Urban rail transit network resilience assessment method and system

Abstract

The present invention relates to the technical field of road traffic network assessment, and in particular to a method and system for assessing the resilience of an urban rail transit network. The present invention is based on data-driven, first extracting the AFC data of the urban rail transit network, processing the AFC data to obtain OD pair data, then cleaning the OD pair data, and integrating them to form an overall data set; then calculating the normal base value of each OD pair in the overall data set; and determining the scope of the tested scene, extracting the tested scene data set from the overall data set; then using the normal base value and the tested scene data set to calculate and obtain the standard score; finally, constructing a resilience index based on the standard score, and evaluating the resilience changes of the network based on the resilience index. Compared with the prior art, the present invention has the advantages of weakening the influence of noise and accurately assessing the resilience characteristics of the urban rail transit network under disturbance events.

Description

Technical Field

[0001] The present invention relates to the technical field of road traffic network assessment, and in particular to a method and system for assessing the resilience of an urban rail transit network. Background Art

[0002] As the main component of urban public transportation, urban rail transit has become a key pillar for the sustainable development of urban transportation with its high capacity, high speed and high efficiency. However, with the continuous expansion of the scale of urban rail transit network, the passenger volume has increased sharply, and the large passenger flow of tens of millions has posed a huge challenge to the safety and reliability of the operation organization. Urban rail transit is extremely vulnerable to disturbance events such as train accidents, equipment failures, extreme weather, major events, etc., which will cause the risk of sudden large passenger flow and impact the normal operation of the network system. In the above context, the resilience assessment of urban rail transit systems with high travel demand intensity, high frequency of disturbance impact, multiple types of risk factors, and deep degree of closure is the key to ensuring urban traffic order and urban development.

[0003] Resilience refers to the ability to effectively resist, absorb, adapt and recover in the face of disturbances. It is an emerging indicator for evaluating the ability of public transportation systems to provide reliable services. Urban rail transit network resilience assessment usually uses two types of indicators: topology-based indicators and performance-based indicators. Topology-based indicators are based on complex network theory, including average node degree, network efficiency, betweenness centrality, relative size of the largest connected graph, and average shortest path length. However, topology-based indicators can only reflect the damage to the network structure, while ignoring the impact of disturbance events on real operational characteristics such as passenger flow and travel time delays. Performance-based indicators can evaluate the impact of disturbance events on real network operational events, including the frequency of disturbance events, average passenger travel time, and total travel time delays.

[0004] However, data noise is inevitable in actual data sets. Current traffic resilience assessment indicators cannot distinguish and weaken the impact of data noise. Their quantitative accuracy can easily be reduced due to the influence of noise in the data set, resulting in a significant deviation between the quantitative results of the indicators and the actual resilience value of the network. For example, in the absence of disturbance events, due to the influence of data noise, the network performance values ​​calculated using the current traffic resilience indicators may still show significant fluctuations, resulting in unreliable network resilience results obtained based on the existing indicators.

[0005] For example, the invention patent with publication number CN117808187A discloses a method for evaluating the resilience of an urban rail transit network. It constructs a dynamic "passenger flow-vehicle flow coupling network" percolation model based on multi-source data of urban rail transit, and uses the model to perform percolation at each moment; however, it cannot identify or weaken the impact of data noise. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for evaluating the resilience of an urban rail transit network.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] According to one aspect of the present invention, a method for evaluating the resilience of an urban rail transit network is provided, the method comprising the following steps:

[0009] S1. Extract the AFC (Automatic Fare Collection System) data of the urban rail transit network, process the AFC data to obtain the OD pair (Original, Destination, referring to the traffic volume) data, then clean the OD pair data, and finally integrate them into the overall data set;

[0010] S2, calculate the normal base value of each OD pair in the overall data set;

[0011] S3, determine the scope of the scene to be tested, and extract the scene data set to be tested from the overall data set;

[0012] S4. Calculate and obtain the standard score using the normal base value and the measured scene data set;

[0013] S5. Construct a resilience index based on the standard score and evaluate the resilience change of the network based on the resilience index.

[0014] As a preferred technical solution, the AFC data field of the urban rail transit network in S1 includes the card number ID (Identity document), card swiping time, card swiping site and entry and exit type; the OD pair data field includes the card number ID, entry card swiping time, entry card swiping site, exit card swiping time and exit card swiping site.

