Disaster response intelligent command system and control method thereof

By designing a disaster response intelligent command system, using a variety of advanced algorithms and modules for real-time data acquisition and decision-making optimization, the problem of insufficient decision-making accuracy and timeliness in the existing technology is solved, and rapid response to disaster response and efficient resource utilization is achieved. , mitigated the impact of the disaster.

CN118941103AActive Publication Date: 2024-11-12SICHUAN YINLIHUA APPL TECH CO LTD

Patent Information

Application Number
CN202410935430.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-12
Estimated Expiration
2044-07-12

AI Technical Summary

Technical Problem

The existing emergency command system has insufficient accuracy and timeliness in disaster response, resulting in the expansion and spread of disasters. It is necessary to build a comprehensive emergency response system to improve management efficiency and speed.

Method used

设计了一种灾害响应智能指挥系统,包括环境信息收集模块、灾害预警模块、智能决策模块、应急响应模块和资源管理模块,利用机器学习、蒙特卡罗模拟、Agent-based模型、遗传算法、粒子群优化算法、规则引擎、动态规划法和大语言模型,实现实时数据采集、决策优化和资源调配。

Benefits of technology

It significantly improves the speed and accuracy of disaster warnings, ensures that emergency response is initiated in the early stages of disasters, reduces losses, improves scientific decision-making and resource utilization efficiency, ensures rapid response and seamless collaboration, and reduces casualties and economic losses caused by disasters.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941103A_ABST
    Figure CN118941103A_ABST
Patent Text Reader

Abstract

The invention discloses a disaster response intelligent command system and a control method thereof, and relates to the field of emergency processing, and the system comprises an environment information collection module which is used for obtaining the environment information of a target area; the disaster early warning module is used for acquiring the environment information and generating disaster early warning information based on the environment information; the intelligent decision module is used for acquiring the environment information and generating an emergency response instruction and a resource management instruction based on the environment information; the emergency response module is used for acquiring the emergency response instruction and generating an emergency response scheme based on the emergency response instruction; the resource management module is used for acquiring the resource management instruction and generating a resource management scheme based on the resource management instruction; and the manual information collection module is used for carrying out intelligent interaction with the fed-back personnel and obtaining manual information, and the manual information is used for correcting the emergency response instruction and the resource management instruction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present specification relates to the field of emergency response, and more specifically, the present application relates to a disaster response intelligent command system and a control method thereof. Background Art

[0002] In today's context of rapid development of information technology, the rapid growth of social economy is also accompanied by the frequent occurrence of emergencies, which often bring serious casualties and economic losses. The relevant emergency command system can integrate various data and uniformly dispatch and regulate terminals of various role types, but the final emergency decision is still mostly made by personnel, which usually reduces the accuracy of risk assessment and the timeliness of plan release, and easily causes the expansion and spread of disasters.

[0003] Therefore, it is necessary to build a comprehensive emergency command system platform to improve the efficiency and speed of emergency event management. Summary of the invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, the present application proposes a disaster response intelligent command system, the system comprising: An environmental information collection module is used to obtain environmental information of the target area, including sensor information, satellite information and meteorological information, and send the environmental information to the disaster warning module and the intelligent decision-making module; A disaster warning module, used to obtain the above environmental information and generate disaster warning information based on the above environmental information, wherein the above disaster warning module generates the above disaster warning information based on a machine learning model; An intelligent decision-making module, used to obtain the above environmental information and generate emergency response instructions and resource management instructions based on the above environmental information, wherein the above intelligent decision-making module generates the above emergency response instructions and the above resource management instructions based on Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; An emergency response module, used to obtain the above emergency response instructions and generate an emergency response plan based on the above emergency response instructions, wherein the above emergency response module generates the emergency response plan based on a rule engine and simulation results; A resource management module, used to obtain the resource management instructions and generate a resource management solution based on the resource management instructions, wherein the resource management module generates the resource management solution based on a dynamic programming method and a heuristic algorithm; The artificial information collection module is used to interact intelligently with the feedback personnel to obtain artificial information, and the above artificial information is used to correct the above emergency response instructions and resource management instructions. The above artificial information collection module completes intelligent interaction with the feedback personnel based on a large language model.

[0006] In a second aspect, the present application embodiment further provides a control method for the above-mentioned intelligent command system for disaster response in the first aspect, comprising: Controlling the environmental information collection module to collect environmental information of the target area, and sending the environmental information to the disaster warning module and the intelligent decision-making module; Control the disaster warning module to generate the disaster warning information through the machine learning model based on the environmental information; Control the intelligent decision-making module to generate the emergency response instructions and resource management instructions based on the environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; Control the emergency response module to generate an emergency response plan based on the emergency response instruction through a rule engine and simulation results; Controlling the resource management module to generate a resource management solution based on the resource management instruction through dynamic programming and heuristic algorithm; Control the above-mentioned artificial information collection module to complete intelligent interaction with feedback personnel based on a large language model to obtain artificial information; The above emergency response instructions and resource management instructions are modified through the above manual information.

[0007] In a feasible implementation manner, the controlling the disaster warning module to generate the disaster warning information through the machine learning model based on the environmental information includes: Performing a preprocessing operation on the environmental information to obtain preprocessed first environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; The first environmental information is input into a machine learning model corresponding to the disaster type to generate the disaster warning information corresponding to different disaster types.

