classification method is used for classification coding and hierarchical coding. Among them, the disease types obtained after classification include three categories: crack disease type, rust disease type and wear disease type. The disease types include cracks, rust and wear. The crack disease types include rail cracks, track plate cracks and turnout microcracks. The rust disease types include rail rust and rail corrugation. The wear disease types include point rail wear. Classification coding is used for disease types, and hierarchical coding is used for the disease name and the corresponding disease damage degree under each disease type. After coding, a 6-digit code number is obtained, the first two digits are the disease type code, the middle two digits are the disease name code, and the last two digits are the disease damage degree code. The connection symbol between the two codes is represented by ".";

[0077] According to the current situation survey and literature search, infrastructure diseases are mainly divided into three categories: cracks, rust, and wear. Through further classification, they can be divided into 8 categories, namely: rail cracks, track plate cracks, turnout micro cracks, roadbed settlement, bridge settlement, rail rust, rail corrugation, and rail point wear. As shown in the following table:

[0078] Common disease types and species names

[0079] Serial number Disease type Disease manifestations 1 Crack disease Rail cracks, track plate cracks, turnout micro cracks 2 Rust disease Rail rust and rail corrugation 3 Wear and tear Point rail wear

[0080] The results of infrastructure disease coding are shown in the following table:

[0081] Disease data coding results

[0082]

[0083] A disease database is constructed based on the coded disease data and the monitoring data of in-service high-speed railway infrastructure. The disease database includes disease type, disease code, discovery time, processing time, monitoring time, damage degree and material structure.

[0084] Coding the defects to coordinate the defect data information and establish a unified defect database to facilitate the storage, organization and retrieval of defect data. At the same time, the database is convenient for data entry and docking with the defect model, which is the basis for realizing parametric modeling. Therefore, it is necessary to construct a knowledge graph of the service status of in-service high-speed railway infrastructure. The existing technology lacks the semantic association relationship of the service status of in-service high-speed railway infrastructure, and lacks a unified semantic understanding and knowledge mapping in terms of defects. Specifically:

[0085] On the basis of the disease database, based on the process of ontology element definition - data screening and cleaning - image acquisition and annotation - image storage and knowledge retrieval, a knowledge graph of the service status of in-service high-speed railway infrastructure with unified knowledge mapping is formed with high-speed railway monitoring data as the drive. Among them, the ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain coding files, graph storage refers to storing graphs in graph databases to form knowledge graphs, knowledge retrieval refers to the management and rapid retrieval of disease types, and ontology elements refer to in-service high-speed railway infrastructure and disease types.

[0086] The disease database analyzes and integrates the diverse information resources of the high-speed railway infrastructure disease data feature system, and constructs a knowledge graph of in-service high-speed railway infrastructure by extracting structured data from the disease database.

[0087] In order to solve the problems of inconsistent cognition of high-speed railway disease feature library and difficulty in organizing and retrieving massive disease samples, based on the overall architecture of knowledge graph and on the basis of disease database, based on ontology element definition-data screening and cleaning-image acquisition and annotation-image storage and knowledge retrieval process, a knowledge graph of service status of in-service high-speed railway infrastructure with unified knowledge mapping is formed with high-speed railway monitoring data as the driving force. Among them, ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain coding files, graph storage refers to storing graphs in graph databases to form knowledge graphs, knowledge retrieval refers to the management and rapid retrieval of disease types, and ontology elements refer to in-service high-speed railway infrastructure and disease types.

[0088] The knowledge graph of the service status of high-speed rail infrastructure is mainly divided into the model layer and the data layer. The model layer first needs to use the ontology system to build a knowledge graph framework, and combine the service status of my country's in-service high-speed rail infrastructure to study the classification of defects, and standardize the definition of the types and manifestations of defects in the defect information, so as to define an accurate and clear conceptual framework for the service status of infrastructure. The data layer cleans, integrates and annotates the massive time-course monitoring text and image data obtained, and maps the annotated data to the constructed service status knowledge ontology concept nodes. In order to realize the mapping between the model layer and the data layer, it is necessary to establish an association relationship between entities. As shown in the figure:

[0089] Constructing a knowledge graph of the service status of in-service high-speed railway infrastructure and forming a unified knowledge mapping and semantic understanding is the basis for building a disease database.

