Village shelter space optimization method based on Monte Carlo method and agent model

Abstract

The present invention provides a method for optimizing village refuge spaces based on the Monte Carlo method and agent models, which relates to the field of space optimization and includes the following steps: collecting and organizing various types of spatial data; taking the attributes of agents as the input of the simulation model, randomly sampling the attribute ranges of agents using the Monte Carlo method, conducting simulation in the simulation model, setting constraint conditions, and obtaining the optimal model output; analyzing the agent attributes corresponding to the optimal model output to obtain the optimal refuge space optimization plan; setting up a visualization model, displaying the simulation process through the visualization model, and at the same time outputting the video of the simulation process leading to the optimal village refuge space optimization plan. By adopting the above method for optimizing village refuge spaces based on the Monte Carlo method and agent models, the present invention can significantly improve the optimization effect and utilization efficiency of village refuge spaces.

Description

Technical Field

[0001] The present invention relates to the field of spatial optimization, and in particular to a method for optimizing village evacuation spaces based on the Monte Carlo method and the agent model. Background Art

[0002] The Monte Carlo method is a simulation method based on random numbers, capable of handling complex stochastic processes and uncertainty problems. It has wide applications in fields such as finance, physics, engineering, and computer science. Its flexibility and robustness make it a powerful tool for solving complex problems. In the optimization of evacuation spaces, the Monte Carlo method can be used to simulate evacuation processes and evacuation effects under different scenarios, providing a scientific basis for optimization plans.

[0003] The agent model can simulate the behavioral characteristics of people during disasters, such as evacuation speed, path selection, etc. By constructing environmental agents and crowd agents, the interaction between people and the environment can be simulated, such as the impact of road capacity on evacuation speed. The simulation results based on the agent model can more accurately predict evacuation processes and evacuation effects, providing precise data support for optimization plans.

[0004] With the rapid development of computer technology and artificial intelligence technology, the applications of the Monte Carlo method and the agent model in disaster response and evacuation space optimization are becoming more and more extensive. The maturity of these technologies provides strong technical support for the implementation of plans.

[0005] In the field of optimizing village evacuation spaces, traditional planning methods often have difficulty fully considering various complex factors and uncertainties. The applications of the Monte Carlo method and the agent model can make up for this deficiency, improving the scientific nature and feasibility of optimization plans. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for optimizing village evacuation spaces based on the Monte Carlo method and the agent model. By means of data-driven, model simulation, multi-round iterative optimization, multiple evaluation indicators, resource optimization, auxiliary decision-making, and enhancing the community's disaster response capabilities, etc., the optimization effect and utilization efficiency of village evacuation spaces can be significantly improved.

[0007] To achieve the above purpose, the present invention provides a method for optimizing village evacuation spaces based on the Monte Carlo method and the agent model, including the following steps:

[0008] Collect and sort various types of spatial data, including road network layers, alternative evacuation shelter layers, and population distribution layers;

[0009] Establish crowd agents according to the population distribution layer, and create environmental agents for road grids and evacuation shelters according to the road network layer and the alternative evacuation shelter layer;

[0010] Set attributes for each agent, define the attribute range of each agent, use the attributes of the agents as the input of the simulation model, randomly sample the attribute range of the agents by the Monte Carlo method, conduct simulation in the simulation model, set constraints, and obtain the optimal model output; the agents include swarm agents and environmental agents;

[0011] Set an evaluation model to evaluate the simulation results, update the optimization plan for the village shelter space according to the evaluation results, and analyze the agent attributes corresponding to the optimal model output to obtain the optimal optimization plan for the shelter space;

[0012] Set a visualization model, display the simulation process through the visualization model, and at the same time output the video of the simulation process to the optimal optimization plan for the village shelter space.

[0013] Preferably, the attribute characteristics of the swarm agents include position, speed, age, gender, and function;

[0014] The behavior rules of the swarm agents include evacuation speed, path selection strategy, and preference for shelter selection;

[0015] The attribute characteristics of the environmental agents include road traffic capacity, shelter capacity, location, and entrance location;

[0016] Among them, the interaction methods between the swarm agents and the environmental agents include entrance flow restriction and evacuation path guidance.