[0008] In a feasible implementation manner, the above-mentioned control intelligent decision module generates the above-mentioned emergency response instructions and the above-mentioned resource management instructions based on the above-mentioned environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm, including: Performing a preprocessing operation on the environmental information to obtain preprocessed second environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; Monte Carlo simulation is used based on the pre-processed second environment information to obtain simulation evaluation effects of multiple possible response schemes; Based on the above pre-processed environmental information, an agent-based model is used to simulate the performance of each responding participant in multiple possible response schemes; Generate the above-mentioned disaster warning information based on the above-mentioned simulation evaluation effect and the above-mentioned simulation performance effect; Randomly generate a population of initialized resource allocation schemes and a population of initialized examples based on the second environment information and available resource information; Based on the above-mentioned initial resource allocation scheme population, multiple iterations are performed through the genetic algorithm to obtain preliminary optimization results; Based on the above preliminary optimization results and the above initialization example group, multiple iterations are performed through the particle swarm algorithm to obtain the final optimization result; The resource management instructions are generated based on the final optimization result.

[0009] In a feasible implementation manner, the emergency response module is controlled to generate an emergency response plan based on the emergency response instruction through a rule engine and simulation results, including: Use the rule engine to process the emergency response instructions to obtain an emergency response plan optimized by the rule engine; The optimized emergency response plan is simulated based on a disaster response simulation tool to obtain simulation results; Generate an emergency response plan based on the above simulation results and the optimization suggestions of the rule engine.

[0010] In a feasible implementation manner, the resource management module controls the resource management instruction to generate a resource management solution through a dynamic programming method and a heuristic algorithm, including: Dynamic programming results of resources planned by applying dynamic programming method in resource knowledge graph based on the resource management instructions, wherein the resource knowledge graph stores information on resource type, location, dependency, historical usage efficiency and emergency response strategy; Based on the above dynamic programming results, the resource allocation is further optimized based on the particle swarm optimization algorithm to obtain the above resource management solution.

[0011] In a feasible implementation, the large language model includes: an embedding layer, a Transformer encoding layer, an output layer and auxiliary components, the Transformer encoding layer includes multiple Transformer units, the output layer includes multiple fully connected layers and activation functions, the activation function is a softmax activation function, the auxiliary components include a Dropout layer and a Layer Normalization layer, and the optimizer of the large language model is an Adam optimizer.

[0012] In a feasible implementation manner, the above-mentioned modification of the above-mentioned emergency response instruction and resource management instruction by the above-mentioned manual information includes: Evaluate the authenticity of the above artificial information based on expert evaluation method and intelligent evaluation model to obtain effective artificial information; The emergency response instructions and resource management instructions are modified based on the effective manual information.

[0013] In a feasible implementation manner, the authenticity of the artificial information is evaluated based on the expert evaluation method and the intelligent evaluation model to obtain effective artificial information, including: Obtain the identity of the reporter corresponding to the manual information; Determine a preliminary authenticity score based on the identity of the reporter mentioned above; Performing preprocessing operations on the above artificial information to obtain key information; Send the above key information to the expert review unit to obtain the expert authenticity score through the expert evaluation method; Send the above key information to the above intelligent evaluation model to obtain the model authenticity score; Obtain the first weight corresponding to the above expert authenticity score; Obtain a second weight corresponding to the authenticity score of the above model; Determining the final authenticity based on the preliminary authenticity score, the expert authenticity score, the first weight, the model authenticity score and the second weight; The above-mentioned valid artificial information is determined based on the final authenticity and authenticity evaluation threshold.

[0014] In a feasible implementation manner, the first weight information is negatively correlated with the disaster level, the second weight information is positively correlated with the disaster level, the first weight information is negatively correlated with the disaster occurrence time, and the second weight information is positively correlated with the disaster occurrence event; The above authenticity assessment threshold is positively correlated with the above disaster level, and the above authenticity assessment threshold is negatively correlated with the time of occurrence of the above disaster.

[0015] In summary, the intelligent command system for disaster response of the embodiment of the present application uses the environmental information collection module to automatically collect key data from multiple sources (such as sensors, satellites, and weather stations) to obtain environmental changes in the target area in real time. This real-time data collection and transmission significantly improves the speed and accuracy of disaster warning, so that emergency response can be initiated in the early stages of disaster formation, greatly reducing potential losses and impacts. The intelligent decision-making module integrates a variety of advanced algorithms such as Monte Carlo simulation, agent-based models, genetic algorithms, and particle swarm optimization algorithms, which can comprehensively consider various situations and variables, simulate and evaluate different emergency response and resource allocation schemes. This not only improves the scientific nature of decision-making, but also ensures that the most effective decision is made in a complex and changing disaster environment. The resource management module dynamically adjusts resource allocation and optimizes resource utilization efficiency through dynamic programming and heuristic algorithms. This ensures that limited resources can be quickly and reasonably allocated to where they are most needed, especially in the case of tight resources, and can maximize the use efficiency of resources. The artificial information collection module interacts intelligently with on-site personnel based on a large language model to collect feedback and information from the front line in real time. This interaction not only improves the breadth and depth of information collection, but also enhances the accuracy and reliability of information through the analysis of intelligent models. In addition, real-time updated artificial information can be used to promptly correct emergency response instructions and resource management instructions, further improving the adaptability and effectiveness of instructions. The efficient information exchange and coordination mechanism between the modules in the system design ensures a rapid response in emergency situations. From early warning to decision-making to resource allocation and emergency disposal, each link is closely connected and seamlessly collaborates, ensuring the smoothness and efficiency of the entire emergency management process. Through these optimization measures, the disaster response intelligent command system of this scheme can not only improve the speed and efficiency of disaster response, but also significantly reduce casualties and economic losses caused by disasters, and improve the overall emergency management capabilities of society.

