A method for identifying key stations of urban rail transit in waterlogging city and evaluating network resilience

Abstract

The present application relates to the field of rail transit technology, and more particularly to a method for identifying key stations and evaluating network resilience of urban rail transit in waterlogging cities, comprising obtaining urban rail transit line station data, station spatial location data, station surrounding POI data and waterlogging point data; constructing a traffic topology network using the line stations; calculating station topology importance, passenger flow load potential, alternative path scarcity and storm waterlogging exposure; constructing a normal comprehensive criticality evaluation model based on station topology importance, passenger flow load potential and alternative path scarcity; and identifying key stations using the normal comprehensive criticality evaluation values in descending order TOP. The present application solves the problem that existing methods are still insufficient in describing the station disaster risk exposure under extreme weather conditions such as storm waterlogging.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a method for identifying key stations and assessing network resilience in urban rail transit systems in flood-prone cities. Background Technology

[0002] With the continuous expansion of urban rail transit construction, rail transit has become an important component of the urban public transportation system. Leveraging its advantages of large capacity, high punctuality, and high operational efficiency, it plays a vital role in alleviating urban congestion and improving residents' travel efficiency. However, with the continuous increase in the number of lines, transfer stations, and operational scale, the urban rail transit system has gradually evolved into a complex network system with multiple interwoven lines and intricate transfer relationships. If key stations are affected by equipment failures, sudden surges in passenger flow, extreme weather, or other external disturbances, it may lead to a decline in station service capacity, disruption of transfer routes, increased detour costs for passengers, and even reduced network connectivity and overall transportation efficiency degradation. Therefore, identifying key stations in the urban rail transit network and analyzing the impact of their failures on network resilience is of significant practical importance.

[0003] In recent years, although existing research has provided an important foundation for the identification of key stations in urban rail transit, certain shortcomings still exist. First, existing studies mostly evaluate station importance from a topological perspective, with relatively insufficient consideration of the functional clustering around the station, potential passenger load, and alternative travel conditions after station failure. Second, related studies mostly focus on normal operation or general disturbance scenarios, and the characterization of station disaster risk exposure under extreme weather conditions such as rainstorms and flooding is still insufficient. In particular, in the absence of real AFC passenger flow data, how to construct a reasonable passenger load substitution index and combine it with network structure, alternative routes, and disaster exposure information remains a problem that needs to be further addressed in rail transit resilience assessment. Summary of the Invention

[0004] To address the shortcomings of existing methods, this invention constructs a comprehensive evaluation model that integrates the importance of station topology, passenger flow load potential, scarcity of alternative routes, and exposure to rainstorms and urban flooding. It introduces an impact coefficient for exposure to rainstorms and urban flooding to identify key stations under rainstorm and urban flooding scenarios. Furthermore, it combines station failure attack simulation to evaluate the resilience degradation characteristics of urban rail transit networks, thereby providing technical support for risk prevention, graded defense, and resilience enhancement of rail transit systems under extreme rainfall conditions.

[0005] The technical solution adopted in this invention is: a method for identifying key stations and assessing network resilience in urban rail transit systems prone to flooding, comprising the following steps: Step 1: Obtain urban rail transit line station data, station spatial location data, station surrounding POI data, and waterlogging point data; Step 2: Construct a traffic topology network using route stations; As a preferred embodiment of the present invention, a traffic topology network is constructed using the Space-L method.

[0006] Step 3: Calculate the station topological importance, passenger flow load potential, alternative route scarcity, and exposure to rainstorm flooding; In a preferred embodiment of the present invention, the site topological importance is a weighted sum of the site's degree centrality, betweenness centrality, proximity centrality, PageRank centrality, and failure loss efficiency.

[0007] In a preferred embodiment of the present invention, the passenger flow load potential is a weighted sum of the POI weighting strength, the number of lines, and the PageRank centrality of the station.

[0008] In a preferred embodiment of the present invention, the scarcity of alternative routes is a weighted sum of the number of alternative stations, the number of alternative routes, and the impact of detours.

[0009] In a preferred embodiment of the present invention, the rainstorm waterlogging exposure is a weighted sum of the distance exposure of the station, the number of waterlogged points within the first range, and the number of waterlogged points within the second range.

[0010] In a preferred embodiment of the present invention, the formula for the distance exposure term is: , ; in, For the site Distance to the nearest flooded area; For the site Waterlogged areas Spatial distance between them; For distance attenuation parameters, This is a collection of areas prone to waterlogging and flooding. For set Any waterlogged or flooded area in the area.

[0011] Step 4: Construct a routine comprehensive criticality evaluation model based on the importance of site topology, passenger flow load potential, and scarcity of alternative routes; use the routine comprehensive criticality evaluation values ​​in descending order to TOP to identify key sites; As a preferred embodiment of the present invention, the formula of the normal comprehensive criticality evaluation model is:

[0012] In the formula, , , These are the weighting coefficients.

[0013] As a preferred embodiment of the present invention, a comprehensive criticality assessment model for rainstorm scenarios is constructed using a normal comprehensive criticality assessment model and a rainstorm flood exposure influence coefficient. , This represents the impact coefficient of rainstorm exposure.