[0017] Preferably, the model output includes the shelter time of the agents, the average speed of the agents, and the covered area.

[0018] Preferably, use the attributes of the agents as the input of the simulation model, randomly sample the attribute range of the agents by the Monte Carlo method, conduct simulation in the simulation model, set constraints, and obtain the optimal model output, including:

[0019] Build a Monte Carlo tree, initialize the root node, and initialize the positions of the swarm agents and environmental agents;

[0020] Set the simulation model, including a dynamics layer, a pathfinding layer, and a behavior layer, set each layer in the simulation model as a Monte Carlo tree node, and the information in each node includes state, quality value, access times, parent node, and child node;

[0021] The state includes the actions of the swarm agents and environmental agents in the dynamics layer, pathfinding layer, and behavior layer respectively;

[0022] Within the set total number of training times, train the Monte Carlo tree to obtain a trained Monte Carlo tree; according to the tree structure of the trained Monte Carlo tree, set a range for each model output, use the range of the model output as the training target, evaluate the model output that reaches the training target each time, and obtain the solution with the highest evaluation index as the optimal Monte Carlo method-based simulation model under the current training conditions.

[0023] Preferably, use the Monte Carlo method-based simulation model to simulate the evacuation behavior and shelter process of people during a disaster:

[0024] First, perform grid division, then identify the positions and areas of obstacles in each sub-grid to obtain the positions and areas of the blank spaces, and then splice the positions of adjacent blank spaces to obtain the first path;

[0025] Match each path in the first path with the basic function, and then use the integral form to solve the length of each path;

[0026] ;

[0027] Among them, s represents the length of the first path, represents the function for path fitting, represents the starting coordinate of the first path, represents the ending coordinate of the first path;

[0028] Set the shelter as the end point and different population gathering places as the starting points, and according to the obtained first path, obtain all path planning solutions from the starting point to the end point;

[0029] According to the length of each path obtained by solving, calculate the lengths of the paths in all path planning solutions, sort all the paths in the path planning solutions in ascending order according to the path length, and select the path with the smallest path length as the evacuation path for the current population gathering point.

[0030] Preferably, when people take shelter according to the evacuation path, count the density of people in each path, and use the fundamental diagram density correction algorithm to gradually correct the simulated test values towards the expected density:

[0031] ;

[0032] Among them, is the calculated density velocity, is the free velocity, is the current density, is the maximum density, that is, the density when the crowd is crowded until it is completely stationary; represents the current scenario.

[0033] Preferably, the correction plan is as follows:

[0034] Set different carrying levels according to the number of people that the crowd gathering point can carry daily: the first carrying crowd, the second carrying crowd, and the third carrying crowd;

[0035] Increase the number of evacuation routes according to the increase of different carrying levels:

[0036] At the level of the first carrying crowd, one evacuation route is set;

[0037] At the level of the second carrying crowd, three evacuation routes are set;

[0038] At the level of the third carrying crowd, six evacuation routes are set;

[0039] In addition, analyze different carrying levels:

[0040] When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 30% and less than 60% of the number of people in this carrying level, add one evacuation route;

[0041] When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 60% of the number of people in this carrying level, add two evacuation routes.

[0042] Preferably, set an evaluation model to evaluate the simulation results, and update the optimized plan for the village evacuation space according to the evaluation results, including

[0043] Select the time consumed in the evacuation passage, the number of casualties, the time to enter the evacuation site, and the proportion of agents in the evacuation site from the simulation results as evaluation indicators;

[0044] ;

[0045] Among them, is the weight of the evacuation time of the agent, is the weight of the average speed of the agent, is the weight of the covered area, is the weight of the proportion of agents in the evacuation site during the simulation, is the weight of the number of casualties during the simulation, ; represents the time consumed in the evacuation passage, represents the average speed of the agent, represents the environment of the model, represents the proportion of agents in the evacuation site during the simulation, represents the number of casualties during the simulation;

[0046] Score according to the weights of the evaluation indicators and the actual simulation of the evaluation indicators to obtain the score of the simulation result. When the score of the simulation result is equal to or higher than the set threshold, directly output the optimized plan for the village refuge space; when the score of the simulation result is lower than the set threshold, analyze each evaluation indicator and adjust the corresponding refuge space setting according to the score of each evaluation indicator.