[0016] The intelligent command system for disaster response proposed in this application and other advantages, objectives and features of this application will be reflected in part through the following description, and in part will be understood by technical personnel in this field through research and practice of this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 A structural schematic diagram of a disaster response intelligent command system provided in an embodiment of the present application; Figure 2A schematic diagram of a control method process provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0019] See also Figure 1 , is a structural diagram of a disaster response intelligent command system 10 provided in an embodiment of the present application, which may specifically include: The environmental information collection module 101 is used to obtain the environmental information of the target area, the environmental information includes sensor information, satellite information and meteorological information, and send the environmental information to the disaster warning module 102 and the intelligent decision-making module 103; A disaster warning module 102 is used to obtain the above environmental information and generate disaster warning information based on the above environmental information, wherein the above disaster warning module 102 generates the above disaster warning information based on a machine learning model; An intelligent decision-making module 103 is used to obtain the above-mentioned environmental information and generate emergency response instructions and resource management instructions based on the above-mentioned environmental information, wherein the above-mentioned intelligent decision-making module 103 generates the above-mentioned emergency response instructions and the above-mentioned resource management instructions based on Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; The emergency response module 104 is used to obtain the emergency response instruction and generate an emergency response plan based on the emergency response instruction, wherein the emergency response module 104 generates the emergency response plan based on the rule engine and the simulation result; The resource management module 105 is used to obtain the resource management instructions and generate a resource management solution based on the resource management instructions, wherein the resource management module 105 generates the resource management solution based on a dynamic programming method and a heuristic algorithm; The manual information collection module 106 is used to intelligently interact with the feedback personnel to obtain manual information, and the above manual information is used to correct the above emergency response instructions and resource management instructions. The above manual information collection module 106 completes intelligent interaction with the feedback personnel based on a large language model.

[0020] Exemplarily, the environmental information collection module 101 is responsible for collecting environmental information from the target area, including sensor information, satellite information and meteorological information. This information is obtained through various sensor devices, satellite systems and meteorological monitoring stations, covering various environmental parameters such as temperature, humidity, rainfall, wind speed, etc. The collected environmental information will be sent to the disaster warning module 102 and the intelligent decision-making module 103 for further processing and analysis.

[0021] The disaster warning module 102 receives data from the environmental information collection module 101 and generates disaster warning information based on the environmental information. This module uses machine learning models to analyze environmental data and predict possible disasters such as floods, typhoons or earthquakes. Machine learning models can identify patterns and trends in data to provide accurate warning information.

[0022] The intelligent decision-making module 103 also receives environmental information and generates emergency response instructions and resource management instructions based on this information. This module uses Monte Carlo simulation, agent-based models, genetic algorithms, and particle swarm optimization algorithms to optimize the decision-making process. These advanced algorithms can simulate different response plans and resource allocation strategies to select the most effective plan to deal with disasters.

[0023] The emergency response module 104 is responsible for receiving the emergency response instructions generated by the intelligent decision module 103 and generating specific emergency response plans based on these instructions. This module uses the rule engine and simulation results to formulate response plans, such as evacuation plans, rescue team deployment, and emergency site settings.

[0024] The resource management module 105 obtains the resource management instructions provided by the intelligent decision-making module 103 and generates a resource management plan based on these instructions. This module uses dynamic programming and heuristic algorithms to optimize resource allocation to ensure that resources are efficiently used in disaster response.

[0025] The artificial information collection module 106 obtains artificial information by intelligently interacting with on-site feedback personnel. This module interacts based on a large language model and can understand and respond to natural language input. The collected artificial information is used to modify emergency response instructions and resource management instructions to ensure the real-time and accuracy of the instructions.

[0026] In summary, this system uses the environmental information collection module to automatically collect key data from multiple sources (such as sensors, satellites, and weather stations) to obtain real-time information about environmental changes in the target area. This real-time data collection and transmission significantly improves the speed and accuracy of disaster warnings, allowing emergency responses to be initiated in the early stages of disaster formation, greatly reducing potential losses and impacts. The intelligent decision-making module integrates a variety of advanced algorithms such as Monte Carlo simulation, agent-based models, genetic algorithms, and particle swarm optimization algorithms, which can comprehensively consider various situations and variables, simulate and evaluate different emergency responses and resource allocation schemes. This not only improves the scientific nature of decision-making, but also ensures that the most effective decisions are made in complex and changing disaster environments. The resource management module dynamically adjusts resource allocation and optimizes resource utilization efficiency through dynamic programming and heuristic algorithms. This ensures that limited resources can be quickly and reasonably allocated to where they are most needed, especially in the case of resource constraints, to maximize the benefits of resource utilization. The artificial information collection module interacts intelligently with on-site personnel based on a large language model to collect feedback and information from the front line in real time. This interaction not only improves the breadth and depth of information collection, but also enhances the accuracy and reliability of information through the analysis of intelligent models. In addition, real-time updated manual information can be used to promptly revise emergency response instructions and resource management instructions, further improving the adaptability and effectiveness of the instructions. The efficient information exchange and coordination mechanism between the modules in the system design ensures a rapid response in emergency situations. From early warning to decision-making to resource allocation and emergency disposal, each link is closely connected and seamlessly collaborates, ensuring the smoothness and efficiency of the entire emergency management process. Through these optimization measures, the disaster response intelligent command system of this scheme can not only improve the speed and efficiency of disaster response, but also significantly reduce casualties and economic losses caused by disasters, and improve the overall emergency management capabilities of society.

[0027] Second, please refer to Figure 2 , a flow chart of a control method is also proposed for the embodiment of the present application, which is used for the above-mentioned intelligent command system for disaster response in the first aspect, including: S110, controlling the environmental information collection module to collect environmental information of the target area, and sending the environmental information to the disaster warning module and the intelligent decision-making module; Exemplarily, the control environment information collection module collects the environment information of the target area, including sensor information, satellite information, and meteorological information, etc. This ensures that the system can obtain accurate real-time data and provide basic support for disaster warning and emergency decision-making. The collected information will be automatically sent to the disaster warning module and the intelligent decision-making module for further analysis and processing.

[0028] S120, controlling the disaster warning module to generate the disaster warning information through the machine learning model based on the environmental information; For example, machine learning technology can be used to analyze environmental data, issue disaster warnings in a timely manner, and predict the possibility and potential impact of disasters. Machine learning models will identify abnormal patterns and trends in data, thereby generating warnings about disasters such as floods, typhoons, and earthquakes.

