An emergency command system and method based on BIM digital base

Abstract

The present invention relates to the technical field of emergency command, and discloses an emergency command system and method based on a BIM digital base. The system comprises the following steps: establishing a unified data model by extending the IFC standard, semantically mapping and entity aligning BIM building information, GIS geospatial data, monitoring data, and emergency plan documents, generating an emergency knowledge graph with BIM building information as nodes, coupling physical simulation with an LSTM prediction algorithm, injecting monitoring data into a BIM model to deduce the disaster process, outputting a risk heat map and an evacuation restricted area, calculating the spatiotemporal accessibility of emergency resources through a reinforcement learning model based on the deduction results, generating an optimization plan including drone paths, personnel evacuation routes, and equipment scheduling instructions, synchronizing the instructions to an augmented reality individual terminal and a mobile command cabin through an edge computing node, annotating resource deployment status in the BIM model in real time, synchronizing multi-terminal operations, significantly improving the timeliness and accuracy of disaster situation awareness, and effectively reducing disaster losses.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency command, and discloses an emergency command system and method based on a BIM digital base. Background Art

[0002] In the areas of smart urban management and emergency response, traditional emergency management models face technical bottlenecks such as difficulty in coordinating multi-source heterogeneous data, lagging disaster situation simulations, and extensive emergency resource scheduling. While BIM (Building Information Modeling) can accurately describe building structures and spatial information, it lacks effective integration with GIS geospatial data and real-time monitoring data. Emergency plan documents are often unstructured, making them difficult to quickly link to actual scenarios. Existing disaster simulations rely on single physical simulations or statistical models, unable to integrate with real-time monitoring data for dynamic predictions. Emergency resource scheduling is often based on empirical decisions and lacks the ability to accurately respond to dynamic changes in time and space. Summary of the Invention

[0003] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0004] In order to solve the above technical problems, the main purpose of the present invention is to provide an emergency command method based on a BIM digital base, comprising:

[0005] By extending the IFC standard to establish a unified data model, semantic mapping and entity alignment are performed on BIM building information, GIS geospatial data, monitoring data, and emergency plan documents to generate an emergency knowledge graph with BIM building information as the node;

[0006] The knowledge graph identifies disaster association chains, couples the physical simulation engine with the LSTM prediction algorithm, injects monitoring data into the BIM model to deduce the disaster process, and outputs risk heat maps and evacuation restricted areas;

[0007] Based on the simulation results, the spatial and temporal accessibility of emergency resources is calculated through a reinforcement learning model, generating an optimization plan that includes drone paths, personnel evacuation routes, and equipment dispatch instructions.

[0008] Through edge computing nodes, commands are synchronized to augmented reality individual terminals and mobile command cabins, resource deployment status is marked in real time in the BIM model, and multi-terminal operations are synchronized.

[0009] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0010] Establishing the unified data model includes adding spatial topological attributes to components in BIM building information, coordinate matching GIS geospatial data with BIM building information, defining exclusive dynamic entity categories for monitoring data, and binding them to the operation and maintenance attribute sets of building components through unique device codes;

[0011] Build a distributed semantic query channel across building models, geographic data, and IoT monitoring, use artificial intelligence to analyze the semantic correlation between emergency plan text keywords and building component functions, and complete alignment after reaching the preset matching standards.

[0012] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0013] The method for generating an emergency knowledge graph includes:

[0014] S11 builds a spatial topological network based on the physical adjacency relationship of components;

[0015] S12 establishes the functions and linkage links between components according to the emergency plan;

[0016] S13 builds a probability prediction model for secondary disasters caused by component failure based on historical disaster data;

[0017] S14 converts real-time IoT monitoring data into status attribute parameters of building components. When the monitoring data is abnormal, it automatically increases the risk level of the associated components in the map.

[0018] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0019] The method of coupling the physical simulation engine and the LSTM prediction algorithm to deduce the disaster process includes:

[0020] S21 establishes a dual-engine collaboration mechanism, wherein the physical simulation engine constructs an initial propagation path based on the physical laws corresponding to the disaster type, and the prediction algorithm dynamically corrects the parameter deviation of the physical simulation engine by analyzing historical disaster time series data;

[0021] S22 real-time data-driven update maps the IoT monitoring data stream into the dynamic state variables of BIM building information in real time, and drives the synchronous iterative optimization of the physical simulation engine and LSTM prediction algorithm through edge computing nodes.