[0047] The adjusted refuge space setting is used as the new optimized plan for the village refuge space, and the relevant data of the new optimized plan for the village refuge space are fed back into the agent model.

[0048] Therefore, the present invention adopts the above-mentioned method for optimizing the village refuge space based on the Monte Carlo method and the agent model, and the technical effects are as follows:

[0049] 1. Improve the scientificity of refuge space planning

[0050] Data-driven: By collecting and sorting out various types of spatial data, including road networks, alternative shelters, and population distributions, etc., a solid data foundation is provided for the optimization process. The accuracy and integrity of these data ensure the scientificity and feasibility of the optimization plan.

[0051] Model simulation: Using the Monte Carlo method and the agent model for simulation, it is possible to simulate the evacuation situation and refuge process under different scenarios, so as to more accurately predict and evaluate the effect of the refuge space.

[0052] 2. Enhance the flexibility of the optimization plan

[0053] Multiple rounds of iteration: By continuously iterating the optimization process, the optimized plan for the refuge space can be adjusted in a timely manner according to the evaluation results until the optimal solution is found. This flexibility enables the optimization plan to better adapt to the changes in the actual situation.

[0054] Multiple evaluation indicators: Set multiple evaluation indicators to comprehensively evaluate the effect of the optimization plan, including evacuation time, shelter utilization rate, crowd safety, etc. These indicators can be weighted according to actual needs to reflect the importance of different aspects.

[0055] 3. Improve the utilization efficiency of the refuge space

[0056] Reasonable layout: By optimizing the location and capacity of the shelters and the traffic capacity of the road network, the refuge space can be more effectively utilized during a disaster. This helps to reduce the evacuation time and improve the safety of the crowd.

[0057] Resource optimization: The optimization plan can ensure that limited resources (such as shelter capacity, rescue forces, etc.) are reasonably allocated and utilized, thereby improving the overall ability to respond to disasters.

[0058] 4. Auxiliary Decision-Making

[0059] Visual display: Presenting the optimization plan to decision-makers in a visual manner helps them more intuitively understand the optimization process and results. This contributes to enhancing the efficiency and accuracy of decision-making.

[0060] Decision support: Introducing the optimization process and results to relevant departments or the public through detailed reports or presentations provides strong support for decision-making.

[0061] 5. Enhancing the Community's Ability to Respond to Disasters

[0062] Improving residents' awareness: By optimizing the shelter space, the awareness and response ability of community residents towards disasters can be enhanced. During the optimization process, means such as publicity and education can be combined to raise residents' awareness of disaster prevention and reduction.

[0063] Promoting community participation: The implementation of the optimization plan requires the active participation and cooperation of community residents. By strengthening community participation, a good atmosphere of co-construction and sharing can be formed, and the overall ability of the community to respond to disasters can be improved. Description of the Drawings

[0064] Figure 1 It is a flowchart of the method for optimizing the village shelter space based on the Monte Carlo method and the agent model of the present invention. Detailed Implementation Manner

[0065] The technical solution of the present invention will be further described below with reference to the drawings and embodiments.

[0066] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention belongs.

[0067] Embodiment 1

[0068] As Figure 1 shown, the present invention provides a method for optimizing the village shelter space based on the Monte Carlo method and the agent model, including the following steps:

[0069] Collect and organize various types of spatial data, including road network layers, alternative shelter layers, and population distribution layers;

[0070] Establish crowd agents according to the population distribution layer, and create environmental agents for road grids and shelters according to the road network layer and alternative shelter layer;

[0071] The attribute characteristics of crowd agents include position, speed, age, gender, and function;

[0072] The behavior rules of crowd agents include evacuation speed, path selection strategy, and preference for shelters;

[0073] The attribute characteristics of environmental agents include road passing capacity, shelter capacity, location, and entrance location;

[0074] Among them, the interaction methods between crowd agents and environmental agents include entrance flow restriction and evacuation path guidance.

