A method for constructing a safety knowledge graph for bridge components based on multi-category monitoring information and hazard information.

Abstract

This invention discloses a method for constructing a bridge component safety knowledge graph based on multi-category monitoring information and hazard information, relating to the field of bridge structural safety monitoring technology. The method includes: determining structural reliability indicators as the characterization indicators of bridge structural safety; establishing a two-dimensional collaborative analysis system based on monitoring items and hazard items; constructing a structural reliability analysis framework that integrates multi-source data, with structural reliability indicators as the ultimate analysis objective; and, based on the structural reliability analysis framework, designing a four-layer ontology model including a structural physical layer, a damage monitoring layer, a reliability assessment layer, and an auxiliary decision-making layer using Protege software to form a knowledge graph, establishing a knowledge representation path for bridge structural safety assessment and decision-making, and realizing timely dynamic updates of the bridge structural safety status and reliable maintenance decisions. This invention constructs a knowledge graph through a four-layer ontology model with a causal transmission mechanism, forming a knowledge representation path capable of systematically analyzing bridge structural safety.

Description

Technical Field

[0001] This invention relates to the field of bridge structural safety monitoring technology, specifically a method for constructing a safety knowledge graph of bridge components based on multi-category monitoring information and hazard information. Background Technology

[0002] In modern transportation infrastructure construction, bridges serve as crucial hubs, and their structural safety and service life directly impact public safety, property security, and socio-economic stability. With the rapid development of the Internet of Things, sensor technology, and intelligent testing equipment, bridge structural health monitoring systems have deployed numerous high-precision monitoring devices, such as strain sensors, displacement gauges, and accelerometers, enabling real-time data collection of structural response and environmental loads. Simultaneously, the inspection and maintenance process accumulates a wealth of unstructured data, including damage records and repair histories. Statistics show that a large bridge can generate gigabytes of monitoring data daily, encompassing various data types such as time-series data, image data, and text reports.

[0003] However, significant technical bottlenecks exist in the current field of bridge health monitoring and digital maintenance. From a data perspective, massive amounts of monitoring data suffer from "information silos": multi-source, heterogeneous data such as equipment monitoring data (e.g., vibration frequency, stress and strain), inspection and maintenance data (e.g., crack width records, corrosion assessment), and environmental data (e.g., temperature, humidity, wind speed and direction) lack unified semantic association and structured integration. For example, real-time vibration data collected by sensors cannot directly correspond to local damage descriptions in inspection reports, leading to the fragmentation of potential causal relationships between data and making it difficult to construct a complete structural safety analysis model. Furthermore, existing data analysis often employs statistical methods based on single data sources, such as threshold alarms based on strain data, failing to comprehensively consider the impact of multiple factors coupled on structural safety. This results in the value of massive dynamic monitoring data not being fully realized, and structural safety hazards being difficult to identify in a timely manner.

[0004] From a structural safety analysis perspective, bridge components, as the basic units of the load-bearing structure, require precise assessment of their safety status to ensure overall structural stability. However, traditional monitoring methods focus only on the overall bridge response, neglecting detailed analysis at the component level. For example, localized damage such as critical welds in steel box girders and crack development in concrete piers is difficult to diagnose directly using overall monitoring data. Furthermore, there are technological gaps in the fusion and interpretation of multi-source monitoring data at the component scale, making it impossible to effectively quantify the degradation process and remaining lifespan of components. This situation leads to a lack of scientific basis for bridge maintenance decisions, often resulting in "over-maintenance" or "delayed maintenance," which wastes resources and increases structural safety risks.

[0005] In the field of digital bridge maintenance, existing systems primarily focus on data storage and simple queries, lacking in-depth correlation analysis between monitoring data and maintenance strategies. For example, maintenance plan formulation still relies on manual experience and periodic inspection results, failing to create a closed-loop management system that integrates real-time monitoring data with structural performance assessments and maintenance plan optimization. Furthermore, the massive amounts of data throughout the bridge's entire lifecycle have not yet formed a structured knowledge system, failing to provide managers with intuitive and interpretable decision support and thus hindering the development of intelligent and scientific bridge operation and maintenance.

