Rural emergency evacuation method and system

Abstract

The invention belongs to the technical field of intelligent disaster prevention, and relates to a rural emergency evacuation method and system, and the method comprises the steps: obtaining the distribution data of emergency evacuation elements in a target region through a GIS platform; according to the distribution data of the emergency evacuation elements, combining construction land constraint conditions, screening shelters, and generating an emergency evacuation path network; emergency evacuation elements in the emergency evacuation path network are analyzed through evaluation indexes of multiple dimensions, and an analysis result is obtained; performing global optimization on the emergency evacuation path network through an NSGA-II multi-objective optimization algorithm in combination with an analysis result; and configuring emergency evacuation space elements in the target area according to a global optimization result. The scheme is based on a public safety perspective, focuses on core elements of rural emergency evacuation space planning, constructs an emergency evacuation scheme which accords with rural reality and has dynamic adjustable capability, and provides theoretical support and practical reference for improving rural disaster protection capability.

Description

TECHNICAL FIELD

[0001] The present application relates to a rural emergency evacuation method and system, belonging to the technical field of intelligent disaster prevention. BACKGROUND

[0002] In sudden disasters such as earthquakes, rapid and orderly evacuation is a key passive defense measure to maximize personnel life safety. The suddenness, strong destructiveness and possible secondary disasters of earthquakes make personnel exposed to dangerous environments face a high risk of injury and death. Timely evacuation to a pre-planned safe area can effectively avoid the direct threat of building collapse, escape from the spread of secondary disasters, and create conditions for subsequent rescue operations. Therefore, scientific and efficient evacuation is an indispensable core link in disaster emergency response.

[0003] Compared with cities, there are several problems in rural areas during emergency evacuation: First, the rural infrastructure is relatively weak, and the construction standards of roads and houses are not uniform, which may cause problems such as narrow roads, houses expanding into roads, many steep slopes, and sudden lane changes.

[0004] Second, self-built houses account for a large proportion of rural houses, which are more prone to collapse and hinder evacuation routes.

[0005] Third, rural areas lack standardized emergency shelters that have been scientifically evaluated, and temporary resettlement sites often lack necessary supplies, and the distribution method is relatively random, generally using equal distribution, which leads to a lack of supplies in some emergency shelters. SUMMARY

[0007] To solve the above problems, the present application provides a rural emergency evacuation method and system, which focuses on the core elements of rural emergency evacuation space planning from the perspective of public safety, and builds an emergency evacuation plan that fits the actual situation of rural areas and has dynamic adjustment ability, providing theoretical support and practical reference for improving the disaster prevention capability of rural areas.

[0008] To achieve the above purpose, the present application provides the following technical scheme: a rural emergency evacuation method, comprising the following steps: obtaining distribution data of emergency evacuation elements in a target area through a GIS platform; according to the distribution data of emergency evacuation elements, combining with the constraint conditions of construction land, screening shelters, and generating an emergency evacuation path network; analyzing emergency evacuation elements in the emergency evacuation path network through multiple evaluation indexes to obtain analysis results; combining the analysis results, performing global optimization on the emergency evacuation path network through an NSGA-II multi-objective optimization algorithm; and configuring emergency evacuation space elements in the target area according to the global optimization results.

[0009] Further, the method for establishing the emergency evacuation path network comprises: calculating spatial distances between path networks and shelters in a map by GIS; calculating average reachable times of each shelter according to the spatial distances; modeling all roads in the countryside according to the spatial distances and the average reachable times, and identifying positions and types of various traffic nodes from the modeled model; dividing homesteads in the map into units by comprehensively considering terrain features and existing household data, and counting spatial distribution densities and aggregation forms of the homesteads; analyzing matching degrees of path networks, traffic nodes, and residential structures of the homesteads, so as to determine potential evacuation organization difficulty areas and key evacuation nodes; selecting areas in the map that can be used as alternative shelters; and calculating service ranges of all shelters and alternative shelters in the map.

[0010] Further, the index types corresponding to the evaluation indexes in the multiple dimensions are obtained according to distribution data of various traffics and shelters in the target area; and the analysis result comprises a land use scheme, a village type, a geological disaster risk, a road network distribution, a shelter distribution, and an evacuation planning analysis basic layer.

[0011] Further, the evaluation indexes in the multiple dimensions comprise: an evaluation index in a dimension of evacuation paths, an evaluation index in a dimension of shelters, an evaluation index in a dimension of geological disaster risks, and an evaluation index in a dimension of population distribution; the evaluation index in the dimension of evacuation paths is used for evaluating connectivity and reachability of evacuation paths; the evaluation index in the dimension of shelters is used for evaluating layout rationality and capacity of shelters; the evaluation index in the dimension of geological disaster risks is used for evaluating geological disaster risks in the region; and the evaluation index in the dimension of population distribution is used for evaluating intensive degree of population distribution and evacuation demand.

[0012] Further, the evaluation index in the dimension of evacuation paths comprises road connectivity, evacuation path length, path width, and traffic congestion degree; the evaluation index in the dimension of shelters comprises shelter capacity, shelter service radius, and shelter reachability; the evaluation index in the dimension of geological disaster risks comprises an area proportion of geological disaster-prone areas, a proportion of low-lying areas, and a potential blocking risk; and the evaluation index in the dimension of population distribution comprises population density, a proportion of old population, and a proportion of children population.

[0013] Further, the method for determining the evaluation result of each dimension of evaluation index comprises: determining initial values of evaluation indexes according to index types corresponding to the evaluation indexes; dividing the evaluation indexes into positive index types and reverse index types; normalizing the initial values of the positive index types and the initial values of the reverse index types to obtain evaluation index values in multiple dimensions; obtaining the evaluation index values in each dimension in the target area, multiplying the evaluation index values by their weights, and then adding all the evaluation indexes to obtain an evaluation result of the dimension.

[0014] Further, according to the evaluation results, the emergency evacuation space in the target area is classified. If the evaluation results of the evaluation indexes of the refuge site dimension in the target area are higher than the preset refuge site optimization index value, the area is divided into a refuge site optimization zone. If the evaluation results of the evaluation indexes of the evacuation path dimension in the target area are higher than the preset evacuation path optimization index value, the area is divided into an evacuation path optimization zone. If the evaluation results of the evaluation indexes of the geological disaster risk dimension in the target area are higher than the preset geological disaster risk prevention and control index value, the area is divided into a geological disaster risk prevention and control zone. If the evaluation results of the evaluation indexes of multiple dimensions in the target area are all lower than the preset optimization index value, the area is divided into a general emergency evacuation zone.