[0090] Since the existing technology lacks semantic constraint rules on diseases and methods for batch parameter modeling of infrastructure service status, the modeling accuracy is low.

[0091] The knowledge in the service status knowledge graph of in-service high-speed railway infrastructure is expressed in a form suitable for parametric modeling, which can accurately describe the key between the defect object and the infrastructure. The knowledge is abstracted into semantic description information to guide and constrain the process of defect construction.

[0092] Through in-depth understanding and logical analysis of semantic description information, we can identify the specific directions where constraints need to be imposed on the disease. Semantic constraints mainly include attribute constraints and geometric constraints. Attribute constraints mainly include constraints on attribute semantic information such as time, category, material structure, object coding, etc. Geometric constraints are mainly constraints on geometric parameter information such as length, width, shape, area, texture, wavelength, peak value, etc. established for different disease manifestations.

[0093] After the semantic constraint rules are designed, the organization and operation of the service status objects of the in-service high-speed railway infrastructure should be guided according to the semantic constraint rules, which includes the guidance of the service status of the infrastructure and the use of geometric parameters for constraints. Among them, the guidance of the disease model objects includes operations such as screening objects and adding objects, and the constraints on the disease model objects mainly include operations such as judging the screening, modifying entity attributes, and deleting objects.

[0094] Attribute information is also of great value in conveying the service status information of in-service high-speed railway infrastructure. Some non-spatial information, such as material structure, is replaced and displayed with the texture closest to the raw material. The other part of attribute information is managed by database according to non-spatial data format, which is used to manage the attribute table and geometric constraint table.

[0095] Semantic Description Attribute Table

[0096] Fields Data Types illustrate value Category text Disease type Crack type disease Code text Disease Code 10.01.01 DiscoveryTime text Discovery time 2023.11.12 ProcessingTime text Processing time 2023.12.02 MonitoringTime text Monitoring time 2023.03.03 Degree text Damage light Structure text Material structure Rail

[0097] Among them, Category represents the field of disease category, Code represents the field of disease code, DiscoveryTime represents the field of discovery time, ProcessingTime represents the field of processing time, MonitoringTime represents the field of monitoring time, Degree represents the field of damage degree, and Structure represents the field of material structure.

[0098] Since different disease models are constrained by different geometric parameters, it is necessary to separate the geometric parameters from the attribute table. The length, width, depth, settlement amplitude, displacement, wavelength, peak value, etc. are all in mm.

[0099] Since the linear morphology is divided into straight lines, broken lines, curves and mixed types, where broken lines include angles and lengths, and curves include curvature and arc length; the surface morphology is divided into rectangles, ellipses and irregular surfaces. By setting parameter thresholds, the surfaces that are close to regular shapes are classified as rectangles and ellipses, and the remaining surfaces are classified as irregular surfaces; the block morphology is divided into cubes, spheres, cylinders, cones and irregular bodies, and the classification method is the same as the surface morphology. The missing rail fasteners are divided into rail connecting plates, rail pads, rail bolts, sleeper clamps, sleeper bolts, sleeper seats, turnout switching machines, turnout locking machines, and turnout contact wire clamps according to the fastener types, and they are numbered and stored one by one. The foreign body intrusion categories are divided into rolling stones, branches, pedestrians, animals, and light objects, and they are also numbered and stored one by one. Therefore, the specific constraint parameters for different types of diseases are shown in the following table:

[0100]

[0101]

[0102] Parametric modeling is a method of building a 3D solid model by using the geometric parameters of the solid model. It does not require manual drawing, but can be built by using parameter control only. The basis of parametric modeling is the geometric semantic description of the solid model.