[0014] As a preferred embodiment of the present invention, a site failure attack strategy is designed, site failure simulation is performed, and the resilience of the normal comprehensive criticality evaluation model and / or the rainstorm scenario comprehensive criticality evaluation model is assessed by using network efficiency retention rate, maximum connected subgraph ratio, reachable OD ratio and potential passenger flow service loss.

[0015] The beneficial effects of this invention are: 1. Construct a comprehensive evaluation model that integrates the importance of station topology, passenger flow load potential, scarcity of alternative routes, and exposure to rainstorms and urban flooding, to overcome the problems of existing methods for identifying key stations in urban rail transit that mainly rely on topological structure indicators and do not adequately consider station service functions and disaster exposure factors; 2. By introducing the impact coefficient of rainstorm and urban flooding exposure, a comprehensive criticality evaluation model for rainstorm scenarios is constructed. This model can simultaneously identify structurally critical stations under normal conditions and disaster-exposed stations under rainstorm and urban flooding scenarios, making the identification results of critical stations more consistent with the actual risk characteristics under extreme rainfall scenarios. 3. In the absence of real AFC passenger flow data, this invention uses the weighted strength of POIs around the station, the number of transfer lines, and PageRank centrality to characterize passenger flow load potential, which can improve the applicability of the method under data-limited conditions. 4. This invention uses four types of site failure strategies—random attack, topological importance attack, normal comprehensive critical attack, and rainstorm scenario critical attack—to simulate and evaluate the resilience of urban rail transit networks. It can comprehensively reveal the characteristics of network resilience degradation from aspects such as network efficiency, connectivity structure, reachability OD relationship, and potential passenger flow service capacity. 5. This invention can classify and defend urban rail transit stations based on the comprehensive criticality ranking results of rainstorm scenarios, providing a basis for prioritizing the configuration of protective measures such as station entrances and exits, ventilation shafts, underground passages, transfer passages, equipment rooms, drainage facilities, backup power supplies and emergency connections, which helps to improve the safety and resilience of urban rail transit systems under rainstorm and flooding scenarios. Attached Figure Description

[0016] Figure 1 This invention illustrates the framework for evaluating and resiliently analyzing key urban rail transit stations under heavy rain scenarios. Figure 2 A flowchart illustrating the site failure simulation of the present invention is shown; Figure 3A schematic diagram of the Shenzhen rail transit network topology is shown in an embodiment of the present invention; Figure 4 This invention illustrates a node degree distribution map of the Shenzhen rail transit network in an embodiment of the present invention. Figure 5 a, b, c, and d are schematic diagrams of random attack, topological importance attack, normal comprehensive critical attack, and critical attack in rainstorm scenario, respectively. Figure 6 The diagram illustrates the changes in the resilience of urban rail transit networks under different attack strategies in an embodiment of the present invention. Figure 7 The diagram shows the variation of urban rail transit network resilience under different rainstorm flooding exposure coefficients in an embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments. The drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0018] like Figure 1 As shown, a method for identifying key stations and assessing network resilience in urban rail transit systems prone to flooding includes the following steps: This includes data acquisition, network construction, indicator calculation, criticality evaluation, critical site identification, and network resilience assessment. Step 1: Obtain urban rail transit line station data, station spatial location data, POI data around the stations, and data on waterlogging and flooding points; In this embodiment, as Figure 3 Taking the Shenzhen urban rail transit network as the research object, we first organize the data of urban rail transit lines and stations. The line and station data includes line name, station name, station order, transfer relationship and connection relationship of adjacent stations. Through the line and station data, the basic relationship of "line-station order-station" can be obtained, which can be used to construct the urban rail transit topology network in the future.

[0019] Further, station spatial location data will be obtained, including the latitude and longitude coordinates of rail transit stations, to characterize the spatial distribution of stations and to be used for subsequent calculation of waterlogging exposure and analysis of alternative station spatial relationships.

[0020] Further acquisition of POI data around the station is used to characterize the degree of functional agglomeration and potential passenger flow attraction around the station; the POI data around the station includes POI data of residential, commercial, office, transportation connection, public service and leisure services.

[0021] Further data on waterlogging points will be obtained, including the spatial coordinates, distribution location, and spatial proximity of waterlogging points to rail transit stations. Waterlogging points serve as sources of rainstorm hazard and are used to characterize the disaster exposure level of rail transit stations under rainstorm and waterlogging scenarios.

[0022] Step 2: Based on the connection relationship between urban rail transit lines and stations, construct the urban rail transit topology network using the Space-L method, abstracting stations as nodes and the interval connection between adjacent stations on the same line as edges. In this embodiment, the Space-L method is used to construct the urban rail transit network. The Space-L method abstracts stations as nodes and the connection relationship between adjacent stations on the same line as edges, which can better reflect the physical adjacency relationship of the rail transit network.

[0023] Specifically, firstly, stations on the same line are sorted according to the station order of different rail transit lines; then, an edge is established between two adjacent stations on the same line; for transfer stations, stations with the same name on different lines are uniformly coded and deduplicated to ensure that the same transfer station corresponds to only one node in the network; finally, the stations and edges of multiple lines are merged to form an undirected topology network of urban rail transit.