[0015] Further, the method for configuring the emergency evacuation space elements in the target area according to the global optimization results is as follows. For the refuge site optimization zone, the number of refuge sites is increased or the area of the refuge site is expanded. For the evacuation path optimization zone, the connectivity and traffic capacity of the road are improved, traffic congestion is reduced, and evacuation efficiency is improved. For the geological disaster risk prevention and control zone, the geological disaster monitoring and early warning system is strengthened, the site selection of the refuge site is optimized, and the site selection in the high-risk area of geological disasters is avoided. For the general emergency evacuation zone, the existing layout of the refuge site and the evacuation path is maintained, and regular maintenance and inspection are carried out to ensure the usability in emergency situations. According to the population distribution and refuge demand, the service radius of the refuge site is adjusted to ensure that residents can reach the nearest refuge site within the specified time. Special evacuation channels and refuge sites are set for the evacuation needs of special groups. Necessary life facilities and emergency material storage points are set inside the refuge site to improve the emergency support capacity of the refuge site. GIS and multi-objective optimization algorithms are used to dynamically optimize the refuge site and the evacuation path.

[0016] Further, the method for dynamically optimizing the refuge site and the evacuation path by the multi-objective optimization algorithm is as follows. An initial refuge site and evacuation path set is randomly generated and used as an initial population. A child population is generated from the initial population through selection, crossover and mutation operations. The initial population and the child population are combined to generate a combined population. The individuals in the combined population are classified according to the Pareto front, and the individuals in the same class are sorted according to the crowding degree. The obtained classes from low to high are filled into a new population. If the number of individuals in the same class exceeds the remaining capacity, the individual with a larger crowding degree is preferentially retained. The above steps are repeated until the Pareto optimal solution set with the smallest Pareto front between the two objective functions is obtained.

[0017] This invention also discloses a rural emergency evacuation system, comprising: a data acquisition module for acquiring distribution data of emergency evacuation elements within a target area through a GIS platform; an emergency evacuation path network generation module for selecting refuge sites and generating an emergency evacuation path network based on the distribution data of emergency evacuation elements and construction land constraints; an evacuation element analysis module for analyzing the emergency evacuation elements in the emergency evacuation path network using multiple dimensions of evaluation indicators to obtain analysis results; a global optimization module for globally optimizing the emergency evacuation path network using the NSGA-II multi-objective optimization algorithm based on the analysis results; and a result output module for configuring emergency evacuation spatial elements within the target area based on the global optimization results. The technical solution of the present invention has at least the following technical effects or advantages: 1. This invention is based on the perspective of public safety and focuses on the core elements of rural emergency evacuation space planning. It constructs an emergency evacuation plan that fits the actual situation in rural areas and has dynamic and adjustable capabilities, providing theoretical support and practical reference for improving rural disaster prevention capabilities.

[0018] 2. This invention analyzes and optimizes emergency evacuation elements within the target area by employing multiple evaluation indicators, thereby achieving the goal of scientifically planning and precisely configuring emergency evacuation routes and refuge sites based on the actual geographical, population, and disaster characteristics of rural areas. This solves the technical problem of low matching degree between rural emergency evacuation planning and actual rural needs in related technologies. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a rural emergency evacuation method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the emergency evacuation route network in one embodiment of the present invention; Figure 3 This is a diagram showing the Pareto distance and iterative evolution of propagation in an embodiment of the NSGAⅡ algorithm model of the present invention; Figure 4 This is a schematic diagram of an evacuation method with the shortest evacuation time but the highest congestion in a Pareto optimal solution set according to an embodiment of the present invention. Figure 5 This is a schematic diagram of an evacuation method in an embodiment of the present invention, which balances the number of evacuees at each evacuation point in the Pareto optimal solution set to reduce the overall congestion. Figure 6 This is a comparison diagram of the location of refuge sites and emergency evacuation routes generated by the GIS spatial analysis model (a) and the NSGA-Ⅱ optimization algorithm (b) in one embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention is described in detail through specific embodiments. However, it should be understood that the specific embodiments are provided only for a better understanding of the present invention and should not be construed as limiting the present invention. In the description of the present invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0021] To address the issue that existing emergency evacuation methods are unsuitable for rural conditions, this invention proposes a rural emergency evacuation method and system. The aim is to develop a novel rural emergency evacuation scheme that combines a GIS spatial analysis model with the NSGA-II optimization algorithm. The scheme consists of three parts: First, ArcGIS is used to perform spatial analysis of the study area, constructing a basic evacuation network, feasible emergency evacuation routes, and emergency shelter locations based on variables. Second, the NSGA-II algorithm is used to obtain the Pareto front solution set for multi-objective emergency evacuation. Finally, based on the optimized equilibrium solution, alternative shelters are added and deployed for disaster victims whose evacuation time has not yet met the requirements. This scheme, grounded in public safety, focuses on the core elements of rural emergency evacuation spatial planning, constructing an emergency evacuation scheme that fits the realities of rural areas and possesses dynamic adjustability, providing theoretical support and practical reference for improving rural disaster prevention capabilities. The following detailed description, with reference to accompanying drawings and embodiments, illustrates the invention in detail.

[0022] Example 1 This embodiment uses Xiji Town, Tongzhou District, Beijing as an example to describe in detail the rural emergency evacuation method. This rural emergency evacuation method, as follows: Figure 1 As shown, it includes the following steps: S1 obtains the distribution data of emergency evacuation elements within the target area through a GIS platform.

[0023] This embodiment utilizes ArcGIS's Network Analyst module, combined with multi-source geographic information data, to select emergency shelter locations and construct rural emergency evacuation route models. The collected data covers 11 core categories of information, including land use, population distribution, refuge space, geological hazard risk, and road structure. After standardization, spatial registration, and layer overlay, a fused database is formed, providing data support for route analysis and site selection.

