A Simulation Method for Evacuation of Rural Refugees Based on a Simulation Model

By constructing a simulation method for rural asylum evacuation personnel evacuation simulation method, the accuracy of emergency evacuation strategies and resource allocation in rural areas are solved, and efficient and flexible evacuation optimization and data support are achieved.

CN120163071BActive Publication Date: 2025-07-22SOUTHWEAT UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510646746.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-20
Publication Date
2025-07-22
Estimated Expiration
2045-05-20

AI Technical Summary

Technical Problem

Traditional emergency evacuation planning and management methods are difficult to meet the complex evacuation needs in rural areas, especially in the case of diverse geographical environment, weak infrastructure and wide distribution of residents, and existing technologies are difficult to provide accurate evacuation strategies and resource allocation support.

Method used

The construction of a simulation method for evacuation of rural refugees based on simulation models includes collecting design indicators and influencing factors, building an agent model, designing an agent interaction mechanism, simulating the evacuation process and evaluating the results, and optimizing the evacuation strategy.

Benefits of technology

It realizes accurate simulation of emergency evacuation in rural areas, optimizes evacuation paths and resource allocation, improves the flexibility and adaptability of emergency response, and provides data-driven decision-making support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120163071B_ABST
    Figure CN120163071B_ABST
Patent Text Reader

Abstract

The present invention provides a method for simulating the evacuation of rural refuge seekers based on a simulation model, belonging to the technical field of evacuation methods, including: S1, collecting the design indicators and influencing factors of rural refuge spaces; S2, constructing a database for the planning and design of rural refuge spaces; S3, constructing four types of agents, namely residents, refuge sites, refuge roads, and environmental dynamic changes; S4, constructing an environmental model according to the actual situation of the target rural area and embedding the four types of agents into it to form a complete rural refuge emergency agent simulation model; S5, designing the interaction mechanism between the four types of agents; S6, setting the simulation scenario according to the actual situation of the target rural area; S7, starting the rural refuge emergency agent simulation model to simulate the evacuation process of rural refuge seekers in case of emergency; S8, evaluating the simulation results. The present invention can optimize the emergency evacuation strategy in rural areas and improve the efficiency and safety in dealing with emergencies such as natural disasters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of evacuation methods, and in particular, to a method for simulating the evacuation of rural refuge seekers based on a simulation model. Background Art

[0002] In the context of the increasing safety needs in rural areas where natural disasters occur frequently, traditional emergency evacuation planning and management methods have been difficult to meet the increasingly complex evacuation needs. Due to the diverse geographical environment, relatively weak infrastructure, and wide distribution of residents in rural areas, the emergency evacuation work in rural areas faces many challenges.

[0003] In order to better address these challenges and improve the emergency response ability and evacuation efficiency in rural areas, using modern simulation technology to simulate the emergency evacuation process has become an effective solution. By constructing a virtual environment, simulation technology simulates the flow and distribution of people and materials in emergency situations, helping decision-makers evaluate the effects of different evacuation strategies and emergency plans before real events occur. This method can not only reduce the risks and uncertainties in actual evacuations, but also provide a scientific basis for optimizing evacuation routes, improving the utilization efficiency of shelters, and strengthening the allocation of emergency resources.

[0004] Therefore, based on the above needs, it is of great significance to research and develop a simulation model that can accurately simulate the evacuation process of rural refuge seekers. Such a model will combine the actual situation in rural areas, consider various factors such as topography, population distribution, and traffic conditions, provide comprehensive and reliable simulation results and data analysis for decision-makers, and provide strong support for emergency management and public safety work in rural areas. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for simulating the evacuation of rural refuge seekers based on a simulation model, which can optimize the emergency evacuation strategy in rural areas and improve the efficiency and safety in dealing with emergencies such as natural disasters.

[0006] To achieve the above purpose, the present invention provides a method for simulating the evacuation of rural refuge seekers based on a simulation model, including the following steps:

[0007] Step S1, collect the design indicators and influencing factors of rural refuge spaces;

[0008] Step S2, construct a database for the planning and design of rural refuge spaces according to the design indicators and influencing factors;

[0009] Step S3, construct resident agents, shelter site agents, shelter road agents, and environmental dynamic change agents;

[0010] Step S4: Construct an environmental model according to the actual situation of the target rural area, and embed the resident agent, shelter site agent, evacuation route agent, and environmental dynamic change agent into it to form a complete rural evacuation emergency agent simulation model;

[0011] Step S5: Design the interaction mechanism among the four types of agents;

[0012] Step S6: Set the simulation scenario according to the actual situation of the target rural area;

[0013] Step S7: Start the rural evacuation emergency agent simulation model, simulate the evacuation process of rural evacuees in case of emergency, and record the behavior performance of each agent;

[0014] Step S8: Evaluate the simulation results, and optimize the agent behavior rules and environmental model according to the evaluation results.

[0015] Preferably, in step S1, the design indicators include shelters and evacuation routes; the influencing factors include resident behavior factors, spatial environmental characteristics, infrastructure characteristics, and natural environmental characteristics.