[0006] In conclusion, constructing a systematic knowledge representation path, with bridge structural components as the core carrier, to achieve deep integration and semantic association of multi-source monitoring data, and thus break down the technical barriers between health monitoring and digital management, has become an urgent need to improve the level of bridge operation and maintenance. Summary of the Invention

[0007] This invention addresses the needs and shortcomings of current technological development by providing a method for constructing a safety knowledge graph for bridge components based on multi-category monitoring information and hazard information. Through data fusion and knowledge representation, a data application and analysis system is formed, from raw dynamic monitoring data to structural safety assessment and decision-making.

[0008] The present invention provides a method for constructing a safety knowledge graph of bridge components based on multi-category monitoring information and hidden danger information. The technical solution adopted to solve the above-mentioned technical problems is as follows:

[0009] A method for constructing a safety knowledge graph for bridge components based on multi-category monitoring information and hazard information, comprising the following steps:

[0010] S1. Determine the structural reliability index as the characterization index of bridge structural safety, establish a two-dimensional collaborative analysis system based on monitoring items and hidden danger items, take the structural reliability index as the final analysis purpose, and build a structural reliability analysis framework that integrates multi-source data.

[0011] S2. Based on the structural reliability analysis framework of multi-source data fusion, a four-layer ontology model including structural physical layer, damage monitoring layer, reliability assessment layer and auxiliary decision-making layer is designed using Protege software to form a knowledge graph, establish a knowledge representation path for bridge structural safety assessment decision-making, and realize timely dynamic updates of bridge structural safety status and reliable maintenance decisions.

[0012] Optionally, perform step S1, a structural reliability study, on the likelihood that the structure will perform its intended function under specified time and conditions;

[0013] When analyzing structural reliability, the probability distribution of random variables in structural resistance and load effects is first determined. The limit state equation Z = RS ≤ 0 is used to determine whether the structure has failed, where R represents structural resistance and S represents load effect. Then, the failure probability of Z ≤ 0 is calculated, and the structural reliability is obtained by subtracting the failure probability of Z ≤ 0 from 1. The structural reliability index β is used to quantify the structural reliability. The larger the index value, the more reliable the structure.

[0014] Step S1 involves analyzing the variables in the structural resistance and load effects of the structure based on the continuously updated monitoring and hazard data, and establishing a structural reliability analysis framework based on multi-source data fusion.

[0015] Optionally, step S2 is executed, where the structural physical layer is designed using Protege software. This involves using components as carriers to realize the force analysis and transfer path of structural resistance and load effects. This process requires creating component type classes and defining and associating geometric attributes and spatial topological relationships. Specifically, this includes:

[0016] S2.1.1 Set geometric attributes, define the geometric attributes of different component types in the component type class, and define the subclass inheritance relationship of the component type class; realize the one-to-one mapping between the geometric attributes of the BIM model and the geometric attributes of the subclass instances of the component type class through the component ID attribute, and obtain unique structural space coordinates for each subclass instance to provide data support for the establishment of spatial topological relationships between different components;

[0017] S2.1.2 Based on the obtained structural spatial coordinates, set spatial topology association attributes for each subclass instance of the component type class; by setting the spatial topology relationship of each subclass instance, realize the linkage effect of stress analysis between different components, and thus provide a data transmission path for the calculation of local structural reliability analysis or overall structural reliability analysis.

[0018] Optionally, step S2 is performed, where a damage monitoring layer is designed using Protege software to associate different components with monitoring items and potential hazards, and the types and degrees of component damage caused by monitoring items or potential hazards are calculated. This process specifically includes:

[0019] S2.2.1 Establish a dual-mode damage triggering mechanism in the damage monitoring layer;

[0020] S2.2.2 Based on the monitoring items and hidden danger items associated with each subclass instance of the component type, the dual-mode damage triggering mechanism calculates the damage type and its damage confidence of each subclass instance, and passes it to the reliability assessment layer through damage attributes.

[0021] Preferably, the damage attribute is a structured data that includes key fields such as damage type, confidence level, location of occurrence, number of involved components, and time of first detection.