[0075] Set attributes for each agent, define the attribute range of each agent, use the attributes of the agents as the input of the simulation model, adopt the Monte Carlo method to randomly sample the attribute range of the agents, conduct simulation in the simulation model, set constraint conditions, and obtain the optimal model output; the agents include crowd agents and environmental agents; the model output includes the shelter time of the agents, the average speed of the agents, and the covered area. Use the attributes of the agents as the input of the simulation model, adopt the Monte Carlo method to randomly sample the attribute range of the agents, conduct simulation in the simulation model, set constraint conditions, and obtain the optimal model output, including:

[0076] Build a Monte Carlo tree, initialize the root node, and initialize the positions of crowd agents and environmental agents;

[0077] Set the simulation model, including the dynamics layer, pathfinding layer, and behavior layer, set each layer in the simulation model as a Monte Carlo tree node, and the information in each node includes state, quality value, access times, parent node, and child nodes;

[0078] The state includes the actions of crowd agents and environmental agents in the dynamics layer, pathfinding layer, and behavior layer respectively;

[0079] Within the set total number of training times, train the Monte Carlo tree to obtain a trained Monte Carlo tree; according to the tree structure of the trained Monte Carlo tree, set the range for each model output, use the range of the model output as the training target, evaluate the model output that reaches the training target each time, and obtain the plan with the highest evaluation index as the optimal Monte Carlo method-based simulation model under the current training situation.

[0080] Use the Monte Carlo method-based simulation model to simulate the evacuation behavior and shelter process of the crowd during a disaster:

[0081] First, conduct grid division, then identify the positions and areas of obstacles in each sub-grid to obtain the positions and areas of the blank spaces, and then splice the positions of adjacent blank spaces to obtain the first path;

[0082] Match each path in the first path with the basic function, and then solve the length of each path in the form of integration;

[0083] ;

[0084] Among them, s represents the length of the first path, represents the function for path fitting, represents the starting coordinate of the first path, represents the ending coordinate of the first path;

[0085] Set the shelter as the end point and different population gathering places as the starting points. According to the obtained first path, obtain all the path planning schemes from the starting point to the end point;

[0086] According to the length of each path solved, calculate the lengths of the paths in all the path planning schemes, sort all the paths in the path planning schemes in ascending order according to the path length, and select the path with the minimum path length as the evacuation path for the current population gathering point.

[0087] When the population evacuates according to the evacuation path, count the density of the population in each path, and use the fundamental diagram density correction algorithm to gradually correct the simulated test values towards the expected density:

[0088] ;

[0089] Among them, is the calculated density speed, is the free speed, is the current density (expected density), is the maximum density, that is, the density when the crowd is crowded until it is completely stationary; represents the current scenario. For example, the values in normal and leisure scenarios are , the value in the commuting scenario is , and the value in the emergency scenario is . The fundamental diagram density correction algorithm is an algorithm that improves the diagram density algorithm using a greedy strategy. The fundamental diagram density correction algorithm can better optimize the village environment during the simulation of the evacuation behavior and shelter process of the crowd in case of disasters, so as to obtain a more perfect village construction plan, reduce the time for evacuees to reach the shelter, and reduce casualties.

[0090] The correction plan is as follows:

[0091] Set different carrying levels according to the number of people that the population gathering point can bear daily: the first carrying population, the second carrying population, and the third carrying population;

[0092] Increase the number of evacuation paths according to the increase of different carrying levels:

[0093] At the level of the first carrying population, a refuge path is set;

[0094] At the level of the second carrying population, three refuge paths are set;

[0095] At the level of the third carrying population, six refuge paths are set;

[0096] In addition, analyze different carrying levels:

[0097] When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 30% and less than 60% of the number of people in this carrying level, one additional refuge path is added;

[0098] When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 60% of the number of people in this carrying level, two additional refuge paths are added.