[0022] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0023] The dynamic correction includes, when the deviation between the real-time monitoring data and the simulation output exceeds a preset threshold, the prediction algorithm adjusts the boundary condition parameters of the physical simulation engine to form a closed-loop feedback;

[0024] The synchronous iterative optimization includes continuously outputting disaster evolution prediction results at a frequency based on the updated BIM building information state variables.

[0025] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0026] The method for calculating the spatiotemporal accessibility of emergency resources by using a reinforcement learning model includes:

[0027] S31 constructs the risk heat map, evacuation restricted area and traffic network status generated by disaster simulation into a three-dimensional spatiotemporal state space;

[0028] S32 defines movement constraint rules for emergency resources in the state space;

[0029] S33 uses the shortest response time and maximum resource coverage as a joint reward function to train multiple targets, trains a reinforcement learning model through historical rescue data, and outputs a probability matrix of resources arriving at the target location.

[0030] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0031] The drone path planning generates a three-dimensional flight corridor that avoids restricted areas based on the spatiotemporal accessibility probability matrix, while dynamically adjusting the flight altitude to avoid interference from the disaster cyclone.

[0032] The personnel evacuation route is combined with the building structure safety level and real-time pedestrian flow data to calculate multi-exit diversion paths and plan low-intensity escape channels for special groups;

[0033] The equipment scheduling instruction is generated by reversely deducing the optimal matching combination of equipment loading points and transportation vehicles based on the probability matrix of resources arriving at the target location, and generating equipment delivery sequence instructions with time windows.

[0034] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0035] The method for implementing instruction synchronization through edge computing nodes includes:

[0036] S41 transmits lightweight BIM model slices and resource positioning instructions to the augmented reality individual terminal, and transmits the full BIM model and a 3D disaster situation sand table for disaster simulation to the mobile command cabin;

[0037] S42 state synchronization maps resource deployment state changes to BIM component attribute datasets in real time, reducing network transmission load through incremental update protocols.

[0038] As a preferred solution of the emergency command method based on the BIM digital base of the present invention, wherein:

[0039] The synchronous multi-terminal operation in the disconnected environment includes edge pre-storage and resume transmission, which is consistent with offline operation;

[0040] The edge pre-storage and retransmission pre-store key BIM component spatial data and resource deployment baseline status through edge computing nodes. If the public network is interrupted, the basic operation synchronization between terminals is maintained based on the pre-stored data;

[0041] If the network is disconnected, the timestamp and spatial coordinates of each terminal operation instruction are recorded. If the network is restored, the operation records are merged through a conflict resolution algorithm to reconstruct the offline operation consistency.

[0042] As a preferred solution of the emergency command system based on the BIM digital base of the present invention, wherein:

[0043] Data fusion module, disaster simulation module, plan generation module and instruction execution module;

[0044] The data fusion module is used to preprocess multi-source data and fuse them into an emergency knowledge graph;

[0045] The disaster simulation module is used to analyze disaster factors and simulate the process, and output a risk map;

[0046] The solution generation module is used to determine the accessibility of emergency resources and generate a scheduling solution;

[0047] The instruction execution module includes an instruction synchronization unit and a resource monitoring unit, wherein the instruction synchronization unit is used for edge computing transmission instructions, and the resource monitoring unit marks the deployment status in real time.

[0048] Beneficial effects of the present invention:

[0049] By constructing a unified data model based on the IFC standard, the present invention realizes the semantic integration and entity alignment of multi-source data such as BIM, GIS and emergency plans, forms an emergency knowledge graph with BIM components as the core, and provides a structured knowledge network for disaster analysis.

[0050] This application uses knowledge graphs to identify disaster association chains and reinforce learning to efficiently inject real-time monitoring data into BIM models, enabling accurate deduction of disaster progress and significantly improving the timeliness and accuracy of disaster situation awareness. By optimizing emergency resources through reinforcement learning, it can quickly generate optimal solutions for drone path planning, personnel evacuation routes, and equipment scheduling, effectively reducing disaster losses.