[0022] After calculating the damage type and damage confidence of each subclass instance, it is encapsulated in the form of damage attributes and transmitted to the reliability assessment layer in real time through a standardized data interface.

[0023] Preferably, the dual-mode damage triggering mechanism establishes attribute associations between monitoring items, hazard items, and damage type confidence levels through a monitoring triggering mechanism and a hazard level assessment mechanism, respectively.

[0024] The monitoring triggering mechanism establishes a judgment logic of "abnormal monitoring value → triggering the over-limit threshold → calculating the confidence level of the damage type":

[0025] If the monitored value X > the set threshold α,

[0026] Then activate the preset damage type confidence index calculation rules;

[0027] The hazard level assessment mechanism establishes a judgment logic of "hazard item exists → triggering hazard level assessment → calculating damage type confidence level":

[0028] If the hazard item Y ∈ a certain hazard level interval

[0029] Then activate the preset damage type confidence index calculation rules.

[0030] Preferably, the construction methods involved include three types of damage: apparent damage, stress risk, and structural displacement.

[0031] Optionally, step S2 is performed, using Protege software to design a reliability assessment layer, thereby calculating reliability indices from individual components to the overall bridge structure. This process specifically includes:

[0032] S2.3.1 Develop a reliability assessment rule base in the reliability assessment layer;

[0033] S2.3.2. Based on the rules in the reliability assessment rule base, calculate the structural resistance and load effect of each individual component; based on the load transfer path of the component, further calculate the local structural resistance and load effect caused by the component and its adjacent components; evaluate the entire bridge, integrate the data of each component, and calculate the overall structural resistance and load effect of the bridge structure.

[0034] S2.3.3 Based on the rules in the reliability assessment rule base, and using professional knowledge in materials mechanics and engineering mechanics, damage evolution rules are designed for three types of damage: apparent damage, stress risk, and structural displacement. Then, the resistance reduction factor and load amplification factor of the component under different damage types are determined to quantify the impact of damage on structural performance.

[0035] S2.3.4 Substitute the calculated structural resistance, load effect, and resistance reduction coefficient and load amplification coefficient determined by the damage evolution rule into the limit state equation for calculation. Use the limit state equation to determine the failure probability of the structure, thereby clarifying the reliability index of the structure and completing the safety assessment of a single component, a local structure, or the overall bridge structure.

[0036] Further optional, step S2 is performed, using Protege software to design an auxiliary decision-making layer, and using dynamically updatable structural reliability indicators to provide suggestions for preventive maintenance decisions for the bridge. This process specifically includes: S2.4.1, comprehensively collecting technical condition indicators related to the structure and determining reliability indicators from them;

[0037] S2.4.2 Based on professional knowledge in materials mechanics and structural mechanics, analyze the influence mechanism of technical condition indicators on reliability indicators, and establish quantitative or qualitative correlation rules between technical condition indicators and reliability indicators through experimental research, numerical simulation or historical data statistical analysis.

[0038] S2.4.3. Classify preventive maintenance recommendations into different types, and clarify the applicable conditions and maintenance objectives of each type; determine the corresponding reliability index range based on different structural technical conditions and the developed correlation rules; match the determined reliability index range with the preventive maintenance recommendation types to establish a one-to-one correspondence.

[0039] S2.4.4 Obtain information on the actual technical condition of the structure through on-site testing or monitoring data collection, and evaluate the technical condition of the structure according to pre-set standards and methods to draw maintenance conclusions; based on the maintenance conclusions, find the correspondence between the established technical condition, reliability indicators and preventive maintenance recommendations, determine preventive maintenance recommendations that match the current technical condition and reliability indicators of the structure, and provide a scientific basis for the maintenance decision of the structure.

