Smart park emergency communication scheduling method based on large model

Abstract

The invention discloses a smart park emergency communication scheduling method based on a large model, and belongs to the technical field of large models, and the method comprises the following steps: S1, collecting data in real time through deploying a sensor network, and constructing a behavior representation model of each task; s2, dynamically adjusting task behavior representation; s3, constructing a behavior representation evolution network, and realizing dynamic evolution of task behavior characteristics in a task execution process; s4, representing tasks and resources as cells through a cellular automaton model, and optimizing a global scheduling effect in a behavior representation evolution network space; s5, dynamically adjusting the task priority, the execution path and the resource allocation strategy according to the real-time environment change in combination with a context awareness mechanism; and S6, optimizing task priority adjustment and resource allocation by using a game theory model and a collective behavior evolution mechanism, and finally generating a global optimal scheduling strategy. According to the method, the emergency communication scheduling of the smart park is optimized through the behavior representation evolution network, the cellular automaton model and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of large models, and in particular to a large-model-based smart park emergency communication dispatching method. Background Art

[0002] With the continuous development and application of smart parks, task management and scheduling within the park have become a key link to improve efficiency and ensure safety. Smart parks are usually composed of various sensors and smart devices, and use the Internet of Things technology to collect and process data within the park, thereby realizing intelligent scheduling and management of resources. However, with the expansion of the scale of the park and the complexity of the types of tasks, traditional task scheduling methods and resource management strategies have gradually exposed some shortcomings, especially in emergency response and resource optimization.

[0003] Existing park scheduling systems usually use rule-based static scheduling algorithms or simple priority sorting mechanisms to handle task execution. The main problem with this type of method is that it lacks flexibility and cannot respond to complex environmental changes in the park and dynamic dependencies between tasks in real time. For example, when encountering emergencies or drastic changes in the environment within the park, traditional methods are often unable to adjust task priorities or resource allocation strategies in a timely manner, resulting in delays in task execution or waste of resources. In addition, traditional scheduling algorithms fail to fully consider the need for collaboration between tasks, and there is no effective mechanism to deal with resource conflicts and sharing issues. With the increase in tasks and the shortage of resources in the park, simple scheduling methods are difficult to meet the requirements of efficient and intelligent management.

[0004] At present, some scheduling methods based on artificial intelligence have begun to enter the field of park management. These methods can better predict task requirements and environmental changes by introducing machine learning and optimization algorithms. However, most of these methods rely on a large amount of historical data for model training, and it is difficult to respond in a timely manner in emergency situations. For example, when the priority of an emergency task changes or the environment changes suddenly, the prediction model based on historical data is difficult to provide a fast and effective scheduling strategy. In addition, some existing intelligent scheduling methods mostly adopt a single optimization goal, such as the optimization of task priority, ignoring the synergy between tasks and the comprehensive consideration of multiple resource requirements. This makes the existing methods show great limitations in the scheduling of complex park tasks.

[0005] In addition to task scheduling, resource management within the park is also a key issue. Traditional resource management systems generally optimize resource usage by setting fixed resource allocation strategies. However, this static allocation method cannot fully maximize the benefits of resources when faced with the increasing complexity of park tasks. For example, when multiple tasks occur simultaneously within the park and there is a strong dependency between them, static resource allocation schemes often lead to resource conflicts and cannot effectively avoid resource contention or delayed execution of tasks. In addition, the tasks and resource requirements within the park are usually dynamically changing, and existing resource management methods often ignore this time-varying characteristic, resulting in insufficient adjustment of scheduling and resource allocation strategies when task and resource requirements change rapidly, thus affecting the operating efficiency of the entire park.

[0006] Emergency task scheduling is another major challenge in smart parks. In emergency situations, task priorities and resource allocation must be quickly adjusted according to real-time environmental changes. Traditional emergency scheduling methods rely on preset task priorities and simple emergency response rules, and are unable to make dynamic decisions based on real-time environmental data. For example, in emergencies such as fires and accidents, the existing scheduling system may not be able to obtain and process relevant data in real time, resulting in inflexible and inefficient scheduling of emergency tasks. In addition, most of the existing emergency scheduling methods lack a flexible task collaboration mechanism and cannot establish an effective collaborative relationship between tasks to improve emergency response efficiency. With the increase in the types of tasks in the park, the existing emergency scheduling strategies have gradually shown their inability to meet the needs of fast, dynamic and efficient emergency scheduling.

[0007] Therefore, existing technologies still have many problems when facing complex task scheduling and resource management in the park. Traditional scheduling methods fail to fully consider the dynamic changes of task behavior and the need for collaboration between tasks, and lack the flexibility to cope with changes in the park environment. Although existing intelligent scheduling systems can optimize task scheduling to a certain extent, they usually rely on historical data and a single optimization goal, and fail to effectively solve the priority changes of emergency tasks, collaborative optimization between tasks, and the management of resource conflicts.