[0051] This application uses edge computing and augmented reality to achieve real-time synchronization and visual presentation of instructions on multiple terminals, ensuring dynamic monitoring and collaborative operation of emergency resource deployment status. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0053] Figure 1 This is a flow chart of an emergency command method based on a BIM digital base according to the present invention;

[0054] Figure 2 This is a composition diagram of an emergency command system based on a BIM digital base in the present invention;

[0055] Figure 3 A method for generating an emergency knowledge graph for an emergency command method based on a BIM digital base of the present invention;

[0056] Figure 4 The present invention provides a method for deducing disaster progress by coupling an emergency command method based on a BIM digital base with a physical simulation engine and an LSTM prediction algorithm. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0060] Example 1

[0061] like Figure 1 As shown, an emergency command method based on a BIM digital base includes:

[0062] By extending the IFC standard to establish a unified data model, semantic mapping and entity alignment are performed on BIM building information, GIS geospatial data, monitoring data, and emergency plan documents to generate an emergency knowledge graph with BIM building information as the node;

[0063] Establishing the unified data model includes adding spatial topological attributes to components in the BIM building information, coordinate matching GIS geospatial data with BIM building information, aligning the coordinate systems of GIS geographic entities and BIM models through a coordinate transformation matrix, defining exclusive dynamic entity categories for monitoring data, and binding them to the operation and maintenance attribute sets of building components through unique device codes;

[0064] Build a distributed semantic query channel across building models, geographic data, and IoT monitoring, use artificial intelligence to analyze the semantic correlation between emergency plan text keywords and building component functions, and complete alignment after reaching the preset matching standards.

[0065] In this application, a specific implementation method for artificial intelligence analysis of the semantic relevance between emergency plan text keywords and building component functions includes:

[0066] First, we can use natural language processing to analyze the text content of the emergency plan, identify keywords, understand the real intention and referent of the keywords in the emergency scenario, and mark the building components in the knowledge graph based on the real intention and referent of the emergency scenario.

[0067] Furthermore, the building components in the knowledge graph of the real intention and the referred object in the emergency scenario are converted into vectors in the same semantic space, and the vectors in the semantic space are converted into points in the mathematical space, so that the semantically similar ones are close in position in the vector space.

[0068] The semantic vectors of the key actions, objects and their context combinations identified in the emergency plan are compared with the semantic vectors of the functional descriptions of all building components in the knowledge graph. By calculating their cosine similarity in a high-dimensional vector space, the semantic association strength between each plan element and each building component function is quantified.

[0069] Furthermore, a preset matching threshold standard is set. When the semantic correlation between a plan element and the functional description of a building component exceeds the matching threshold standard, the two are highly semantically related, and a semantic link is automatically established in the emergency knowledge graph, binding the plan steps or requirements to the corresponding building component nodes, ensuring that in emergency scenarios, the plan instructions can be accurately associated with the actual facilities in the physical space.

[0070] Specifically, the IFC standard was established to solve the heterogeneity problems of BIM, GIS, and data formats, and to establish a foundation for cross-domain data interoperability.

[0071] A new set of spatial topology attributes has been added to the IFC standard architecture. The attribute set includes components’ spatial adjacency relationships, boundary connection points, etc., enabling BIM components to describe complex spatial relationships, create dynamic entity classes such as sensor entities and emergency plan entities, and incorporate unstructured data such as streaming data and text documents into the IFC entity framework.

[0072] Since there are scale and direction deviations between the BIM local coordinate system and the GIS global coordinate system, spatial alignment is performed through multi-source data.

[0073] Specifically, a spatial transformation matrix is ​​constructed by combining translation / rotation / scaling parameters, and the latitude, longitude and elevation coordinates of the GIS geographic entity are matched to the local coordinates of the BIM component through the spatial transformation matrix;

[0074] The topological properties of BIM components are used to establish geometric consistency associations with GIS network nodes.

[0075] Dynamic data binding is used to achieve real-time mapping between physical devices and digital models.

[0076] Specifically, a globally unique code is assigned to each device: device ID + location tag; the device code is associated with the functional properties of the BIM component through the IFC attribute set.

[0077] Semantic mapping and entity alignment Specifically, lightweight query agents are deployed in original data storage nodes such as BIM servers, GIS platforms, and databases. The agent nodes are coordinated through a learning framework to complete joint semantic retrieval without moving the original data.

[0078] Cross-modal semantic associations are used to parse keywords in emergency plan documents, extract elements such as action subjects and operation instructions, identify components with corresponding functions in the BIM model, analyze the behavior patterns of components in historical events through graph neural networks, and calculate the semantic similarity between text instructions and component functions. When the similarity exceeds the preset threshold, the binding relationship between text instructions and BIM components is automatically established.

[0079] Emergency knowledge graph generation includes graph architecture and dynamic knowledge;

[0080] Among them, the graph architecture includes spatial topology chain, functional collaboration chain and risk transmission chain;

[0081] The spatial topology chain generates a component adjacency network based on IFC spatial relationship attributes;

[0082] The functional coordination chain builds functional linkage relationships between components based on the alignment results of the emergency plan, such as the pressure supply chain of fire pumps, water pipes, and sprinkler heads;

[0083] The risk transmission chain integrates historical disaster cases to establish probabilistic impact paths of component failure and secondary disasters, such as the probability of electrical short circuit caused by pipeline leakage.