[0040] The present invention provides a method for constructing a safety knowledge graph of bridge components based on multi-category monitoring information and hazard information. Compared with the prior art, the beneficial effects are as follows:

[0041] 1. This invention can form a data application and analysis system from raw dynamic monitoring data (including equipment monitoring data and inspection hazard data) to structural safety assessment and decision-making through data fusion and knowledge representation;

[0042] 2. This invention integrates multi-source heterogeneous data and constructs a knowledge graph through a four-layer ontology model with a causal transmission mechanism, forming a knowledge representation path that can systematically analyze the safety of bridge structures, thereby achieving timely and dynamic assessment of bridge structural safety and ultimately achieving the goal of intelligent bridge management and maintenance. Attached Figure Description

[0043] Appendix Figure 1 This is a flowchart illustrating the method implementation of an embodiment of the present invention;

[0044] Appendix Figure 2 This is a structural reliability analysis framework diagram built using a beam bridge as an example in an embodiment of the present invention;

[0045] Appendix Figure 3 This is a schematic diagram of the core classification system of the structural physical layer designed using a beam bridge as an example in an embodiment of the present invention;

[0046] Appendix Figure 4 This is a schematic diagram of the associated attributes of a structural physical layer subclass instance designed using a beam bridge as an example in an embodiment of the present invention;

[0047] Appendix Figure 5 This is a schematic diagram of the core classification system for damage monitoring layer designed using a beam bridge as an example in an embodiment of the present invention;

[0048] Appendix Figure 6 This is a schematic diagram of the associated attributes of a damage monitoring layer example designed using a beam bridge as an example in an embodiment of the present invention;

[0049] Appendix Figure 7 This is a schematic diagram of the core classification system for reliability assessment layer designed using a beam bridge as an example in an embodiment of the present invention;

[0050] Appendix Figure 8 This is a schematic diagram of the correlation attributes of key items in the reliability index analysis of a single component in the reliability assessment layer of this invention, taking a beam bridge as an example.

[0051] Appendix Figure 9 This is a schematic diagram of the correlation attributes of key items in the reliability index analysis of the local or overall structure of the reliability assessment layer designed in an embodiment of the present invention, taking a beam bridge as an example;

[0052] Appendix Figure 10 This is a schematic diagram of the core classification system for the auxiliary decision-making layer designed using a beam bridge as an example in an embodiment of the present invention;

[0053] Appendix Figure 11This is a schematic diagram of the associated attributes of the auxiliary decision-making layer preventive maintenance suggestion subclass designed using a beam bridge as an example in an embodiment of the present invention. Detailed Implementation

[0054] To make the technical solution, the technical problem solved, and the technical effect of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments.

[0055] Example:

[0056] Combined with appendix Figure 1 This embodiment proposes a method for constructing a safety knowledge graph for bridge components based on multi-category monitoring information and hazard information, which includes the following steps:

[0057] S1. Determine the structural reliability index β as the characterization index of bridge structural safety, establish a two-dimensional collaborative analysis system based on monitoring items and hidden danger items, and build a structural reliability analysis framework with multi-source data fusion with the structural reliability index β as the final analysis objective.

[0058] Structural reliability studies the likelihood that a structure will perform its intended function under specified time and conditions.

[0059] When analyzing structural reliability, first determine the probability distribution of random variables in structural resistance R and load effect S. Use the limit state equation Z = RS ≤ 0 to determine whether the structure has failed. Then calculate the failure probability of Z ≤ 0 and subtract the failure probability of Z ≤ 0 from 1 to obtain the structural reliability. Use the structural reliability index β to measure the structural reliability. The larger the value of the structural reliability index β, the more reliable the structure.

[0060] When performing step S1, based on the continuously updated monitoring data and potential hazard data, the variables in the structural resistance R and load effect S are sorted out, and a structural reliability analysis framework based on multi-source data fusion is built.

[0061] S2. Based on the structural reliability analysis framework of multi-source data fusion, a four-layer ontology model including structural physical layer, damage monitoring layer, reliability assessment layer and auxiliary decision-making layer is designed using Protege software to form a knowledge graph, establish a knowledge representation path for bridge structural safety assessment decision-making, and realize timely dynamic updates of bridge structural safety status and reliable maintenance decisions.

