Method and system for constructing allocation strategy of post-earthquake first-aid repair resources

Abstract

The invention provides a post-earthquake first-aid repair resource allocation strategy construction method and system, and relates to the technical field of post-earthquake first-aid repair scheduling, and the method comprises the steps: firstly, obtaining earthquake disaster situation, manpower, equipment and material information; then, dividing first-aid repair areas based on disaster situation information and analyzing task priorities; meanwhile, constructing a digital twinborn model by relying on human resources, equipment and material information, and dynamically updating and generating evolution data; further, fusing the digital twinborn evolution data, the first-aid repair area and the task priority, and constructing and optimizing a resource allocation strategy; and finally, transmitting the optimization strategy to a deployment center, dynamically updating the strategy according to execution data fed back in real time, forming closed-loop optimization, and realizing efficient and accurate scheduling of post-earthquake resources.

Description

Technical Field

[0001] This invention relates to the field of post-earthquake emergency repair and dispatch technology, and more specifically, to a method and system for constructing a post-earthquake emergency repair resource allocation strategy. Background Technology

[0002] Earthquakes are sudden and destructive natural disasters. The "golden 72 hours" after an earthquake are a critical window for rescue and relief efforts, requiring efficient resource allocation and operational organization. Currently, emergency response and dispatch systems mainly rely on manual command, traditional communication, and pre-planned scheduling, which suffers from slow response, resource misallocation, and inefficient dispatching. This is especially problematic in extreme environments such as communication disruptions, frequent aftershocks, and traffic congestion, severely hindering rescue efficiency and safety.

[0003] Meanwhile, post-earthquake repair involves multiple types of resources, including personnel (repair teams), machinery (engineering equipment), and materials (supplies and equipment). These resources are characterized by complex status, limited quantity, and dispersed location. Existing systems generally lack dynamic perception and centralized intelligent coordination capabilities, making it impossible to accurately match task requirements. This leads to frequent occurrences of redundant scheduling, resource idleness, and bottleneck resource waste. Furthermore, regional task allocation is often based on administrative divisions or experience-based decisions, failing to scientifically reflect the severity of disaster damage and recovery priorities, making it difficult to guarantee the accuracy and timeliness of resource allocation.

[0004] Therefore, there is an urgent need for a method and system for constructing a post-earthquake emergency repair resource allocation strategy to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for constructing a post-earthquake emergency repair resource allocation strategy to improve the aforementioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows: Firstly, this application provides a method for constructing a post-earthquake emergency repair resource allocation strategy, including: Obtain information on earthquake disaster, human resources, repair equipment, and repair materials; Based on the earthquake disaster information, emergency repair areas were divided and repair priorities were analyzed to obtain multiple emergency repair areas and their corresponding emergency repair task priority scores. Based on the aforementioned human resources information, emergency repair equipment information, and emergency repair material information, twin modeling processing is performed, and dynamic evolution and status updates are carried out based on the established twin digital model to obtain digital twin evolution data of emergency repair resources. Based on the digital twin evolution data of the emergency repair resources, the emergency repair area and its corresponding emergency repair task priority score, a resource allocation strategy is constructed and optimized to obtain an optimized resource allocation strategy. The optimized resource allocation strategy is transmitted to the allocation center, and the strategy execution data fed back by the allocation center is obtained in real time. Based on the strategy execution data, the resource allocation strategy is updated to obtain the updated resource allocation strategy.

[0006] Secondly, this application also provides a system for constructing a post-earthquake emergency repair resource allocation strategy, characterized in that it includes: The acquisition unit is used to acquire earthquake disaster information, human resource information, emergency repair equipment information, and emergency repair material information. The analysis unit is used to divide the emergency repair areas and analyze the emergency repair priorities based on the earthquake disaster information, and obtain multiple emergency repair areas and their corresponding emergency repair task priority scores. The evolution unit is used to perform twin modeling processing based on the human resources information, emergency repair equipment information and emergency repair material information, and to perform dynamic evolution and state update based on the established twin digital model to obtain digital twin evolution data of emergency repair resources. The construction unit is used to construct and optimize the resource allocation strategy based on the digital twin evolution data of the emergency repair resources, the emergency repair area and its corresponding emergency repair task priority score, so as to obtain the optimized resource allocation strategy. The allocation unit is used to transmit the optimized resource allocation strategy to the allocation center, and to obtain the strategy execution data fed back by the allocation center in real time. Based on the strategy execution data, the resource allocation strategy is updated to obtain the updated resource allocation strategy.

