Semantic path network modeling and application method based on multi-source data integration

Abstract

The invention discloses a semantic path network modeling and application method based on multi-source data integration, and relates to the technical field of path network construction and application, and the method comprises the steps: obtaining multi-source data of a target scene; according to a preset mapping rule, performing ontology mapping on the BIM data, the GIS data and the target sensor data to construct a semantic path ontology oriented to disaster rescue; constructing a knowledge graph according to the semantic path ontology to obtain a semantic path network; and realizing disaster rescue related information query and semantic path planning by using the semantic path network, thereby realizing improvement of the accuracy of the planning path and improvement of the evacuation / rescue efficiency.

Description

Technical Field

[0001] This application relates to the field of path network construction and application technology, and in particular to a semantic path network modeling and application method based on multi-source data integration. Background Technology

[0002] With the acceleration of urbanization, high-density building complexes such as subway hubs, underground shopping streets, and super high-rise buildings are increasing. Their internal structures are complex and densely populated. Once disasters such as fires, floods, and structural collapses occur, the traditional emergency command model that relies on human experience and static drawings is difficult to achieve efficient and accurate personnel evacuation and rescue resource scheduling.

[0003] In recent years, the development of smart city and digital twin technologies has provided new approaches to disaster relief. Building Information Modeling (BIM) can meticulously describe a building's interior geometry, spatial topology, and component attributes; Geographic Information Systems (GIS) excel at managing large-scale outdoor geospatial data and network relationships; and Internet of Things (IoT) technology, through the deployment of various sensors, can perceive disaster conditions (such as smoke and water levels) and the dynamic locations of people in real time. Theoretically, integrating BIM, GIS, and IoT data can construct a digital environment that covers both indoor and outdoor areas, combining static and dynamic data, providing support for route planning.

[0004] However, BIM, GIS and IoT data are fragmented in the data layer, resulting in significant semantic heterogeneity, which may lead to inaccurate path planning. Summary of the Invention

[0005] The main purpose of this application is to provide a semantic path network modeling and application method based on multi-source data integration, which aims to solve the technical problem of low accuracy when using existing path networks for path planning.

[0006] To achieve the above objectives, this application proposes a semantic path network modeling and application method based on multi-source data integration, including: Acquire multi-source data of the target scene, including BIM data, GIS data, and target sensor data; According to preset mapping rules, the BIM data, the GIS data, and the target sensor data are ontology mapped to construct a semantic path ontology for disaster relief. A knowledge graph is constructed based on the semantic path ontology to obtain a semantic path network; The semantic path network is used to realize disaster relief related information query and semantic path planning.

[0007] In one embodiment, the semantic path planning includes indoor path planning and outdoor path planning, and the path includes at least one of rescue path and evacuation path.

[0008] In one embodiment, constructing a knowledge graph based on the semantic path ontology to obtain a semantic path network includes: Construct triples based on the semantic path ontology and its corresponding attributes; Each triple is stored in a knowledge graph database to obtain a semantic path network; When the data from the target sensor changes, the corresponding triplet is updated to obtain the updated semantic path network. The process of utilizing the semantic path network to query disaster relief-related information and plan semantic paths includes: By utilizing the updated semantic path network, disaster relief-related information can be queried and semantic path planning can be performed.

[0009] In one embodiment, updating the corresponding triples to obtain the updated semantic path network when the data from the target sensor changes includes: When the change value of the target sensor is greater than the change threshold, the corresponding triplet is updated to obtain the updated semantic path network.

[0010] In one embodiment, the attributes of the semantic path ontology include path length, risk value, and distance to rescue resources; The step of utilizing the updated semantic path network to achieve semantic path planning includes: Semantic path planning is achieved based on path length, risk value, and distance to rescue resources.

[0011] In one embodiment, after the step of acquiring multi-source data of the target scene, the method further includes: The multi-source data is spatiotemporally aligned.