[0008] Therefore, how to provide a smart park emergency communication dispatching method based on a large model is an urgent problem that technicians in this field need to solve. Summary of the invention

[0009] One purpose of the present invention is to propose a smart park emergency communication dispatching method based on a large model. Based on a large model and a situational awareness mechanism, the present invention proposes an efficient smart park emergency communication dispatching method. Through innovative technologies such as real-time data stream analysis, dynamic adjustment of task behavior, behavior representation evolutionary network and cellular automaton model, it can accurately adjust task priority, execution path and resource allocation strategy. This method makes full use of environmental changes and collaboration between tasks, optimizes the global dispatching strategy through game theory and collective behavior mechanism, and has the advantages of accurate task dispatching, efficient resource utilization, strong adaptability and system adaptive adjustment, which significantly improves the efficiency and reliability of emergency response.

[0010] The smart park emergency communication dispatching method based on a large model according to an embodiment of the present invention comprises the following steps: S1. By deploying a sensor network to collect real-time environmental data, task status, resource usage and equipment operation status in the park, a behavior representation model for each task is constructed, wherein the behavior representation model includes static resource requirements and dynamic execution behavior characteristics of the task; S2, extract the resource consumption rate, priority changes, response speed to environmental changes and coordination requirements of tasks through data flow analysis, and dynamically adjust the task behavior representation; S3. Construct a behavior representation evolution network and use information dissemination, resource flow and behavior interaction mechanisms to achieve dynamic evolution of task behavior characteristics during task execution; S4. Represent tasks and resources as cells through the cellular automaton model. In the behavior representation evolution network space, simulate the interaction between tasks according to local evolution rules, and reduce resource conflicts and optimize the global scheduling effect by sharing resources or adjusting execution timing. S5. Combined with the context-aware mechanism, dynamically adjust the task priority, execution path and resource allocation strategy according to real-time environmental changes to ensure that the task behavior is adapted to environmental changes and generate a context-aware driven task scheduling solution; S6. A task scheduling solution driven by contextual awareness uses game theory models and collective behavior evolution mechanisms to optimize task priority adjustment and resource allocation through collaboration and resource sharing between tasks, and ultimately generates a global optimal scheduling strategy.

[0011] Optionally, the S2 specifically includes: S21. Set tasks based on real-time data flow analysis At the moment The resource consumption rate is ,Task At the moment The priority is ,Task At the moment The speed of response to environmental changes is , and tasks With the task The collaborative demand is : ; in, Indicates the task In the time interval Resource consumption within ; in, Indicates the task the rate at which the priority changes over time, Indicates the task At the moment Priority; ; in, Indicates the task At the moment The priority of Indicates the task With the task The strength of the dependency relationship between them; S22. Build a dynamic adjustment mechanism for task behavior based on tasks. The resource consumption rate, priority changes, response speed to environmental changes, and coordination requirements between tasks are used to define the task behavior adjustment function: ; in, Indicates the task At the moment The behavior adjustment function, Indicates the task The rate of change of resource consumption rate, , , and represents the adjustment factor; S23. Adjust the function according to the task behavior and adjust the task in real time The resource requirements, priorities and execution paths of When, the task The resource demand increases; when When, the task The priority of When, the task Collaboration with other tasks is enhanced, including Indicates the preset resource demand threshold. Indicates the preset priority threshold, represents a preset collaboration intensity threshold; S24. Optimize resource sharing and collaboration among tasks through inter-task coordination mechanism. When adjusting the task With the task execution time.

[0012] Optionally, the S3 specifically includes: S31, constructing a behavior representation evolution network, wherein the behavior representation evolution network is composed of a plurality of nodes, wherein each node represents a task, the nodes are connected by weighted edges, indicating the interdependence between the tasks, and the node states evolve dynamically over time in the network; S32. Define the behavior feature evolution rules based on the neighborhood tasks for each task, and describe the changes of the behavior features of the task over time through the dynamic evolution formula: ; in, Representation Task At the moment behavioral characteristics, Representation Task At the moment behavioral characteristics, Representation Task With the task The mutual influence weight between Representation Task At the moment behavioral characteristics, Representation Task At the moment The neighborhood task set, Representation Task At the moment resource requirements, , and represents the adjustment parameter; S33, introduce information dissemination mechanism, with the coordination and information exchange between tasks as the core. Task nodes disseminate information through weighted connections according to the priority, resource requirements and execution status between tasks. Each task node updates its behavior characteristics according to the information of neighboring tasks. S34. Introduce a resource flow mechanism. When the resource demand of a task exceeds the available resources, the task obtains the required resources through the resource flow mechanism with neighboring tasks: ; in, Representation Task At the moment resource requirements, Representation Task With the task The resource flow coefficient between Representation Task At the moment The change in resource demand; S35. Through the combined effects of information dissemination mechanism, resource flow mechanism and behavioral interaction, task behavior characteristics are dynamically evolved in the behavior representation evolution network.