[0007] The beneficial effects of this invention are as follows: This invention comprehensively understands the post-disaster environmental state by integrating multi-source data, improving information timeliness and decision-making accuracy. It employs graph convolutional neural networks for spatial partitioning and task scoring, enabling more precise and rational disaster area division and task prioritization. A digital twin human-machine-object model is constructed to dynamically reflect resource status and perform predictions, achieving precise resource allocation. This invention also integrates ant colony and genetic algorithm scheduling strategies to rapidly generate high-quality scheduling schemes under complex constraints, improving repair efficiency. Furthermore, by introducing a multimodal communication mechanism, this invention ensures stable command issuance and information feedback even when network infrastructure in the disaster area is damaged.

[0008] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the method for constructing a post-earthquake emergency repair resource allocation strategy as described in an embodiment of the present invention; Figure 2 This is a schematic diagram of the system structure for constructing the post-earthquake emergency repair resource allocation strategy as described in this embodiment of the invention.

[0011] In the diagram: 701, Acquisition Unit; 702, Analysis Unit; 703, Evolution Unit; 704, Construction Unit; 705, Allocation Unit. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0013] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0014] Example 1: This embodiment provides a method for constructing a post-earthquake emergency repair resource allocation strategy.

[0015] See Figure 1 The figure shows that the method includes steps S1, S2, S3, S4 and S5.

[0016] Step S1: Obtain earthquake disaster information, human resource information, emergency repair equipment information, and emergency repair material information; Understandably, this step, in extreme or complex environments following an earthquake (such as mountainous areas, plateaus, or areas with communication disruptions), comprehensively acquires key environmental and resource status information affecting emergency repair scheduling through heterogeneous sensing devices and data sources, providing a data foundation for subsequent modeling and scheduling algorithms. The data sources include seismic networks, remote sensing imagery (satellites / drones), edge terminals (wearable devices, operational equipment sensors), and road monitoring nodes. The sensed content includes road accessibility, weather conditions, aftershock information, personnel / equipment location status, and material inventory status. This step overcomes the bottleneck of traditional disaster information reliance on manual reporting, constructing a dynamic database supporting millisecond-level updates, providing a high-precision, high-reliability data foundation for subsequent intelligent decision-making.

[0017] Step S2: Based on the earthquake disaster information, divide the emergency repair areas and analyze the emergency repair priorities to obtain multiple emergency repair areas and their corresponding emergency repair task priority scores; Understandably, this step, given the scarcity of resources and the wide, uneven distribution of tasks in post-earthquake emergency repairs, aims to spatially divide the affected areas and prioritize tasks. This supports a shift in the subsequent scheduling algorithm from "globally optimal" to "locally priority-driven" distributed collaborative optimization. By combining multi-dimensional disaster factors, population impact, and infrastructure criticality indicators, a regional grading evaluation system is constructed. Then, a graph neural network (GNN) is used for spatial relationship modeling and clustering to achieve scientifically sound and rational emergency repair allocation and zoning. In this step, step S2 includes steps S21, S22, S23, S24, and S25.

[0018] Step S21: Construct a geospatial map based on the earthquake disaster information, wherein villages and towns within the earthquake area are used as nodes of the geospatial map, and connecting roads between villages and towns are used as edges. Understandably, in this step, each small geographical unit (such as a 1km×1km grid, village / town boundary) serves as a node in the graph, and edges are established based on its neighboring units (such as adjoining, road accessibility, and continuous communication coverage). Each node carries a multi-dimensional attribute feature vector:

[0019] in, It is a multi-dimensional attribute feature vector. For road accessibility, As a communication availability indicator, As the disaster damage area factor, To influence the population density factor, To restore the importance factors.