[0012] In one embodiment, when the target scenario is a fire scenario, the target sensor includes a smoke sensor, and the method further includes: Based on the concentration information collected by each smoke sensor, the smoke diffusion trajectory is predicted; Based on the semantic path network and smoke diffusion trajectory, smoke avoidance evacuation routes and / or backsmoke rescue routes are calculated.

[0013] In one embodiment, when the target scenario is a flood scenario, the target sensor includes a water level sensor, and the method further includes: Based on the water level information collected by each water level sensor, links with water levels higher than the water level threshold are deleted.

[0014] In one embodiment, the semantic path ontology includes pedestrian links, emergency exits, trapped individuals, rescuers, fire stations, and temporary shelters; The attributes of the pedestrian network include length, width, evacuation capacity, and damage status; The attributes of the emergency exit include location coordinates, opening status, service range, and evacuation efficiency, wherein the evacuation efficiency is dynamically calculated based on the population density. The attributes of the trapped individuals include their location, number, health status, and trajectory trends; The attributes of the rescue personnel include their team affiliation, equipment type, and mission status; The attributes of the fire station include its location, number of fire trucks, and available resources; The attributes of the temporary shelter include capacity, supplies, and occupancy rate.

[0015] The semantic path network modeling and application method based on multi-source data integration proposed in this application has at least the following technical effects: A semantic path network modeling and application method based on multi-source data integration is proposed. This method acquires multi-source data of the target scenario; according to preset mapping rules, the BIM data, GIS data, and target sensor data are mapped ontology to construct a semantic path ontology for disaster relief; a knowledge graph is constructed based on the semantic path ontology to obtain a semantic path network; and the semantic path network is used to realize disaster relief-related information query and semantic path planning, thereby improving the accuracy of planned paths and the efficiency of evacuation / rescue. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the semantic path network modeling and application method based on multi-source data integration provided in this application.

[0019] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application. To better understand the technical solutions of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0021] This application provides a method for semantic path network modeling and application based on multi-source data integration.

[0022] In the first embodiment of the semantic path network modeling and application method based on multi-source data integration in this application, referring to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the semantic path network modeling and application method based on multi-source data integration of this application. The semantic path network modeling and application method based on multi-source data integration may include steps S10 to S40: Step S10: Obtain multi-source data of the target scene, including BIM data, GIS data and target sensor data.

[0023] It should be noted that the target scenario can include one of the following: a subway hub, an underground shopping street, or a building; BIM data can include one or more of the following: geometric parameters (width, height), functional attributes (fire resistance rating, evacuation capacity), and equipment relationships (such as the linkage status between exits and fire doors) of the target scenario's indoor spatial topology units (rooms, corridors, staircases, and emergency exits); GIS data can include the spatial coordinates and connectivity of the outdoor pedestrian network (sidewalks, overpasses, underpasses), as well as the location and resource availability of surrounding rescue resources (fire stations, hospitals); target sensors can include positioning sensors and disaster monitoring sensors corresponding to the target scenario, such as smoke sensors in a fire scenario and water level sensors in a flood scenario. Accordingly, target sensor data can include the location of trapped persons, the location of rescue personnel, and the location and monitoring data of disaster monitoring sensors.

[0024] In one feasible implementation, after the step of acquiring multi-source data of the target scene, the multi-source data can be preprocessed to make the multi-source data spatiotemporally aligned.

[0025] Step S20: According to the preset mapping rules, the BIM data, the GIS data and the target sensor data are mapped ontology to construct a semantic path ontology for disaster relief.

[0026] It should be noted that the semantic path ontology can include three main categories: personnel and rescue resources.

[0027] In one feasible implementation, the semantic path ontology includes pedestrian links, emergency exits, trapped individuals, rescue personnel, fire stations, and temporary shelters; the attributes of the pedestrian network include length, width, evacuation capacity, and damage status; the attributes of the emergency exits include location coordinates, open status, service area, and evacuation efficiency, wherein the evacuation efficiency is dynamically calculated based on population density; the attributes of the trapped individuals include location, number, health status, and trajectory trend; the attributes of the rescue personnel include their team affiliation, equipment type, and mission status; the attributes of the fire station include location, number of fire trucks, and resource reserves; and the attributes of the temporary shelters include capacity, material reserves, and occupancy rate.