[0024] In evacuation simulation, a simulation environment model incorporating terrain, buildings, road networks, and refuge points is constructed. Parameters such as population distribution, road accessibility, and vulnerability are set to simulate disaster scenarios such as floods and earthquakes, analyzing path accessibility and bottleneck nodes. Network Analyst is used to establish a topology network, assign road attributes, and clarify path connectivity and traffic levels. The "building density on both sides of the road" index is introduced to identify potential blockage risks, and low-lying and flood-prone areas are identified based on DEM data to assess evacuation safety. Through community emergency drills and drone / GPS tracking, crowd behavior data is collected, individualized simulation parameters are set, and path load and facility accessibility are tested under different scenarios to identify high-risk road sections and key bottlenecks.

[0025] Based on simulation results, the selection of refuge site locations and evacuation routes are optimized, and diversion measures, alternative routes, and route signage adjustment schemes are proposed to improve route flexibility and evacuation efficiency. Peak congestion points and their formation mechanisms are analyzed using "person-time curves," supporting spatial optimization of evacuation strategies. Finally, a basic evacuation network layer with high responsiveness and safety is constructed, providing key inputs and decision-making basis for subsequent NSGA-II multi-objective optimization. The specific methods are as follows: Xiji Town, Tongzhou District, Beijing, is located in the southeastern part of Tongzhou District. It borders Jiangxintun Town, Xianghe County, Hebei Province to the east (separated by the Chaobai River), Huoxian Town to the south, Zhangjiawan Town across the North Canal to the southwest, Lucheng Town to the north, and Qigezhuang Town, Dachang Hui Autonomous County, Hebei Province to the northeast. The town covers an area of ​​94.28 square kilometers. The Chaobai River and the Beijing-Hangzhou Grand Canal flow through the town, and the Beijing-Shenyang Expressway, Beijing-Tianjin Highway (National Highway 103), and Tongxiang Highway pass through it. Natural disasters in the Tongzhou area mainly include floods, droughts, strong winds, earthquakes, ground subsidence, and hail, with floods and droughts occurring most frequently, and earthquakes causing the greatest damage (see "Tongzhou District Geological Conditions and Geological Disaster Prevention Requirements - Geological Disaster Investigation and Assessment," Beijing Municipal Commission of Planning and Natural Resources). Tongzhou District is located at the lowest point in Beijing, and the area between the North Canal and the Chaobai River in the east has a terrain characterized by higher elevations near the riverbed and lower elevations further away due to the deposition of sediment from recent river floods. This creates a strip-shaped depression extending along the riverbed, making it extremely prone to flooding. These disasters are characterized by their suddenness and destructive power, significantly impacting agricultural production and infrastructure in the area.

[0026] This embodiment uses a GIS platform to integrate multi-source data such as the current land use status of Xiji Town, village types, geological disaster risks, villages to be relocated as a whole, road network and distribution of refuge sites, to construct distribution data of emergency evacuation elements within the target area.

[0027] Based on the village type map, villages within the region are categorized into four types: urban-integrated villages, relocated villages, improved and renovated villages, and villages with distinctive features. Statistical analysis of their type distribution, quantity, and spatial coverage is conducted. Combined with the geological hazard risk map, the spatial relationship between the location and distribution of high-risk villages and evacuation sites is identified, clarifying evacuation blind spots and priority prevention and control areas. Simultaneously, the spatial distribution of relocated villages is processed, their post-relocation land use potential is assessed, and their feasibility as backup evacuation spaces or emergency material reserve points is proposed.

[0028] Analysis of the village type map reveals that urban-integrated villages have high population densities and should be prioritized for the development of large-scale comprehensive refuge sites; villages undergoing improvement and upgrading require additional infrastructure to enhance their refuge capabilities; and villages undergoing overall relocation should optimize their refuge site layout in conjunction with relocation plans. The northeastern part of Xiji Town, located in the flood storage area of ​​the North Canal (outside the embankment road), primarily houses villages undergoing overall relocation; the central part, near the Xiji Town government, mainly contains urban-integrated villages; and the southern edge, near the canal, is primarily composed of villages undergoing characteristic upgrading. The population of Xiji Town in Tongzhou District is mainly concentrated in the town center, specifically in Xiji Village, Zhanggezhuang Village, and Langdong Village. Most villages along the outer boundary of the town exhibit high population density, while those on the right side of the town have lower density. This is somewhat correlated with the village types mentioned above in Xiji Town, where urban-integrated villages have the highest population density, followed by villages undergoing characteristic upgrading.

[0029] Based on the distribution data of emergency evacuation elements and combined with the constraints of construction land, S2 selects refuge sites and generates an emergency evacuation route network.

[0030] The distribution data of emergency evacuation elements include path speed, shelter capacity, shelter location allocation, pedestrian size, and pedestrian speed. Specifically, regarding path speed settings: In terms of road network and speed settings, reasonable road speed settings are the foundation for optimizing emergency evacuation routes. Assigning different speed values ​​according to road grade can effectively reflect the traffic capacity of different roads during evacuation. This study classifies roads into four grades: highways, arterial roads, secondary arterial roads, and local roads / narrow roads, with corresponding speed values ​​of 100 km / h, 70 km / h, 50 km / h, and 30 km / h, respectively. Route planning is based on the assumption of a pedestrian walking speed of 1.25 m / s (Niu Jufen, 2025). Considering the actual needs of slower-moving groups such as the elderly and children, the shortest path is generated by analyzing the existing rural road network. Arterial roads are prioritized as the core channels of the evacuation routes, supplemented by secondary roads connecting the internal and peripheral areas of villages, forming this evacuation route system. The service radius of the emergency shelters is set at 500-1000 meters (according to DB11 / T 2141-2023, "Classification and Grading of Emergency Shelters"), and is marked with red dotted lines to ensure that residents can reach the shelters on foot within 10 minutes. This graded speed setting fully considers road conditions and safety requirements, providing a scientific basis for subsequent route planning.

[0031] Constructing a network dataset is a crucial step in optimizing emergency evacuation routes. During the preparation phase, basic data such as settlements (point layer, including population attributes), shelters (point layer, including capacity attributes), and road networks (line layer, including grade and speed attributes) were established. In the network dataset construction phase, road line features were selected as the network source, and the connectivity strategy was set to "End Point." Simultaneously, a cost attribute named "Time," with the type selected as "Cost," was added and associated with the road's Length and Speed ​​fields. The calculation formula is as follows:

[0032] In this way, the study incorporates the travel time of the road network as a cost factor into the emergency evacuation route model, providing an accurate basis for cost calculation for subsequent analysis.