[0016] Preferably, in step S2, construct a rural evacuation space planning and design database according to the design indicators and influencing factors. The specific operation is as follows: conduct a Pearson correlation analysis on the design indicators and influencing factors, identify the design indicators and influencing factors that have a significant impact on rural evacuation space planning and design, and construct a rural evacuation space planning and design database.

[0017] Preferably, in step S3, the basic attributes of the resident agent include age, gender, mobility, and mental state; the behavior rules of the resident agent include environmental perception strategy, evacuation strategy, path selection strategy, exit selection strategy, and action execution strategy;

[0018] The environmental perception strategy is as follows:

[0019] ;

[0020] where represents the perception distance; represents the position of the resident agent; represents the position of the shelter site or obstacle;

[0021] The evacuation strategy is as follows:

[0022] ;

[0023] where represents the risk assessment; represents the risk threshold; represents the evacuation decision;

[0024] ;

[0025] Among them, represents the distance from the resident agent to the nearest disaster point; represents the score of the current traffic condition; represents the impact score of environmental obstacles; , , all represent weights;

[0026] The path selection strategy is as follows:

[0027] ;

[0028] Among them, represents the optimal path; represents the weight of the th segment in the path; represents the length of the path segment; represents the total number of all segments in the path;

[0029] The exit selection strategy is as follows:

[0030] ;

[0031] Among them, represents the best exit; represents the maximum number of people that the exit of the shelter site can accommodate; represents the proportion of the number of people currently using this exit;

[0032] The action execution strategy is as follows:

[0033] ;

[0034] Among them, represents the action speed; represents the remaining distance to the shelter site; represents the remaining time;

[0035] The basic attributes of the shelter road agent include road width, road length, and road traffic capacity; the behavioral rules of the shelter road agent include simulating traffic conditions, calculating travel time, and obstacle avoidance and path optimization;

[0036] Simulating traffic conditions is as follows:

[0037] ;

[0038] Among them, represents the traffic flow; represents the th vehicle's speed; Indicates the proportion of the road occupied by the vehicle;

[0039] The passing time is calculated as follows:

[0040] ;

[0041] Wherein, Indicates the passing time; Indicates the road length; Indicates the average vehicle speed;

[0042] Obstacle avoidance and path optimization are as follows:

[0043] ;

[0044] Wherein, Indicates the optimal obstacle avoidance path; Indicates the influence degree of the obstacle; Indicates the weight coefficient;

[0045] The basic attributes of the shelter site agent include location, capacity, and accessible roads; the behavior rules of the shelter site agent include the strategy for receiving shelter seekers, the assessment of accommodation capacity, and the material distribution strategy;

[0046] The strategy for receiving shelter seekers is as follows:

[0047] ;

[0048] Wherein, Indicates the capacity allocation; Indicates the number of resident agents attempting to enter the shelter site; Indicates the number of people that the shelter site can accommodate;

[0049] The assessment of accommodation capacity is as follows:

[0050] ;

[0051] Wherein, Indicates the accommodation capacity; Indicates the number of people already in the shelter site; Indicates the maximum accommodation capacity of the shelter site;

[0052] The material distribution strategy is as follows:

[0053] ;

[0054] Wherein, Indicates the material allocation; Indicates the total amount of materials in the shelter site; Indicates the number of resident agents in the shelter site;

[0055] The basic attributes of the environmental dynamic change agent include perception ability, data processing ability, prediction ability, and communication ability; the behavior rules of the environmental dynamic change agent include real-time monitoring of the external environment, updating environmental data, analyzing environmental data, and sharing environmental data.

[0056] Preferably, a learning mechanism is also introduced into the resident agent, specifically:

[0057] Define the state space : Define various situations encountered by the resident agent as states, and various situations include the distance to the evacuation site, road congestion, and environmental obstacles;

[0058] Design the action space : Design possible actions for the resident agent, including choosing different evacuation routes and adjusting the walking speed;

[0059] Set the reward function : Give corresponding rewards or punishments according to the behavior results of the resident agent, and the behavior results include the time to successfully reach the evacuation site and the number of collisions avoided;

[0060] ;

[0061] Among them, represents the reward for reaching the evacuation site, represents the reward for avoiding collisions;

[0062] Training process: Let the resident agent continuously try different actions in the simulation environment and adjust the strategy according to the reward function until the optimal evacuation path and behavior pattern are found;

[0063] ;

[0064] Among them, represents the value of taking action in state , represents the learning rate, represents the discount factor, represents the new state after taking action , represents the action taken in the new state , represents the updated value.

[0065] Preferably, in step S4, an environmental model is constructed according to the actual situation of the target rural area, and the resident agent, the refuge site agent, the refuge road agent, and the environmental dynamic change agent are embedded therein to form a complete rural refuge emergency agent simulation model. The specific operations are as follows:

[0066] Integrate the influencing factors collected in step S1 and perform processing operations, including data cleaning, data transformation, and data verification;

[0067] Use 3D modeling software to construct the geographical environment model of the target rural area based on the processed data, including terrain and landform, building structure, and road network;

[0068] Based on the geographical environment model, layout the refuge sites and refuge roads;

[0069] Design an interaction interface for the interaction between each agent and the environmental model. The interaction interface includes a data interface, a control interface, and a visualization interface. The data interface defines the format and protocol of data transmission between each agent and the environmental model. The control interface provides an interface for control commands for each agent. The visualization interface is used to display the real-time status of the environmental model and each agent; after the environmental model is constructed, perform model verification and adjustment, including logical verification, parameter adjustment, and test run.