[0029] S130, controlling the intelligent decision-making module to generate the emergency response instructions and the resource management instructions based on the environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; For example, the control intelligent decision-making module generates emergency response instructions and resource management instructions based on environmental information through Monte Carlo simulation, agent-based model, genetic algorithm and particle swarm optimization algorithm. These advanced algorithms are used to simulate and evaluate different emergency response and resource allocation schemes and select the optimal strategy. The algorithm will comprehensively consider various environmental factors and resource conditions, optimize the decision-making process, and improve the efficiency of disaster response.

[0030] S140, controlling the emergency response module to generate an emergency response plan based on the emergency response instruction through a rule engine and simulation results; Exemplarily, the control emergency response module generates an emergency response plan based on the emergency response instructions through the rule engine and simulation results. According to the instructions provided by the intelligent decision-making module, specific operation steps and strategies are formed, including evacuation plans and rescue deployment. The rule engine will determine the most effective response measures, such as evacuation routes and allocation of rescue resources, in combination with the simulation results.

[0031] S150, controlling the resource management module to generate a resource management solution based on the resource management instruction through a dynamic programming method and a heuristic algorithm; Exemplarily, the control resource management module generates resource management solutions based on resource management instructions through dynamic programming and heuristic algorithms. This ensures that resources can be reasonably and effectively configured and used, especially when resources are limited. These algorithms will optimize resource allocation to ensure that relief supplies, personnel, and equipment can quickly reach where they are most needed.

[0032] S160, controlling the above-mentioned artificial information collection module to complete intelligent interaction with the feedback personnel based on the large language model to obtain artificial information; Exemplarily, the artificial information collection module is controlled to acquire artificial information by completing intelligent interaction with feedback personnel based on the large language model. Real-time feedback from frontline rescuers and disaster-stricken people is collected to adjust and optimize emergency response. The large language model allows the module to understand natural language input and collect key information quickly and accurately.

[0033] S170. Modify the emergency response instructions and resource management instructions through the manual information.

[0034] Exemplarily, emergency response instructions and resource management instructions are revised based on the collected manual information. Ensure that all operational instructions are updated according to the latest on-site information to improve the adaptability and effectiveness of the response. The revision process will re-evaluate previous decisions and make necessary adjustments based on the latest information.

[0035] In summary, by acquiring data in real time through the automated environmental information collection module and quickly generating accurate warning information in combination with the disaster warning module, this system can respond quickly and ensure that effective measures can be taken at the early stage of a disaster. The intelligent decision-making module uses advanced decision support tools such as Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm, which can simulate multiple emergency response and resource allocation schemes in a complex disaster environment, so as to select the optimal scheme. The resource management module uses dynamic programming and heuristic algorithms to optimize the allocation and use of resources. This not only ensures the best use of limited resources in disaster response, but also reduces resource waste and improves resource utilization efficiency. The manual information collection module interacts intelligently with on-site personnel based on a large language model, and can collect feedback and update information from the front line in real time, and use this information to correct emergency response instructions and resource management instructions in a timely manner. The design of the system ensures the rapid circulation and efficient coordination of information among various modules, whether it is the release of warning information, the transmission of decision information, or the execution of resource allocation, it can achieve seamless docking and rapid implementation. By improving the accuracy of warnings and the timeliness of responses, this system can significantly reduce economic losses and casualties caused by disasters. The system ensures that effective measures are taken before or in the early stages of a disaster, thereby mitigating the overall impact of the disaster.

[0036] In some examples, the controlling the disaster warning module to generate the disaster warning information based on the environmental information through the machine learning model includes: Performing a preprocessing operation on the environmental information to obtain preprocessed first environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; The first environmental information is input into a machine learning model corresponding to the disaster type to generate the disaster warning information corresponding to different disaster types.

[0037] Exemplarily, the purpose of the preprocessing operation is to improve the quality of the data and ensure that the data input to the machine learning model is accurate and reliable, thereby enhancing the accuracy and credibility of disaster warnings. The noise filtering operation is intended to remove random noise in environmental information. Noise may be caused by sensor failure, data transmission errors, or external environmental factors.

[0038] Noise filtering operations include, but are not limited to, smoothing data using digital filters (eg, low-pass filters, high-pass filters) or statistical methods (eg, moving averages).

[0039] Outlier detection operations are used to identify and handle outliers in the data that may distort the results of the predictive model. Outlier detection operations include, but are not limited to, statistical tests (such as Z-score or IQR methods) and distance- or density-based algorithms (such as local outlier factors).

[0040] Data smoothing operations are designed to reduce short-term fluctuations in the data in order to reveal longer-term trends. Data smoothing operations include, but are not limited to, moving averages and exponential smoothing.

[0041] Through specially trained machine learning models, early warning information for different disaster types is generated based on pre-processed environmental information. This information helps to quickly respond to potential disaster events and take necessary preventive measures in advance.

[0042] First, a dedicated machine learning model is trained for each disaster type (such as floods, typhoons, earthquakes, etc.). These models are fed with historical environmental data and corresponding disaster records during the training phase to learn the association between data patterns and disasters.

[0043] During the operation phase, the preprocessed first environmental information is input into the corresponding machine learning model. Each model will evaluate the degree of match between the current data and the historical disaster pattern and output the probability or risk level of the disaster.

[0044] Based on the output of the model, the system generates disaster warning information and releases it to government agencies, emergency response teams, and the public through various channels (e.g., mobile applications, social media, broadcast systems).

[0045] Through this highly automated and intelligent processing flow, this solution can effectively utilize environmental monitoring data to quickly and accurately generate disaster warning information, which can not only significantly improve the timeliness of disaster prevention and response, but also reduce the potential losses caused by disasters.