[0024] Assume the Shenzhen Metro Space-L network is as follows:

[0025] In the formula, For a set of nodes, The number of nodes in the network; Let it be the set of edges.

[0026] If node and If two objects are adjacent on the same line, then there is an edge between them. The corresponding adjacency matrix Defined as:

[0027] Since urban rail transit lines typically operate in both directions, this embodiment abstracts the Shenzhen Metro network as an undirected network; if the spatial distance between adjacent stations is further considered, the edge weights can be expressed as:

[0028] In the formula, For the site and Edge weights between them This represents the spatial distance between the two stations.

[0029] In this embodiment, the processed Shenzhen Metro network consists of 343 nodes and 400 undirected edges, exhibiting a clear chain-like structure. Further calculations of the network's basic topology indicators are shown in Table 1.

[0030] Table 1. Topological characteristics of Shenzhen's rail transit network

[0031] Step 3: Calculate the station topological importance, passenger flow load potential, alternative route scarcity, and exposure to rainstorm flooding; In this embodiment, to comprehensively characterize the differences between sites in terms of network structure, service functions, alternative conditions, and disaster exposure, a site topological importance is constructed. Passenger flow capacity potential Scarcity of alternative paths and exposure to rainstorms and flooding Four categories of evaluation indicators.

[0032] First, calculate the topological importance of the site. The topological importance of a station is used to characterize its connectivity and overall transportation impact within the urban rail transit network structure. Since a single topological index cannot fully reflect the structural function of a station, this embodiment selects degree centrality. Betweenness centrality Proximity centrality PageRank centrality and node failure efficiency loss As an evaluation indicator; like Figure 4 Node degree Indicates the relationship with the site The number of directly connected neighboring sites is calculated using the following formula:

[0033] Degree centrality The formula used to eliminate the influence of network size on node degree comparison is:

[0034] Betweenness centrality The formula used to measure the bridging effect of a station on the shortest path is as follows:

[0035] In the formula, Represents a node arrive The total number of shortest paths; This indicates the nodes it passes through. The number of shortest paths.

[0036] Proximity centrality It is used to measure the average distance from a node to other nodes in the network, reflecting the global reachability of a site. The calculation formula is as follows:

[0037] In the formula, For nodes arrive The shortest path length.

[0038] PageRank centrality The formula used to assess a site's overall influence on network connectivity is as follows:

[0039] In the formula, The damping coefficient; Represents a node The proportion of influence is distributed equally among each of its neighboring nodes.

[0040] Efficiency loss Network efficiency is used to measure the impact of a site failure on the overall network transportation efficiency, and is defined as:

[0041] When two sites are unreachable, Let the efficiency of the original network be... Delete node The network efficiency after that is Then the node The failure loss efficiency is:

[0042] Dimensionless processing is applied to each indicator to determine the importance of site topology. Represented as:

[0043] In the formula, , , , and These are the standardized indicators; The weights of each indicator are, and .

[0044] Calculate passenger flow load potential Because real AFC passenger flow data is often difficult to obtain, involves privacy concerns, has inconsistent time granularity, and varies in the level of data development across different cities, this embodiment uses a weighted intensity of POIs surrounding the station. Number of stations passed and PageRank centrality It approximates the station's potential passenger flow attraction capacity and transfer capacity pressure.

[0045] Among them, the weighted intensity of POIs around the site Used to characterize the intensity of urban functional activities such as commerce, residence, office, public services, and transportation hubs within the service area, addressing the problem that relying solely on network topology indicators is insufficient to reflect the ability to attract passenger flow around a station; number of lines passing through the station. It is used to characterize the station's ability to handle cross-line transfers and passenger flow conversions, and to solve the problem that it is difficult to reflect the difference in passenger flow carrying capacity between ordinary stations and transfer stations; PageRank centrality It is used to characterize the global distribution potential of a station in the rail transit network and solves the problem that simply using the number of local connections is insufficient to reflect the network propagation influence of a station.

[0046] Site The POI weighted strength is:

[0047] In the formula, For the site Surrounding area Number of POIs; For the first Customer attraction weight corresponding to POI type; Number of POI types.

[0048] Site The number of routes traversed is:

[0049] In the formula, This represents the total number of lines in the urban rail transit network. To determine the number of lines to belong to; when the station Belongs to the When there are multiple lines, When the site Not belonging to the first When there are multiple lines, ;therefore, Indicates the station visited The number of rail transit lines; for ordinary stations, For transfer stations between two lines, For transfer stations of three lines or above, The value is determined based on the actual number of routes it passes through.

[0050] To eliminate differences in the dimensions of different indicators, the weighted strength of POIs, the number of traversed paths, and PageRank centrality were standardized respectively, resulting in... , and The standardized form of the number of passing lines is as follows:

[0051] In the formula, This refers to the standardized number of routes traversed. and These are the minimum and maximum values ​​for the number of times all stations have been visited, respectively.

[0052] Further insights into the station's passenger flow capacity potential:

[0053] In the formula, , and These are the standardized POI weighted strength, number of traversed lines, and PageRank, respectively. .