[0103] This case is based on the knowledge graph of the service status of in-service high-speed railway infrastructure, abstracting the knowledge into semantic descriptive information, and establishing new constraint rules for various types of diseases based on the knowledge description information. Appropriate constraint rules are the key step in establishing a disease model. All diseases correspond to the same attribute template, and different types of diseases correspond to different geometric constraint parameter templates. Because different types of diseases are different, the specific geometric constraint parameters required are also different. For example, for cracks, its disease is manifested as rail cracks, so the geometric constraint parameters are length, width, and shape; and for rust, the specific geometric constraint parameters are area, shape, and texture.

[0104] Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and semantic constraint rules, a rapid infrastructure service status modeling method based on semantic constraints is constructed. The specific steps are as follows:

[0105] Based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the attribute semantic rule constraints, the same semantic description attribute table of different defect types is constructed, wherein the semantic description attribute table includes an attribute table and semantic description attribute information, the attribute table includes fields, defect types, descriptions and value fields, and the semantic description attribute information includes the values ​​of each field in the attribute table corresponding to different defect types;

[0106] Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and the geometric parameter rule constraints, a geometric parameter table of different disease types is constructed, wherein the geometric parameter table includes a geometric table and geometric parameter information, the geometric table includes the disease type, disease manifestation and specific geometric constraint parameter fields, and the geometric parameter information includes the values ​​of each field in the geometric table corresponding to different disease types, wherein the specific geometric constraint parameters of the manifestation of rail cracks and track plate crack damage in the crack disease type include length, width and shape, the specific geometric constraint parameters of the manifestation of turnout microcrack disease include length, the specific geometric constraint parameters of the manifestation of rail rust disease in the rust disease type include surface area, shape and texture, the specific geometric constraint parameters of the manifestation of rail corrugation disease in the wear disease type include wavelength, peak value and length, and the specific geometric constraint parameters of the manifestation of point rail wear disease include area, shape and texture;

[0107] Based on the above two steps and the parametric modeling method, a rapid modeling method of infrastructure service status based on semantic constraints is obtained;

[0108] The infrastructure service status rapid modeling method based on semantic constraints is used to model the service status of high-speed railway infrastructure, obtain the damage model, and update it in real time. The specific steps are as follows:

[0109] (1) Demand analysis. Identify the objects that need to be parametrically modeled, determine that the parametric modeling objects are defects, analyze the attribute characteristics and geometric characteristics of the defect model, and identify the key constraint directions. That is, first, by checking the high-speed railway defect data, collect the defect data that needs to be parametrically modeled based on the service status knowledge graph of the in-service high-speed railway infrastructure, and integrate various expression modes of the defect data based on the defect data and store them in the defect database, where the expression modes include normal, under maintenance, or faulty.

[0110] (2) Establish attribute constraints and geometric constraints. According to the conclusions of the demand analysis, define the semantic description attribute information and geometric parameter information required for the disease model. For example, the semantic description attribute information includes the type, code, time, etc. of the disease. For example, the time of disease discovery is used as an attribute constraint to limit the types of diseases that exist in the disease model. Based on the conclusions of the analysis, extract the attribute table and geometric table of each disease type from the disease data, and abstract the knowledge in the service status knowledge graph of in-service high-speed railway infrastructure into semantic description information. According to the attribute table and geometric table, the semantic description attribute information and geometric parameter information are extracted respectively to obtain the semantic description attribute table and geometric parameter table;

[0111] (3) Model generation. Generate a disease model using modeling software based on the defined attribute constraints and geometric constraints. During the model generation process, it is necessary to ensure that the model meets the actual requirements of semantic constraints and ensure the accuracy of the disease model. That is, the disease data that requires parametric modeling is guided and constrained based on the semantic description attribute table and geometric parameter table, and the parametric modeling method is used to construct the disease model corresponding to each disease in each expression mode.