[0084] Specifically, dynamic knowledge includes real-time driving and graph optimization;

[0085] Furthermore, the real-time driver converts the monitoring data stream into component node state variables. When the state variables exceed the safety threshold, the risk propagation calculation is triggered along the relationship chain:

[0086] Furthermore, new disaster event data is automatically accumulated into a historical case library to optimize the risk transmission map.

[0087] By establishing a unified data model and applying multi-source data spatial alignment and dynamic data binding, semantic mapping and entity alignment of BIM building information, GIS geospatial data, monitoring data and emergency plan documents can be achieved; through natural language processing and semantic vector calculation, the semantic correlation between emergency plan text and building component functions is analyzed and binding is completed; an emergency knowledge graph including spatial topology chain, functional collaboration chain and risk transmission chain is constructed, and combined with real-time data drive and historical case optimization mechanism, cross-domain data interoperability, intelligent emergency response and dynamic risk assessment can be achieved.

[0088] like Figure 3 As shown, the method for generating an emergency knowledge graph includes:

[0089] S11 builds a spatial topological network based on the physical adjacency relationship of components;

[0090] Analyze the physical adjacency relationships between components, such as the connection between walls and floors, and the interface between pipes and equipment, and construct a three-dimensional spatial relationship network based on topological principles;

[0091] Furthermore, the three-dimensional relationship network explicitly expresses the adjacency relationship of "component A is directly connected to component B";

[0092] The implicit deduction of the three-dimensional relationship network is the transmission path of "component C indirectly affects component E through component D".

[0093] S12 establishes the functions and linkage links between components according to the emergency plan;

[0094] Functional linkage link generation includes emergency plan analysis, function mapping matching, and linkage link generation;

[0095] Specifically, emergency plan parsing is used to identify action subjects and operation instructions in documents;

[0096] Functional mapping matching searches for entities with corresponding functions in the BIM component library and verifies their execution capabilities based on the component operation and maintenance history data;

[0097] The linkage link generation establishes a functional transfer chain of instructions, execution components, and benefit components, such as starting a fire pump, pressurizing a water pipe, and discharging water from a sprinkler head.

[0098] S13 builds a probability prediction model for secondary disasters caused by component failure based on historical disaster data;

[0099] Disaster transmission probability models include data-driven modeling and dynamic prediction;

[0100] Among them, data-driven modeling is used to include component failure cases, such as short circuits caused by pipeline ruptures, count the frequency of secondary disasters, introduce environmental variable weights, and detect in real time the probability changes of disasters caused by environmental changes.

[0101] Dynamic prediction: When a component is in an abnormal state, conduction calculations are triggered along the spatial topological network to predict the probability of a disaster. For example, the probability of a pipeline leak is 92%, which will cause an electrical short circuit and an 85% probability of causing a transformer explosion.

[0102] S14 converts real-time IoT monitoring data into state attribute parameters of building components. When monitoring data is abnormal, it automatically increases the risk level of related components in the map, realizing the dynamic spread of the disaster impact chain.

[0103] Specifically, the IoT monitoring values ​​are converted into component status parameters, and parameter safety thresholds are preset;

[0104] Furthermore, the dynamic diffusion of risk includes primary response and secondary transmission;

[0105] The primary response is an increase in the risk level of the abnormal component itself;

[0106] Secondary transmission increases the risk of adjacent components along the spatial topological network, increases the risk of function-dependent components along the functional linkage chain, and increases the risk of secondary disaster target components according to the disaster probability model.

[0107] The knowledge graph identifies disaster association chains, couples physical simulation with the LSTM prediction algorithm, injects monitoring data into the BIM model to deduce the disaster process, and outputs risk heat maps and evacuation restricted areas;

[0108] like Figure 4 As shown, the method of coupling the physical simulation engine and the LSTM prediction algorithm to deduce the disaster process includes:

[0109] S21 establishes a dual-engine collaboration mechanism, wherein the physical simulation engine constructs an initial propagation path based on the physical laws corresponding to the disaster type, and the prediction algorithm dynamically corrects the parameter deviation of the physical simulation engine by analyzing historical disaster time series data;

[0110] Physical simulation and prediction algorithms collaborate to select physical laws based on the type of disaster to build the basic propagation, including gas and liquid leakage, and calculate the diffusion path and concentration gradient based on fluid mechanics; structural damage applies material mechanics to simulate the chain reaction caused by the failure of load-bearing components, and the initial path generation is combined with BIM spatial topology data to set the initial boundary conditions.