[0020] This step transforms the complex spatial relationships of the physical world into a computational model that can be understood and processed by machines by constructing a geospatial map that integrates multi-dimensional dynamic factors. This provides input for subsequent graph neural network analysis that conforms to real-world geographical constraints and contains rich semantic information, effectively solving the fundamental defect of traditional grid partitioning methods that cannot accurately represent actual spatial correlation and dynamic accessibility.

[0021] Step S22: Based on the earthquake disaster information, feature extraction is performed to obtain multi-dimensional features of the earthquake disaster. The multi-dimensional features of the earthquake disaster are the disaster area feature, the population density impact feature, the recovery importance factor, the road accessibility feature, and the communication availability feature obtained through semantic recognition. Understandably, this step uses semantic recognition methods to identify multidimensional features in the uploaded earthquake disaster information, including population density features, recovery importance factors, road accessibility features, and communication availability features. Population density features include population density itself, and the recovery importance factor represents the relative importance weight of each region in the overall post-disaster recovery. Road accessibility features are the probabilities of passability at different time scales. Communication availability features are obtained by fusing base station damage reports, satellite communication link quality indicators, and UAV relay network coverage data.

[0022] Step S23: Based on a preset graph convolutional neural network, learn the nodes of the geospatial map and historical emergency repair tasks, wherein the multidimensional features of the earthquake disaster and its corresponding emergency repair tasks are updated by node embedding representation to obtain the updated node embedding representation vector. It is understandable that in this step, the graph convolutional neural network (GCN) learns the representation of the nodes and updates the node embedding representation as shown below:

[0023] in, For node embedding representation, It is a multi-dimensional attribute feature vector. For nodes The adjacent nodes.

[0024] In this step, the generated node embeddings contain geospatial disaster information, enabling subsequent regional division and priority assessment to simultaneously consider real-time disaster conditions and historical task execution patterns, significantly improving the accuracy and adaptability of emergency repair decisions. Particularly in response to the uncertainty of the post-earthquake environment, this embedding representation is robust to the loss of certain features. When some real-time data is unavailable, reasonable inferences can be made through historical patterns and neighborhood features, ensuring the reliability of the system's decisions under incomplete data conditions. The Graph Convolutional Neural Network (GCN) for node representation learning is a common method for training existing neural networks and will not be elaborated upon here.

[0025] Step S24: Send the updated node embedding representation vector to the preset K-means clustering model for clustering, and divide the emergency repair area based on the multiple clusters obtained by clustering to obtain at least two emergency repair areas. It is understandable that the emergency repair area division results generated in this step maintain the homogeneity of disaster characteristics and emergency repair needs within the region, while ensuring the heterogeneity and relative independence between regions, providing a scientific basis for the formulation of subsequent differentiated resource allocation strategies. In particular, through semantic embedding vector clustering, the division results can simultaneously consider the static characteristics, spatial correlation, and historical task patterns of nodes. Compared with traditional division methods based on administrative boundaries or simple geographic grids, this is more in line with the actual needs of post-earthquake emergency repairs and effectively improves the accuracy and efficiency of subsequent resource scheduling. In particular, this step automatically determines the optimal number of region divisions by calculating the weighted objective function of the contour coefficient and the average geographic distance between nodes in the cluster under different k values, avoiding the subjectivity of human pre-setting. The geographic distance is calculated using the Euclidean distance of vectors.

[0026] Step S25: Output the embedded representation vector of each emergency repair area and its contained nodes to the preset multi-factor priority scoring formula for scoring, and obtain the emergency repair task priority corresponding to each emergency repair area.

[0027] It is understandable that the multi-factor priority scoring formula in this step is as follows:

[0028] in, As the disaster damage area factor, To influence the population density factor, To restore the importance factor, and All scores are adjustable weighting coefficients. The higher the score, the more urgent the task, and the more priority will be given to allocating human, machine, and material resources for response.

[0029] Step S3: Perform twin modeling processing based on the human resources information, emergency repair equipment information and emergency repair material information, and perform dynamic evolution and status updates based on the established twin digital model to obtain digital twin evolution data of emergency repair resources; It is understandable that this step, by deeply mining the semantic information of the disaster situation implicit in the node embedding vectors, enables priority assessment to not only consider apparent disaster indicators but also capture deeper characteristics such as the urgency, spread risk, and feasibility of rescue efforts, thereby achieving accurate classification of repair tasks. In particular, the model possesses online learning capabilities; when receiving new disaster feedback data, it can automatically correct the parameters of the scoring formula through backpropagation, allowing priority assessment to continuously adapt to the evolving disaster situation and effectively improve the allocation efficiency of large-scale post-disaster relief resources. In this step, step S3 includes steps S31, S32, and S33.