[0062] S2.1 Using Protege software to design the structural physical layer, and using components as carriers to realize the force analysis and transfer path of structural resistance and load effects, this process requires creating component type classes and defining and binding geometric attributes and spatial topological relationship attributes, specifically including:

[0063] S2.1.1 Set geometric attributes, define the geometric attributes of different component types in the component type class, and define the subclass inheritance relationship of the component type class; realize the one-to-one mapping between the geometric attributes of the BIM model and the geometric attributes of the subclass instances of the component type class through the component ID attribute, and obtain unique structural space coordinates for each subclass instance to provide data support for the establishment of spatial topological relationships between different components;

[0064] S2.1.2 Based on the obtained structural spatial coordinates, set spatial topology association attributes for each subclass instance of the component type class, such as component adjacency relationship, load transfer path, etc.; by setting the spatial topology relationship of each subclass instance, realize the linkage effect of stress analysis between different components, and thus provide data transfer path for the calculation of local structural reliability analysis or overall structural reliability analysis.

[0065] S2.2 Using Protege software, a damage monitoring layer is designed to associate different components with monitoring items and potential hazards, and to calculate the type and extent of component damage caused by monitoring items or potential hazards. This process specifically includes:

[0066] S2.2.1. A dual-mode damage triggering mechanism is established in the damage monitoring layer. This mechanism establishes attribute associations between monitoring items, hazard items, and damage type confidence levels through a monitoring triggering mechanism and a hazard level assessment mechanism, respectively.

[0067] The monitoring triggering mechanism establishes a judgment logic of "abnormal monitoring value → triggering the over-limit threshold → calculating the confidence level of the damage type":

[0068] If the monitored value X > the set threshold α,

[0069] Then activate the preset damage type confidence index calculation rules;

[0070] The hazard level assessment mechanism establishes a judgment logic of "hazard item exists → triggering hazard level assessment → calculating damage type confidence level":

[0071] If the hazard item Y ∈ a certain hazard level interval

[0072] Then activate the preset damage type confidence index calculation rules;

[0073] S2.2.2 Based on the monitoring items and hidden danger items associated with each subclass instance of the component type, the dual-mode damage triggering mechanism calculates the damage type and its damage confidence of each subclass instance, and passes it to the reliability assessment layer through damage attributes.

[0074] Damage attributes are a type of structured data that includes key fields such as damage type, confidence level, location of occurrence, component number involved, and first detection time. After calculating the damage type and its damage confidence level for each subclass instance, the data is encapsulated in the form of damage attributes and transmitted to the reliability assessment layer in real time through a standardized data interface.

[0075] Damage types include three categories: apparent damage, stress risk, and structural displacement.

[0076] It should be added that, based on the above two-layer ontological model design of the structural physical layer and the damage monitoring layer, the following can be clarified:

[0077] (1) By using the core classification system of the damage monitoring layer and the associated attribute settings under different instances, data can be provided for the calculation of structural resistance R and load effect S in the reliability assessment layer;

[0078] (2) Based on the setting of spatial topological attributes such as load transfer path and adjacency relationship of each component instance in the structural physical layer, the structural transfer path of stress data can be provided for the analysis and calculation of structural reliability in the reliability assessment layer.

[0079] S2.3. Using Protege software to design a reliability assessment layer, the reliability index β is calculated from individual components to the overall bridge structure. This process specifically includes:

[0080] S2.3.1 Develop a reliability assessment rule base in the reliability assessment layer;

[0081] S2.3.2. Based on the rules in the reliability assessment rule base, calculate the structural resistance R and load effect S of each individual component; based on the load transfer path of the component, further calculate the local structural resistance R and load effect S caused by the component and its adjacent components; evaluate the entire bridge, integrate the data of each component, and calculate the overall structural resistance R and load effect S of the bridge structure.

[0082] S2.3.3 Based on the rules in the reliability assessment rule base, and using professional knowledge in materials mechanics and engineering mechanics, damage evolution rules are designed for three types of damage: apparent damage, stress risk, and structural displacement. Then, the resistance reduction factor and load amplification factor of the component under different damage types are determined to quantify the impact of damage on structural performance.

[0083] S2.3.4 Substitute the calculated structural resistance R, load effect S, and resistance reduction coefficient and load amplification coefficient determined by the damage evolution rule into the limit state equation for calculation. Use the limit state equation to determine the failure probability of the structure, and then clarify the reliability index β of the structure to complete the safety assessment of a single component, a local structure, or the overall bridge structure.