[0028] In one feasible implementation, mapping rules are defined using SWRL, which may include, for example, the following mapping rules: 1. Semantic rules for BIM and GIS static data: If the Door attribute in BIM is FireRating=A and Function=Escape, then it is mapped to the ontology EmergencyExit. Here, Door is defined as a door, FireRating is defined as the fire rating, Function=Escape is defined as the function of escape, and EmergencyExit is defined as an emergency exit. 2. IoT dynamic data rules: If Concentration ≥ 5% VOL and is associated with "Corridor" in BIM data, then PedestrianLink's DamageState will be mapped to partial blockage, and RiskValue will be mapped to 8. Here, Concentration is defined as smoke concentration, Corridor is defined as corridor or passage, DamageState is defined as damage state, and RiskValue is defined as risk value.

[0029] Step S30: Construct a knowledge graph based on the semantic path ontology to obtain a semantic path network.

[0030] In one feasible implementation, step S30 may include steps A11-A12: Step A11: Construct triples based on the semantic path ontology and its corresponding attributes.

[0031] For example, "Evacuee (the trapped person)", "locatedAt (location)" and "PedestrianLink 005 (one of the walking links)" can be constructed as a triple.<Evacuee,locatedAt,PedestrianLink 005> ; Construct a triple from "PedestrianLink 004 (one of the walking links)", "RiskValue (the risk value of PedestrianLink004)" and "8>".<PedestrianLink 004,RiskValue,8> >

[0032] Step A12: Store each triple in the knowledge graph database to obtain the semantic path network.

[0033] In one feasible implementation, each triple is stored in a knowledge graph database that supports spatiotemporal indexing, and a SPARQL query template is constructed to obtain a semantic path network.

[0034] For example, the query template "Query PedestrianLink with current RiskValue<3 and Capacity≥50 people" is used to query pedestrian links with a risk value of less than 3 and an evacuation capacity of more than 50 people.

[0035] In another feasible implementation, step S30 may also include step A13 following step A12: when the data of the target sensor changes, update the corresponding triplet to obtain the updated semantic path network.

[0036] Accordingly, step S40 is: using the updated semantic path network to realize disaster relief related information query and semantic path planning.

[0037] In one specific implementation, step A13 may include step A131: Step A131: When the change value of the target sensor is greater than the change threshold, update the corresponding triplet to obtain the updated semantic path network.

[0038] For example, the target sensors include a water level sensor and a trapped person location sensor. When the water level changes by more than 10 cm, or the trapped person's location changes by more than 5 m, the corresponding triples are updated to obtain an updated semantic path network.

[0039] It should be noted that this implementation adopts an event-triggered update mechanism to complete local triple updates without requiring a full reconstruction of the knowledge graph.

[0040] Step S40: Utilize the semantic path network to realize disaster relief related information query and semantic path planning.

[0041] It should be noted that the relevant information query includes, but is not limited to: querying the real-time risk level of a specific location or area in the target scene based on the semantic path ontology, the distribution and status of available rescue resources (such as fire-fighting equipment, medical points, and temporary shelters), the accessibility of safety exits, and the approximate distribution of trapped personnel. This query function provides dynamic and multi-dimensional decision-making basis for subsequent semantic path planning. Semantic path planning includes indoor path planning and outdoor path planning, and the path includes at least one of rescue paths and evacuation paths. Rescue paths are used to guide rescue personnel to carry out rescue operations, and evacuation paths are used to guide trapped personnel to evacuate from the target scene.

[0042] In one feasible implementation, the attributes of the semantic path ontology include path length, risk value, and distance to rescue resources; step S40 may include step S401: Semantic path planning is achieved based on path length, risk value, and distance to rescue resources.

[0043] It should be noted that this embodiment provides a multi-objective weighted path algorithm, which is used to calculate the optimal evacuation and rescue path based on path length, risk value and distance to rescue resources.