[0030] Step S31: Perform digital twin modeling on the human resources information, emergency repair equipment information, and emergency repair material information to construct digital twins of personnel, equipment, and materials, respectively. Understandably, this step involves several steps: for personnel digital twins, by integrating BeiDou positioning data, wearable device physiological parameters (heart rate variability, blood oxygen saturation), and qualification certification information, a three-dimensional personnel model is constructed, including dynamic performance evaluation coefficients (calculated based on a fatigue accumulation model) and spatial mobility capability tags (combining terrain adaptability and transportation tools); for equipment digital twins, by integrating airborne sensor data (vibration spectrum, hydraulic pressure, fuel consumption rate) and historical maintenance records, an equipment health status model is generated, integrating aftershock condition stability indicators (such as tilt angle safety thresholds) and fault probability prediction values ​​(based on vibration characteristics and equipment age correlation analysis); for material digital twins, by integrating RFID inventory data, environmental sensor readings (temperature and humidity), and road accessibility information, an intelligent material model is established, including available quantity within the validity period, supply route reliability score (real-time road access probability), and airdrop adaptability parameters (determining the impact resistance characteristics of material packaging).

[0031] Step S32: Obtain real-time information on changes in human resources, equipment, and materials during the emergency repair process, and perform scenario evolution on the personnel digital twins, equipment digital twins, and material digital twins based on these information and preset evolution rules to obtain data information of all digital twins after scenario evolution. The evolutionary data sources preset in this step include real-time status (fatigue, positioning, operation feedback) uploaded by personnel / equipment wearable terminals; UAV / remote sensing images to assist in judging area accessibility, equipment location, etc.; dispatch system feedback of allocation results, used to correct paths and predict task completion time; and external systems (such as oil depots, medical warehouses) pushing inventory changes through APIs. Dynamic evolution scenarios include fatigued personnel transitioning to a "rest state" after continuous work, with the virtual twin's state synchronized; faulty equipment entering the maintenance process, with its operational capability score weighting reduced; some access routes being blocked by aftershocks, automatically replanning delivery routes and updating material arrival times; and multi-person team combined tasks, with the twin system performing "combined behavior modeling and effect prediction." The state synchronization strategy includes using time-triggered and state change-triggered synchronization mechanisms: time periodic synchronization (such as updating a batch of twins every 3 minutes); state change synchronization (such as immediately refreshing the twin state when equipment is disconnected or supplies are interrupted).

[0032] This step enables the digital twin to not only passively reflect changes in resource status, but also proactively predict potential risks (such as the remaining service life of equipment and the probability of supply disruptions) through closed-loop interaction between real-time data and evolution rules.

[0033] Step S33: Based on the preset update rules and the data information of all digital twins after the scene evolution, update the corresponding data of each digital twin to obtain the digital twin evolution data of the emergency repair resources.

[0034] It is understandable that this step updates the data of the digital twin through the state synchronization strategy of the previous step. The resulting digital twin evolution data not only has high timeliness and low redundancy, but also effectively resists post-earthquake environmental interference. It ensures that the availability and consistency of the core twin state can still be maintained in extreme cases such as communication interruption and partial data loss, providing real-time, reliable and lightweight data support for subsequent resource scheduling decisions, and significantly improving the robustness and response efficiency of the system in harsh environments.

[0035] Step S4: Based on the digital twin evolution data of the emergency repair resources, the emergency repair area and its corresponding emergency repair task priority score, construct and optimize the resource allocation strategy to obtain the optimized resource allocation strategy. It is understandable that this step, through the deep coupling of digital twin evolutionary data and optimization algorithms, enables the scheduling strategy to not only statically fit the current disaster situation but also proactively address the dynamic evolution of the post-earthquake environment (such as equipment performance degradation and secondary disaster risks), significantly improving resource utilization efficiency and the reliability of repair tasks. Simultaneously, the hybrid optimization algorithm ensures that even in large-scale, multi-constraint post-earthquake scheduling problems, feasible and optimized strategies can still be efficiently generated, meeting the real-time requirements of repair and rescue. Step S4 includes steps S41 and S42.