[0099] Set up an evaluation model to evaluate the simulation results, update the village refuge space optimization plan according to the evaluation results, and analyze the corresponding agent attributes of the optimal model output to obtain the optimal refuge space optimization plan;

[0100] Set up an evaluation model to evaluate the simulation results, and update the village refuge space optimization plan according to the evaluation results, including

[0101] Select the time consumed in the refuge passage, the number of casualties, the time to enter the refuge place, and the proportion of agents in the refuge place from the simulation results as evaluation indicators;

[0102] ;

[0103] Among them, is the weight of the agent's refuge time, is the weight of the agent's average speed, is the weight of the covered area, is the weight of the proportion of agents in the refuge place during the simulation, is the weight of the number of casualties during the simulation, ; represents the time consumed in the refuge passage, represents the average speed of the agent, represents the environment of the model, represents the proportion of agents in the refuge place during the simulation, represents the number of casualties during the simulation;

[0104] Score according to the weight of the evaluation index and the actual simulation of the evaluation index to obtain the score of the simulation result. When the score of the simulation result is equal to or higher than the set threshold, directly output the optimized plan for the village refuge space; when the score of the simulation result is lower than the set threshold, analyze each evaluation index and adjust the corresponding refuge space setting according to the score of each evaluation index.

[0105] The adjusted refuge space setting is used as the new optimized plan for the village refuge space, and the relevant data of the new optimized plan for the village refuge space are fed back into the agent model.

[0106] Set up a visualization model, and the simulation process is displayed through the visualization model. At the same time, output the video of the simulation process to the optimized plan for the optimal village refuge space.

[0107] Transmit the output video to the corresponding village. The corresponding village plays the video repeatedly in a place where all villagers can see it, and organizes the villagers to conduct repeated drills according to the evacuation route in the video to improve the evacuation time and proficiency during evacuation and reduce the casualty rate during actual evacuation.

[0108] This application also provides a storage medium on which a computer program is stored. When the computer program is executed, the steps provided in the above embodiments can be implemented. The storage medium may include: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical discs that can store program codes.

[0109] This application also provides an electronic device, which may include a memory and a processor. When the processor calls the computer program stored in the memory, the steps provided in the above embodiments can be implemented. Of course, the electronic device may also include various network interfaces, power supplies and other components.

[0110] Therefore, the present invention adopts the above-mentioned method for optimizing the village refuge space based on the Monte Carlo method and the agent model. Through means such as data-driven, model simulation, multi-round iterative optimization, multiple evaluation indexes, resource optimization, auxiliary decision-making, and enhancing the community's disaster response ability, the optimization effect and utilization efficiency of the village refuge space can be significantly improved.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing the village refuge space based on the Monte Carlo method and the agent model, characterized in that, It includes the following steps: Collect and organize various types of spatial data, including road network layers, alternative shelter layers, and population distribution layers; Establish crowd agents based on the population distribution layer, and create environmental agents for road grids and shelters according to the road network layer and alternative shelter layer; The agents include crowd agents and environmental agents; Set attributes for each agent, define the attribute range of each agent, use the attributes of the agents as the input of the simulation model, randomly sample the attribute range of the agents using the Monte Carlo method, and perform simulation in the simulation model, set constraints, and obtain the optimal model output; Use the simulation model based on the Monte Carlo method to simulate the evacuation behavior and shelter process of the crowd during a disaster: First, perform grid division, then identify the positions and areas of obstacles in each sub-grid to obtain the positions and areas of the blank spaces, and then splice the positions of adjacent blank spaces to obtain the first path; Match each path in the first path with the basic function, and then solve it in the form of integration to obtain the length of each path; ； Among them, represents the length of the first path, represents the function for path fitting, represents the starting coordinate of the first path, represents the ending coordinate of the first path; Set the shelter as the end point, and different crowd gathering places as the starting points, and obtain all path planning schemes from the starting point to the end point according to the obtained first path; Calculate the lengths of the paths in all path planning schemes according to the lengths of the solved paths, sort all the paths in the path planning schemes in ascending order of path length, and select the path with the smallest path length as the shelter path for the current crowd gathering point; When the crowd takes shelter according to the shelter path, count the density of the crowd in each path, and use the basic map density correction algorithm to gradually correct the simulated test values towards the expected density: ； Among them, is the calculated density velocity, is the free velocity, is the current density, is the maximum density, that is, the density when the crowd is so crowded that it is completely stationary; represents the value in the current scenario. The values in the normal and leisure scenarios are , and the value in the commuting scenario is , and the value in the emergency scenario is ; Set up an evaluation model to evaluate the simulation results, update the village shelter space optimization plan according to the evaluation results, and analyze the agent attributes corresponding to the optimal model output to obtain the optimal shelter space optimization plan; Set up a visualization model, display the simulation process through the visualization model, and at the same time output the video of the simulation process to the optimal village shelter space optimization plan.