[0054] Through the above design, even in the absence of AFC passenger flow data, the potential passenger flow load level of a station can be estimated by comprehensively utilizing the intensity of urban functional activities around the station, transfer carrying capacity pressure, and network distribution potential, thereby improving the applicability of the key station identification method in different urban rail transit networks.

[0055] Calculate the scarcity of alternative paths , Used to characterize sites After a station fails, the difficulty for passengers to complete alternative travel through nearby stations or other rail transit lines is considered. Existing key station identification methods mostly focus on the station's topological importance in the network, but do not adequately consider the conditions for alternative routes to emerge after a station fails. In reality, when there are many alternative stations and routes around a station, passengers can transfer to other stations by walking, bus connections, or transfers, and the impact of station failure is relatively small. When there are few alternative stations and routes around a station, station failure is more likely to lead to local rail transit service interruptions, increased detour distances for passengers, and decreased network accessibility. Therefore, this embodiment constructs an alternative route scarcity index from three aspects: the number of alternative stations, the number of alternative routes, and the impact of detours.

[0056] Let the set of rail transit stations be... Site The set of nearby alternative sites is Its definition is:

[0057] In the formula, To remove the site Other rail transit stations besides those mentioned above; For the site With the site Spatial distance between them; The search radius for nearby alternative sites; Indicates location at the site A set of nearby stations within a certain spatial range that passengers can transfer to.

[0058] Site Number of alternative sites Represented as:

[0059] In the formula, Indicates site Number of nearby alternative sites; Represents a set The number of elements in the middle. The larger the size, the better for the site. The more rail transit stations nearby that passengers can transfer to, the better the alternative conditions.

[0060] Site Number of alternative lines Represented as:

[0061] In the formula, Indicates site Number of alternative routes in the surrounding area; Indicates nearby alternative sites The collection of rail transit lines to which it belongs; Indicates site The collection of rail transit lines covered by all nearby alternative stations; The larger the size, the better for the site. The more alternative routes available in the surrounding area, the better the alternative travel conditions.

[0062] Site Detour impact Used to characterize deleted sites Then, the degree of shortest path growth or accessibility decrease for each OD pair in the rail transit network; assuming the original rail transit network is... Delete site The network following the edges connected to it is For any starting station and the final stop Composition of OD sites Let it be in the original network The shortest path length in is In deleting the site The shortest path length in the post-network is ;in, and All of them are station nodes in the rail transit network, and ,at the same time and Excluding deleted sites .

[0063] Not including sites OD pair set Represented as:

[0064] Site The impact of the detour can be expressed as:

[0065] In the formula, Not including sites The set of OD pairs; For the number of OD pairs; To delete the site Post-OD The detour impact coefficient.

[0066] in, It can be represented as:

[0067] In the formula, when OD is paired with Deleting the site When connectivity is maintained, the impact of detours is represented by the relative growth rate of the shortest path; when OD to Deleting the site If the connection is no longer established, it means that the OD pair cannot complete the trip through the rail transit network, and its detour impact coefficient is set to 1.

[0068] Using a minimum value function for truncation can prevent individual ODs from growing too much and having an excessive impact on the overall result.

[0069] To eliminate the differences in the units of measurement of different indicators, the number of alternative sites was adjusted. Number of alternative lines and detour impact Dimensionless processing was performed separately to obtain , and ;in,

[0070] In the formula, and These represent the minimum and maximum values ​​for the number of sites that can be replaced by all sites, respectively. and These are the minimum and maximum values ​​for the number of alternative lines for all stations, respectively. and These represent the minimum and maximum values ​​of the impact of detours on all stations, respectively.

[0071] in, and The larger the value, the better the site. The more alternative stations and routes available in the vicinity, the better the alternative conditions, and the lower the scarcity of alternative routes should be; The larger the value, the better the site. The greater the impact of detours caused by failure, the higher the scarcity of alternative routes should be.

[0072] Therefore, the site Scarcity of alternative paths Represented as:

[0073] In the formula, Indicates site The scarcity of alternative paths; This represents the dimensionless number of replacement sites. This represents the number of alternative routes after dimensionless conversion. The effect of detour after dimensionless transformation; , and These are the weighting coefficients for the scarcity of alternative stations, the scarcity of alternative routes, and the impact of detours, respectively, and they satisfy the following conditions: .

[0074] From the above formula, it can be seen that when the site The fewer the number of alternative stations and alternative routes in the vicinity, and the greater the detour impact caused by deleting the station, the worse the alternative travel conditions after the station fails, and the higher its vulnerability and criticality. This indicator can solve the problem that relying solely on topological centrality is insufficient to reflect the differences in alternative travel conditions after a station fails.

[0075] Calculate the exposure degree of rainstorm and urban flooding , Used to depict the site The spatial proximity between a site and a waterlogged area and the degree of clustering of surrounding waterlogged areas characterize the likelihood of a site being affected by waterlogging under a rainstorm and waterlogging scenario. Existing methods usually only use the distance from the site to the nearest waterlogged area or only count the number of waterlogged areas around the site, which makes it difficult to simultaneously reflect the two spatial characteristics of higher risk the closer the distance and higher risk the more concentrated the surrounding waterlogged areas. Therefore, this embodiment constructs a site rainstorm and waterlogging exposure index from two aspects: the attenuation effect of the distance to the nearest waterlogged area and the degree of clustering of neighboring waterlogged areas.