[0033] Shelter capacity settings: When setting the attributes of shelters and settlements, add a "Capacity" field to the shelter attribute table to record the maximum capacity of each shelter. The calculation formula is as follows:

[0034] Meanwhile, the study added a "Population" field to the settlement attribute table to indicate the number of people requiring evacuation. Considering that settlements are generally grouped into groups of 5-8 households, the average number of people per settlement was set to 12. Through these attribute settings, the study can accurately grasp the carrying capacity of shelters and the evacuation needs of settlements, providing crucial data support for route optimization.

[0035] Location Allocation Settings: This study utilizes ArcGIS's Network Analyst module to conduct Location-Allocation Analysis, setting corresponding analysis parameters to achieve optimal service matching between shelters and residential areas. This is a core element in optimizing emergency evacuation routes. During implementation, the study loads facility points (optional shelter locations), selects the shelter point layer, and sets field mappings, mapping the capacity field to the Capacity attribute. It also loads demand points (residential areas), selects the residential area layer, and sets field mappings, mapping the Population field to the Weight attribute (representing demand). The search range is set to 5000 meters (the service radius of Class I shelters). Finally, in the Analysis Type tab, "Maximize Coverage with Capacity Limitations" is selected, and the solution is obtained, yielding the optimal allocation result. This process fully considers the capacity limitations of shelters and the evacuation needs of residential areas, achieving optimal allocation between residential areas and shelters through scientific model calculations. After obtaining the allocation results from the Location-Allocation Analysis, the model needs to further generate optimal routes. The study exported the generated distribution lines and used the "Discounted Feature to Point" tool to export the starting and ending points by selecting the "start" and "end" points respectively. A new path was created in the network analysis, adding stop points, loading the starting and ending features, setting the corresponding fields, and then clicking "Solve" to generate the path, resulting in the final optimal path rendering.

[0036] This process clearly demonstrates the optimal evacuation routes from settlements to nearby shelters, providing intuitive guidance for shelter location selection and route planning in emergency evacuations. The allocation lines represent the relocation of villagers from villages whose shelters cannot accommodate them to nearby shelters with sufficient capacity, taking into account the maximum capacity of each existing shelter.

[0037] Pedestrian Dimensioning: Suitability of Escape Routes In the simulation, the model can dynamically adjust the movement behavior of each agent. When an individual is stationary, the space it occupies has certain geometric and physical characteristics; its projection is roughly elliptical, encompassing dimensions such as shoulder width, body width, and arm width. Referring to the book "Microscopic Simulation of Pedestrians in Urban Rail Transit Stations," statistical data from 2812 pedestrians were used to determine the shoulder width and body width of different pedestrian types. These data were used as the dimensional parameters of the agents in the simulation, as shown in Table 1.

[0038] Table 1. Size parameters of the agent

[0039] Pedestrian speed is generally defined as the arithmetic mean of the instantaneous speeds of pedestrians passing through a specific cross-section. It is influenced by various factors, including the pedestrian's age, gender, mobility, purpose of travel, weather conditions, items carried, surrounding environment, and pedestrian traffic. Among these factors, age and gender are the primary influencing factors. Under free walking conditions, pedestrian speed distribution typically follows a normal (Gaussian) distribution. Li Dewei et al., through field observation, statistical analysis, and comparison with early domestic data, concluded that the average walking speed of pedestrians in China is 1.39 m / s. In emergency situations, the movement speed of crowds increases; therefore, in the simulation, the average walking speed of adult male disaster victims was set at 1.5 m / s, with its speed distribution conforming to a normal distribution range of 1.2 to 1.7 m / s. For female disaster victims, their walking speed was set to 5 / 6 of that of adult males, while the walking speed of the elderly was set to 2 / 3 of that of adult males. Children's walking speed was set to 1 / 3 of that of adult males, i.e., an average walking speed of approximately 0.5 m / s.

[0040] Crowding Degree Setting: Individuals are influenced by their surroundings and other pedestrians while walking. When the external repulsive force is too great, individuals may transition from free walking to slow movement, or even stop completely. The interaction of numerous individuals during movement can lead to traffic congestion. As the flow of people accumulates, the speed of movement gradually decreases until it stops, and the flow of people also ceases. Accidents can occur in various locations related to the number of pedestrians. Using pedestrian density as a key indicator for measuring evacuation efficiency is of great significance. In areas with excessively high pedestrian density, safety incidents such as crowding and stampedes are prone to occur, which not only cause secondary injuries to the evacuation of affected people but also significantly reduce evacuation efficiency. A review of a large amount of representative evacuation data shows a logarithmic relationship between pedestrian flow speed and density; that is, as pedestrian density increases, pedestrian flow speed gradually decreases. When the pedestrian density ρ exceeds 2.0 people / m², the pedestrian flow speed decreases significantly; and when the pedestrian density ρ exceeds 4 people / m², the pedestrian flow speed drops to a low level, almost stagnating. Therefore, ρ = 2.0 people / m² and 4.0 people / m² were set as the thresholds for pedestrian congestion and pedestrian collapse, respectively. When the pedestrian density ρ in the simulation reaches 2.0 people / m², the site space is more likely to experience congestion; while when the pedestrian density ρ exceeds 4.0 people / m², the congestion becomes very severe, easily leading to safety risks. Based on the above analysis, potential problems in the evacuation routes were identified, such as bottleneck locations, congestion points, and areas where movement speed is hindered.

[0041] The emergency evacuation route network is a core element of emergency evacuation route planning, directly determining the efficiency and safety of evacuation operations. The method for establishing an emergency evacuation route network is as follows: During data processing, the spatial distance between path network nodes in the GIS map and the refuge site is calculated to assess the accessibility of the refuge site.