[0013] Optionally, the S4 specifically includes: S41, mapping each task and resource to a cell in a cellular automaton model, wherein each task cell represents an independent task, including resource requirements, execution priority, and coordination requirements; each resource cell represents the resource status in the park, including the availability and current occupancy of the resource, and the relationship between tasks and resources is represented by the edge connection between cells; S42. Define the local evolution rules of the cellular automaton, specifically: In the behavioral representation evolution network space, the behavioral characteristics of each task cell are adjusted according to its own resource requirements, execution status, and interaction with neighboring tasks and resources. When a task shares resources with neighboring tasks, the behavior of the task cell is dynamically updated. S43. Based on the local evolution rules, the task cells and the adjacent resource cells adjust their states through local interactions: when task A and task B share the same resources, the priority of one of the tasks is adjusted or the execution of one of the tasks is delayed to avoid resource conflicts; S44. The behavior characteristics of each task cell are self-optimized and adjusted in the local environment over time. The interaction between task cells and resource cells gradually affects the global scheduling effect through the local evolution mechanism. S45. Through the local evolution mechanism of the cellular automaton model, the entire behavior representation evolution network continuously performs adaptive adjustments, ultimately achieving optimization of the global scheduling effect.

[0014] Optionally, the S5 specifically includes: S51. Based on real-time environmental data and combined with the context awareness mechanism, the task execution environment is dynamically modeled, where the impact of environmental changes on task execution is represented by a weighted comprehensive model: ; in, Representation Task At the moment The environmental perception value, Indicates The weight of the environmental factors affecting the task, represents the number of environmental factors, Indicates Environmental factors at the moment The real-time value of Indicates The expected value of an environmental factor, Indicates The standard deviation of each environmental factor; S52, according to the environmental perception value , for each task Define an environmental sensitivity factor to reflect the task The sensitivity under different environmental conditions, the sensitivity factor calculation formula is: ; in, Representation Task The environmental sensitivity factor, Representation Task Environmental response sensitivity, Representation Task Environmental response thresholds; S53, according to the task Environmental sensitivity factor , dynamically adjust task priorities, execution paths, and resource allocation: ; in, Indicates the task At the moment The priority of Representation Task The initial priority of represents the path adjustment coefficient, Indicates The weight of resource factors, Indicates the number of resource types. Indicates the task At the moment For The demand for a resource; S54. Based on the adjusted task priority, the task execution path is further optimized and adjusted in combination with the impact of environmental changes: ; in, Indicates the task At the moment The execution path of Represents the initial execution path of the task, represents the exponential function, represents the adjustment factor; S55. Generate a context-awareness-driven task scheduling solution based on the dynamically adjusted task priority, execution path, and resource allocation.

[0015] Optionally, the S6 specifically includes: S61. Based on the task scheduling scheme driven by context awareness, the collaborative relationship between tasks is modeled through a game theory model. The game theory model considers resource competition and sharing between tasks. The goal is to achieve optimal resource allocation through the game between tasks. The objective function of the game model is: ; in, represents maximization, Representation Task The utility function of Representation Task From the resources The resource benefits obtained, Representation Task At the moment resource The consumption cost, Indicates the number of resource types; S62. Combined with the collective behavior evolution mechanism, by evolving the strategies between tasks in the game theory model, the optimal resource allocation strategy for each task is calculated. The evolution of the optimal resource allocation strategy follows the Nash equilibrium condition: for all tasks When the utility function reaches its maximum value, the strategy will not be changed. The Nash equilibrium formula is: ; in, Representation Task The optimal utility under Nash equilibrium is Representation Task Optimal utility under Nash equilibrium; S63. Based on the optimal resource allocation strategy obtained by the game theory model, the global task scheduling is optimized by combining task priority and resource consumption; S64. Through the game and collaboration optimization between tasks, the global optimal scheduling strategy is finally generated.

[0016] The beneficial effects of the present invention are: First, the present invention adopts a task behavior characterization model based on a large model, which can reflect the resource requirements, priority changes, and coordination requirements of tasks in real time. This innovative method can dynamically adjust the behavior characterization of tasks according to real-time sensor data and task status, so as to more accurately predict changes in the task execution process. Compared with traditional methods, it can more efficiently deal with complex tasks and emergencies, thereby improving the adaptability and efficiency of park task scheduling, especially in terms of changes in task priority and resource requirements, and achieving accurate real-time scheduling.