[0044] Specifically, a 30% weight is assigned to path length, a 40% weight to risk value, and a 30% weight to the distance to rescue resources. This approach differs from static planning methods based on path length and can improve the accuracy of path planning.

[0045] In one feasible implementation, the target scenario is a fire scenario, and the target sensor includes a smoke sensor; the semantic path network modeling and application method based on multi-source data integration further includes step S501: Based on the concentration information collected by each smoke sensor, the smoke diffusion trajectory is predicted; based on the semantic path network and the smoke diffusion trajectory, smoke avoidance evacuation paths and / or backsmoke rescue paths are calculated.

[0046] It should be noted that the semantic path network modeling and application method based on multi-source data integration in this embodiment can also be used to guide trapped personnel to smoke-proof stairwells.

[0047] In one feasible implementation, the target scenario is a flood scenario, and the target sensor includes a water level sensor; the semantic path network modeling and application method based on multi-source data integration further includes step S502: Based on the water level information collected by each water level sensor, links with water levels higher than the water level threshold are deleted.

[0048] It should be noted that the semantic path network modeling and application method based on multi-source data integration in this embodiment can also be used to plan "water-crossing capability-adaptive paths" for rescue teams (such as links where fire trucks can only pass through water levels ≤ 50cm) and link temporary shelters to issue "capacity warnings".

[0049] This embodiment presents a semantic path network modeling and application method based on multi-source data integration. It acquires multi-source data of the target scene; according to preset mapping rules, it performs ontology mapping on the BIM data, GIS data, and target sensor data to construct a semantic path ontology for disaster relief; based on the semantic path ontology, it constructs a knowledge graph to obtain a semantic path network with spatiotemporal reasoning capabilities; using the semantic path network, it enables disaster relief-related information querying and semantic path planning, improving the accuracy of planned paths and evacuation / rescue efficiency. It should be noted that it can also support the collaborative scheduling of rescue resources.

[0050] For example, to help understand the implementation process of the semantic path network modeling and application method based on multi-source data integration in the embodiments of this application, an application example of the semantic path network modeling and application method based on multi-source data integration is provided. Specifically, the application example is a subway hub fire rescue scenario, and the semantic path network modeling and application method based on multi-source data integration may include steps S1001 to S1006: Step S1001: Extract the geometric and attribute data of the station hall (area 2000㎡, evacuation capacity 1500 people), platforms (2, each with 3 corridors), emergency staircases (4, width 1.8m), and fire exits (8, fire resistance rating A) from the BIM model (LOD400) of the subway hub; obtain the spatial coordinates of the pedestrian walkways (3), overpasses (1), and fire station (1, 800m from the hub, equipped with 3 fire trucks) within a 500m radius of the subway hub; extract the UWB positioning sensors (deployed in the station hall / platform, 50 in total): collect the real-time location of trapped personnel (initially 200 people), and achieve a positioning accuracy of 0.3m after Kalman filtering; smoke sensors (deployed on the top of the corridor, 30 in total). Step S1002: According to the SWRL rules, map the BIM “fire exit” to the ontology “EmergencyExi”, and the attribute “fire resistance rating A” corresponds to “DamageResistance=high”; associate the smoke sensor data “concentration 6%VOL” to the GIS “east corridor of the station hall”, and map it to the ontology “PedestrianLink_003” with “RiskValue=8” and “DamageState=partial blockage”. Step S1003, generate triples such as <Evacuee 001, locatedAt, PedestrianLink 003>, <PedestrianLink 003, RiskValue risk value, 8>, <EmergencyExit 001, connectedTo, PedestrianLink 004>, etc., and store them in the Neo4j database to obtain the semantic path network.

[0051] Step S1004, at T0 + 30s, when the smoke sensor detects that the concentration rises to 8%VOL, the system automatically updates <PedestrianLink 003, RiskValue, 10> and <PedestrianLink 003, DamageState>, and deletes "connectedTo between PedestrianLink 003 and EmergencyExit 001"; Step S1005, for the 50 trapped people on PedestrianLink 003, call the relevant information to query the "multi - target weighted algorithm": exclude the pedestrian link (PedestrianLink 003) with a RiskValue risk value ≥ 8; Prefer the path with "RiskValue risk value = 2 (western corridor), Capacity = 100 people (meeting the evacuation demand), and only 30m away from EmergencyExit 002".