[0070] Preferably, in step S5, the interaction mechanisms between the four types of agents include: the interaction mechanism between resident agents, the interaction mechanism between resident agents and refuge site agents, the interaction mechanism between resident agents and refuge road agents, the interaction mechanism between resident agents and environmental dynamic change agents, the interaction mechanism between refuge site agents and refuge road agents, the interaction mechanism between refuge site agents and environmental dynamic change agents, and the interaction mechanism between refuge road agents and environmental dynamic change agents.

[0071] Preferably, in step S6, set the simulation scenario according to the actual situation of the target rural area. The specific operations are as follows:

[0072] First, set the simulation scenario parameters, including rural area characteristics, refuge site layout, resident distribution, refuge road network, emergency event type, and occurrence time;

[0073] Then, import geographical data into the simulation system, and set the attributes, location layout, and interaction mechanism of each agent;

[0074] Finally, set the simulation conditions, including the simulation time length, data recording frequency, and output format.

[0075] Preferably, in step S7, the rural evacuation emergency agent simulation model is started to simulate the evacuation process of rural evacuees in an emergency, and the performance behaviors of the agents are recorded, including the following steps:

[0076] Load the rural evacuation emergency agent simulation model constructed in step S4 and initialize all agents;

[0077] Start the engine of the rural evacuation emergency agent simulation model, trigger the emergency time, and make the resident agents perceive the emergency according to the type and occurrence time of the emergency event set in step S6;

[0078] Conduct simulation runs, including resident agent behavior simulation, shelter site agent behavior simulation, evacuation route agent behavior simulation, and environmental dynamic change agent behavior simulation;

[0079] Data recording and monitoring, including recording the evacuation time, path selection, and movement speed of resident agents; recording the traffic conditions and passing time of evacuation routes; monitoring the accommodation situation, utilization rate, and possible congestion points of shelter sites;

[0080] When all resident agents reach the shelter site or the preset simulation duration is reached, the simulation ends.

[0081] Preferably, in step S7, during the simulation process, an abnormal situation detection mechanism is also set up, and the detection mechanism is as follows:

[0082] During the simulation process, the system monitors the states of each agent, environmental changes, and traffic flow data in real time;

[0083] Set thresholds according to historical data to judge the occurrence of abnormal situations. When the analyzed real-time data exceeds the threshold, trigger the abnormal situation detection mechanism;

[0084] Calculate the Z-score of each index to judge the abnormal threshold:

[0085] ;

[0086] Wherein, represents the observed value; represents the historical mean; represents the historical standard deviation.

[0087] Therefore, the present invention adopts the above-mentioned rural evacuee evacuation simulation method based on a simulation model, and the beneficial technical effects are as follows:

[0088] (1) Precise simulation of complex environments: By constructing detailed environmental models, including terrain, building structures, road networks, and dynamically changing environmental factors (such as weather, natural disaster progress, etc.), the simulation model can accurately simulate the complex environments in rural areas. This high-precision simulation enables decision-makers to more accurately evaluate the feasibility of evacuation strategies in different environments, providing a basis for formulating adaptable emergency plans.

[0089] (2) Optimization of agent behavior simulation: The introduction of resident agents, shelter site agents, evacuation route agents, and environmental dynamic change agents enables the simulation model to simulate the behavioral responses of various agents in emergency situations. In particular, the learning mechanism in resident agents can find the optimal evacuation routes and behavioral patterns through continuous trial and error and optimization, improving evacuation efficiency and reducing risks. This agent behavior simulation technology helps to identify and optimize key issues at the individual behavior level.

[0090] (3) Data-driven decision support: A large amount of data collected during the simulation process, such as evacuation time, route selection, traffic conditions, shelter site utilization rate, etc., provides rich data support for decision-makers. Through data analysis, bottlenecks and key nodes in the evacuation process can be identified, and targeted improvement measures can be formulated. In addition, the simulation model can also conduct sensitivity analysis and scenario comparison to help decision-makers evaluate the effects of different strategies and select the optimal solution.

[0091] (4) Real-time feedback and dynamic adjustment: During the simulation process, the set abnormal situation detection mechanism can monitor the states of various agents and environmental changes in real time. Once an abnormal situation is detected, the feedback mechanism is immediately triggered. This real-time feedback ability enables decision-makers to adjust strategies in a timely manner during the simulation process, optimize resource allocation, and improve the flexibility and adaptability of emergency response. Description of the Drawings

[0092] Figure 1 is a flowchart of a method for simulating the evacuation of rural refuge seekers based on a simulation model according to the present invention;

[0093] Figure 2 is a schematic diagram of the interaction mechanism of four types of agents. Specific Embodiments

[0094] The technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0095] 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.

[0096] Example 1

[0097] As Figure 1As shown in the figure, it is a flowchart of a method for simulating the evacuation of rural refuge seekers based on a simulation model according to the present invention, including the following steps:

[0098] Step S1: Collect the design indicators and influencing factors of rural refuge spaces.