[0017] Secondly, the present invention combines the behavior characterization evolution network with information dissemination and resource flow mechanisms to effectively simulate the collaboration and resource sharing between tasks. In traditional scheduling methods, the dependencies between tasks are often ignored, resulting in resource conflicts and scheduling conflicts. The present invention introduces a dynamic evolution network and uses weighted connections to represent the interdependencies between tasks, thereby realizing the dynamic evolution of task behavior characteristics. During the execution of tasks, task nodes update their behavior characteristics based on the information of neighboring tasks, so that tasks can be adjusted in real time during execution, thereby optimizing collaboration and resource allocation between tasks. This not only improves the global effect of task scheduling, but also effectively reduces resource conflicts and execution delays.

[0018] In addition, by utilizing the context-aware mechanism, the present invention can capture environmental changes within the park in real time, and dynamically adjust the priority, execution path, and resource allocation strategy of the task according to these changes. Traditional methods are usually unable to respond to environmental changes in a timely manner, resulting in a lag in the scheduling strategy. However, through the context-aware mechanism, the present invention can accurately adjust the task scheduling plan according to the real-time environmental changes within the park. Especially in emergency situations, the scheduling plan can be quickly adjusted according to sudden environmental changes and real-time changes in task requirements to ensure that emergency tasks are given priority, thereby improving the park's emergency response capabilities.

[0019] Finally, the present invention introduces a task collaboration and resource sharing mechanism based on the game theory model, which optimizes the priority adjustment and resource allocation of tasks through collaboration between tasks. This mechanism avoids the single task optimization problem in traditional methods, and can establish a more flexible and efficient collaboration relationship between tasks, ensuring that resources are used more balanced and reasonably in a complex task environment, while also improving the overall efficiency of multi-task collaboration in the park. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is an overall flow chart of the smart park emergency communication dispatching method based on a large model proposed in the present invention. DETAILED DESCRIPTION

[0021] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0022] refer to Figure 1 ,The emergency communication dispatching method for a smart park ,based on a large model includes the following steps: S1. By deploying a sensor network to collect real-time environmental data, task status, resource usage and equipment operation status in the park, a behavior representation model for each task is constructed, wherein the behavior representation model includes static resource requirements and dynamic execution behavior characteristics of the task; S2, extract the resource consumption rate, priority changes, response speed to environmental changes and coordination requirements of tasks through data flow analysis, and dynamically adjust the task behavior representation; S3. Construct a behavior representation evolution network and use information dissemination, resource flow and behavior interaction mechanisms to achieve dynamic evolution of task behavior characteristics during task execution; S4. Represent tasks and resources as cells through the cellular automaton model. In the behavior representation evolution network space, simulate the interaction between tasks according to local evolution rules, and reduce resource conflicts and optimize the global scheduling effect by sharing resources or adjusting execution timing. S5. Combined with the context-aware mechanism, dynamically adjust the task priority, execution path and resource allocation strategy according to real-time environmental changes to ensure that the task behavior is adapted to environmental changes and generate a context-aware driven task scheduling solution; S6. A task scheduling solution driven by contextual awareness uses game theory models and collective behavior evolution mechanisms to optimize task priority adjustment and resource allocation through collaboration and resource sharing between tasks, and ultimately generates a global optimal scheduling strategy.

[0023] In this implementation, S2 specifically includes: S21. Set tasks based on real-time data flow analysis At the moment The resource consumption rate is ,Task At the moment The priority is ,Task At the moment The speed of response to environmental changes is , and tasks With the task The collaborative demand is : ; in, Indicates the task In the time interval Resource consumption within ; in, Indicates the task the rate at which the priority changes over time, Indicates the task At the moment Priority; ; in, Indicates the task At the moment The priority of Indicates the task With the task The strength of the dependency relationship between them; S22. Build a dynamic adjustment mechanism for task behavior based on tasks. The resource consumption rate, priority changes, response speed to environmental changes, and coordination requirements between tasks are used to define the task behavior adjustment function: ; in, Indicates the task At the moment The behavior adjustment function, Indicates the task The rate of change of resource consumption rate, , , and represents the adjustment factor; S23. Adjust the function according to the task behavior and adjust the task in real time The resource requirements, priorities and execution paths of When, the task The resource demand increases; when When, the task The priority of When, the task Collaboration with other tasks is enhanced, including Indicates the preset resource demand threshold. Indicates the preset priority threshold, represents a preset collaboration intensity threshold; S24. Optimize resource sharing and collaboration among tasks through inter-task coordination mechanism. When adjusting the task With the task execution time.