[0052] Step S1006, push the dynamic path of "western corridor → EmergencyExit emergency exit_002" to the terminals of the trapped people (updated every 10s); Push "the area where the trapped people gather (PedestrianLink 004)" and "the opening status of the emergency exit" to the terminals of the rescue team; Display "real - time personnel evacuation progress (120 people have been evacuated), location of rescue resources (the fire truck is 500m away from the hub)" on the large - screen of the command center to support global scheduling.

[0053] It should be noted that the above examples are only used to understand this application and do not constitute a limitation on the semantic path network modeling and application method based on multi - source data integration of this application. Any simple transformation in more forms based on this technical concept is within the protection scope of this application.

[0054] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for semantic path network modeling and application based on multi-source data integration, characterized in that, include: Acquire multi-source data of the target scene, including BIM data, GIS data, and target sensor data; According to preset mapping rules, the BIM data, the GIS data, and the target sensor data are ontology mapped to construct a semantic path ontology for disaster relief. A knowledge graph is constructed based on the semantic path ontology to obtain a semantic path network; The semantic path network is used to realize disaster relief related information query and semantic path planning.

2. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, The semantic path planning includes indoor path planning and outdoor path planning, and the path includes at least one of rescue path and evacuation path.

3. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, The step of constructing a knowledge graph based on the semantic path ontology to obtain a semantic path network includes: Construct triples based on the semantic path ontology and its corresponding attributes; Each triple is stored in a knowledge graph database to obtain a semantic path network; When the data from the target sensor changes, the corresponding triplet is updated to obtain the updated semantic path network. The process of utilizing the semantic path network to query disaster relief-related information and plan semantic paths includes: By utilizing the updated semantic path network, disaster relief-related information can be queried and semantic path planning can be performed.

4. The semantic path network modeling and application method based on multi-source data integration as described in claim 3, characterized in that, The step of updating the corresponding triplet to obtain the updated semantic path network when the data of the target sensor changes includes: When the change value of the target sensor is greater than the change threshold, the corresponding triplet is updated to obtain the updated semantic path network.

5. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, The attributes of the semantic path ontology include path length, risk value, and distance to rescue resources; The step of utilizing the updated semantic path network to achieve semantic path planning includes: Semantic path planning is achieved based on path length, risk value, and distance to rescue resources.

6. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, After the step of acquiring multi-source data of the target scene, the method further includes: The multi-source data is spatiotemporally aligned.

7. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, When the target scenario is a fire scenario, the target sensor includes a smoke sensor, and the method further includes: Based on the concentration information collected by each smoke sensor, the smoke diffusion trajectory is predicted; Based on the semantic path network and smoke diffusion trajectory, smoke avoidance evacuation routes and / or backsmoke rescue routes are calculated.

8. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, When the target scenario is a flood scenario, the target sensor includes a water level sensor, and the method further includes: Based on the water level information collected by each water level sensor, links with water levels higher than the water level threshold are deleted.

9. The semantic path network modeling and application method based on multi-source data integration as described in claim 1, characterized in that, The semantic path ontology includes pedestrian links, emergency exits, trapped individuals, rescuers, fire stations, and temporary shelters; The attributes of the pedestrian network include length, width, evacuation capacity, and damage status; The attributes of the emergency exit include location coordinates, opening status, service range, and evacuation efficiency, wherein the evacuation efficiency is dynamically calculated based on the population density. The attributes of the trapped individuals include their location, number, health status, and trajectory trends; The attributes of the rescue personnel include their team affiliation, equipment type, and mission status; The attributes of the fire station include its location, number of fire trucks, and available resources; The attributes of the temporary shelter include capacity, supplies, and occupancy rate.

Images