[0036] Step S41: Construct a resource-task-region mapping space based on the digital twin evolution data of the emergency repair resources, the emergency repair area and its corresponding emergency repair task priority score, wherein emergency repair tasks and resources are randomly assigned to each region to obtain at least one initial resource allocation strategy. Understandably, this step first establishes a three-dimensional correlation matrix (resource dimension × task dimension × region dimension) based on the results of the emergency repair area division and task priority scoring. Each dimension is infused with dynamic evolutionary features: the resource dimension includes real-time efficiency coefficients of digital twins (such as operational capacity after personnel fatigue correction and remaining equipment availability); the task dimension integrates priority scores and spatiotemporal constraints (such as task deadline windows and geographical coverage); and the region dimension integrates dynamic environmental parameters (such as aftershock risk probability and road accessibility reliability). Subsequently, constraint satisfaction techniques are used to generate initial strategies: by defining hard constraints (such as matching resource skills with task types and resource location with regional accessibility) and soft constraints (such as prioritizing high-efficiency resources for high-priority tasks), a random sampling algorithm is used to generate feasible initial strategies that satisfy all constraints in the solution space.

[0037] Step S42: Optimize and update all the initial resource allocation strategies based on the preset ant colony algorithm and genetic algorithm. The ant colony algorithm is used to optimize the path of each emergency repair task in the initial resource allocation strategy, and the genetic algorithm is used to perform multi-objective optimization selection on the optimized initial resource allocation strategy to obtain the optimized resource allocation strategy.

[0038] Understandably, this step first uses all initial resource allocation strategies as input parameters. Then, it optimizes the path selection for each repair task in the initial resource allocation strategies using the ant colony algorithm, obtaining the optimized path for each repair task. Finally, it models the objective function based on the optimized path for each repair task, where it assumes... The task number for the set of all scheduled tasks is: ; A collection of human-machine-material resources; For resources Execute the task The matching scoring function; The estimated response time for the resource to travel from its current location to the task point; To estimate the efficiency of operations per unit time, the following objective function is obtained:

[0039] in: , For decision variables, it represents whether to allocate resources. Assigned to task ; All parameters are adjustable, balancing the weights of response speed, efficiency, and matching degree; Task and resources The matching degree takes into account the matching of personnel types and task types, and the matching of mechanical equipment performance and task difficulty. The optimal allocation strategy; The estimates are based on a map planning system, taking into account route accessibility, road closures, and the threat of aftershocks. and Both are two different nodes.

[0040] Step S5: Transmit the optimized resource allocation strategy to the allocation center, and obtain the strategy execution data fed back by the allocation center in real time. Update the resource allocation strategy based on the strategy execution data to obtain the updated resource allocation strategy.

[0041] It is understandable that the communication modes and adaptation logic supported in this step include: The BeiDou / Maritime Satellite Communication Link (Wide Area Command Distribution Main Channel) is suitable for: high-altitude epicenters, signal blind spots, and areas where ground base stations are ineffective; it supports data transmission such as SMS, location backhaul, and voice SMS; the BeiDou module supports bidirectional links and has positioning and scheduling command capabilities; marine satellites are used for communication in mobile communication vehicles and unmanned aerial vehicles (UAVs) carriers. Mesh self-organizing network communication (local node networking collaboration) is suitable for: communication of emergency rescue teams when communication base stations are paralyzed; terminal devices such as wearable radios, unmanned vehicles, and edge computing nodes can achieve multi-hop forwarding without a central node through the Mesh protocol; it supports task collaboration, file transfer, and status synchronization between nodes; each node can dynamically become a "relay node" to improve network coverage and resilience.

[0042] Shortwave radio communication (extreme disaster recovery plan channel) is suitable for: complete network outage and extreme isolated operation scenarios; ultra-long distance low bandwidth information transmission via HF / VHF bands; can be used as a "last hop" voice / coded command channel; and is compatible with the communication frequencies of traditional emergency rescue teams (armed police, firefighters).