[0024] In this implementation, S3 specifically includes: S31, constructing a behavior representation evolution network, wherein the behavior representation evolution network is composed of a plurality of nodes, wherein each node represents a task, the nodes are connected by weighted edges, indicating the interdependence between the tasks, and the node states evolve dynamically over time in the network; S32. Define the behavior feature evolution rules based on the neighborhood tasks for each task, and describe the changes of the behavior features of the task over time through the dynamic evolution formula: ; in, Representation Task At the moment behavioral characteristics, Representation Task At the moment behavioral characteristics, Representation Task With the task The mutual influence weight between Representation Task At the moment behavioral characteristics, Representation Task At the moment The neighborhood task set, Representation Task At the moment resource requirements, , and represents the adjustment parameter; S33, introduce information dissemination mechanism, with the coordination and information exchange between tasks as the core. Task nodes disseminate information through weighted connections according to the priority, resource requirements and execution status between tasks. Each task node updates its behavior characteristics according to the information of neighboring tasks. S34. Introduce a resource flow mechanism. When the resource demand of a task exceeds the available resources, the task obtains the required resources through the resource flow mechanism with neighboring tasks: ; in, Representation Task At the moment resource requirements, Representation Task With the task The resource flow coefficient between Representation Task At the moment The change in resource demand; S35. Through the combined effects of information dissemination mechanism, resource flow mechanism and behavioral interaction, task behavior characteristics are dynamically evolved in the behavior representation evolution network.

[0025] In this implementation manner, the S4 specifically includes: S41, mapping each task and resource to a cell in a cellular automaton model, wherein each task cell represents an independent task, including resource requirements, execution priority, and coordination requirements; each resource cell represents the resource status in the park, including the availability and current occupancy of the resource, and the relationship between tasks and resources is represented by the edge connection between cells; S42. Define the local evolution rules of the cellular automaton, specifically: In the behavioral representation evolution network space, the behavioral characteristics of each task cell are adjusted according to its own resource requirements, execution status, and interaction with neighboring tasks and resources. When a task shares resources with neighboring tasks, the behavior of the task cell is dynamically updated. S43. Based on the local evolution rules, the task cells and the adjacent resource cells adjust their states through local interactions: when task A and task B share the same resources, the priority of one of the tasks is adjusted or the execution of one of the tasks is delayed to avoid resource conflicts; S44. The behavior characteristics of each task cell are self-optimized and adjusted in the local environment over time. The interaction between task cells and resource cells gradually affects the global scheduling effect through the local evolution mechanism. S45. Through the local evolution mechanism of the cellular automaton model, the entire behavior representation evolution network continuously performs adaptive adjustments, ultimately achieving optimization of the global scheduling effect.

[0026] In this implementation manner, S5 specifically includes: S51. Based on real-time environmental data and combined with the context awareness mechanism, the task execution environment is dynamically modeled, where the impact of environmental changes on task execution is represented by a weighted comprehensive model: ; in, Representation Task At the moment The environmental perception value, Indicates The weight of the environmental factors affecting the task, represents the number of environmental factors, Indicates Environmental factors at the moment The real-time value of Indicates The expected value of an environmental factor, Indicates The standard deviation of each environmental factor; S52, according to the environmental perception value , for each task Define an environmental sensitivity factor to reflect the task The sensitivity under different environmental conditions, the sensitivity factor calculation formula is: ; in, Indicates the task The environmental sensitivity factor, Indicates the task Environmental response sensitivity, Indicates the task Environmental response thresholds; S53, according to the task Environmental sensitivity factor , dynamically adjust task priorities, execution paths, and resource allocation: ; in, Indicates the task At the moment The priority of Indicates the task The initial priority of represents the path adjustment coefficient, Indicates The weight of resource factors, Indicates the number of resource types. Indicates the task At the moment For The demand for a resource; S54. Based on the adjusted task priority, the task execution path is further optimized and adjusted in combination with the impact of environmental changes: ; in, Indicates the task At the moment The execution path of Represents the initial execution path of the task, represents the exponential function, represents the adjustment factor; S55. Generate a context-awareness-driven task scheduling solution based on the dynamically adjusted task priority, execution path, and resource allocation.