[0099] The design indicators include refuge sites and refuge roads; the influencing factors include residents' behavioral factors, spatial environmental characteristics, infrastructure characteristics, and natural environmental characteristics.

[0100] More specific influencing factors include the location of the refuge site, the capacity of the refuge site, the topology of the road network, the width of the road, the accessibility of the road, the individual characteristics of residents, the refuge response, the evacuation action mode, the main disaster sources, the topography and landforms, the building distribution, the weather conditions, etc.

[0101] Step S2: Construct a database for the planning and design of rural refuge spaces based on the design indicators and influencing factors.

[0102] Specifically: Conduct statistical analysis and correlation analysis on the design indicators and influencing factors, and construct a database for the planning and design of rural refuge spaces.

[0103] Step S3: Construct resident agents, refuge site agents, refuge road agents, and environmental dynamic change agents.

[0104] The basic attributes of resident agents include age, gender, mobility, and mental state; the behavioral rules of resident agents include environmental perception strategies, evacuation strategies, path selection strategies, exit selection strategies, and action execution strategies.

[0105] Environmental perception strategy: Define how resident agents perceive the location and status of emergency events, refuge sites, and refuge roads:

[0106] ;

[0107] Among them, represents the perception distance; represents the position of the resident agent; represents the position of the refuge site or obstacle.

[0108] Evacuation strategy: Determine whether to evacuate and the timing of evacuation according to the severity of the emergency event and the individual's position:

[0109] ;

[0110] Among them, represents the risk assessment; represents the risk threshold; represents the evacuation decision.

[0111] Path selection strategy: Considering factors such as the shortest path, obstacle avoidance, and congestion avoidance, select the best evacuation path:

[0112] ;

[0113] Among them, represents the optimal path; represents the weight of the th segment in the path; represents the length of the path segment; represents the total number of all segments in the path.

[0114] Exit selection strategy: Considering the distance, capacity, and congestion of the exit comprehensively, select the most suitable exit for the shelter site:

[0115] ;

[0116] Among them, represents the best exit; represents the maximum number of people that the exit of the shelter site can accommodate; represents the proportion of the number of people currently using this exit.

[0117] Action execution strategy: According to the decision result, execute the movement operation and go to the shelter site:

[0118] ;

[0119] Among them, represents the action speed; represents the remaining distance to the shelter site; represents the remaining time.

[0120] The basic attributes of the shelter road agent include road width, road length, and road passing capacity; the behavior rules of the shelter road agent include simulating traffic conditions, calculating passing time, and obstacle avoidance and path optimization.

[0121] Simulating traffic conditions: According to the movement of the resident agent and the road passing capacity, simulate the traffic congestion situation:

[0122] ;

[0123] Among them, represents the traffic flow; represents the th vehicle's speed; represents the proportion of the road occupied by this vehicle.

[0124] Calculating passing time: Provide the estimated passing time from the starting point to the ending point for the resident agent:

[0125] ;

[0126] Among them, represents the passing time; represents the road length; represents the average vehicle speed.

[0127] Obstacle avoidance and path optimization: Considering the obstacles and congestion points on the road, provide the best path suggestions for the resident agents:

[0128] ;

[0129] Among them, represents the optimal obstacle avoidance path; represents the influence degree of the obstacle; represents the weight coefficient.

[0130] The basic attributes of the shelter site agent include location, capacity, and accessible roads; the behavioral rules of the shelter site agent include the strategy of receiving shelter seekers, the evaluation of accommodation capacity, and the material distribution strategy.

[0131] Strategy of receiving shelter seekers: According to the capacity limit, receive and manage the resident agents entering the shelter site:

[0132] ;

[0133] Among them, represents the capacity allocation; represents the number of resident agents attempting to enter the shelter site; represents the number of people that the shelter site can accommodate.

[0134] Evaluation of accommodation capacity: Real-time evaluate the remaining capacity of the shelter site and provide information for the evacuation decision of the resident agents:

[0135] ;

[0136] Among them, represents the accommodation capacity; represents the number of people already in the shelter site; represents the maximum accommodation capacity of the shelter site.

[0137] The material distribution strategy is as follows:

[0138] ;

[0139] Among them, represents the material distribution; represents the total amount of materials in the shelter site; represents the number of resident agents in the shelter site.

[0140] The basic attributes of the environmental dynamic change agent include perception ability, data processing ability, prediction ability, and communication ability; the behavior rules of the environmental dynamic change agent include real-time monitoring of the external environment, updating environmental data, analyzing environmental data, and sharing environmental data.

[0141] Real-time monitoring of the external environment: Continuously monitor changes in the external environment, including the occurrence and development of disasters, as well as factors such as traffic and weather that may affect evacuation.

[0142] Updating environmental data: Regularly or real-time update environmental data to ensure the accuracy and timeliness of information.

[0143] Analyzing environmental data: Based on the collected data and analysis results, predict the future change trend of the environment and send early warning information to other agents in a timely manner.

[0144] Sharing environmental data: Establish a communication connection with other agents to share environmental information and evacuation status, and promote collaborative decision-making and actions.