[0043] To ensure the successful implementation of the communication mode, terminal equipment with multi-mode communication capabilities is configured, possessing the following characteristics: Offline task execution capability: Supports task caching and local execution engine; Even when communication is interrupted, the job can continue through voice broadcast and map guidance; The instruction content is encrypted and cached locally, and the execution log is sent back when communication is restored.

[0044] Multi-mode module access: integrates Beidou, 4G / 5G, Mesh node, Wi-Fi / AP, and shortwave interface; can automatically switch communication modes by priority to ensure uninterrupted tasks; supports Mesh network hotspot function, allowing people to share network signals when they are close by.

[0045] Intelligent voice broadcast and task prompts: Supports TTS voice module to broadcast scheduling instructions; operators can confirm tasks through voice control devices; supports simple gesture / button operation confirmation (prevents accidental operation due to fatigue).

[0046] This step includes a communication and feedback protocol mechanism, as shown in the table below: Table 1 Communication and Feedback Protocol Mechanisms

[0047] In this step, step S5 includes steps S51 and S52.

[0048] Step S51: Generate allocation instructions based on the resource allocation strategy, encrypt and transmit the allocation instructions to the allocation center, wherein the allocation instructions are transmitted by attaching a local cache copy and a voice broadcast instruction, and real-time execution data fed back by the allocation center is obtained in real time; It is understandable that this step, through redundant encoding of instructions and multimodal delivery, effectively resists post-earthquake communication link interruptions, high error rates, and interference from harsh environments, ensuring the reliable transmission and correct understanding of scheduling intentions. At the same time, through lightweight encryption and multi-source perception feedback, a panoramic depiction of the execution process is achieved with extremely low bandwidth usage, providing a high-fidelity and strong real-time data foundation for subsequent dynamic policy updates, and completely solving the pain points of "loss of connection after instruction is issued" and "blind spots in execution status" in traditional scheduling.

[0049] Step S52: Send the real-time execution data fed back by the allocation center to the twin digital model for data update, and regenerate the allocation strategy based on the updated data to obtain the updated real-time resource allocation strategy.

[0050] Understandably, in this step, the real-time updates and rapid simulations of the digital twin enable the scheduling system to have online optimization capabilities of "foresight-decision-execution-feedback". This can effectively cope with typical post-earthquake uncertainties such as sudden equipment failures, road interruptions caused by secondary disasters, and sharp declines in personnel efficiency. It significantly improves the dynamic adaptability and system resilience of resource allocation, ensuring that the emergency repair task can still be carried out efficiently and reliably in a highly uncertain and harsh environment.

[0051] Example 2: like Figure 2 As shown, this embodiment provides a system for constructing a post-earthquake emergency repair resource allocation strategy. See [link to documentation]. Figure 2 The system includes an acquisition unit 701, an analysis unit 702, an evolution unit 703, a construction unit 704, and a coordination unit 705.

[0052] The acquisition unit 701 is used to acquire earthquake disaster information, human resources information, emergency repair equipment information, and emergency repair material information. Analysis unit 702 is used to divide the emergency repair area and analyze the emergency repair priority based on the earthquake disaster information, and obtain multiple emergency repair areas and their corresponding emergency repair task priority scores. Evolution unit 703 is used to perform twin modeling processing based on the human resources information, emergency repair equipment information and emergency repair material information, and to perform dynamic evolution and state update based on the established twin digital model to obtain digital twin evolution data of emergency repair resources. The construction unit 704 is used to construct and optimize the resource allocation strategy based on the digital twin evolution data of the emergency repair resources, the emergency repair area and its corresponding emergency repair task priority score, so as to obtain the optimized resource allocation strategy. The allocation unit 705 is used to transmit the optimized resource allocation strategy to the allocation center, and to obtain the strategy execution data fed back by the allocation center in real time. Based on the strategy execution data, the resource allocation strategy is updated to obtain the updated resource allocation strategy.