[0042] Based on spatial distance and considering the design speed and service capacity of different roads, the average reachability time of each evacuation site is calculated, identifying efficient passages and inefficient nodes in evacuation routes. The connectivity performance of highways and provincial roads is analyzed in particular, assessing their capacity in emergency situations and whether they can meet the evacuation needs of large numbers of people and vehicles during a disaster. Combined with existing evacuation site distribution data, the service area of ​​evacuation sites around highways and provincial roads is clarified, thereby optimizing the resource allocation of major evacuation routes. Figure 2 As shown, all roads in Xiji Town, Tongzhou District, are marked with different shades of color according to road classification, mainly divided into: expressways, provincial roads, county roads, and rural roads. Within the town's area, expressways form a cross shape, and provincial roads crisscross from west to southeast.

[0043] Based on spatial distance and average reachability, all roads in rural areas were modeled, and the locations and types of various traffic nodes (road intersections), such as main road junctions, secondary connections, and village entrances and exits, were identified from the models. The study revealed weak connections between some high-density residential villages and main evacuation routes in Xiji Town, particularly the lack of efficient connections between some internal village paths and main roads, potentially leading to prolonged evacuation routes or even congestion. The spatial distribution and accessibility of traffic intersections have a decisive impact on overall evacuation time and efficiency during disasters. Therefore, optimizing the traffic corridor system, especially the reconstruction and expansion of key nodes, is a core element in ensuring the efficient operation of village emergency evacuation systems.

[0044] Homestead Land Division: Homestead land distribution data reflects the residential pattern and evacuation units within villages, serving as a crucial foundation for emergency evacuation planning. Density statistics of homestead land units allow for further detailed analysis of their spatial distribution characteristics, revealing the intensity of residential clustering in different areas. Initially, a division method of 5-6 households per homestead land unit was adopted, simplifying it into regular square dots, with each dot representing a residential settlement unit. Statistical analysis was then conducted on their spatial distribution density, clustering patterns, and proximity relationships. This method helps to quickly establish a spatial model of the village's residential pattern at a macro level, thus reflecting the internal residential structure of the village more precisely and identifying high-density areas that may become evacuation obstacles or key nodes that are difficult to organize, providing a foundation for subsequent evacuation route planning and refuge site layout. However, actual research revealed limitations in the homestead land division method based on a uniform number of households, particularly in simulating the actual resident population distribution in Xiji Town. The spatial accuracy exhibited deviated from reality, with some areas having a large number of homestead land plots but a low actual resident population, or sparsely distributed homestead land plots but still high population density. Therefore, subsequent studies improved the method for dividing residential land into micro-level subdivisions based on geographical units. This method comprehensively considers topographic features (such as plot shape, natural boundaries, river and road networks) as well as existing household registration data and field survey results, simulating actual living spaces through a more flexible spatial unit delineation approach.

[0045] By analyzing the matching degree between the route network, traffic nodes and the residential structure of homesteads, potential areas of difficulty in evacuation organization and key evacuation nodes can be identified.

[0046] Select areas on the map that can serve as alternative shelters, such as unused land, green spaces, parks, squares, and public facility land. Public facility land includes: school playgrounds, hospital vacant lots, village committee squares, sports stadiums, and squares surrounding community service centers.

[0047] Calculate the service area of ​​all shelters and alternative shelters on the map. Perform detailed observations of the area within 1000m of the service area of ​​the alternative shelters.

[0048] S3 analyzes the emergency evacuation elements in the emergency evacuation route network using multiple evaluation indicators and obtains the analysis results.

[0049] The evaluation indicators for multiple dimensions are obtained based on the distribution data of various transportation and refuge elements within the target area; the analysis results include land use plans, village types, geological disaster risks, road network distribution, refuge site distribution, and evacuation planning analysis base layers.

[0050] The evaluation indicators across multiple dimensions include: evaluation indicators for evacuation routes, evaluation indicators for refuge sites, evaluation indicators for geological disaster risk, and evaluation indicators for population distribution. The evaluation indicators for evacuation routes are used to evaluate the connectivity and accessibility of evacuation routes; the evaluation indicators for refuge sites are used to evaluate the rationality and capacity of the layout of refuge sites; the evaluation indicators for geological disaster risk are used to evaluate the geological disaster risk in the region; and the evaluation indicators for population distribution are used to evaluate the density of population distribution and evacuation needs.

[0051] Evaluation indicators for the evacuation route dimension include road connectivity, evacuation route length, route width, and traffic congestion; evaluation indicators for the refuge location dimension include refuge location capacity, refuge location service radius, and refuge location accessibility; evaluation indicators for the geological disaster risk dimension include the proportion of areas prone to geological disasters, the proportion of low-lying areas, and potential blocking risks; and evaluation indicators for the population distribution dimension include population density, the proportion of elderly population, and the proportion of children population.

[0052] In this embodiment of the invention, population density is statistically analyzed using administrative villages as the basic unit. The data comes from population census data or village-level population data in township statistical yearbooks. The calculation method is to divide the total village population by the village area to obtain the population density of each administrative village. This method does not involve personal information or specific addresses, but only uses administrative villages as the overall calculation unit. The evaluation of the refuge site dimension includes three aspects: capacity, service radius, and accessibility. The capacity is directly taken as the designed or approved capacity of each refuge site. The service radius is set according to the size of the refuge site. For example, the service radius of a small refuge (square type) is 300-500 meters, that of a medium-sized refuge (school playground type) is 800-1000 meters, and that of a large refuge (stadium type) is 1500 meters. The classification standard can refer to the "Urban Disaster Prevention and Refuge Site Planning Code" GB51143-2015. Accessibility is defined as the shortest path time from a residential point (such as the center of an administrative village) to each refuge site. The evaluation of geological hazard risk is based on the classification standards of the current "Regulations on Geological Hazard Prevention and Control" and the "Technical Specification for Geological Hazard Risk Assessment (BD11 / T893-2012)". The region is divided into four risk areas: extremely high, high, medium, and low. The proportion of high and extremely high risk areas is used as the risk indicator. Simultaneously, the proportion of low-lying areas below the flood warning line is extracted using a digital elevation model (DEM). Further consideration is given to factors such as the density and height of buildings on both sides of roads or slopes greater than 35° to determine potential blockage risks. The evaluation of evacuation routes includes road connectivity, route length, width, and traffic congestion. Road connectivity can be calculated using graph theory indices such as average nodal degree, α / β connectivity, and network redundancy coefficient. Route length is taken as the average shortest path length from a residential point to the nearest shelter. Route width is obtained through remote sensing or field surveying and assigned a standard width value according to road type. Traffic congestion can be represented by the ratio of population density to road capacity. The calculation methods for each evaluation dimension all involve normalization and weighted summation to obtain a comprehensive evaluation value. Based on the evaluation results of each dimension, the study area is divided into zones: if the result of the refuge site dimension is higher than the preset threshold, it is classified as a refuge site optimization zone; if the result of the evacuation route dimension is higher than the preset threshold, it is classified as an evacuation route optimization zone; if the result of the geological disaster risk dimension is higher than the prevention and control threshold, it is classified as a geological disaster prevention and control zone; if the results of multiple dimensions are all lower than the threshold, it is classified as a general emergency evacuation zone.