[0111] The dynamic correction principle of the prediction algorithm analyzes historical disaster time series data, such as temperature change curves and structural deformation records, to identify uncovered actual interference factors, such as sudden wind direction changes and equipment resonance. When the deviation between real-time monitoring data and the simulation output path is greater than the threshold, the physical engine parameters are dynamically adjusted.

[0112] S22 real-time data-driven update maps the IoT monitoring data stream into the dynamic state variables of BIM building information in real time, and drives the synchronous iterative optimization of the physical simulation engine and LSTM prediction algorithm through edge computing nodes.

[0113] When public network communication is interrupted, the system continuously monitors the network status through the dual-mode link detection module, immediately triggers the activation of the Ka satellite communication channel, and the edge computing node calls the pre-stored key BIM component spatial data package and resource deployment baseline status, and pushes lightweight incremental data packages to the augmented reality individual terminal and mobile command cabin through the satellite link; the terminal builds a local emergency command environment based on the pre-stored data, and integrates the newly received incremental package with the local baseline status through the spatiotemporal consistency engine, maintaining the synchronization of basic operation instructions during the public network interruption, such as updating restricted area markings and fine-tuning resource locations. Through the triple mechanism of seamless link switching, key data preloading, and baseline status anchoring, it ensures that command instructions continue to take effect in extreme environments.

[0114] Specifically, the IoT monitoring values ​​are converted into BIM component status parameters, for example, pipeline pressure sensor data is mapped into the stress load attributes of BIM pipeline components;

[0115] Complete cleaning and format conversion at edge nodes close to the data source to avoid cloud transmission delays and ensure data freshness.

[0116] The dynamic correction includes, when the deviation between the real-time monitoring data and the simulation output exceeds a preset threshold, the prediction algorithm adjusts the boundary condition parameters of the physical simulation engine to form a closed-loop feedback;

[0117] The synchronous iterative optimization includes continuously outputting disaster evolution prediction results at a frequency based on the updated BIM building information state variables.

[0118] Based on the simulation results, the spatial and temporal accessibility of emergency resources is calculated through a reinforcement learning model, generating an optimization plan that includes drone paths, personnel evacuation routes, and equipment dispatch instructions.

[0119] The method for calculating the spatiotemporal accessibility of emergency resources by using a reinforcement learning model includes:

[0120] S31 constructs the risk heat map, evacuation restricted area and traffic network status generated by disaster simulation into a three-dimensional spatiotemporal state space;

[0121] The risk probability values ​​output by disaster simulation are mapped to three-dimensional spatial grid attributes. For example, the high-risk area grid is marked in red, and the prohibited areas are marked in the spatial grid. The road network capacity data is integrated in real time, and finally a time window attribute is added to each spatial grid.

[0122] S32 defines the movement constraint rules of emergency resources in the state space. For example, rescue vehicles cannot pass through water depths greater than 30 cm, and drones are prohibited from entering low-altitude areas when the wind speed is greater than 10 m / s.

[0123] S33 uses the shortest response time and maximum resource coverage as a joint reward function to train multiple targets, trains a reinforcement learning model through historical rescue data, and outputs a probability matrix of resources arriving at the target location.

[0124] Specifically, real-time methods for reinforcement learning models include multi-objective joint optimization and probability matrix generation;

[0125] Furthermore, multi-objective joint optimization includes the setting of reward functions and penalty terms;

[0126] Specifically, the shortest response time reward in the reward function increases the positive score each time the resource arrival time is advanced, and the maximum coverage rate reward is converted into a reward value by the proportion of the high-risk area covered per unit time.

[0127] In the present application, a preferred solution includes: setting penalty items including deducting huge points for entering restricted areas, triggering negative rewards due to excessive resource idle rates, etc.

[0128] The drone path planning generates a three-dimensional flight corridor that avoids restricted areas based on the spatiotemporal accessibility probability matrix, while dynamically adjusting the flight altitude to avoid interference from the disaster cyclone.

[0129] The drone path planning constructs a three-dimensional obstacle avoidance mechanism: horizontal obstacle avoidance generates a detour path based on the restricted area probability matrix, and vertical obstacle avoidance dynamically adjusts the flight altitude layer to avoid disaster cyclones.

[0130] The corridor is generated from the logical starting point to [safety height H] and finally reaches the end point of [bypass restricted area boundary].