[0046] In some examples, the control of the intelligent decision-making module generates the emergency response instructions and the resource management instructions based on the environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm, including: Performing a preprocessing operation on the environmental information to obtain preprocessed second environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; Monte Carlo simulation is used based on the pre-processed second environment information to obtain simulation evaluation effects of multiple possible response schemes; Based on the above pre-processed environmental information, an agent-based model is used to simulate the performance of each responding participant in multiple possible response schemes; Generate the above-mentioned disaster warning information based on the above-mentioned simulation evaluation effect and the above-mentioned simulation performance effect; Randomly generate a population of initialized resource allocation schemes and a population of initialized examples based on the second environment information and available resource information; Based on the above-mentioned initial resource allocation scheme population, multiple iterations are performed through the genetic algorithm to obtain preliminary optimization results; Based on the above preliminary optimization results and the above initialization example group, multiple iterations are performed through the particle swarm algorithm to obtain the final optimization result; The resource management instructions are generated based on the final optimization result.

[0047] Exemplarily, the collected environmental information is preprocessed, including noise filtering, outlier detection and data smoothing, to obtain the preprocessed second environmental information, thereby cleaning and optimizing the data to ensure the quality and accuracy of the data in subsequent algorithm processing.

[0048] Monte Carlo simulation is used to evaluate multiple possible response scenarios based on pre-processed secondary environmental information. Monte Carlo simulation estimates the effects of various response scenarios through random sampling techniques, helping decision makers understand how different scenarios might perform under different circumstances.

[0049] Agent-based models are used to simulate the performance of various response participants in multiple possible response scenarios. By simulating the behavior of various participants (such as rescue teams, medical personnel, and disaster victims), agent-based models can provide detailed performance of each scenario in actual operation.

[0050] Disaster warning information is generated based on the simulation evaluation effect and simulation performance effect. This step converts the simulation results into practical warning information to guide actual disaster response activities.

[0051] The resource allocation scheme population and particle swarm are randomly generated using the preprocessed second environment information and available resource information. This step is the basis for the initial iteration of the optimization algorithm and provides a variety of potential resource management schemes for optimization.

[0052] The genetic algorithm is used to perform multiple iterations on the initialized resource allocation scheme population to obtain preliminary optimization results. The genetic algorithm searches for the optimal resource management scheme through selection, crossover and mutation operations to improve resource utilization efficiency.

[0053] The particle swarm optimization algorithm is used to further iterate the above preliminary optimization results to obtain the final optimization results. The particle swarm optimization algorithm adjusts the solution by tracking the historical optimal solutions of individuals and groups, which is an effective method for finding the global optimal solution.

[0054] Generate resource management instructions based on the final optimization results. Convert the calculation results into actual executable resource management operation instructions to ensure that resources are most effectively allocated in disaster response.

[0055] The method proposed in the embodiment of the present application ensures that the disaster response intelligent command system can make scientific and accurate decisions in actual operations, maximize the response speed and resource utilization efficiency, and thus effectively reduce the impact of disasters.

[0056] In some examples, the controlling the emergency response module generates an emergency response plan based on the emergency response instruction through a rule engine and simulation results, including: Use the rule engine to process the emergency response instructions to obtain an emergency response plan optimized by the rule engine; The optimized emergency response plan is simulated based on a disaster response simulation tool to obtain simulation results; Generate an emergency response plan based on the above simulation results and the optimization suggestions of the rule engine.

[0057] Exemplarily, the emergency response module receives emergency response instructions from the intelligent decision-making module and uses the rule engine to process these instructions. The rule engine analyzes and processes the received instructions according to preset rules and logic, with the aim of preliminarily optimizing these instructions to ensure their logical consistency and feasibility of implementation. The rule engine includes various rules for condition judgment, priority sorting, and resource allocation, which are designed based on past experience and best practices in emergency management.

[0058] The emergency response plan optimized by the rule engine is simulated in detail using the disaster response simulation tool. Through simulation, the implementation effect of the plan can be observed in a virtual environment, and possible execution difficulties and effects can be evaluated, so as to further optimize the response plan. The simulation tool can involve complex models, such as crowd evacuation models, traffic flow models, or resource allocation models, which can simulate the performance and impact of specific plans in actual operations.

[0059] The emergency response plan is finally determined by combining the simulation results and the optimization suggestions provided by the rule engine. By combining the actual data of the simulation and the logical processing of the rule engine, it is ensured that the final emergency response plan is feasible, optimal, and can be quickly and effectively executed in actual situations. A detailed analysis of the collected data may include a comparison of different simulation scenarios to select the emergency measures that best meet the current needs. At the same time, it is necessary to ensure that the plan meets regulatory requirements, resource constraints, and time sensitivity.

[0060] By integrating the rule engine and the simulation tool, the method proposed in the embodiment of the present application not only optimizes the design of the emergency response plan, but also improves the operability and effectiveness of the plan in actual disaster situations. This method allows decision makers to understand the potential effects and risks of various emergency measures in detail before implementation, so as to make more informed decisions.

[0061] In some examples, the controlling the resource management module generates a resource management solution based on the resource management instruction by a dynamic programming method and a heuristic algorithm, including: Dynamic programming results of resources planned by applying dynamic programming method in resource knowledge graph based on the resource management instructions, wherein the resource knowledge graph stores information on resource type, location, dependency, historical usage efficiency and emergency response strategy; Based on the above dynamic programming results, the resource allocation is further optimized based on the particle swarm optimization algorithm to obtain the above resource management solution.