[0053] The method for determining the evaluation results of each dimension's evaluation indicators is as follows: Based on the evaluation indicators corresponding to the preset indicator types, determine the initial values ​​of the evaluation indicators; divide the evaluation indicators into positive indicator types and negative indicator types; normalize the initial values ​​of the positive indicator types and the initial values ​​of the negative indicator types to obtain the evaluation indicator values ​​for multiple dimensions; obtain the evaluation indicator values ​​for each dimension within the target area, multiply the evaluation indicator values ​​by their weights, and then add all the evaluation indicators together to obtain the evaluation result for that dimension.

[0054] Based on the evaluation results, emergency evacuation spaces within the target area are classified. If the evaluation results for the refuge site dimension are higher than the preset optimization index value, the area is designated as a refuge site optimization zone. Similarly, if the evaluation results for the evacuation route dimension are higher than the preset optimization index value, the area is designated as an evacuation route optimization zone. Furthermore, if the evaluation results for the geological disaster risk dimension are higher than the preset geological disaster risk prevention and control index value, the area is designated as a geological disaster risk prevention and control zone. Analysis of the village type map indicates that urban cluster villages have high population densities, necessitating the development of large-scale comprehensive refuge sites. If the evaluation results for multiple dimensions within the target area are lower than the preset optimization index value, the area is designated as a general emergency evacuation zone. Simultaneously, the population distribution dimension evaluation results serve primarily as a corrective indicator. When the population density or the proportion of elderly and vulnerable groups in the target area is high, the optimization requirements for refuge sites and evacuation routes will be correspondingly increased, thereby guiding population-oriented optimization in key areas.

[0055] Based on the analysis results, S4 uses the NSGA-II multi-objective optimization algorithm to perform global optimization of the emergency evacuation route network.

[0056] In the existing optimization results for refuge sites and emergency evacuation routes, some village refuge sites cannot accommodate the number of people allocated according to the current computer simulation routes, rendering the optimized emergency evacuation routes infeasible. Furthermore, some existing optimization models assign one or two disaster victims to separate refuge sites, which does not meet the needs of emergency evacuation and is also infeasible. Existing optimization methods, such as GIS spatial analysis, are mainly based on Dijkstra's algorithm. However, for multi-source road planning and the comprehensive weighting of multiple variables (such as traffic flow and congestion levels), the basic Dijkstra algorithm cannot generate perfectly suitable routes during simulation. When calculating using the weight matrix, Dijkstra's algorithm uses a breadth-first search algorithm to find neighboring points, which cannot meet the multivariate and multi-source route optimization requirements of this embodiment, and may also lead to an exponential increase in the time complexity of the breadth-first algorithm.

[0057] To address the shortcomings of existing optimization methods, this embodiment introduces a novel optimization algorithm model: the NSGA II algorithm. NSGA II (Non-dominated sorting genetic algorithm II) is a multi-objective optimization algorithm widely used to solve optimization problems with multiple conflicting objectives. Based on genetic algorithms (GA) and mathematical models, it uses non-dominated sorting, congestion measurement, and an elite strategy to find the Pareto optimal solution set. As an algorithm for solving traffic and route optimization problems, NSGA II can calculate the maximization of emergency response efficiency, the minimization of rescue time, and the minimization of resource consumption in disaster management and emergency response.

[0058] The multi-objective optimization algorithm dynamically optimizes refuge sites and evacuation routes as follows: An initial set of refuge sites and evacuation routes is randomly generated and used as the initial population; offspring populations are generated from the initial population through selection, crossover, and mutation operations; the initial population and offspring populations are merged to generate a combined population; individuals in the combined population are classified according to the Pareto front, and individuals in the same level are sorted according to their crowding level; the levels obtained from the classification are filled into the new population from low to high, and if the number of individuals in the same level exceeds the remaining capacity, individuals with higher crowding levels are retained first; the above steps are repeated until the Pareto optimal solution set with the smallest Pareto front between the two objective functions is obtained.

[0059] In this embodiment, each administrative village in Xiji Town is used as an evacuation point, and multiple designated evacuation exits (traffic nodes) are used as target points. A multi-source, multi-objective evacuation network is constructed based on a GIS spatial analysis model. By analyzing the resident and registered population data of each village in Xiji Town in 2023, population distribution weights are established. This embodiment achieves multi-objective optimization through the NSGA-II algorithm, balancing the two objectives of total evacuation time and the balance of the number of people at evacuation points.

[0060] This embodiment uses the gamultiobj function in MATLAB based on the NSGA-II algorithm to perform multi-objective optimization of evacuation route allocation schemes, aiming to find the Pareto optimal solution set between two objective functions (e.g., total evacuation time and evacuation point load). The optimization problem contains 460 decision variables, all of which are continuous variables and satisfy certain boundary constraints. Regarding algorithm parameter settings, the initial population is uniformly generated using the gacreationuniform method, selection tourrnament is used for tournament selection, crossover intermediate is used for intermediate crossover, and mutation is performed by mutationadaptfeasible to adapt to the constraint boundaries. The maximum number of generations is set to 100, the population size is 100, and the total number of function evaluations reaches 10,000.

[0061] During the optimization process, the algorithm recorded the average distance and distribution breadth of the Pareto front after each iteration. The results show that the average distance rapidly decreased from the initial value of 1, dropping to approximately 0.0034 by the 100th generation, indicating that the non-dominated solution set had converged well to the Pareto front. Simultaneously, the final solution set distribution breadth was 0.1523, demonstrating that overall, the algorithm achieved good convergence under the set parameters, validating the applicability of NSGA-II in multi-objective path optimization problems.