[0112] (4) Based on the semantic description attribute table and geometric parameter table, the disease data that need to be parametrically modeled are guided and constrained, and the disease model corresponding to each disease in each expression mode is constructed using the parametric modeling method;

[0113] (5) Compare the real-time acquired damage data with the damage data to see if they are consistent. If not, update the service status knowledge graph of the in-service high-speed railway infrastructure. Analyze the updated service status knowledge graph of the in-service high-speed railway infrastructure and update the semantic description information and geometric parameter information. Update the parameters in the damage model to generate a new damage model.

[0114] In this case, faced with a high-speed railway environment with complex infrastructure object disease relationships, semantic constraint rules can be used to effectively guide and constrain the complex modeling process, ensure the accuracy of infrastructure service status modeling, reduce the difficulty of infrastructure service status modeling, and solve the problems of complex traditional modeling operations and poor standardization.

[0115] The above are only representative embodiments of the present invention in many specific application scopes, and do not constitute any limitation on the protection scope of the present invention. Any technical solutions formed by transformation or equivalent replacement fall within the protection scope of the present invention.

Claims

1. A method for rapid modeling of the service status of high-speed railway infrastructure based on semantic constraints, characterized in that: The steps include: Step 1: Encode the acquired disease data of the in-service high-speed railway infrastructure, and construct a disease database based on the monitoring data of the in-service high-speed railway infrastructure, wherein the disease database is a relational database; Step 2: Based on the structured data in the disease database, a knowledge graph of the service status of in-service high-speed railway infrastructure is constructed to form a unified knowledge mapping. The specific steps are as follows: On the basis of the damage database, based on the process of ontology element definition - data screening and cleaning - image acquisition and annotation - image storage and knowledge retrieval, a knowledge graph of the service status of in-service high-speed railway infrastructure with unified knowledge mapping is formed with in-service high-speed railway infrastructure as the driver. Among them, ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain coding files, graph storage refers to storing the graph in the graph database to form a knowledge graph, knowledge retrieval refers to the management and rapid retrieval of damage types, and ontology elements refer to in-service high-speed railway infrastructure and damage types; Step 3: A rapid infrastructure service status modeling method based on semantic constraints, a knowledge graph of the service status of in-service high-speed railway infrastructure, and a defect database are used to semantically guide and constrain defect data to rapidly model and update the service status of high-speed railway infrastructure.

2. According to claim 1, a method for rapid modeling of high-speed railway infrastructure service status based on semantic constraints is characterized in that: The specific steps of step 1 are: Step 1.1, classify the acquired disease data of in-service high-speed railway infrastructure, and use the Omniclass classification method for classification coding and hierarchical coding after classification. Among them, the disease types obtained after classification include three categories: crack disease type, rust disease type and wear disease type. The disease types include cracks, rust and wear. The crack disease types include rail cracks, track plate cracks and turnout microcracks. The rust disease types include rail rust and rail corrugation. The wear disease types include point rail wear. Classification coding is used for the disease types, and hierarchical coding is used for the disease name and the corresponding disease damage degree under each disease type. After coding, a 6-digit code number is obtained, the first two digits are the disease type code, the middle two digits are the disease name code, and the last two digits are the disease damage degree code. The connection symbol between the two codes is represented by ". "; Step 1.2: Based on the coded disease data and the monitoring data of the in-service high-speed railway infrastructure, a disease database is constructed. The disease database includes disease type, disease code, discovery time, processing time, monitoring time, damage degree and material structure.