[0131] The evacuation routes are calculated by combining the building structure safety level with real-time pedestrian flow data to calculate multi-exit diversion paths and plan low-intensity escape routes for special groups (injured people / elderly people);

[0132] The equipment scheduling instruction is generated by reversely deducing the optimal matching combination of equipment loading points and transportation vehicles based on the probability matrix of resources arriving at the target location, and generating equipment delivery sequence instructions with time windows.

[0133] Equipment dispatch instructions are generated by reversely inferring the resource demand time window. For example, if a fire extinguisher needs to arrive in area B within t minutes, the feasibility requirement of the transport vehicle and equipment loading point is reversed to [location B, time window t]. The output vehicle speed is matched to [loading point P1, departure time t+1min], and the equipment loading time is to [warehouse W1, dispatch instruction tmin]. The reverse calculation is carried out one by one to the fire extinguisher loading time, and finally a complete equipment dispatch instruction is generated.

[0134] The instructions are synchronized to the augmented reality individual terminal and mobile command cabin through the edge computing node, the resource deployment status is marked in real time in the BIM model, and synchronous multi-terminal operations in a disconnected environment are achieved based on the Ka satellite link coordinated with the edge computing node.

[0135] The method for implementing instruction synchronization through edge computing nodes includes:

[0136] S41 transmits lightweight BIM model slices and resource positioning instructions to the augmented reality individual terminal, and transmits the full BIM model and a 3D disaster situation sand table for disaster simulation to the mobile command cabin;

[0137] The differentiated transmission strategy augments the reality individual terminal to receive lightweight BIM model slicing synchronization resource positioning instructions.

[0138] The mobile command cabin obtains the full BIM model and loads the three-dimensional disaster situation sand table.

[0139] In this application, a preferred specific implementation method of the three-dimensional disaster sandbox includes the integration of risk heat maps and resource deployment status.

[0140] Dynamic data clipping includes spatial range clipping and semantic level filtering; specifically, spatial range clipping intercepts BIM components within a radius of M meters based on location coordinates; the semantic level retains key structural components and filters decorative components.

[0141] S42 state synchronization maps resource deployment state changes to BIM component attribute datasets in real time, reducing network transmission load through incremental update protocols.

[0142] State change mapping events drive updates, and resource state changes trigger BIM component property updates.

[0143] The synchronous multi-terminal operation in the disconnected environment includes edge pre-storage and resume transmission, which is consistent with offline operation;

[0144] The guarantee of synchronization during network disconnection includes pre-stored data strategy and network disconnection resuming process;

[0145] The pre-stored data strategy includes safety-related entities such as key BIM components, load-bearing structures, escape routes, and a snapshot of the global resource status at the last moment before the network is disconnected.

[0146] The Ka satellite link is activated the moment the public network is interrupted, and the edge node pushes pre-stored data packets to the terminal, maintaining basic operation instruction synchronization from the baseline state.

[0147] Furthermore, the activation of the Ka satellite link is configured according to actual needs using existing technology.

[0148] The edge pre-storage and retransmission pre-store key BIM component spatial data and resource deployment baseline status through edge computing nodes. If the public network is interrupted, the basic operation synchronization between terminals is maintained based on the pre-stored data;

[0149] If the network is disconnected, the timestamp and spatial coordinates of each terminal operation instruction are recorded. If the network is restored, the operation records are merged through a conflict resolution algorithm to reconstruct the offline operation consistency.

[0150] If a disconnected operation occurs, the conflict is resolved by aligning the timestamps. If the conflict is at the same location, a high-risk operation is adopted. If it is at different locations, there is no conflict and the operations are merged.

[0151] Example 2

[0152] like Figure 2 As shown, an emergency command system based on a BIM digital base includes:

[0153] The data fusion module includes a data processing unit for pre-processing BIM building information, GIS geospatial data, monitoring data and emergency plan documents. Specifically, for BIM building information, it analyzes data such as building structure and component attributes; for GIS geospatial data, it extracts geographic information such as terrain and roads; it collects and cleans monitoring data in real time and filters abnormal data; and it structures unstructured emergency plan documents and extracts key emergency processes and disposal measures.

[0154] The data fusion module also includes a semantic mapping and entity alignment unit. Based on the extended IFC standard, it establishes a unified data model and uses semantic mapping technology to uniformly convert the semantics of data from different sources to eliminate semantic ambiguity. It uses the entity alignment algorithm to identify and match information describing the same entity in multi-source data, generate an emergency knowledge graph with BIM building information as the node, and realize the deep fusion and association of multi-source data.