[0145] A learning mechanism is also introduced into the resident agent, specifically as follows:

[0146] Defining the state space : Define various situations encountered by the resident agent as states, and various situations include the distance to the shelter, road congestion, and environmental obstacles;

[0147] Designing the action space : Design possible actions for the resident agent, including choosing different evacuation routes and adjusting walking speeds;

[0148] Setting the reward function : Give corresponding rewards or punishments according to the behavior results of the resident agent, and the behavior results include the time to successfully reach the shelter and the number of collisions avoided;

[0149] ;

[0150] Among them, represents the reward for reaching the shelter, represents the reward for avoiding collisions;

[0151] Training process: Let the resident agent continuously try different actions in the simulated environment and adjust the strategy according to the reward function until the optimal evacuation path and behavior pattern are found;

[0152] ;

[0153] Among them, represents taking action in state 's value, represents the learning rate, represents the discount factor, represents the new state after taking an action, represents the action taken in the new state and represents the updated value.

[0154] Step S4: Construct an environmental model according to the actual situation of the target rural area, and embed the resident agent, shelter site agent, evacuation route agent, and environmental dynamic change agent into it to form a complete rural evacuation emergency agent simulation model.

[0155] Specifically: Integrate the influencing factors collected in Step S1 and perform processing operations, including data cleaning, data transformation, and data verification;

[0156] Use 3D modeling software to construct the geographical environment model of the target rural area based on the processed data, including terrain and landform, building structure, and road network;

[0157] Terrain and landform: According to elevation data or satellite imagery, simulate the terrain undulations, rivers, lakes, and other natural features of the rural area.

[0158] Building structure: Based on the CAD drawings of buildings or on-site measurement data, construct 3D models of buildings, including buildings such as houses, warehouses, schools, etc. that may be used as shelters.

[0159] Road network: According to road planning data or on-site surveys, draw the road network of the rural area, including main roads, secondary roads, country roads, etc., and mark the attributes such as width, length, and traffic capacity of the roads.

[0160] On the basis of the geographical environment model, layout the shelter sites and evacuation routes;

[0161] Shelter site location: Precisely locate the shelter sites (such as school playgrounds, open spaces, etc.) in the model and mark their attributes such as capacity and facilities.

[0162] Road connection: Ensure that all shelter sites are connected to the road network, and the road width and traffic capacity can meet the evacuation requirements.

[0163] Obstacle setting: Set obstacles that may affect evacuation in the model, such as trees, fences, bridges, etc., and mark their locations and attributes.

[0164] Design an interaction interface for the interaction between each intelligent agent and the environment model. The interaction interface includes a data interface, a control interface, and a visualization interface. The data interface defines the format and protocol for data transmission between each intelligent agent and the environment model. The control interface provides an interface for control commands for each intelligent agent. The visualization interface is used to display the real-time status of the environment model and each intelligent agent;

[0165] The data interface is responsible for defining the format and protocol for data transmission between each intelligent agent and the environment model. Specifically, the data interface includes the following types of data transmission:

[0166] Status data:

[0167] The position, speed, health status, etc. of the resident intelligent agent.

[0168] The capacity, current occupancy, etc. of the shelter site intelligent agent.

[0169] The traffic flow, congestion situation, etc. of the evacuation road intelligent agent.

[0170] The weather conditions, disaster progress, etc. of the environmental dynamic change intelligent agent.

[0171] Perception data:

[0172] The perception information of the resident intelligent agent about the surrounding environment, such as the distance to the nearest shelter site, the position of surrounding obstacles, etc.

[0173] The perception information of the shelter site intelligent agent about the inflow and outflow of evacuees.

[0174] Request / response data:

[0175] The evacuation request sent by the resident intelligent agent to the shelter site intelligent agent.

[0176] The response of the shelter site intelligent agent to the resident intelligent agent, such as accepting or rejecting the evacuation request.

[0177] Control Interface

[0178] The control interface provides an interface for control commands for each intelligent agent. The control commands include:

[0179] Movement command:

[0180] The command received by the resident intelligent agent to move in a specific direction, such as "move 100 meters north".

[0181] Evacuation command:

[0182] The command received by the resident intelligent agent to find a shelter site, such as "find the nearest shelter site and go there".

[0183] Resource allocation command:

[0184] The allocation commands of the shelter site agent for resources (such as food, water).

[0185] Traffic control commands:

[0186] The control commands of the shelter road agent for traffic flow, such as "road closure" or "changing traffic lights".

[0187] Visualization Interface

[0188] The visualization interface is used to display the environmental model and the real-time status of each agent. The specific information to be displayed includes:

[0189] Environmental status view:

[0190] Show the map of the entire environmental model, including terrain, buildings, road network, etc.

[0191] Agent status view:

[0192] Real-time display of the location, moving direction and speed of the resident agent.

[0193] Show the current capacity of the shelter site agent and the number of shelter seekers.

[0194] Event view:

[0195] Show the occurrence location and influence range of emergency events, such as fires, floods, etc.

[0196] Traffic view:

[0197] Show the traffic flow and congestion on the shelter roads.

[0198] After the environmental model is constructed, conduct model verification and adjustment, including logical verification, parameter adjustment and test run.

[0199] Step S5: Design the interaction mechanism between the four types of agents.