[0084] It should be added that the reliability assessment layer obtains the apparent damage confidence, stress damage confidence, and displacement damage confidence of each component based on the damage monitoring layer. Combining the adjacency relationship and load transfer path between different components, the layer analyzes the structural resistance R and load effect S of all components that trigger the damage confidence calculation rule through reliability assessment rules. Furthermore, based on the limit state equation Z = RS and the load transfer path of each component, the layer calculates the reliability index β of the damaged component and its local structure, as well as the overall bridge structure, to achieve the reliability analysis of the bridge structure.

[0085] S2.4. Using Protege software to design an auxiliary decision-making layer, and through dynamically updated structural reliability indicators, to provide suggestions for preventive maintenance decisions for bridges. This process specifically includes:

[0086] S2.4.1. Collect all technical condition indicators related to the structure and determine the reliability indicators from them;

[0087] S2.4.2 Based on professional knowledge in materials mechanics and structural mechanics, analyze the influence mechanism of technical condition indicators on reliability indicators, and establish quantitative or qualitative correlation rules between technical condition indicators and reliability indicators through experimental research, numerical simulation or historical data statistical analysis.

[0088] S2.4.3. Classify preventive maintenance recommendations into different types, and clarify the applicable conditions and maintenance objectives of each type; determine the corresponding reliability index range based on different structural technical conditions and the developed correlation rules; match the determined reliability index range with the preventive maintenance recommendation types to establish a one-to-one correspondence.

[0089] S2.4.4 Obtain information on the actual technical condition of the structure through on-site testing or monitoring data collection, and evaluate the technical condition of the structure according to pre-set standards and methods to draw maintenance conclusions; based on the maintenance conclusions, find the correspondence between the established technical condition, reliability index β and preventive maintenance recommendations, determine preventive maintenance recommendations that match the current technical condition and reliability index β of the structure, and provide a scientific basis for the maintenance decision of the structure.

[0090] It should be added that the auxiliary decision-making level, referring to the "Specifications for Maintenance of Highway Bridges and Culverts" (JTG 5120-2021), describes and defines the preventive maintenance recommendations corresponding to different technical condition index value ranges. Based on the correlation rules between technical condition indicators and reliability index β, and combined with project realities and engineering experience, it sets correlations between the technical condition index and the reliability index of individual components, the reliability index of their respective local structures, and the overall structural reliability index. Furthermore, based on the component ID, the component type (e.g., superstructure, substructure, bridge deck system) of different components is clarified, thereby obtaining preventive maintenance recommendations based on the reliability of individual components, the reliability of local structures, and the overall structural reliability, achieving intelligent maintenance at the component level for this beam bridge.

[0091] Taking a beam bridge as an example, the above technical solutions are as follows:

[0092] When specifically executing step S1, taking the structural resistance R (including material properties R1, geometric parameters R2, and structural connection method R3) and the load effect S (including dead load effect S1 and live load effect S2) as an example, the structural reliability analysis framework is constructed as follows: Figure 2 As shown.

[0093] In the specific execution step S2.1, the geometric properties of component types such as main beams, suspended beams, piers, abutments, and bridge decks are defined to obtain the following: Figure 3 The diagram shows the core classification system of the structure-physical layer, and the corresponding diagram shows the associated attributes of the structure-physical layer subclass instances. Figure 4 As shown, Figure 4 Medium gray represents instances of the subclass itself, and white represents the attributes of the instance.

[0094] When implementing step S2.2, taking the main beam as an example, the core classification system of the designed damage monitoring layer is as follows: Figure 5 As shown in the diagram, the associated attributes of the corresponding damage monitoring layer instance are illustrated below. Figure 6 As shown, Figure 6 Medium gray represents instances of the subclass itself, and white represents the attributes of the instance.