[0155] The disaster simulation module includes an association chain identification and simulation calculation unit. The disaster association chain identification is based on the emergency knowledge graph, analyzes the logical relationship between various types of data, identifies the key elements and association chains in the occurrence and development of disasters, sorts out the disaster propagation path and impact range, and provides knowledge support for disaster simulation.

[0156] Furthermore, the deduction and calculation unit couples physical simulation with the LSTM prediction algorithm to inject monitoring data into the BIM model in real time. Physical simulation simulates the physical process of disaster occurrence, and the LSTM prediction algorithm dynamically predicts the development trend of disasters. By combining the two, the disaster process can be deduced, and an intuitive risk heat map and clear evacuation restricted areas can be output to help command personnel grasp the disaster situation.

[0157] The solution generation module includes a spatiotemporal calculation and solution optimization unit, which is used to determine the accessibility of emergency resources and generate a scheduling plan; the spatiotemporal calculation uses a reinforcement learning model, combined with disaster simulation results and real-time environmental data, to calculate the accessibility of emergency resources under different time and space conditions, and analyze key indicators such as the time and path for various emergency resources to arrive at the disaster site.

[0158] Furthermore, the solution optimization generation unit generates an optimization solution that includes drone path planning, personnel evacuation route design, and equipment scheduling instructions based on the spatiotemporal calculation results, taking into account factors such as personnel safety and resource utilization efficiency, to ensure that emergency resources are reasonably allocated.

[0159] The instruction execution module includes an instruction synchronization unit and a resource monitoring unit, wherein the instruction synchronization unit is used for edge computing transmission instructions, and the resource monitoring unit marks the deployment status in real time;

[0160] Specifically, the command synchronization unit synchronizes the generated emergency plan instructions to the enhanced terminal and mobile command cabin through the edge computing node, realizing fast and stable command transmission under normal network conditions; in a network disconnected environment, the Ka satellite link coordinated with the edge computing node is used to ensure reliable synchronization of instructions.

[0161] The resource status monitoring unit marks the resource deployment status in the BIM model in real time, tracks the allocation and use of emergency resources, and synchronizes multi-terminal operations, so that command personnel and executors can grasp the progress of emergency response in real time and achieve efficient collaborative work.

[0162] It is important to note that the configuration and arrangement of the present application, as illustrated in various exemplary embodiments, are illustrative only. Although only two embodiments are described in detail in this disclosure, those reading this disclosure should readily understand that the dimensions, scales, structures, shapes, and proportions of various components, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, colors, and orientations, can be varied without materially departing from the subject matter described herein. For example, components shown as integrally formed may be comprised of multiple parts or components, the positions of components may be inverted or otherwise altered, and the nature, number, or position of discrete components may be modified or changed. Therefore, all such modifications are intended to be encompassed within the scope of this invention. The order or sequence of any process or method steps may be altered or reordered according to alternative embodiments. Any "means-plus-function" clause is intended to cover structures that perform the functions described herein, and not only structural equivalence but also structural equivalents. Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of this invention. Therefore, the invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0163] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention).

[0164] It will be appreciated that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort is complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0165] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An emergency command method based on BIM digital base, characterized in that: include: Establish a unified data model, perform semantic mapping and entity alignment on BIM building information, GIS geospatial data, monitoring data, and emergency plan documents, and generate an emergency knowledge graph with BIM building information as the node; The knowledge graph identifies disaster association chains, couples physical simulation with the LSTM prediction algorithm, injects monitoring data into the BIM model to deduce the disaster process, and outputs risk heat maps and evacuation restricted areas; The method of coupling the physical simulation engine and the LSTM prediction algorithm to deduce the disaster process includes: S21 establishes a dual-engine collaboration mechanism, wherein the physical simulation engine constructs an initial propagation path based on the physical laws corresponding to the disaster type, and the prediction algorithm dynamically corrects the parameter deviation of the physical simulation engine by analyzing historical disaster time series data; S22 real-time data-driven update, mapping IoT monitoring data streams into dynamic state variables of BIM building information in real time, and driving the synchronous iterative optimization of the physical simulation engine and LSTM prediction algorithm through edge computing nodes; Based on the simulation results, the spatial and temporal accessibility of emergency resources is calculated through a reinforcement learning model, generating an optimization plan that includes drone paths, personnel evacuation routes, and equipment dispatch instructions. Through edge computing nodes, commands are synchronized to augmented reality individual terminals and mobile command cabins, resource deployment status is marked in real time in the BIM model, and multi-terminal operations are synchronized.