[0200] Such as Figure 2 As shown, the interaction mechanisms between the four types of agents include: the interaction mechanism between resident agents, the interaction mechanism between resident agents and shelter site agents, the interaction mechanism between resident agents and shelter road agents, the interaction mechanism between resident agents and environmental dynamic change agents, the interaction mechanism between shelter site agents and shelter road agents, the interaction mechanism between shelter site agents and environmental dynamic change agents, and the interaction mechanism between shelter road agents and environmental dynamic change agents.

[0201] Resident agent and resident agent:

[0202] Information exchange: Resident agents can exchange information such as evacuation directions, shelter conditions, and personal status through wireless communication or visual perception. This helps to form group behaviors such as following leaders and avoiding congestion.

[0203] Collision avoidance mechanism: During movement, resident agents need to avoid collisions and congestion. This can be achieved by setting a safe distance, predicting path conflicts, and adjusting the movement direction.

[0204] Resident agents and shelter agents:

[0205] Matching and allocation: Resident agents select appropriate shelters for evacuation based on the information provided by shelter agents (such as capacity, location, facilities, etc.). Shelter agents then allocate and schedule resources according to the received requests.

[0206] Status feedback: Shelter agents provide feedback on the current status (such as the number of people already accommodated, remaining capacity, and facility usage) to resident agents, helping resident agents make more reasonable decisions.

[0207] Resident agents and evacuation route agents:

[0208] Path planning: Resident agents plan the optimal evacuation route based on the road network information and real-time traffic conditions provided by evacuation route agents. Evacuation route agents are responsible for updating the road status (such as congestion level and closure situation) to ensure the effectiveness of path planning.

[0209] Traffic control: During emergency evacuation, evacuation route agents may need to exert a certain degree of control over traffic, such as setting temporary traffic signals and adjusting lane directions, to optimize the evacuation efficiency. Resident agents need to follow these traffic control instructions to move.

[0210] Resident agents and environmental dynamic change agents:

[0211] Environmental perception: Resident agents obtain real-time environmental change information (such as fire spread, weather changes, and seismic wave propagation) through sensors or simulated environmental perception functions. This information affects the decisions and actions of resident agents.

[0212] Adaptation and response: Based on the environmental change information, resident agents need to adjust their behavioral strategies, such as changing the evacuation direction, accelerating the movement speed, and finding temporary shelters. Environmental dynamic change agents are responsible for simulating and providing this change information.

[0213] Shelter agents and evacuation route agents:

[0214] Collaborative Optimization: The shelter agent and the evacuation route agent need to be collaboratively optimized to ensure the efficiency and smoothness of the evacuation process. For example, when a certain shelter is approaching saturation, the evacuation route agent can adjust the traffic flow to guide the resident agent to other shelters.

[0215] Information Sharing: Real-time information such as the capacity change of shelters and road congestion is shared between the two to make more accurate decisions and scheduling.

[0216] Shelter Agent and Environment Dynamic Change Agent:

[0217] Risk Assessment: The shelter agent evaluates the safety and sustainability of the shelter based on the information provided by the environment dynamic change agent (such as natural disaster threats, environmental pollution, etc.).

[0218] Emergency Response: In the face of serious environmental threats, the shelter agent needs to activate the emergency response mechanism, such as strengthening protective measures, evacuating the internal residents, etc., and maintaining close communication with the environment dynamic change agent to obtain the latest information.

[0219] Evacuation Route Agent and Environment Dynamic Change Agent:

[0220] Road Condition Update: The environment dynamic change agent provides real-time road condition information (such as damage degree, waterlogging situation, etc.) to the evacuation route agent, so that the evacuation route agent can update the road network diagram and adjust the path planning algorithm.

[0221] Traffic Control Suggestions: According to the environmental change information, the evacuation route agent may need to put forward traffic control suggestions (such as blocking dangerous sections, opening temporary channels, etc.) to the traffic management department to ensure the safety and smoothness of the evacuation process.

[0222] Step S6: Set up the simulation scenario according to the actual situation of the target rural area.

[0223] First, set the simulation scenario parameters, including rural area characteristics, shelter layout, resident distribution, evacuation road network, emergency event type and occurrence time;

[0224] Then, import the geographical data into the simulation system, and set the attributes, location layout and interaction mechanism of each agent;

[0225] Finally, set the simulation conditions, including the simulation time length, data recording frequency and output format.

[0226] Step S7: Start the rural evacuation emergency agent simulation model, simulate the evacuation process of rural evacuees in case of emergency, and record the performance behaviors of each agent.

[0227] Load the rural refuge emergency intelligent agent simulation model constructed in step S4 and initialize all intelligent agents;

[0228] Start the engine of the rural refuge emergency intelligent agent simulation model, trigger an emergency event, and enable the resident intelligent agents to perceive the emergency situation according to the emergency event type and occurrence time set in step S6;

[0229] Conduct simulation runs, including simulation of resident intelligent agent behavior, refuge site intelligent agent behavior, refuge road intelligent agent behavior, and environmental dynamic change intelligent agent behavior;

[0230] Data recording and monitoring, including recording the evacuation time, path selection, and moving speed of resident intelligent agents; recording the traffic conditions and passing time of refuge roads; monitoring the accommodation situation, utilization rate, and possible congestion points of refuge sites;

[0231] When all resident intelligent agents reach the refuge site or the preset simulation duration is reached, the simulation ends.