[0027] In this implementation manner, S6 specifically includes: S61. Based on the task scheduling scheme driven by context awareness, the collaborative relationship between tasks is modeled through a game theory model. The game theory model considers resource competition and sharing between tasks. The goal is to achieve optimal resource allocation through the game between tasks. The objective function of the game model is: ; in, represents maximization, Indicates the task The utility function of Indicates the task From the resources The resource benefits obtained, Indicates the task At the moment resource The consumption cost, Indicates the number of resource types; S62. Combined with the collective behavior evolution mechanism, by evolving the strategies between tasks in the game theory model, the optimal resource allocation strategy for each task is calculated. The evolution of the optimal resource allocation strategy follows the Nash equilibrium condition: for all tasks When the utility function reaches its maximum value, the strategy will not be changed. The Nash equilibrium formula is: ; in, Indicates the task The optimal utility under Nash equilibrium is Indicates the task Optimal utility under Nash equilibrium; S63. Based on the optimal resource allocation strategy obtained by the game theory model, the global task scheduling is optimized by combining task priority and resource consumption; S64. Through the game and collaboration optimization between tasks, the global optimal scheduling strategy is finally generated.

[0028] Embodiment 1: In order to verify the feasibility of the present invention in implementation, the present invention is applied to the emergency communication dispatch system of a smart park, which is located in the high-tech development zone of a certain city. The park is mainly divided into several functional areas, including office buildings, scientific research buildings, production workshops, and public service areas. There are a large number of equipment and sensors in the park, among which the role of the sensors is to monitor and collect various environmental data (such as temperature, humidity, air pressure, etc.) and equipment operating status (such as power consumption, equipment failure, etc.) in the park in real time. In addition, there are also many tasks in the park that need to be coordinated and executed between different areas, such as equipment maintenance, resource scheduling, security inspections, etc.

[0029] Traditional park scheduling methods usually rely on manual scheduling and a fixed task priority system. This method has problems such as slow response time and uneven resource scheduling. Especially in emergency situations, the inflexibility of task scheduling and resource conflicts are particularly prominent. Through the large model-based emergency communication scheduling method of the present invention, we can quickly perform dynamic scheduling optimization according to real-time environmental changes, task requirements and resource status when an emergency event occurs in the park, thereby improving the overall emergency response efficiency of the park.

[0030] In specific applications, the present invention collects environmental data, task status, resource usage and equipment operation status in real time by deploying a sensor network in the park. These data are transmitted to the central control platform of the park. The platform performs data flow analysis based on the computing power of the large model, extracts information such as resource consumption rate, priority change, and reaction speed to environmental changes of each task, and then realizes task priority adjustment, resource allocation and execution path optimization by dynamically adjusting task behavior representation.

[0031] In an emergency situation of power failure, a power system failure occurred in a laboratory building in the park, causing the experimental equipment in the building to fail to work properly. Based on the environmental data (such as voltage changes, equipment shutdown information) and task data (such as equipment maintenance tasks, emergency response tasks) collected by sensors in real time, the scheduling system immediately generated a new scheduling plan based on the urgency of the task and the availability of resources. In this plan, the equipment maintenance task was promoted to a priority task. At the same time, the scheduling system adjusted the allocation of repair resources (such as technicians and maintenance tools) to avoid resource conflicts, and adjusted the execution order between tasks through the cellular automation model, thereby achieving rapid response.

[0032] In this emergency dispatch, the park's dispatch system was able to generate a new dispatch plan within 10 minutes, and through the information dissemination mechanism and resource flow mechanism, all tasks were reallocated in the shortest possible time. Compared with the traditional manual dispatch method, the task that originally took 30 minutes to complete was completed in less than 15 minutes after implementation, and the dispatch efficiency was increased by nearly 50%.

[0033] In order to further verify the effect of the present invention, we recorded and analyzed the data of 10 emergency dispatch cases of different scales in the park within a week. These cases include different types of emergency events such as power system failure, equipment abnormality, fire alarm, etc.

[0034] Table 1 Data comparison table of traditional scheduling method and scheduling method of the present invention

[0035] Based on the above Table 1, we compared the scheduling efficiency, task completion time and resource utilization of the traditional scheduling method and the method of the present invention under different types of emergency events.

[0036] First, from the dispatching time, the traditional dispatching method has a long response time, especially in emergency events such as power system failure and equipment abnormality, the dispatching time is 30 minutes and 25 minutes respectively. After using the method of the present invention, the dispatching time is greatly shortened, the dispatching time for power system failure is 10 minutes, and the dispatching time for equipment abnormality is 8 minutes, which shows that the method of the present invention can generate task scheduling plans more quickly in emergency situations. Obviously, the present invention significantly improves the speed of emergency response through the application of real-time environmental data analysis and intelligent dispatching models.

[0037] Secondly, the comparison of task completion time further verifies the advantages of the present invention. In the traditional scheduling method, the task completion time in power system failure and equipment abnormality tasks is 60 minutes and 50 minutes respectively, while the method of the present invention shortens the task completion time to 15 minutes and 12 minutes, and the task completion speed is increased by nearly 50%. This shows that the present invention not only provides optimization in scheduling time, but also greatly improves efficiency at the task execution level, especially when the emergency response time is relatively tight.