2. The emergency command method based on BIM digital base according to claim 1, characterized in that: Establishing the unified data model includes adding spatial topological attributes to components in BIM building information, coordinate matching GIS geospatial data with BIM building information, defining exclusive dynamic entity categories for monitoring data, and binding them to the operation and maintenance attribute sets of building components through unique device codes; Build a distributed semantic query channel across building models, geographic data, and IoT monitoring, use artificial intelligence to analyze the semantic correlation between emergency plan text keywords and building component functions, and complete alignment after reaching the preset matching standards.

3. The emergency command method based on BIM digital base according to claim 1, characterized in that: The method for generating an emergency knowledge graph includes: S11 builds a spatial topological network based on the physical adjacency relationship of components; S12 establishes the functions and linkage links between components according to the emergency plan; S13 builds a probability prediction model for secondary disasters caused by component failure based on historical disaster data; S14 converts real-time IoT monitoring data into status attribute parameters of building components. When the monitoring data is abnormal, it automatically increases the risk level of the associated components in the map.

4. The emergency command method based on BIM digital base according to claim 1 is characterized in that: The dynamic correction includes, when the deviation between the real-time monitoring data and the simulation output exceeds a preset threshold, the prediction algorithm adjusts the boundary condition parameters of the physical simulation engine to form a closed-loop feedback; The synchronous iterative optimization includes continuously outputting disaster evolution prediction results at a frequency based on the updated BIM building information state variables.

5. The emergency command method based on BIM digital base according to claim 1 is characterized in that: The method for calculating the spatiotemporal accessibility of emergency resources through a reinforcement learning model includes: S31 constructs the risk heat map, evacuation restricted area and traffic network status generated by disaster simulation into a three-dimensional spatiotemporal state space; S32 defines movement constraint rules for emergency resources in the state space; S33 uses the shortest response time and maximum resource coverage as a joint reward function to train multiple targets, trains a reinforcement learning model through historical rescue data, and outputs a probability matrix of resources arriving at the target location.

6. The emergency command method based on BIM digital base according to claim 1 is characterized in that: The drone path planning generates a three-dimensional flight corridor that avoids restricted areas based on the spatiotemporal accessibility probability matrix, while dynamically adjusting the flight altitude to avoid interference from the disaster cyclone. The personnel evacuation route is calculated by combining the building structure safety level with real-time pedestrian flow data to calculate multi-exit diversion paths and plan low-intensity escape channels for special groups; The equipment dispatch instruction is generated by reversely deducing the optimal matching combination of equipment loading points and transportation vehicles based on the probability matrix of resources arriving at the target location, and generating equipment delivery sequence instructions with time windows.

7. The emergency command method based on BIM digital base according to claim 6 is characterized in that: The method for implementing instruction synchronization through edge computing nodes includes: S41 transmits lightweight BIM model slices and resource positioning instructions to the augmented reality individual terminal, and transmits the full BIM model and a 3D disaster situation sand table for disaster simulation to the mobile command cabin; S42 state synchronization maps resource deployment state changes to BIM component attribute datasets in real time, reducing network transmission load through incremental update protocols.

8. The emergency command method based on BIM digital base according to claim 7 is characterized in that: Synchronous multi-terminal operations in disconnected environments, including edge pre-storage and resume, are consistent with offline operations; The edge pre-storage and retransmission pre-store key BIM component spatial data and resource deployment baseline status through edge computing nodes. If the public network is interrupted, the basic operation synchronization between terminals is maintained based on the pre-stored data; If the network is disconnected, the timestamp and spatial coordinates of each terminal operation instruction are recorded. If the network is restored, the operation records are merged through a conflict resolution algorithm to reconstruct the offline operation consistency.

9. An emergency command system based on a BIM digital base, which is implemented based on an emergency command method based on a BIM digital base according to any one of claims 1 to 8, characterized in that: The system includes: a data fusion module, a disaster simulation module, a plan generation module and an instruction execution module; The data fusion module is used to preprocess multi-source data and fuse them into an emergency knowledge graph; The disaster simulation module is used to analyze disaster factors and simulate the process, and output a risk map; The solution generation module is used to determine the accessibility of emergency resources and generate a scheduling solution; The instruction execution module includes an instruction synchronization unit and a resource monitoring unit, wherein the instruction synchronization unit is used for edge computing transmission instructions, and the resource monitoring unit marks the deployment status in real time.

Images