[0232] During the simulation process, an abnormal situation detection mechanism is also set up.

[0233] The detection mechanism is as follows:

[0234] During the simulation process, the system monitors the states of each intelligent agent, environmental changes, and traffic flow data in real time;

[0235] Set thresholds based on historical data to judge the occurrence of abnormal situations. When the analyzed real-time data exceeds the threshold, trigger the abnormal situation detection mechanism;

[0236] Calculate the Z - score of each index to judge the abnormal threshold:

[0237] ;

[0238] Among them, represents the observed value; represents the historical mean; represents the historical standard deviation.

[0239] Step S8: Evaluate the simulation results and optimize the intelligent agent behavior rules and environmental models according to the evaluation results.

[0240] It should be noted that the content not elaborated in detail in the present invention is all prior art and is well known to those skilled in the art.

[0241] Therefore, by adopting the above - mentioned rural refuge personnel evacuation simulation method based on a simulation model, the present invention can optimize the emergency evacuation strategy in rural areas and improve the efficiency and safety in dealing with emergencies such as natural disasters.

[0242] 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 the technical solutions of the present invention or make equivalent replacements, 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 simulating the evacuation of rural refuge seekers based on a simulation model, characterized in that, It includes the following steps: Step S1: Collect the design indicators and influencing factors of rural evacuation spaces; Step S2: Construct a rural evacuation space planning and design database based on the design indicators and influencing factors; Step S3: Construct resident agents, evacuation site agents, evacuation road agents, and environmental dynamic change agents; Step S4: Construct an environmental model according to the actual situation of the target rural area, and embed the resident agents, evacuation site agents, evacuation road agents, and environmental dynamic change agents into it to form a complete rural evacuation emergency agent simulation model; Step S5: Design the interaction mechanism between the four types of agents; Step S6: Set the simulation scenario according to the actual situation of the target rural area; Step S7: Start the rural evacuation emergency agent simulation model, simulate the evacuation process of rural evacuees in an emergency, and record the performance behaviors of each agent; Step S8: Evaluate the simulation results, and optimize the agent behavior rules and environmental model according to the evaluation results; In Step S3, the basic attributes of the resident agent include age, gender, mobility, and mental state; the behavior rules of the resident agent include environmental perception strategy, evacuation strategy, path selection strategy, exit selection strategy, and action execution strategy; The environmental perception strategy is as follows: ; Among them, represents the perception distance; represents the position of the resident agent; represents the position of the shelter or obstacle; The evacuation strategy is as follows: ; Among them, represents risk assessment; represents the risk threshold; represents the evacuation decision; ; Among them, represents the distance from the resident agent to the nearest disaster point; represents the score of the current traffic condition; represents the impact score of environmental obstacles; , , all represent weights; The path selection strategy is as follows: ; Among them, represents the optimal path; represents the weight of the th segment in the path; represents the length of the path segment; represents the total number of all segments in the path; The exit selection strategy is as follows: ; Among them, represents the best exit; represents the maximum number of people that the exit of the evacuation site can accommodate; represents the proportion of the number of people currently using this exit; The action execution strategy is as follows: ; Among them, represents the movement speed; represents the remaining distance to the shelter site; represents the remaining time; The basic attributes of the evacuation road agent include road width, road length, and road traffic capacity; the behavior rules of the evacuation road agent include simulating traffic conditions, calculating travel time, and obstacle avoidance and path optimization; Simulating traffic conditions is as follows: ; Among them, represents the traffic flow; represents the speed of the th vehicle; represents the road proportion occupied by the vehicle Calculating travel time is as follows: ; Among them, represents the passing time; represents the road length; represents the average vehicle speed; Obstacle avoidance and path optimization are as follows: ; Among them, represents the optimal obstacle avoidance path; represents the influence degree of the obstacle; represents the weight coefficient; The basic attributes of the evacuation site agent include location, capacity, and accessible roads; the behavior rules of the evacuation site agent include receiving evacuees strategy, evaluating accommodation capacity, and material distribution strategy; The receiving evacuees strategy is as follows: ; Among them, represents capacity allocation; represents the number of resident agents attempting to enter the shelter site; represents the number of people that the shelter site can accommodate; Evaluating accommodation capacity is as follows: ; Among them, represents the accommodation capacity; represents the number of people already in the shelter area; represents the maximum accommodation capacity of the shelter area; The material distribution strategy is as follows: ; Among them, represents material distribution; represents the total amount of materials in the shelter site; represents the number of resident agents in the shelter site; The basic attributes of the environmental dynamic change agent include perception ability, data processing ability, prediction ability, and communication ability; the behavior rules of the environmental dynamic change agent include real-time monitoring of the external environment, updating environmental data, analyzing environmental data, and sharing environmental data.

2. The rural refuge personnel evacuation simulation method based on a simulation model according to claim 1, wherein In Step S1, the design indicators include evacuation shelters and evacuation roads; the influencing factors include resident behavior factors, spatial environment characteristics, infrastructure characteristics, and natural environment characteristics.