[0038] Finally, the comparison of resource utilization further highlights the advantages of the present invention in resource allocation and collaborative optimization. In traditional scheduling methods, resource utilization is generally low, especially in the case of power system failure and experimental equipment failure, the resource utilization is 65% and 60% respectively. However, after applying the present invention, the resource utilization rate has been significantly improved, reaching 90% in the case of power system failure and 85% in the case of equipment abnormality. This improvement shows that the present invention can not only respond quickly to emergency events, but also maximize the utilization efficiency of resources in the park through precise resource allocation and optimization.

[0039] From the above data comparison, it can be seen that the present invention has significant advantages in handling emergency events in the park. Whether in terms of dispatch response speed, task completion efficiency or resource utilization efficiency, it is superior to the traditional dispatching method. The invention provides an efficient and intelligent solution that can achieve more accurate task scheduling and resource management in the park, significantly improving the park's emergency response capabilities and resource optimization efficiency.

[0040] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A smart park emergency communication dispatching method based on a large model, characterized in that: The steps include: S1. By deploying a sensor network to collect real-time environmental data, task status, resource usage and equipment operation status in the park, a behavior representation model for each task is constructed, wherein the behavior representation model includes static resource requirements and dynamic execution behavior characteristics of the task; S2, extract the resource consumption rate, priority changes, response speed to environmental changes and coordination requirements of tasks through data flow analysis, and dynamically adjust the task behavior representation; S3. Construct a behavior representation evolution network and use information dissemination, resource flow and behavior interaction mechanisms to achieve dynamic evolution of task behavior characteristics during task execution; S4. Represent tasks and resources as cells through the cellular automaton model. In the behavior representation evolution network space, simulate the interaction between tasks according to local evolution rules, and reduce resource conflicts and optimize the global scheduling effect by sharing resources or adjusting execution timing. S5. Combined with the context-aware mechanism, dynamically adjust the task priority, execution path and resource allocation strategy according to real-time environmental changes to ensure that the task behavior is adapted to environmental changes and generate a context-aware driven task scheduling solution; S6. A task scheduling solution driven by contextual awareness uses game theory models and collective behavior evolution mechanisms to optimize task priority adjustment and resource allocation through collaboration and resource sharing between tasks, and ultimately generates a global optimal scheduling strategy.

2. The large model-based smart park emergency communication dispatching method according to claim 1 is characterized in that: The S2 specifically includes: S21. Set tasks based on real-time data flow analysis At the moment The resource consumption rate is ,Task At the moment The priority is ,Task At the moment The speed of response to environmental changes is , and tasks With the task The collaborative demand is : ； in, Indicates the task In the time interval Resource consumption within ； in, Indicates the task the rate at which the priority changes over time, Indicates the task At the moment Priority; ； in, Indicates the task At the moment The priority of Indicates the task With the task The strength of the dependency relationship between them; S22. Build a dynamic adjustment mechanism for task behavior based on tasks. The resource consumption rate, priority changes, response speed to environmental changes, and coordination requirements between tasks are used to define the task behavior adjustment function: ； in, Indicates the task At the moment The behavior adjustment function of Indicates the task The rate of change of resource consumption rate, , , and represents the adjustment factor; S23. Adjust the function according to the task behavior and adjust the task in real time The resource requirements, priorities and execution paths of When, the task The resource demand increases; when When, the task The priority of When, the task Collaboration with other tasks is enhanced, including Indicates the preset resource demand threshold. Indicates the preset priority threshold, represents a preset collaboration intensity threshold; S24. Optimize resource sharing and collaboration among tasks through inter-task coordination mechanism. When adjusting the task With the task execution time.

3. The smart park emergency communication dispatching method based on a large model according to claim 1 is characterized in that: The S3 specifically includes: S31, constructing a behavior representation evolution network, wherein the behavior representation evolution network is composed of a plurality of nodes, wherein each node represents a task, the nodes are connected by weighted edges, indicating the interdependence between the tasks, and the node states evolve dynamically over time in the network; S32. Define the behavior feature evolution rules based on the neighborhood tasks for each task, and describe the changes of the behavior features of the task over time through the dynamic evolution formula: ； in, Indicates the task At the moment behavioral characteristics, Indicates the task At the moment behavioral characteristics, Indicates the task With the task The mutual influence weight between Indicates the task At the moment behavioral characteristics, Indicates the task At the moment The neighborhood task set, Indicates the task At the moment resource requirements, , and represents the adjustment parameter; S33, introduce information dissemination mechanism, with the coordination and information exchange between tasks as the core. Task nodes disseminate information through weighted connections according to the priority, resource requirements and execution status between tasks. Each task node updates its behavior characteristics according to the information of neighboring tasks. S34. Introduce a resource flow mechanism. When the resource demand of a task exceeds the available resources, the task obtains the required resources through the resource flow mechanism with neighboring tasks: ； in, Indicates the task At the moment resource requirements, Indicates the task With the task The resource flow coefficient between Indicates the task At the moment The change in resource demand; S35. Through the combined effects of information dissemination mechanism, resource flow mechanism and behavioral interaction, task behavior characteristics are dynamically evolved in the behavior representation evolution network.