[0076] Let the set of waterlogged points be . Site Waterlogged areas The spatial distance between them is Then the site Distance to the nearest flooded area , is represented as:

[0077] In the formula, This is a collection of areas prone to waterlogging and flooding. For set Any waterlogged or flooded area in the area; For the site Waterlogged areas The spatial distance between them; Site Distance to the nearest flooded area; The smaller the value, the closer the station is to areas prone to flooding and waterlogging, and the higher the likelihood of being affected by heavy rain and flooding.

[0078] To convert distance into rainfall intensity, a distance attenuation coefficient is introduced to obtain the station data. Distance Exposure Item :

[0079] In the formula, This is the distance attenuation parameter.

[0080] As can be seen from this formula, the closer the station is to the waterlogged area, the better. The closer the value is to 1, the higher the level of exposure; the farther the station is from the waterlogged area, the higher the level of exposure. The closer to 0, the lower the level of distance exposure.

[0081] Further statistics on sites The number of waterlogging points within a certain spatial range is used to characterize the spatial concentration of waterlogging risk around a station; Number of waterlogging points within 800m Represented as:

[0082] Site Number of waterlogging points within 1000m Represented as:

[0083] In the formula, This indicates the number of elements in the set; the number of water accumulation points within 800m is used to characterize the direct water accumulation within the vicinity of the station, and the number of water accumulation points within 1000m is used to characterize the water accumulation within the extended influence range around the station; if the number of water accumulation points in both ranges is relatively large, it indicates that there is a significant water accumulation risk clustering characteristic around the station.

[0084] To eliminate differences in the units of measurement of different indicators, the distance exposure item was... Number of waterlogged areas within 800m and the number of waterlogged areas within a 1000m radius After performing dimensionless processing, we obtain , and ;in, The larger the value, the closer the station is to the waterlogged area and the higher the exposure level. and The larger the value, the more waterlogging points there are around the station, and the higher the degree of waterlogging risk.

[0085] Therefore, the site Exposure to rainstorms and urban flooding Represented as:

[0086] In the formula, , and These are the weighting coefficients for the distance exposure term, the number of water accumulation points in the nearest range, and the number of water accumulation points in the extended range, respectively, and they satisfy the following conditions: .

[0087] From the above formula, it can be seen that when the site The closer to the waterlogged area, and the greater the number of waterlogged areas within 800m and 1000m, the better. The larger the value, the higher the spatial exposure level of the site under rainstorm and waterlogging scenarios. This indicator can solve the problem of one-sided evaluation results when using only the distance to the nearest waterlogging point or the number of waterlogging points in a single neighborhood, so that the site's rainstorm and waterlogging exposure can simultaneously reflect the proximity and the clustering of waterlogging points.

[0088] Step 4: Construct a comprehensive criticality evaluation model for normal conditions based on the importance of station topology, passenger flow load potential, and scarcity of alternative routes. On this basis, introduce the impact coefficient of rainstorm and urban flooding exposure to construct a comprehensive criticality evaluation model for rainstorm scenarios. Under normal conditions, the overall criticality of a station is jointly determined by its topological importance, passenger flow potential, and the scarcity of alternative routes. The evaluation model for overall criticality under normal conditions is as follows:

[0089] In the formula, As a normal, comprehensive, and critical process, , , Let be the weight coefficient, and satisfy... .

[0090] Under the scenario of rainstorm and urban flooding, exposure to waterlogging amplifies the risk of station failure. This embodiment introduces rainstorm and urban flooding exposure into the normal comprehensive criticality assessment model, constructing a comprehensive criticality evaluation model for rainstorm scenarios:

[0091] In the formula, For the site The comprehensive key aspects of the rainstorm scenario; This represents the influence coefficient of rainstorm exposure. When... At that time, the model degenerates into a normal comprehensive criticality evaluation model.

[0092] Step 5: Sort the rail transit stations according to the comprehensive criticality score of the rainstorm scenario and identify the key stations under the rainstorm and flooding scenario; In this embodiment, based on the comprehensive key aspects of the rainstorm scenario The urban rail transit stations were ranked from highest to lowest. The higher the ranking of the stations, the greater their impact on network connectivity, service functions, and disaster risk under rainstorm and flooding scenarios, and they should be identified as key targets.

[0093] Based on the comprehensive key aspects of the rainstorm scenario The key stations for the Shenzhen Metro during heavy rain scenarios were identified by sorting them from largest to smallest. The top 20 stations are shown in Table 2.

[0094] Table 2. Top 20 stations ranked by comprehensive criticality in rainstorm scenarios.

[0095] As shown in Table 2, the stations with the highest rankings are Buji, Dayun, Chegongmiao, Gushu, Wuhe, Shenzhen North, Lingzhi, Gangxia North, Xili, and Shangwu. Among them, some stations are multi-line transfer stations and network hub stations, which have high topological importance and passenger flow potential. Although some stations may not be located in the core of the network topology, they also show high scenario criticality due to their high exposure to rainstorms and flooding.