2. The method for optimizing the village shelter space based on the Monte Carlo method and the agent model according to claim 1, characterized in that The attribute characteristics of the crowd agent include position, speed, age, gender, and function; The behavior rules of the crowd agent include evacuation speed, path selection strategy, and preference for shelters; The attribute characteristics of the environmental agent are road traffic capacity, shelter capacity, position, and entrance position; Among them, the interaction methods between the crowd agent and the environmental agent include entrance flow restriction and evacuation path guidance.

3. The optimization method of village shelter space based on Monte Carlo method and agent model according to claim 2, characterized in that The model output includes the shelter time of the agent, the average speed of the agent, and the covered area.

4. The method for optimizing the village shelter space based on the Monte Carlo method and the agent model according to claim 2, wherein Use the attributes of the agents as the input of the simulation model, randomly sample the attribute range of the agents using the Monte Carlo method, perform simulation in the simulation model, set constraints, and obtain the optimal model output, including: Build a Monte Carlo tree, initialize the root node, and initialize the positions of the crowd agents and environmental agents; Set up a simulation model, including a dynamics layer, a pathfinding layer, and a behavior layer. Set each layer in the simulation model as a Monte Carlo tree node. The information within each node includes state, quality value, visit count, parent node, and child nodes; The state includes the actions of swarm agents and environmental agents in the dynamics layer, pathfinding layer, and behavior layer respectively; Within the set total number of training times, train the Monte Carlo tree to obtain a trained Monte Carlo tree; according to the tree structure of the trained Monte Carlo tree, set a range for each model output. Use the range of the model output as the training target, evaluate the model output that reaches the training target each time, and obtain the plan with the highest evaluation index as the optimal simulation model based on the Monte Carlo method under the current training conditions.

5. The optimized method for village shelter space based on the Monte Carlo method and the agent model according to claim 1, characterized in that The correction plan is as follows: Set different carrying levels according to the number of people the crowd gathering point can carry daily: the first carrying crowd, the second carrying crowd, and the third carrying crowd; Increase the number of evacuation paths according to the increase in different carrying levels: At the level of the first carrying crowd, one evacuation path is set; At the level of the second carrying crowd, three evacuation paths are set; At the level of the third carrying crowd, six evacuation paths are set; In addition, analyze different carrying levels: When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 30% and less than 60% of the number of people in this carrying level, add one evacuation path; When the sum of the number of people under 5 years old and over 60 years old in the carrying level accounts for more than 60% of the number of people in this carrying level, add two evacuation paths.

6. The method for optimizing the village shelter space based on the Monte Carlo method and the agent model according to claim 1, characterized in that, Set up an evaluation model to evaluate the simulation results, and update the village evacuation space optimization plan according to the evaluation results, including Select the time consumed in the evacuation passage, the number of casualties, the time to enter the evacuation site, and the proportion of agents in the evacuation site from the simulation results as evaluation indicators; ； Among them, is the weight of the agent's shelter time, is the weight of the agent's average speed, is the weight of the covered area, is the weight of the proportion of agents in the shelter during the simulation, is the weight of the number of casualties during the simulation, represents the time consumed by the evacuation route, represents the average speed of the agent, represents the environment of the model, represents the proportion of agents in the shelter during the simulation, represents the number of casualties during the simulation; Score according to the weights of the evaluation indicators and the actual simulation situation of the evaluation indicators to obtain the score of the simulation result. When the score of the simulation result is equal to or higher than the set threshold, directly output the village evacuation space optimization plan; when the score of the simulation result is lower than the set threshold, analyze each evaluation indicator and adjust the corresponding evacuation space setting according to the score of each evaluation indicator; The adjusted evacuation space setting is used as the new village evacuation space optimization plan, and the relevant data of the new village evacuation space optimization plan are fed back into the agent model.

Images