[0062] Exemplarily, dynamic programming is performed to plan resource allocation using the information stored in the resource knowledge graph. The resource knowledge graph contains key information such as resource type, location, dependencies, historical usage efficiency, and emergency response strategies. Dynamic programming is an optimization technique that is suitable for solving multi-stage decision-making problems. With this approach, the system is able to plan optimal or near-optimal resource allocation in multiple time steps based on resource availability and demand, as well as other factors such as dependencies and historical data. The dynamic programming process obtains necessary input data from the resource knowledge graph, such as the location, type, historical usage frequency, etc. of each resource. Then, based on this information, the system constructs state transition equations and decision-making processes to minimize costs or maximize benefits.

[0063] The particle swarm optimization algorithm is used to further optimize resource allocation. Based on dynamic programming, the particle swarm optimization algorithm is used to further optimize the resource allocation scheme. The particle swarm optimization algorithm initializes a set of solutions (particles), each particle represents a potential resource allocation scheme. Each particle adjusts its position and speed according to the historical optimal solution of the individual and the group. Through the iterative process, the particles move towards a better resource allocation scheme until the stop condition or the preset number of iterations is reached, and finally the optimal resource management scheme is determined.

[0064] By combining dynamic programming and particle swarm optimization, the method proposed in the embodiment of the present application can not only take into account complex resource dependencies and constraints, but also use group wisdom to find the best resource allocation method. This method significantly improves the scientificity, accuracy and response speed of resource management, and is particularly suitable for dynamically changing disaster response environments, ensuring that resources can be quickly and effectively deployed to key areas, thereby maximizing their utility and reducing the impact of disasters.

[0065] In some examples, the large language model includes: an embedding layer, a Transformer encoding layer, an output layer and auxiliary components, the Transformer encoding layer includes multiple Transformer units, the output layer includes multiple fully connected layers and activation functions, the activation function is a softmax activation function, the auxiliary components include a Dropout layer and a LayerNormalization layer, and the optimizer of the large language model is an Adam optimizer.

[0066] Exemplarily, the large language model structure includes an embedding layer, a Transformer encoding layer, an output layer, and auxiliary components, and is optimized using an Adam optimizer.

[0067] The main function of the embedding layer is to convert the input natural language text (usually words or characters) into dense vector representations. These vectors are numerical forms that the model can process and can capture and represent the semantic and grammatical properties of words. The embedding layer usually uses pre-trained word embeddings such as GloVe or Word2Vec, or learns these representations during model training.

[0068] Transformer encoding layer This layer includes multiple Transformer units, each of which contains a self-attention mechanism and a feed-forward neural network. The Transformer encoding layer uses the self-attention mechanism to process the relationship between each part of the input data and all other parts, which enables the model to capture long-distance dependencies. This layer is the core part of understanding and processing input data. Each Transformer unit calculates the attention score of each element in the input sequence and updates its representation accordingly to better capture the contextual information of the text.

[0069] The output layer consists of multiple fully connected layers and an activation function. The softmax activation function is used to convert the output of the neural network into a probability distribution, which is suitable for classification tasks. The main function of the output layer is to make the final task prediction based on the deep features passed from the Transformer encoding layer, such as sentiment analysis, text classification, etc.

[0070] Auxiliary components include Dropout layer and Layer Normalization layer. Dropout layer is used to prevent model overfitting and increase the generalization ability of the model by randomly "discarding" the activation of a part of neurons during training. Layer Normalization layer is used to stabilize the training of neural network and accelerate the training process and improve model performance by normalizing layer input.

[0071] The optimizer uses the Adam optimizer, which can help large language models converge faster during training and improve the performance of the model on complex data sets.

[0072] In some examples, the modifying of the emergency response instruction and resource management instruction by the manual information includes: Evaluate the authenticity of the above artificial information based on expert evaluation method and intelligent evaluation model to obtain effective artificial information; The emergency response instructions and resource management instructions are modified based on the effective manual information.

[0073] Exemplarily, the purpose of the assessment is to screen out authentic and reliable artificial information so as to be used to correct and update emergency response instructions and resource management instructions, because in an emergency, the information received may include inaccurate or misleading data, which may lead to inappropriate response measures.

[0074] Expert evaluation method: Relies on the judgment of domain experts on information. Experts evaluate the authenticity of information based on their experience and knowledge, which may involve verifying the source of information, checking the logic of the content, etc.

[0075] Intelligent evaluation models: Automatically evaluate the authenticity of information using pre-trained machine learning models. These models may learn to identify features of real information based on historical data, such as consistency of information, relevance to known facts, etc.

[0076] Modify emergency response instructions and resource management instructions based on verified and effective manual information. Ensure that emergency response and resource allocation instructions reflect the latest and most accurate scenario conditions, thereby improving the effectiveness and efficiency of response measures.

[0077] Adjustments based on real manual information, such as evacuation routes, rescue team deployment, emergency medical service arrangements, etc. For example, if the manual information received shows that the impact of a disaster in a certain area is more serious than expected, it may be necessary to send more rescue resources or adjust the evacuation instructions.

[0078] Optimize the allocation and dispatch of resources such as disaster relief supplies and human resources based on real artificial information. For example, if valid information shows that a certain area is in urgent need, resource management instructions will prioritize the allocation of resources to that area.

[0079] By combining the intuitive judgment of experts with the data-driven analysis of the intelligent evaluation model, the method proposed in the embodiment of the present application not only improves the accuracy of judging the authenticity of information, but also can adjust and optimize emergency response measures and resource allocation plans in real time. This flexible and accurate adjustment mechanism greatly improves the efficiency of disaster response, reduces potential losses and impacts, and ensures the scientificity and effectiveness of response measures.

[0080] In some examples, the authenticity of the artificial information is evaluated based on the expert evaluation method and the intelligent evaluation model to obtain effective artificial information, including: Obtain the identity of the reporter corresponding to the manual information; Determine a preliminary authenticity score based on the identity of the reporter mentioned above; Performing preprocessing operations on the above artificial information to obtain key information; Send the above key information to the expert review unit to obtain the expert authenticity score through the expert evaluation method; Send the above key information to the above intelligent evaluation model to obtain the model authenticity score; Obtain the first weight corresponding to the above expert authenticity score; Obtain a second weight corresponding to the authenticity score of the above model; Determining the final authenticity based on the preliminary authenticity score, the expert authenticity score, the first weight, the model authenticity score and the second weight; The above-mentioned valid artificial information is determined based on the final authenticity and authenticity evaluation threshold.