[0096] This invention, by coupling normal comprehensive criticality with rainstorm and urban flooding exposure, can simultaneously identify structurally critical stations and disaster-exposed stations. In rainstorm and urban flooding scenarios, critical stations are not entirely equivalent to transfer hubs or highly central stations under normal conditions, and some stations with high exposure to urban flooding should also be included in the key protection scope.

[0097] Based on the above key site identification results, sites can be classified into different levels of defense according to the comprehensive criticality ranking results of rainstorm scenarios. The top 5% of sites are classified as Level 1 defense sites, the 5% to 15% of sites are classified as Level 2 defense sites, the 15% to 30% of sites are classified as Level 3 defense sites, and the remaining sites are classified as sites of general concern. Level 1 defense sites should focus on areas that are susceptible to backflow of water, such as site entrances and exits, ventilation shafts, underground passages, transfer passages, and equipment rooms. Level 2 defense sites should strengthen the inspection of drainage facilities, the deployment of waterproof materials, and the preparation for passenger flow management. Level 3 defense sites can carry out enhanced monitoring and patrols during key periods in conjunction with the rainstorm warning level.

[0098] Step 6: Set up a site failure attack strategy, simulate site failures in the urban rail transit network, and calculate the network efficiency retention rate, the proportion of the largest connected subgraph, the reachable OD proportion, and the potential passenger flow service loss to evaluate the resilience degradation characteristics of the urban rail transit network under the scenario of rainstorm and flooding.

[0099] In this embodiment, to verify the effectiveness of the integrated critical model for rainstorm scenarios, based on the urban rail transit topology network constructed in step two, different station failure strategies are set, and station failure simulation is performed on the urban rail transit network. The station failure simulation process is as follows: Figure 2 As shown.

[0100] like Figure 2As shown, the site failure simulation process includes the following steps: First, import the rail transit network data and set the attack strategy and simulation parameters according to the research needs; then initialize the network state and resilience evaluation parameters; next, determine the failed sites in the current round according to the attack strategy, delete the target sites and their associated edges, and update the remaining network topology; then calculate resilience indicators such as network efficiency retention rate, maximum connected subgraph ratio, reachable OD ratio, and potential passenger flow service loss; finally, determine whether the set failure ratio or network function degradation termination condition has been met; if the termination condition has not been met, continue to select the next round of failed sites according to the attack strategy and repeat the above process; if the termination condition has been met, output the network resilience evaluation results under each attack strategy.

[0101] To further illustrate the meaning of different attack strategies, this embodiment provides a schematic diagram of urban rail transit network attack patterns, such as... Figure 5 As shown.

[0102] Figure 5 'a' represents a random attack, which means that the site is randomly selected as the failure node without considering the site's topological location, functional attributes and disaster exposure level in the network, in order to simulate non-directed failure scenarios such as equipment failure and occasional disturbances. Figure 5 The 'b' represents a topology importance attack, which involves deleting stations in descending order of their topology importance scores. This is used to simulate the impact of the failure of core transfer stations, bridging stations, or network hub stations on the connectivity of the rail transit network. Figure 5 (c) indicates a normal comprehensive critical attack, which means that sites are deleted in descending order of their normal comprehensive criticality scores. This is used to assess the impact of the failure of comprehensive critical sites under normal conditions on the overall network operation capability. Figure 5 The 'd' represents a critical attack under the rainstorm scenario, which means deleting sites in descending order of their comprehensive criticality score under the rainstorm scenario. This is used to simulate the site failure process under the combined effects of rainstorm flooding disaster exposure and network structural vulnerability.

[0103] In the site failure simulation process, each round determines the site to be deleted according to the corresponding attack strategy, and simultaneously deletes the site and its connected edges to form a new remaining network; assuming a certain attack strategy is used to delete... The remaining network after each station is The original network is After each round of station deletion, the resilience indicators of the remaining network are recalculated, including network efficiency retention rate, maximum connected subgraph ratio, reachable OD ratio, and potential passenger flow service loss, in order to characterize the resilience degradation features of the urban rail transit network during the continuous failure of stations.

[0104] The network efficiency retention rate is used to characterize the degree to which the overall operational efficiency of the rail transit network is maintained after station failures. Its calculation formula is as follows:

[0105] In the formula, For the original network efficiency; To remove in a certain attack strategy Remaining network efficiency after each site; The smaller the value, the more significant the decrease in overall network efficiency after a site fails.

[0106] The maximum connected subgraph ratio is used to measure the degree to which the network's main connectivity structure is preserved after a site failure. The calculation formula is as follows:

[0107] In the formula, For the first The maximum number of nodes in a connected subgraph after a station fails; This represents the original total number of network nodes; The smaller the value, the lower the degree to which the network's main connectivity structure is maintained, and the higher the degree of network fragmentation.