4. The large model-based smart park emergency communication dispatching method according to claim 1 is characterized in that: The S4 specifically includes: S41, mapping each task and resource to a cell in a cellular automaton model, wherein each task cell represents an independent task, including resource requirements, execution priority, and coordination requirements; each resource cell represents the resource status in the park, including the availability and current occupancy of the resource, and the relationship between tasks and resources is represented by the edge connection between cells; S42. Define the local evolution rules of the cellular automaton, specifically: In the behavioral representation evolution network space, the behavioral characteristics of each task cell are adjusted according to its own resource requirements, execution status, and interaction with neighboring tasks and resources. When a task shares resources with neighboring tasks, the behavior of the task cell is dynamically updated. S43. Based on the local evolution rules, the task cells and the adjacent resource cells adjust their states through local interactions: when task A and task B share the same resources, the priority of one of the tasks is adjusted or the execution of one of the tasks is delayed to avoid resource conflicts; S44. The behavior characteristics of each task cell are self-optimized and adjusted in the local environment over time. The interaction between task cells and resource cells gradually affects the global scheduling effect through the local evolution mechanism. S45. Through the local evolution mechanism of the cellular automaton model, the entire behavior representation evolution network continuously performs adaptive adjustments, ultimately achieving optimization of the global scheduling effect.

5. The large model-based smart park emergency communication dispatching method according to claim 1 is characterized in that: The S5 specifically includes: S51. Based on real-time environmental data and combined with the context awareness mechanism, the task execution environment is dynamically modeled, where the impact of environmental changes on task execution is represented by a weighted comprehensive model: ； in, Indicates the task At the moment The environmental perception value, Indicates The weight of the impact of environmental factors on the task, represents the number of environmental factors, Indicates Environmental factors at the moment The real-time value of Indicates The expected value of an environmental factor, Indicates The standard deviation of each environmental factor; S52, according to the environmental perception value , for each task Define an environmental sensitivity factor to reflect the task The sensitivity under different environmental conditions, the sensitivity factor calculation formula is: ； in, Indicates the task The environmental sensitivity factor, Indicates the task Environmental response sensitivity, Indicates the task Environmental response thresholds; S53, according to the task Environmental sensitivity factor , dynamically adjust task priorities, execution paths, and resource allocation: ； in, Indicates the task At the moment The priority of Indicates the task The initial priority of represents the path adjustment coefficient, Indicates The weight of resource factors, Indicates the number of resource types. Indicates the task At the moment For The demand for a resource; S54. Based on the adjusted task priority, the task execution path is further optimized and adjusted in combination with the impact of environmental changes: ； in, Indicates the task At the moment The execution path of Represents the initial execution path of the task, represents the exponential function, represents the adjustment factor; S55. Generate a context-awareness-driven task scheduling solution based on the dynamically adjusted task priority, execution path, and resource allocation.

6. The large model-based smart park emergency communication dispatching method according to claim 1 is characterized in that: The S6 specifically includes: S61. Based on the task scheduling scheme driven by context awareness, the collaborative relationship between tasks is modeled through a game theory model. The game theory model considers resource competition and sharing between tasks. The goal is to achieve optimal resource allocation through the game between tasks. The objective function of the game model is: ； in, represents maximization, Indicates the task The utility function of Indicates the task From the resources The resource benefits obtained, Indicates the task At the moment resource The consumption cost, Indicates the number of resource types; S62. Combined with the collective behavior evolution mechanism, by evolving the strategies between tasks in the game theory model, the optimal resource allocation strategy for each task is calculated. The evolution of the optimal resource allocation strategy follows the Nash equilibrium condition: for all tasks When the utility function reaches its maximum value, the strategy will not be changed. The Nash equilibrium formula is: ； in, Indicates the task The optimal utility under Nash equilibrium is Indicates the task Optimal utility under Nash equilibrium; S63. Based on the optimal resource allocation strategy obtained by the game theory model, the global task scheduling is optimized by combining task priority and resource consumption; S64. Through the game and collaboration optimization between tasks, the global optimal scheduling strategy is finally generated.

Images