[0081] Exemplarily, the identity of the person who reports the artificial information is first confirmed, which is an initial step in assessing the authenticity of the information. Knowing the identity of the source of the information can help assess the credibility of the information. For example, information from an experienced and reputable source may be considered more reliable, and the reliability of information reported by rescuers is higher.

[0082] A preliminary authenticity score is given based on the identity background and historical record of the reporter. This score provides a reference starting point for subsequent detailed evaluation and reflects the initial credibility of the information.

[0083] Clean and extract the collected artificial information, remove noise, highlight key information, improve the quality of data, and ensure the accuracy of subsequent evaluation.

[0084] The pre-processed key information is submitted to the expert review unit for review. Experts use their professional knowledge and experience to evaluate the authenticity of the information and provide an authenticity score based on professional judgment.

[0085] At the same time, key information is also fed into a pre-trained intelligent evaluation model, which automatically evaluates the authenticity of the information based on historical data and learned patterns, providing a data-driven authenticity score.

[0086] Determine the relative importance of expert scores (first weight) and model scores (second weight) in the final authenticity decision. The weight distribution reflects the degree of trust and reliance on different evaluation methods and is adjusted according to the needs of the situation.

[0087] A comprehensive final authenticity score is calculated by combining the preliminary authenticity score, expert authenticity score, model authenticity score and corresponding weights. All evaluation results are integrated to obtain a comprehensive judgment that takes into account multiple factors.

[0088] The final authenticity score is compared with the set authenticity assessment threshold to determine whether the information is authentic enough to be used for the modification of emergency response and resource management instructions. Only when the authenticity of the information is above a certain threshold will it be considered valid to ensure that the decision of the emergency command system is based on reliable information.

[0089] Through this series of comprehensive and systematic steps, the method proposed in the embodiment of the present application ensures that the disaster response command system can make fast and accurate decisions based on highly reliable artificial information, greatly improving the efficiency and effectiveness of responding to emergencies.

[0090] In some examples, the first weight information is negatively correlated with the disaster level, the second weight information is positively correlated with the disaster level, the first weight information is negatively correlated with the disaster occurrence time, and the second weight information is positively correlated with the disaster occurrence event; The above authenticity assessment threshold is positively correlated with the above disaster level, and the above authenticity assessment threshold is negatively correlated with the time of occurrence of the above disaster.

[0091] Exemplarily, the first weight is negatively correlated with the disaster level, that is, the higher the disaster level, the lower the weight of the expert score. In addition, the first weight is also negatively correlated with the time of the disaster, which means that the weight of the expert score gradually decreases from the beginning of the disaster to the later stage of the disaster. In the early stage of a disaster and for high-level disasters, the situation on the scene may change rapidly, and the expert's judgment may not be accurate or timely due to the lack of the latest on-the-ground information. Therefore, as the disaster level increases and time passes, the expert's score weight decreases relative to the data-driven model.

[0092] The second weight is positively correlated with the disaster level and the time of the disaster, that is, as the disaster level increases and time passes, the weight of the model score increases. As time goes by, more data is collected and analyzed by the model, and the model's predictions may become more accurate and reliable. Especially in high-level disasters, the model is able to quickly process large amounts of data and provide real-time and accurate assessments.

[0093] The authenticity assessment threshold is positively correlated with the disaster level, that is, as the disaster level increases, the authenticity requirements for information are higher and the assessment threshold increases. It is negatively correlated with the time of the disaster, meaning that the threshold gradually decreases over time. In severe disaster situations, the cost of false information is very high, so the authenticity requirements for information are more stringent. As the disaster continues, the emergency may require a faster response, so the rigor of the information may be reduced to allow for a quick response.

[0094] Through this dynamic adjustment mechanism, the disaster response system can flexibly adjust the information evaluation criteria according to the severity and development stage of different disasters. This not only improves the flexibility and adaptability of information processing, but also ensures that at different stages and different types of disasters, the decision-making basis is based on the most reliable information at the moment. This strategy helps to improve the efficiency and accuracy of emergency response, thereby more effectively managing disasters and mitigating their impacts.

[0095] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0096] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0097] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of the processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0098] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0100] An embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the process of predicting the decomposition rate in alumina production in the corresponding embodiment.

[0101] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0103] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0105] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0106] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0107] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A disaster response intelligent command system, characterized in that: include: An environmental information collection module is used to obtain environmental information of a target area, the environmental information including sensor information, satellite information and meteorological information, and send the environmental information to a disaster warning module and an intelligent decision-making module; A disaster warning module, used to obtain the environmental information and generate disaster warning information based on the environmental information, wherein the disaster warning module generates the disaster warning information based on a machine learning model; An intelligent decision-making module, used for acquiring the environmental information and generating emergency response instructions and resource management instructions based on the environmental information, wherein the intelligent decision-making module generates the emergency response instructions and the resource management instructions based on Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; An emergency response module, used to obtain the emergency response instruction and generate an emergency response plan based on the emergency response instruction, wherein the emergency response module generates the emergency response plan based on a rule engine and simulation results; A resource management module, used to obtain the resource management instruction and generate a resource management solution based on the resource management instruction, wherein the resource management module generates the resource management solution based on a dynamic programming method and a heuristic algorithm; The artificial information collection module is used to intelligently interact with the feedback personnel to obtain artificial information, and the artificial information is used to correct the emergency response instructions and resource management instructions. The artificial information collection module completes the intelligent interaction with the feedback personnel based on a large language model.