[0108] The reachability-to-OD ratio measures the proportion of all site pairs that remain connected after a site failure, assuming the network is broken down into [number] sites after the failure. The connected subgraph, the th ... A connected subgraph contains The nth node, then the nth The reachable OD ratio after a round attack is defined as:

[0109] In the formula, This represents the number of all ordered OD site pairs in the original network. This represents the number of ordered OD (Original Demand) site pairs that remain within the same connected subgraph and are reachable from each other after a failure. The smaller the value, the smaller the range of stations a passenger can complete their journey after a station fails, and the more severe the network accessibility degradation.

[0110] Potential passenger flow service loss is used to characterize the degree of loss of network potential passenger flow service capacity after a site failure; since it is difficult to obtain actual AFC passenger flow, this embodiment uses passenger flow load potential to characterize the site's potential service capacity; let the site... The potential passenger flow capacity is The total potential load of the original network is After a site fails, the potential load retained in the maximum connected subgraph is: Then the first The potential loss of customer traffic after a round attack is defined as:

[0111] In the formula, For the first The set of nodes in the largest connected subgraph after a station fails. The larger the value, the more severe the potential loss of network passenger flow service capacity after failure.

[0112] During the simulation, target sites were gradually removed according to a set failure rate, and the aforementioned network resilience indicators were recalculated at each step. Considering that targeted attacks typically cause significant damage to the network at a lower failure rate, this embodiment focuses on analyzing the network resilience changes within the first 30% of site failures. The results are as follows: Figure 6 As shown.

[0113] Depend on Figure 6 It can be seen that within the range of the first 30% of stations failing, the impact of random attacks on the Shenzhen Metro network is relatively limited. As the proportion of failed stations increases, the network efficiency retention rate, the proportion of the largest connected subgraph, and the proportion of reachable OD gradually decrease, while the potential loss of passenger flow services gradually increases. However, the overall change is relatively gradual, indicating that the Shenzhen Metro network has a certain degree of robustness to random station failures.

[0114] In comparison, topology importance attacks, normal comprehensive criticality attacks, and rainstorm scenario criticality attacks all cause more significant network performance degradation. After a critical site is removed, the shortest paths between some site pairs are forced to lengthen or even be interrupted, leading to a decrease in overall network transmission efficiency. The decrease in the proportion of the largest connected subgraph indicates that critical site failures accelerate the network fragmentation process, causing the originally connected rail transit network to gradually split into multiple smaller connected subgraphs. The decrease in the reachable OD ratio indicates that the range of site pairs that passengers can complete their journeys is significantly reduced. The rapid increase in potential passenger flow service loss indicates that critical site failures not only affect the network structure and topology connectivity but also weaken the potential service capacity of the rail transit system.

[0115] Further comparison of the three types of targeted attacks reveals that critical attacks in rainstorm scenarios have the most significant destructive effects, with network efficiency retention rate, the proportion of the largest connected subgraph, and the proportion of reachable OD decreasing more rapidly, and potential passenger flow service losses increasing more significantly. This indicates that incorporating rainstorm and urban flooding exposure into the critical site evaluation model can more effectively identify sites that have a greater impact on network resilience under extreme rainfall conditions.

[0116] To further quantitatively compare the destructive extent of different attack strategies, this embodiment selects a 10% site failure scenario for analysis; at this time, the actual number of deleted sites is 34, and the network resilience index comparison under different attack strategies is shown in Table 3.

[0117] Table 3 Comparison of network resilience indicators under different attack strategies when the top 10% of sites fail.

[0118] As shown in Table 3, when the top 10% of sites are deleted, the network efficiency retention rate under random attacks is 0.5477, the network efficiency loss is 0.4523, the maximum connected subgraph ratio is 0.7094, the reachable OD ratio is 0.5139, and the potential passenger flow service loss is 0.2564, indicating that the impact of random failures on the overall network performance is relatively limited.

[0119] In comparison, all three types of targeted attacks caused more damage to the network; among them, the critical attacks in the rainstorm scenario caused the most significant damage; under this strategy, the network efficiency retention rate dropped to 0.1253, the network efficiency loss reached 0.8747, the maximum connected subgraph ratio dropped to 0.1370, the reachable OD ratio was only 0.0369, and the potential passenger flow service loss reached 0.8299, indicating that the failure of critical sites in the rainstorm scenario would seriously weaken network connectivity and service capabilities.

[0120] Further comparisons show that topology importance attacks, normal comprehensive criticality attacks, and rainstorm scenario criticality attacks all have stronger destructive effects than random attacks, with rainstorm scenario criticality attacks being the most prominent. This indicates that by introducing rainstorm flood exposure, the constructed rainstorm scenario comprehensive criticality model can more effectively identify sites that have a greater impact on network resilience under disaster scenarios. Therefore, subsequent defense and resilience enhancement should prioritize risk prevention and emergency support for critical sites under rainstorm scenarios.

[0121] To test the impact coefficient of rainstorm and urban flooding exposure This embodiment further sets different impacts on the key site identification results. Sensitivity analysis was performed on the selected values. As a baseline scenario, other values ​​are used to analyze the changes in the ranking of key sites and network resilience loss results after the weighting of rainstorm and urban flooding exposure changes. The overlap between the Top 20 key sites and the baseline scenario is shown in Table 4.