3. A method for simulating the evacuation of rural refuge seekers based on a simulation model according to claim 1, characterized in that, In Step S2, construct a rural evacuation space planning and design database according to the design indicators and influencing factors. The specific operation is as follows: conduct a Pearson correlation analysis on the design indicators and influencing factors, identify the design indicators and influencing factors that have a significant impact on rural evacuation space planning and design, and construct a rural evacuation space planning and design database.

4. The rural refuge personnel evacuation simulation method based on a simulation model according to claim 3, characterized in that A learning mechanism is also introduced into the resident agent, specifically: Define the state space : Define various situations encountered by resident agents as states. Various situations include the distance to the shelter, road congestion, and environmental obstacles; Design action space : Design actions for resident agents, including choosing different evacuation routes and adjusting walking speeds; Set the reward function : Give corresponding rewards or punishments according to the behavior results of the resident agent. The behavior results include the time to successfully reach the shelter and the number of collisions avoided; ; Among them, represents the reward for reaching the evacuation site, represents the reward for avoiding collisions; Training process: Let the resident agent continuously try different actions in the simulated environment and adjust the strategy according to the reward function until the optimal evacuation path and behavior pattern are found; ; Among them, represents taking an action in state of value, represents the learning rate, represents the discount factor, represents the new state after taking action and represents taking an action in the new state and represents the updated value.

5. A method for simulating the evacuation of rural refuge seekers based on a simulation model according to claim 4, characterized in that, In step S4, an environmental model is constructed according to the actual situation of the target rural area, and the resident agent, the shelter site agent, the evacuation route agent, and the environmental dynamic change agent are embedded therein to form a complete rural evacuation emergency agent simulation model. The specific operations are as follows: Integrate the influencing factors collected in step S1 and perform processing operations, including data cleaning, data transformation, and data verification; Use 3D modeling software to construct the geographical environment model of the target rural area based on the processed data, including terrain, building structure, and road network; Based on the geographical environment model, layout the shelter sites and evacuation routes; Design a set of interaction interfaces for the interaction between each agent and the environmental model. The interaction interfaces include data interfaces, control interfaces, and visualization interfaces. The data interfaces define the formats and protocols for data transmission between each agent and the environmental model. The control interfaces provide interfaces for control commands for each agent. The visualization interfaces are used to display the real-time status of the environmental model and each agent. After the environmental model is constructed, perform model verification and adjustment, including logic verification, parameter adjustment, and test run.

6. The rural refuge personnel evacuation simulation method based on a simulation model according to claim 5, wherein In step S5, the interaction mechanisms between the four types of agents include: the interaction mechanism between resident agents, the interaction mechanism between resident agents and shelter site agents, the interaction mechanism between resident agents and evacuation route agents, the interaction mechanism between resident agents and environmental dynamic change agents, the interaction mechanism between shelter site agents and evacuation route agents, the interaction mechanism between shelter site agents and environmental dynamic change agents, and the interaction mechanism between evacuation route agents and environmental dynamic change agents.

7. A method for simulating the evacuation of rural refuge seekers based on a simulation model according to claim 6, characterized in that, In step S6, set the simulation scenario according to the actual situation of the target rural area. The specific operations are as follows: First, set the simulation scenario parameters, including rural area characteristics, shelter site layout, resident distribution, evacuation route network, emergency event types, and occurrence times; Then, import geographical data into the simulation system, and set the attributes, location layouts, and interaction mechanisms of each agent; Finally, set the simulation conditions, including the simulation time length, data recording frequency, and output format.

8. A method for simulating the evacuation of rural refuge seekers based on a simulation model according to claim 7, characterized in that, In step S7, start the rural evacuation emergency agent simulation model, simulate the evacuation process of rural evacuees in an emergency, and record the performance behaviors of the agents, including the following steps: Load the rural evacuation emergency agent simulation model constructed in step S4 and initialize all agents; Start the engine of the rural evacuation emergency agent simulation model, trigger the emergency event, and make the resident agents perceive the emergency according to the emergency event type and occurrence time set in step S6; Perform simulation runs, including resident agent behavior simulation, shelter site agent behavior simulation, evacuation route agent behavior simulation, and environmental dynamic change agent behavior simulation; Data recording and monitoring, including recording the evacuation time, path selection, and moving speed of resident agents; recording the traffic conditions and passing times of evacuation routes; monitoring the accommodation situation, utilization rate, and congestion points that appear in shelter sites; When all resident agents reach the shelter site or the preset simulation duration is reached, the simulation ends.

9. A method for simulating the evacuation of rural refuge seekers based on a simulation model according to claim 8, characterized in that In step S7, during the simulation process, an abnormal situation detection mechanism is also set up, and the detection mechanism is as follows: During the simulation process, the system monitors the states of each agent, environmental changes, and traffic flow data in real time; Set thresholds according to historical data to judge the occurrence of abnormal situations. When the analyzed real-time data exceeds the threshold, the abnormal situation detection mechanism is triggered; Calculate the Z-socre of each index to judge the abnormal threshold: ; Among them, represents the observed value; represents the historical mean; represents the historical standard deviation.

Citation Information

Patent Citations

  • Multi-agent-based virtual miner safety behavior modeling and emergency simulation system

    CN104317637A

  • Urban traffic emergency evacuation simulation system and control method

    CN104765924A