[0053] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0055] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing a post-earthquake repair resource allocation strategy, characterized in that, The method comprises the following steps: obtaining earthquake disaster information, human resource information, repair equipment information and repair material information; based on the earthquake disaster information, the repair area is divided and the repair priority is analyzed, and a plurality of repair areas and their corresponding repair task priority scores are obtained; based on the human resource information, the repair equipment information and the repair material information, a twin modeling process is carried out, and dynamic evolution and state updating are carried out based on the established twin digital model to obtain digital twin evolution data of repair resources; based on the digital twin evolution data of the repair resources, the repair area and the corresponding repair task priority score, a resource allocation strategy is constructed and optimized to obtain an optimized resource allocation strategy; the optimized resource allocation strategy is transmitted to the allocation center, and the strategy execution data fed back by the allocation center is obtained in real time, and the resource allocation strategy is updated based on the strategy execution data to obtain an updated resource allocation strategy.

2. The method of claim 1, wherein based on the earthquake disaster information, the repair area is divided and the repair priority is analyzed, which comprises: constructing a geographical space graph according to the earthquake disaster information, wherein the villages and towns in the earthquake area are taken as the nodes of the geographical space graph, and the connecting roads of the villages and towns are taken as the edges; based on the earthquake disaster information, the feature is extracted to obtain the multi-dimensional feature of the earthquake disaster, wherein the multi-dimensional feature of the earthquake disaster is the disaster area feature, the influence population density feature, the recovery importance factor, the road accessibility feature and the communication availability feature obtained by semantic recognition; based on the preset graph convolutional neural network, the nodes of the geographical space graph and the historical repair tasks are learned, wherein the multi-dimensional feature of the earthquake disaster and its corresponding repair task are updated by node embedding representation to obtain an updated node embedding representation vector; the updated node embedding representation vector is sent to the preset K-means clustering model for clustering, and the repair area is divided based on the plurality of clustering clusters obtained by clustering to obtain at least two repair areas; each repair area and its contained node embedding representation vector are output to the preset multi-factor priority scoring formula for scoring to obtain the corresponding repair task priority of each repair area.

3. The method of claim 1, wherein based on the human resource information, the repair equipment information and the repair material information, a twin modeling process is carried out, and dynamic evolution and state updating are carried out based on the established twin digital model to obtain digital twin evolution data of repair resources, which comprises: the human resource information, the repair equipment information and the repair material information are modeled to construct personnel digital twin, equipment digital twin and material digital twin respectively; real-time acquisition of human resource change information, repair equipment change information and repair material change information in the repair process, and scene evolution of personnel digital twin, equipment digital twin and material digital twin based on the preset evolution rule, to obtain data information of all digital twins after scene evolution; based on the preset updating rule and the data information of all digital twins after scene evolution, the corresponding data of each digital twin is updated to obtain the digital twin evolution data of repair resources.

4. The method of claim 1, wherein ,construct and optimize a resource allocation strategy based on the digital twin evolution data of the repair resources, the repair area, and the corresponding repair task priority score, including: construct a resource-task-area mapping space based on the digital twin evolution data of the repair resources, the repair area, and the corresponding repair task priority score, wherein a repair task and a resource are randomly allocated to each area to obtain at least one initial resource allocation strategy; update and optimize all the initial resource allocation strategies based on a preset ant colony algorithm and a genetic algorithm, wherein the path of each repair task in the initial resource allocation strategy is optimized by the ant colony algorithm, and the optimized initial resource allocation strategy is optimized and selected by the genetic algorithm, to obtain an optimized resource allocation strategy.

5. The method of claim 1, wherein transmit the optimized resource allocation strategy to the allocation center and obtain real-time strategy execution data fed back by the allocation center in real time, and update the resource allocation strategy based on the strategy execution data, including: generate an allocation instruction based on the resource allocation strategy, and transmit the allocation instruction to the allocation center after encryption, wherein a local cache copy and a voice broadcast instruction are attached to the allocation instruction for transmission, and real-time execution data fed back by the allocation center is obtained in real time; send the real-time execution data fed back by the allocation center to the twin digital model for data updating, and regenerate the allocation strategy based on the updated data to obtain an updated real-time resource allocation strategy.