[0122] Table 4 Differences Top 20 Key Sites Overlap

[0123] As shown in Table 4, when At that time, 18 of the Top 20 key sites were consistent with the baseline scenario, with a Jaccard overlap of 0.818; At that time, the number of overlapping stations was 16, and the Jaccard overlap was 0.667; this indicates that as the rainstorm exposure influence coefficient increased, the ranking of some stations with higher waterlogging exposure increased, but the core key stations still maintained a high degree of consistency, indicating that the model identification results have a certain degree of stability.

[0124] For further analysis The impact of different values ​​on network resilience assessment results is discussed in this example, which selects a 10% site failure scenario to compare different... The network resilience loss caused by critical attacks during heavy rain scenarios is shown in Table 5.

[0125] Table 5. Different situations when 10% of sites are unavailable Comparison of network resilience indicators

[0126] As shown in Table 5, different Under the given values, network efficiency loss exceeds 85% when 10% of stations fail, indicating that critical attacks during heavy rain scenarios can cause significant damage to the rail transit network; among them, when At that time, the proportion of the largest connected subgraph decreased to 0.0845, and the potential loss of passenger flow services reached 0.9044. This indicates that after appropriately increasing the impact coefficient of rainstorm and waterlogging exposure, the model is more likely to identify sites that are both structurally critical and exposed to waterlogging, and the network damage effect is more obvious.

[0127] To further showcase different The network resilience index changes under various values. This embodiment plots the resilience change curves under attacks on critical sites during a rainstorm scenario, as shown below. Figure 7 As shown in the figure; considering that critical site attacks usually cause significant damage during the low-percentage site failure phase, the figure focuses on showing the changes within the first 30% of site failure range.

[0128] Depend on Figure 7 It can be seen that the differences Under the given values, critical attacks during rainstorm scenarios all lead to a rapid decline in the resilience of the Shenzhen Metro network. Among them, the network efficiency retention rate, the proportion of the largest connected subgraph, and the reachable OD proportion decrease significantly at lower failure rates, while the potential passenger flow service loss increases rapidly, indicating that critical stations under rainstorm scenarios have a significant impact on network connectivity and service functions. The overall trend of the three curves is relatively consistent, indicating that the model has a certain degree of stability under different parameter values.

[0129] From the perspective of differences, a moderate increase Afterwards, the model is more likely to identify sites that simultaneously possess structural criticality and waterlogging exposure, making the network disruption effect more pronounced; however, when As the attack continues to increase, some sites with high local exposure but weak global control may be selected into the attack sequence in advance, resulting in a slight decrease in overall network efficiency.

[0130] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for identifying key stations and assessing network resilience in urban rail transit systems prone to flooding, characterized in that, Includes the following steps: Step 1: Obtain urban rail transit line station data, station spatial location data, station surrounding POI data, and waterlogging point data; Step 2: Construct a traffic topology network using route stations; Step 3: Calculate the station topological importance, passenger flow load potential, alternative route scarcity, and exposure to rainstorm flooding; Step 4: Construct a normalized comprehensive criticality evaluation model based on the importance of station topology, passenger flow load potential, and scarcity of alternative routes; Key sites are identified by ranking them in descending order of their routine comprehensive criticality evaluation values.

2. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, The exposure degree to urban flooding caused by rainstorms is a weighted sum of the distance exposure of the station, the number of waterlogged points within the first range, and the number of waterlogged points within the second range.

3. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 2, characterized in that, The formula for the distance exposure term is: ， ； in, For the site Distance to the nearest flooded area; For the site Waterlogged areas Spatial distance between them; For distance attenuation parameters, This is a collection of areas prone to waterlogging and flooding. For set Any waterlogged or flooded area in the area.

4. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, Site topological importance It is a weighted sum of the site's degree centrality, betweenness centrality, proximity centrality, PageRank centrality, and failure loss efficiency.

5. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, Passenger load potential It is a weighted sum of the site's POI weight strength, number of lines, and PageRank centrality.

6. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, Scarcity of alternative paths It is a weighted sum of the number of alternative stations, the number of alternative routes, and the impact of detours.

7. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, The formula for the routine comprehensive criticality assessment model is: In the formula, , , These are the weighting coefficients.

8. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 7, characterized in that, A comprehensive criticality assessment model for rainstorm scenarios is constructed using a normal comprehensive criticality assessment model and the impact coefficient of rainstorm and urban flooding exposure. , This represents the impact coefficient of rainstorm exposure.

9. The method for identifying critical stations and assessing network resilience of urban rail transit in flood-prone cities according to any one of claims 7 or 8, characterized in that, Design site failure attack strategies, conduct site failure simulations, and assess the resilience of the normal comprehensive criticality evaluation model and / or the rainstorm scenario comprehensive criticality evaluation model by utilizing network efficiency retention rate, maximum connected subgraph ratio, reachable OD ratio, and potential passenger flow service loss.

10. The method for identifying key stations and assessing network resilience of urban rail transit in flood-prone cities according to claim 1, characterized in that, Construct a traffic topology network using the Space-L method.

Images