[0095] It should be added that: when performing step S2.2, the monitoring items are not limited to concrete component temperature, component lateral displacement, component vertical displacement, beam end longitudinal displacement, key section static strain, component vibration acceleration, pier settlement, structural cracks, vehicle load, wind speed and pressure, wind direction, etc.; the hidden danger items are not limited to beam cracks, beam end displacement, abutment cracks, cap beam cracks, pier cracks, exposed rebar corrosion, etc.; the damage types refer to the damage type information in the "Technical Condition Assessment Standard for Highway Bridges" (JTG / TH21-2011) and the "Technical Standard for Urban Bridge Maintenance" (CJJ 99-2017), and further sort and summarize them into three categories: apparent damage, stress risk, and structural displacement. The damage confidence level can be set according to the actual situation of the project and the standard provisions.

[0096] In the specific execution of step S2.3, based on the damage monitoring layer, a reliability assessment layer is further implemented using structural resistance R and load effect S to assess the reliability from individual components to local structures and then to the overall bridge structure. For example, its core classification system is as follows: Figure 7 As shown in the diagram, the correlation attributes of the key items in the reliability index analysis of a single component are illustrated in the following figure. Figure 8 As shown in the diagram, the correlation attributes of key items in the reliability index analysis of local or overall structures are illustrated in the figure below. Figure 9 As shown, Figure 8 and 9 Medium gray represents key analysis items, and white represents the attributes of key items.

[0097] In the specific execution step S2.4, based on the damage monitoring layer and the reliability assessment layer, an auxiliary decision-making layer is further designed, whose core classification system is as follows: Figure 10 As shown in the diagram, the associated attributes of the corresponding preventative maintenance suggestion subclass are illustrated below. Figure 11 As shown, Figure 11 Medium gray represents subclass items, and white represents the attributes of subclass items.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a safety knowledge graph for bridge components based on multi-category monitoring information and hazard information, characterized in that, Includes the following steps: S1. Determine the structural reliability index as the characterization index of bridge structural safety, establish a two-dimensional collaborative analysis system based on monitoring items and hidden danger items, take the structural reliability index as the final analysis purpose, and build a structural reliability analysis framework that integrates multi-source data. S2. Based on the structural reliability analysis framework of multi-source data fusion, a four-layer ontology model including structural physical layer, damage monitoring layer, reliability assessment layer and auxiliary decision layer is designed using Protege software to form a knowledge graph, establish a knowledge representation path for bridge structural safety assessment decision, and realize timely dynamic updates of bridge structural safety status and reliable maintenance decisions. Specifically, the damage monitoring layer is designed using Protege software to associate different components with monitoring items and potential hazards, and to calculate the type and extent of component damage caused by the monitoring items or potential hazards. This process includes: S2.2.

1. Establish a dual-mode damage triggering mechanism in the damage monitoring layer; the dual-mode damage triggering mechanism establishes attribute associations between monitoring items, hazard items, and damage type confidence levels through a monitoring triggering mechanism and a hazard level assessment mechanism, respectively. The monitoring trigger mechanism establishes the following judgment logic: "Abnormal monitoring value → triggering over-limit threshold → calculating damage type confidence level". If the monitored value X > the set threshold α, Then activate the preset damage type confidence index calculation rules; The hazard level assessment mechanism establishes the following judgment logic: "Hazard item exists → Hazard level assessment is triggered → Damage type confidence is calculated": If the hazard item Y ∈ a certain hazard level interval Then activate the preset damage type confidence index calculation rules; S2.2.2 Based on the monitoring items and hidden danger items associated with each subclass instance of the component type, the dual-mode damage triggering mechanism calculates the damage type and its damage confidence of each subclass instance, and passes it to the reliability assessment layer through damage attributes. Damage attributes are structured data containing key fields such as damage type, confidence level, location of occurrence, component number involved, and first detection time. After calculating the damage type and damage confidence level of each subclass instance, the data is encapsulated in the form of damage attributes and transmitted to the reliability assessment layer in real time through a standardized data interface. The method includes three damage types: apparent damage, stress risk, and structural displacement. It uses Protege software to design a reliability assessment layer, enabling the calculation of reliability indices from individual components to the overall bridge structure. This process specifically includes: S2.3.1 Develop a reliability assessment rule base in the reliability assessment layer; S2.3.