6. A post-earthquake repair resource deployment strategy construction system, characterized in that, including: an acquisition unit configured to acquire earthquake disaster information, human resource information, repair equipment information, and repair material information; an analysis unit configured to divide a repair area and analyze a repair priority based on the earthquake disaster information, to obtain a plurality of repair areas and corresponding repair task priority scores; an evolution unit configured to perform twin modeling based on the human resource information, repair equipment information, and repair material information, and to perform dynamic evolution and state updating based on the established twin digital model, to obtain digital twin evolution data of repair resources; a construction unit configured to construct and optimize a resource allocation strategy based on the digital twin evolution data of the repair resources, the repair area, and the corresponding repair task priority score, to obtain an optimized resource allocation strategy; an allocation unit configured to transmit the optimized resource allocation strategy to the allocation center and obtain real-time strategy execution data fed back by the allocation center in real time, and update the resource allocation strategy based on the strategy execution data, to obtain an updated resource allocation strategy.

7. The post-earthquake repair resource deployment strategy construction system according to claim 6, characterized in that, The analysis unit includes: a first analysis subunit configured to construct a geographic space graph based on the earthquake disaster information, wherein villages and towns in the earthquake area are taken as nodes of the geographic space graph, and connecting roads of the villages and towns are taken as edges; a second analysis subunit configured to extract features based on the earthquake disaster information, to obtain multi-dimensional features of the earthquake disaster, wherein the multi-dimensional features of the earthquake disaster are disaster damage area features, influence population density features, recovery importance factors, road accessibility features, and communication availability features obtained through semantic recognition. The third analysis subunit is configured to learn the nodes of the geospatial graph and the historical repair tasks based on a preset graph convolutional neural network, wherein the multi-dimensional features of the earthquake disaster and the corresponding repair tasks are subjected to node embedding representation updating to obtain an updated node embedding representation vector; The fourth analysis subunit is configured to send the updated node embedding representation vector to a preset K-means clustering model for clustering, and divide the repair area based on the obtained multiple clustering clusters to obtain at least two repair areas; The fifth analysis subunit is configured to output each repair area and the node embedding representation vector contained therein to a preset multi-factor priority scoring formula for scoring to obtain a repair task priority corresponding to each repair area.

8. The post-earthquake repair resource deployment strategy construction system according to claim 6, characterized in that, The evolution unit comprises: The first evolution subunit is configured to perform twin modeling on the human resource information, the repair equipment information and the repair material information to respectively construct a personnel digital twin, a device digital twin and a material digital twin; The second evolution subunit is configured to acquire human resource change information, repair equipment change information and repair material change information in real time during the repair process, and perform scene evolution on the personnel digital twin, the device digital twin and the material digital twin based on the preset evolution rules and the data information of all the digital twins after the scene evolution to obtain the data information of all the digital twins after the scene evolution; The third evolution subunit is configured to perform corresponding data updating on each digital twin based on a preset updating rule and the data information of all the digital twins after the scene evolution to obtain digital twin evolution data of the repair resources.

9. The post-earthquake repair resource deployment strategy construction system according to claim 6, characterized in that, The construction unit comprises: The first construction subunit is configured to construct a resource-task-area mapping space based on the digital twin evolution data of the repair resources, the repair area and the corresponding repair task priority score, wherein the repair tasks and resources are randomly allocated to each area to obtain at least one initial resource allocation strategy; The second construction subunit is configured to optimize and update all the initial resource allocation strategies based on a preset ant colony algorithm and a genetic algorithm, wherein the path of each repair task in the initial resource allocation strategy is optimized by the ant colony algorithm, and the optimized initial resource allocation strategy is subjected to multi-objective optimization selection by the genetic algorithm to obtain an optimized resource allocation strategy.

10. The post-earthquake repair resource deployment strategy construction system of claim 6, wherein, The allocation unit comprises: The first allocation subunit is configured to generate an allocation instruction based on the resource allocation strategy, and to encrypt and transmit the allocation instruction to an allocation center, wherein the allocation instruction is transmitted by attaching a local cache copy and a voice broadcast instruction thereto, and real-time execution data fed back by the allocation center is acquired in real time; The second allocation subunit is configured to send the real-time execution data fed back by the allocation center to the twin digital model for data updating, and to regenerate the allocation strategy based on the updated data to obtain an updated real-time resource allocation strategy.

Images