3. The method for rapid modeling of service status of high-speed railway infrastructure based on semantic constraints according to claim 2 is characterized in that: The specific steps of step 3 are: Step 3.1, setting semantic constraint rules, including attribute semantic rule constraints and geometric parameter rule constraints, wherein the attribute semantic rule includes constraint time, category, material structure and object encoding parameter attribute semantic information, and the geometric parameter rule constraints include length, width, shape, texture, wavelength and peak parameter geometric parameter information; Step 3.2: Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and semantic constraint rules, a rapid infrastructure service status modeling method based on semantic constraints is constructed. The specific steps are as follows: Step 3.21, constructing a semantic description attribute table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the attribute semantic rule constraints, wherein the semantic description attribute table includes an attribute table and semantic description attribute information, the attribute table includes fields, data types, descriptions and value fields, and the semantic description attribute information includes the values ​​of each field in the attribute table corresponding to different types of defects; Step 3.22, constructing a geometric parameter table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the geometric parameter rule constraints, wherein the geometric parameter table includes a geometric table and geometric parameter information, the geometric table includes the defect type, defect manifestation and specific geometric constraint parameter fields, and the geometric parameter information includes the values ​​of each field in the geometric table corresponding to different defect types; Step 3.23, based on steps 3.21-3.22 and the parametric modeling method, a rapid modeling method of infrastructure service status based on semantic constraints is obtained; Step 3.3: The infrastructure service status rapid modeling method based on semantic constraints is used to model the service status of high-speed railway infrastructure, obtain a damage model, and update it in real time.

4. According to claim 3, a method for rapid modeling of high-speed railway infrastructure service status based on semantic constraints is characterized in that: The specific steps of step 3.3 are: Step 3.31, firstly, by checking the in-service high-speed railway infrastructure disease data, the disease data that needs to be parameterized modeling in the in-service high-speed railway infrastructure service status knowledge graph is collected, and various expression modes of the disease data are integrated based on the disease data, where the expression mode includes normal, under maintenance or fault; Step 3.32, storing the data obtained in step 3.31 in the disease database, and extracting the attribute table and geometry table of each disease type from the disease data; Step 3.33, abstract the knowledge in the service status knowledge graph of the in-service high-speed railway infrastructure into semantic description information, and extract the semantic description attribute information and the geometric parameter information according to the attribute table and the geometric table, respectively, to obtain the semantic description attribute table and the geometric parameter table; Step 3.34, guiding and constraining the disease data that need parametric modeling based on the semantic description attribute table and the geometric parameter table, and using the parametric modeling method to construct the disease model corresponding to each disease in each expression mode; Step 3.35: Compare the damage data obtained in real time with the damage data to see if they are consistent. If they are inconsistent, update the service status knowledge graph of the in-service high-speed railway infrastructure, analyze the updated service status knowledge graph of the in-service high-speed railway infrastructure, update the semantic description information and geometric parameter information, and update the parameters in the damage model to generate a new damage model.

5. A rapid modeling system for high-speed railway infrastructure service status based on semantic constraints, characterized in that: include: Disease database construction module: encode the acquired disease data of in-service high-speed railway infrastructure, and construct a disease database based on the monitoring data of in-service high-speed railway infrastructure, where the disease database is a relational database; Knowledge graph construction module: Based on the structured data in the disease database, a knowledge graph of the service status of in-service high-speed railway infrastructure is constructed to form a unified knowledge mapping. The specific implementation steps are as follows: On the basis of the damage database, based on the process of ontology element definition - data screening and cleaning - image acquisition and annotation - image storage and knowledge retrieval, a knowledge graph of the service status of in-service high-speed railway infrastructure with unified knowledge mapping is formed with in-service high-speed railway infrastructure as the driver. Among them, ontology element definition is to build a hierarchical structure through ontology, data screening and cleaning refers to the selection and redundant processing of classified data, image acquisition and annotation refers to the acquisition of useful image processing data and annotation processing to obtain coding files, graph storage refers to storing the graph in the graph database to form a knowledge graph, knowledge retrieval refers to the management and rapid retrieval of damage types, and ontology elements refer to in-service high-speed railway infrastructure and damage types; Defect modeling module: A rapid modeling method for infrastructure service status based on semantic constraints, a knowledge graph of the service status of in-service high-speed railway infrastructure, and a defect database to semantically guide and constrain defect data to rapidly model and update the service status of high-speed railway infrastructure.