2. Based on the rules in the reliability assessment rule base, calculate the structural resistance and load effect of each individual component; based on the load transfer path of the component, further calculate the local structural resistance and load effect caused by the component and its adjacent components; evaluate the entire bridge, integrate the data of each component, and calculate the overall structural resistance and load effect of the bridge structure. S2.3.3 Based on the rules in the reliability assessment rule base, and using professional knowledge in materials mechanics and engineering mechanics, damage evolution rules are designed for three types of damage: apparent damage, stress risk, and structural displacement. Then, the resistance reduction factor and load amplification factor of the component under different damage types are determined to quantify the impact of damage on structural performance. S2.3.4 Substitute the calculated structural resistance, load effect, and resistance reduction coefficient and load amplification coefficient determined by the damage evolution rule into the limit state equation for calculation. Use the limit state equation to determine the failure probability of the structure, thereby clarifying the reliability index of the structure and completing the safety assessment of a single component, a local structure, or the overall bridge structure.

2. The method for constructing a bridge component safety knowledge graph based on multi-category monitoring information and hidden danger information according to claim 1, characterized in that, Structural reliability studies the likelihood that a structure will perform its intended function under specified time and conditions. When analyzing structural reliability, the probability distribution of random variables in structural resistance and load effects is first determined. The limit state equation Z=RS≤0 is used to determine whether the structure has failed, where R represents structural resistance and S represents load effect. Then, the failure probability of Z≤0 is calculated, and the structural reliability is obtained by subtracting the failure probability of Z≤0 from 1. The structural reliability index β is used to quantify the structural reliability. The larger the index value, the more reliable the structure. Step S1 involves analyzing the variables in the structural resistance and load effects of the structure based on the continuously updated monitoring and hazard data, and establishing a structural reliability analysis framework based on multi-source data fusion.

3. The method for constructing a bridge component safety knowledge graph based on multi-category monitoring information and hidden danger information according to claim 1, characterized in that, Step S2 involves designing the structural physical layer using Protege software. This involves using structural components as carriers to analyze the force transfer path of structural resistance and load effects. This process requires creating component type classes and defining and associating geometric attributes and spatial topological relationships. Specifically, this includes: S2.1.1 Set geometric attributes, define the geometric attributes of different component types in the component type class, and define the subclass inheritance relationship of the component type class; realize the one-to-one mapping between the geometric attributes of the BIM model and the geometric attributes of the subclass instances of the component type class through the component ID attribute, and obtain unique structural space coordinates for each subclass instance to provide data support for the establishment of spatial topological relationships between different components; S2.1.2 Based on the obtained structural spatial coordinates, set spatial topology association attributes for each subclass instance of the component type class; by setting the spatial topology relationship of each subclass instance, realize the linkage effect of stress analysis between different components, and thus provide a data transmission path for the calculation of local structural reliability analysis or overall structural reliability analysis.

4. The method for constructing a bridge component safety knowledge graph based on multi-category monitoring information and hidden danger information according to claim 3, characterized in that, Step S2 involves designing an auxiliary decision-making layer using Protege software. This layer provides recommendations for preventive maintenance decisions for bridges through dynamically updated structural reliability indicators. Specifically, this process includes: S2.4.1, comprehensively collecting technical condition indicators related to the structure and determining reliability indicators from them. S2.4.2 Based on professional knowledge in materials mechanics and structural mechanics, analyze the influence mechanism of technical condition indicators on reliability indicators, and establish quantitative or qualitative correlation rules between technical condition indicators and reliability indicators through experimental research, numerical simulation or historical data statistical analysis. S2.4.

3. Classify preventive maintenance recommendations into different types, and clarify the applicable conditions and maintenance objectives of each type; determine the corresponding reliability index range based on different structural technical conditions and the developed correlation rules; match the determined reliability index range with the preventive maintenance recommendation types to establish a one-to-one correspondence. S2.4.4 Obtain information on the actual technical condition of the structure through on-site testing or monitoring data collection, and evaluate the technical condition of the structure according to pre-set standards and methods to draw maintenance conclusions; based on the maintenance conclusions, find the correspondence between the established technical condition, reliability indicators and preventive maintenance recommendations, determine preventive maintenance recommendations that match the current technical condition and reliability indicators of the structure, and provide a scientific basis for the maintenance decision of the structure.

Images