[0062] like Figure 3 As shown, the points on the Pareto front are distributed in different locations, indicating a trade-off between evacuation time and congestion. The points in the upper left represent schemes with shorter evacuation times but higher congestion, while the points in the lower right represent schemes with longer evacuation times but lower congestion. The following is based on... Figure 3 The process yields a Pareto optimal solution set, which includes multiple emergency evacuation plans. Based on the actual parameters, the most suitable emergency evacuation plan for the current situation can be selected.

[0063] like Figure 4 As shown, Figure 4 This is a schematic diagram of the evacuation method with the shortest evacuation time but the highest crowding level in the Pareto optimal solution set. Its goal is to evacuate people in an orderly manner in the shortest possible time. Figure 4 As can be seen, the path connections are more concentrated, and the selection of refuge points tends to be closer to residential areas. By optimizing path length and refuge point allocation, the total evacuation time is reduced. However, this concentration may lead to overload of some refuge points, thereby increasing local congestion. It is suitable for emergency evacuation scenarios with extremely high time requirements, but it will cause congestion problems.

[0064] like Figure 5 As shown, Figure 5 This is a schematic diagram illustrating an evacuation method that balances the number of evacuees at each evacuation point within the Pareto optimal solution set, thereby reducing overall congestion. From Figure 5 As can be seen in the diagram, the green-marked refuge points are distributed relatively evenly, and the emergency evacuation route network is also relatively dispersed. The red-marked residential plots are connected to the refuge points via blue paths, and the paths are evenly distributed, avoiding congestion problems caused by excessive concentration of some refuge points. This evacuation method is suitable for use in scenarios where it is necessary to reduce the risk of congestion during evacuation, especially when the capacity of evacuation points is limited, as it can effectively disperse the flow of people and improve evacuation efficiency.

[0065] A graph model was constructed using information on villages and roads in Xiji Town, and evacuation requirements were initialized based on population data. During the simulated evacuation process, the NSGA-II algorithm effectively generated multiple efficient evacuation routes. Experimental results show that the population distribution at evacuation points is more balanced; the total evacuation time is reduced by approximately 15% compared to the traditional static allocation strategy; and the load at each exit is controlled within 90%, significantly alleviating local congestion.

[0066] Figure 6 This is a comparison chart of the location selection of refuge sites and emergency evacuation routes generated by the GIS spatial analysis model and the NSGA-II optimization algorithm, respectively. There are significant differences in the generation of emergency refuge routes in the northern part of Duliuke Village. During the initial survey, when path length was used as a single variable, villagers in the northern part of Duliuke Village should evacuate south (because the path is shorter). However, when considering multiple variables (speed, flow rate, refuge site capacity, etc.), villagers in the northern part of Duliuke Village should evacuate north to the refuge site in Mafang Village. There are some differences in the road selection generated for emergency refuge routes in Huzhuang Village, Yueshang Village, and Wangshang Village. Some emergency refuge routes in the southwest corner of Huzhuang Village have changed after variable optimization. In villages such as Qiandongyi Village, the optimized emergency refuge routes have stronger connections, more intersections, and greater complexity, resulting in a more systematic overall emergency refuge route network. Jingezhuang Village, Hegezhuang Village, Fenggezhuang Village, Laozhuanghu Village, Xiji Town, Genglou Village, Chenheng Village, Langxi Village, and Langdong Village also saw changes in road selection after optimization.

[0067] Each village contains at least one refuge site. The spatial location of each refuge site meets emergency evacuation requirements, and the number of people it can accommodate largely matches the village's permanent resident population. A small number of village refuge sites are linked to those of neighboring villages, thus sharing the burden of disaster victims from neighboring villages and reducing overcrowding and evacuation time. Therefore, the optimized model results of NSGA-II meet the emergency evacuation needs of Xiji Town.

[0068] S5 configures the emergency evacuation space elements within the target area based on the global optimization results.

[0069] The method for configuring emergency evacuation space elements within the target area based on global optimization results is as follows: For the refuge optimization zone, increase the number of refuges or expand their area; for the evacuation route optimization zone, improve road connectivity and traffic capacity, reduce traffic congestion, and improve evacuation efficiency; for the geological disaster risk prevention and control zone, strengthen the geological disaster monitoring and early warning system, optimize the location of refuges, and avoid selecting locations in high-risk areas of geological disasters; for the general emergency evacuation zone, maintain the existing layout of refuges and evacuation routes, and conduct regular maintenance and inspections to ensure their availability in emergency situations; adjust the service radius by adjusting the area and capacity of refuges according to population distribution and evacuation needs to ensure that residents can reach the nearest refuge within a specified time; consider the evacuation needs of special groups and set up dedicated evacuation channels and refuges (e.g., wheelchair ramps and some facilities for the disabled); set up necessary living facilities and emergency material reserve points inside refuges to improve their emergency support capabilities; and use GIS and multi-objective optimization algorithms to dynamically optimize refuges and evacuation routes.

[0070] Example 2 Based on the same inventive concept, this embodiment discloses a rural emergency evacuation system, including: The data acquisition module is used to acquire distribution data of emergency evacuation elements within the target area through the GIS platform; The emergency evacuation route network generation module is used to select refuge sites and generate an emergency evacuation route network based on the distribution data of emergency evacuation elements and the constraints of construction land. The evacuation element analysis module is used to analyze emergency evacuation elements in the emergency evacuation route network through multiple dimensions of evaluation indicators, and obtain analysis results. The global optimization module is used to perform global optimization of the emergency evacuation route network by combining the analysis results and using the NSGA-II multi-objective optimization algorithm. The results output module is used to configure the emergency evacuation space elements within the target area based on the global optimization results. Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0072] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0073] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific embodiments of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention. The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims

1. A rural emergency evacuation method, characterized in that, Includes the following steps: Obtain distribution data of emergency evacuation elements within the target area through a GIS platform; Based on the distribution data of emergency evacuation elements and combined with the constraints of construction land, refuge sites are selected and an emergency evacuation route network is generated. The emergency evacuation elements in the emergency evacuation route network are analyzed using multiple evaluation indicators to obtain the analysis results. Based on the analysis results, the emergency evacuation route network is globally optimized using the NSGA-II multi-objective optimization algorithm. Based on the global optimization results, the emergency evacuation space elements within the target area are configured.