2. A control method for the intelligent command system for disaster response according to claim 1, characterized in that: include: Controlling the environmental information collection module to collect environmental information of the target area, and sending the environmental information to the disaster warning module and the intelligent decision-making module; Controlling the disaster warning module to generate the disaster warning information through the machine learning model based on the environmental information; Controlling the intelligent decision-making module to generate the emergency response instructions and the resource management instructions based on the environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm; Control the emergency response module to generate an emergency response plan based on the emergency response instruction through a rule engine and simulation results; Controlling the resource management module to generate a resource management solution based on the resource management instruction by a dynamic programming method and a heuristic algorithm; Control the artificial information collection module to complete intelligent interaction with the feedback personnel based on the large language model to obtain artificial information; The emergency response instructions and resource management instructions are modified by the manual information.

3. The control method according to claim 2, characterized in that: The controlling the disaster warning module to generate the disaster warning information through the machine learning model based on the environmental information includes: Performing a preprocessing operation on the environmental information to obtain preprocessed first environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; The first environmental information is input into a machine learning model corresponding to the disaster type to generate the disaster warning information corresponding to different disaster types.

4. The control method according to claim 2, characterized in that: The controlling the intelligent decision-making module generates the emergency response instruction and the resource management instruction based on the environmental information through Monte Carlo simulation, Agent-based model, genetic algorithm and particle swarm optimization algorithm, including: Performing a preprocessing operation on the environmental information to obtain preprocessed second environmental information, wherein the preprocessing operation includes a noise filtering operation, an outlier detection operation, and a data smoothing operation; Using Monte Carlo simulation based on the pre-processed second environmental information to obtain simulation evaluation effects of multiple possible response schemes; Based on the preprocessed environmental information, an agent-based model is used to simulate the simulated performance effects of each responding participant in multiple possible response schemes; Generate the disaster warning information based on the simulation evaluation effect and the simulation performance effect; Randomly generate an initialization resource allocation scheme population and an initialization example population based on the second environment information and available resource information; Based on the initial resource allocation scheme population, multiple iterations are performed through a genetic algorithm to obtain a preliminary optimization result; Based on the preliminary optimization result and the initialization example group, a particle swarm algorithm is used to perform multiple iterations to obtain a final optimization result; The resource management instruction is generated based on the final optimization result.

5. The control method according to claim 2, characterized in that: The controlling the emergency response module to generate an emergency response plan based on the emergency response instruction through a rule engine and simulation results includes: Processing the emergency response instruction using a rule engine to obtain an emergency response plan optimized by the rule engine; Simulating the optimized emergency response plan based on a disaster response simulation tool to obtain simulation results; Generate an emergency response plan based on the simulation results and the optimization suggestions of the rule engine.

6. The control method according to claim 2, characterized in that: The controlling the resource management module to generate a resource management solution based on the resource management instruction by a dynamic programming method and a heuristic algorithm includes: Dynamic programming results of resources planned by applying a dynamic programming method in a resource knowledge graph based on the resource management instructions, wherein the resource knowledge graph stores information on resource types, locations, dependencies, historical usage efficiency, and emergency response strategies; Based on the dynamic programming result, resource allocation is further optimized based on a particle swarm optimization algorithm to obtain the resource management solution.

7. The control method according to claim 2, characterized in that: The large language model includes: an embedding layer, a Transformer encoding layer, an output layer and an auxiliary component, the Transformer encoding layer includes multiple Transformer units, the output layer includes multiple fully connected layers and an activation function, the activation function is a softmax activation function, the auxiliary components include a Dropout layer and a Layer Normalization layer, and the optimizer of the large language model is an Adam optimizer.

8. The control method according to claim 2, characterized in that: The modifying of the emergency response instruction and the resource management instruction by the manual information includes: Evaluate the authenticity of the artificial information based on an expert evaluation method and an intelligent evaluation model to obtain effective artificial information; The emergency response instructions and resource management instructions are modified based on the valid manual information.

9. The control method according to claim 8, characterized in that: The authenticity of the artificial information is evaluated based on the expert evaluation method and the intelligent evaluation model to obtain effective artificial information, including: Obtain the identity of the reporter corresponding to the manual information; Determining a preliminary authenticity score based on the identity of the reporter; Performing preprocessing operations on the artificial information to obtain key information; Sending the key information to the expert review unit to obtain an expert authenticity score through an expert evaluation method; Sending the key information to the intelligent evaluation model to obtain a model authenticity score; Obtaining a first weight corresponding to the expert authenticity score; Obtaining a second weight corresponding to the model authenticity score; determining a final authenticity based on the preliminary authenticity score, the expert authenticity score, the first weight, the model authenticity score, and the second weight; The valid artificial information is determined based on the final authenticity and the authenticity evaluation threshold.

10. The control method according to claim 9, characterized in that: The first weight information is negatively correlated with the disaster level, the second weight information is positively correlated with the disaster level, the first weight information is negatively correlated with the disaster occurrence time, and the second weight information is positively correlated with the disaster occurrence event; The authenticity assessment threshold is positively correlated with the disaster level, and the authenticity assessment threshold is negatively correlated with the disaster occurrence time.

Citation Information

Patent Citations

  • Emergency communication intelligent emergency decision support system

    CN110210697A

  • User instruction processing method based on AI system

    CN113127827A

  • Geological disaster emergency management and decision-making system and method and readable storage medium

    CN114399210A

  • Resource allocation method under icing emergency working condition and related equipment

    CN116307581A

  • Intelligent natural disaster early warning system and method based on big data

    CN116704708A

Cited By

  • Emergency expert resource overall management system and method

    CN119904208A

  • Intelligent emergency command system and method based on artificial intelligence

    CN120387705A

  • Intelligent generation method, device and equipment for disaster response plan and medium

    CN121544083A