6. The semantic constraint-based rapid modeling system for high-speed railway infrastructure service status according to claim 5 is characterized in that: The specific implementation steps of the disease database construction module are: Step 1.1, classify the acquired disease data of in-service high-speed railway infrastructure, and use the Omniclass classification method for classification coding and hierarchical coding after classification. Among them, the disease types obtained after classification include three categories: crack disease type, rust disease type and wear disease type. The disease types include cracks, rust and wear. The crack disease types include rail cracks, track plate cracks and turnout microcracks. The rust disease types include rail rust and rail corrugation. The wear disease types include point rail wear. Classification coding is used for the disease types, and hierarchical coding is used for the disease name and the corresponding disease damage degree under each disease type. After coding, a 6-digit code number is obtained, the first two digits are the disease type code, the middle two digits are the disease name code, and the last two digits are the disease damage degree code. The connection symbol between the two codes is represented by ". "; Step 1.2: Based on the coded disease data and the monitoring data of the in-service high-speed railway infrastructure, a disease database is constructed. The disease database includes disease type, disease code, discovery time, processing time, monitoring time, damage degree and material structure.

7. The semantic constraint-based rapid modeling system for high-speed railway infrastructure service status according to claim 6 is characterized in that: The specific implementation steps of the disease model building module are: Step 3.1, setting semantic constraint rules, including attribute semantic rule constraints and geometric parameter rule constraints, wherein the attribute semantic rule includes constraint time, category, material structure and object encoding parameter attribute semantic information, and the geometric parameter rule constraints include length, width, shape, texture, wavelength and peak parameter geometric parameter information; Step 3.2: Based on the service status knowledge graph of in-service high-speed railway infrastructure, the disease database and semantic constraint rules, a rapid infrastructure service status modeling method based on semantic constraints is constructed. The specific steps are as follows: Step 3.21, constructing a semantic description attribute table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the attribute semantic rule constraints, wherein the semantic description attribute table includes an attribute table and semantic description attribute information, the attribute table includes fields, data types, descriptions and value fields, and the semantic description attribute information includes the values ​​of each field in the attribute table corresponding to different types of defects; Step 3.22, constructing a geometric parameter table of different types of defects based on the service status knowledge graph of in-service high-speed railway infrastructure, the defect database and the geometric parameter rule constraints, wherein the geometric parameter table includes a geometric table and geometric parameter information, the geometric table includes the defect type, defect manifestation and specific geometric constraint parameter fields, and the geometric parameter information includes the values ​​of each field in the geometric table corresponding to different defect types; Step 3.23, based on steps 3.21-3.22 and the parametric modeling method, a rapid modeling method of infrastructure service status based on semantic constraints is obtained; Step 3.3: The infrastructure service status rapid modeling method based on semantic constraints is used to model the service status of high-speed railway infrastructure, obtain a damage model, and update it in real time.

8. The semantic constraint-based rapid modeling system for high-speed railway infrastructure service status according to claim 7 is characterized in that: The specific steps of step 3.3 are: Step 3.31, firstly, by checking the in-service high-speed railway infrastructure disease data, the disease data that needs to be parameterized modeling in the in-service high-speed railway infrastructure service status knowledge graph is collected, and various expression modes of the disease data are integrated based on the disease data, where the expression mode includes normal, under maintenance or fault; Step 3.32, storing the data obtained in step 3.31 in the disease database, and extracting the attribute table and geometry table of each disease type from the disease data; Step 3.33, abstract the knowledge in the service status knowledge graph of the in-service high-speed railway infrastructure into semantic description information, and extract the semantic description attribute information and the geometric parameter information according to the attribute table and the geometric table, respectively, to obtain the semantic description attribute table and the geometric parameter table; Step 3.34, guiding and constraining the disease data that need parametric modeling based on the semantic description attribute table and the geometric parameter table, and using the parametric modeling method to construct the disease model corresponding to each disease in each expression mode; Step 3.35: Compare the damage data obtained in real time with the damage data to see if they are consistent. If they are inconsistent, update the service status knowledge graph of the in-service high-speed railway infrastructure, analyze the updated service status knowledge graph of the in-service high-speed railway infrastructure, update the semantic description information and geometric parameter information, and update the parameters in the damage model to generate a new damage model.

Images