2. The rural emergency evacuation method as described in claim 1, characterized in that, The method for establishing the emergency evacuation route network is as follows: Calculate the spatial distance between the path network and refuge sites in the GIS map; The average reachability time to each refuge location is calculated based on the spatial distance. Based on the spatial distance and average reach time, all roads in the rural area are modeled, and the location and type of various traffic nodes are identified from the model. Taking into account both topographic features and existing household registration data, the homesteads on the map are divided into several units, and the spatial distribution density and clustering patterns of the homesteads are statistically analyzed. Analyze the matching degree between the route network, traffic nodes and the residential structure of homesteads to identify potential areas of difficulty in evacuation organization and key evacuation nodes; Select areas on the map that can serve as alternative refuge sites; Calculate the service area of ​​all shelters and alternative shelters on the map.

3. The rural emergency evacuation method as described in claim 1, characterized in that, The evaluation indicators for the multiple dimensions are obtained based on the distribution data of various transportation and refuge elements within the target area; the analysis results include land use schemes, village types, geological disaster risks, road network distribution, refuge site distribution, and evacuation planning analysis base layers.

4. The rural emergency evacuation method as described in claim 3, characterized in that, The evaluation indicators across multiple dimensions include: evaluation indicators for evacuation routes, evaluation indicators for refuge sites, evaluation indicators for geological disaster risk, and evaluation indicators for population distribution. The evaluation indicators for evacuation routes are used to evaluate the connectivity and accessibility of evacuation routes; the evaluation indicators for refuge sites are used to evaluate the rationality and capacity of the layout of refuge sites; the evaluation indicators for geological disaster risk are used to evaluate the geological disaster risk in the region; and the evaluation indicators for population distribution are used to evaluate the density of population distribution and evacuation needs.

5. The rural emergency evacuation method as described in claim 4, characterized in that, The evaluation indicators for the evacuation route dimension include road connectivity, evacuation route length, route width, and traffic congestion level; the evaluation indicators for the refuge location dimension include refuge location capacity, refuge location service radius, and refuge location accessibility; the evaluation indicators for the geological disaster risk dimension include the proportion of geological disaster-prone areas, the proportion of low-lying areas, and potential blocking risks; and the evaluation indicators for the population distribution dimension include population density, the proportion of elderly population, and the proportion of child population.

6. The rural emergency evacuation method as described in claim 5, characterized in that, The method for determining the evaluation results for each dimension's evaluation indicators is as follows: The initial values ​​of the evaluation indicators are determined according to the evaluation indicators corresponding to the preset indicator types. The evaluation indicators are divided into positive indicator types and negative indicator types; The initial values ​​of the positive and negative indicator types are normalized to obtain multi-dimensional evaluation indicator values. Obtain the evaluation index value for each dimension within the target area, multiply the evaluation index value by its weight, and then sum all the evaluation indices to obtain the evaluation result for that dimension.

7. The rural emergency evacuation method as described in claim 6, characterized in that, Based on the evaluation results, the emergency evacuation spaces within the target area are classified. If the evaluation results of the evaluation indicators for the refuge area dimension within the target area are higher than the preset refuge area optimization indicator values, then the area is classified as a refuge area optimization zone. If the evaluation results of the evacuation route dimension evaluation index in the target area are higher than the preset evacuation route optimization index value, then the area is divided into the evacuation route optimization zone. If the evaluation results of the geological hazard risk dimension evaluation indicators in the target area are higher than the preset geological hazard risk prevention and control indicator values, then the area will be designated as a geological hazard risk prevention and control zone. If the evaluation results of multiple dimensions of evaluation indicators in the target area are all lower than the preset optimization index values, then the area will be classified as a general emergency evacuation zone.

8. The rural emergency evacuation method as described in claim 7, characterized in that, The method for configuring emergency evacuation space elements within the target area based on the global optimization results is as follows: For the refuge optimization zone, increase the number of refuges or expand their area; for the evacuation route optimization zone, improve road connectivity and traffic capacity, reduce traffic congestion, and improve evacuation efficiency; for the geological disaster risk prevention and control zone, strengthen the geological disaster monitoring and early warning system, optimize the location of refuges, and avoid selecting locations in high-risk areas of geological disasters; for the general emergency evacuation zone, maintain the existing layout of refuges and evacuation routes, and conduct regular maintenance and inspections to ensure their availability in emergency situations; adjust the service radius of refuges according to population distribution and evacuation needs to ensure that residents can reach the nearest refuge within a specified time; consider the evacuation needs of special groups and set up dedicated evacuation channels and refuges; set up necessary living facilities and emergency material reserve points inside refuges to improve their emergency support capabilities; and use GIS and multi-objective optimization algorithms to dynamically optimize refuges and evacuation routes.

9. The rural emergency evacuation method as described in claim 8, characterized in that, The method of using multi-objective optimization algorithms to dynamically optimize refuge sites and evacuation routes is as follows: Randomly generate an initial set of refuge sites and evacuation routes, and use them as the initial population; A progeny population is generated from the initial population through selection, crossover, and mutation operations. The initial population is merged with the offspring population to generate a combined population; Based on the Pareto front, individuals in the composite population are classified into groups, and individuals in the same group are sorted according to their crowding. The levels obtained from the grading are filled into the new population from low to high. If the number of individuals of the same level exceeds the remaining capacity, individuals with greater crowding are retained first. Repeat the above steps until you obtain the Pareto optimal solution set that minimizes the Pareto front between the two objective functions.

10. A rural emergency evacuation system, characterized in that, include: The data acquisition module is used to acquire distribution data of emergency evacuation elements within the target area through the GIS platform; The emergency evacuation route network generation module is used to select refuge sites and generate an emergency evacuation route network based on the distribution data of emergency evacuation elements and the constraints of construction land. The evacuation element analysis module is used to analyze the emergency evacuation elements in the emergency evacuation route network through multiple dimensions of evaluation indicators, and obtain the analysis results. The global optimization module is used to perform global optimization of the emergency evacuation route network by combining the analysis results and using the NSGA-II multi-objective optimization algorithm. The result output module is used to configure the emergency evacuation space elements in the target area based on the global optimization results.

Images