[0015] As a preferred technical solution, the specific process of processing AFC data in S1 to obtain OD pair data is: sorting AFC data according to card swiping time, matching the inbound card swiping data of the same card number with its adjacent outbound card swiping data to form OD pair data.

[0016] As a preferred technical solution, the process of data cleaning for OD pair data in S1 is specifically as follows: delete data records in which the travel time in the OD pair is less than or equal to 0; delete data with repeated card number ID and entry card swiping time, and only retain OD pairs formed by the entry and exit connections with the closest time; delete data with the same entry and exit stations.

[0017] As a preferred technical solution, the normal base value in S2 includes the average travel time of the OD pair and the standard deviation of the travel time, and the specific calculation formula is:

[0018]

[0019]

[0020] in, The overall data set of urban rail transit network i Tiandi t In the time interval p Mean travel time of group OD pairs; w p,t,i For the overall data set i Tiandi t In the time interval p Traffic of group OD pairs; N is the number of days covered by the overall data set; v p,t,i For the overall data set t In the time interval p The travel time of the OD pair of the group; σ p,t is the standard deviation of the travel time of the pth group of OD pairs in the tth time interval in the entire data set.

[0021] As a preferred technical solution, the scope of the measured scene in S3 includes the scope of the time dimension and the scope of the space dimension, wherein the scope of the time dimension includes the time period covered by the measured scene, and the scope of the space dimension includes the sites covered by the measured scene.

[0022] As a preferred technical solution, the method for extracting the measured scene data set in S3 is: from the OD pair data set of the entire urban rail transit network, extract the OD pairs whose entry card swiping time is within the time dimension of the measured scene, and whose entry card swiping stations are within the spatial dimension of the measured scene, thereby forming a measured scene data set.

[0023] As a preferred technical solution, the standard score in S4 is the standard score between the travel time of each OD pair in the measured scene data set and the normal base value, and its specific calculation formula is:

[0024]

[0025] in, For the overall data set t In the time interval p The average travel time of the group OD pairs; σ p,t For the overall data set t In the time intervalp Standard deviation of travel time for group OD pairs; u p,t The first t In the time interval p The average travel time of the OD pairs in the group.

[0026] As a preferred technical solution, the specific calculation formula of the traffic network resilience index of the measured scenario in S5 is:

[0027]

[0028] in, Q(t) is an indicator of resilience; p is the ordinal number of the OD pair; z p,t It is the standard score between the travel time of each OD pair in the tested scenario data set and the normal base value; I is an indicative function, β is the threshold value, if z p,t >β ,but I is equal to 1, otherwise it is equal to 0; q p,t The first t In the time interval p Traffic of the OD pair of the group.

[0029] According to one aspect of the present invention, a system for evaluating the resilience of an urban rail transit network is provided. The system applies the urban rail transit network resilience evaluation method as described above, and the system includes a data processing module, a normal base value calculation module, a measured scenario data set extraction module, a standard score calculation module and a resilience evaluation module;

[0030] Among them, the data processing module is used to extract and process the AFC data of the urban rail transit network, obtain the passenger OD pair data, perform data cleaning on it, and integrate it to obtain the overall data set of the urban rail transit network; the normal base value calculation module is used to calculate the normal base value of each OD pair in the overall data set; the measured scenario data set extraction module is used to determine the scope of the measured scenario, and extract the measured scenario data set from the overall data set to form a measured scenario data set; the standard score calculation module is used to use the normal base value and the measured scenario data set to calculate the standard score; the resilience assessment module is used to construct a resilience index based on the standard score, and evaluate the resilience changes of the network based on the resilience index.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. The present invention uses AFC data to evaluate resilience. Based on data drive, the AFC data of the urban rail transit network is first extracted, and the AFC data is processed to obtain OD pair data. Then the OD pair data is cleaned and integrated to form an overall data set. Then the normal base value of each OD pair in the overall data set is calculated. The scope of the tested scene is determined, and the tested scene data set is extracted from the overall data set to form a tested scene data set. The normal base value and the tested scene data set are then used to calculate the standard score. Finally, the resilience index is constructed by the standard score, and the resilience change of the network is evaluated based on the resilience index. The present invention extracts and calculates the entry and exit stations, flow, and time characteristics of the OD pair data of passengers in each period from the AFC data, and quantifies the resilience of the tested scene in any time dimension and space dimension based on the OD pair data characteristics and the resilience index. This allows the present invention to not only characterize the degree of influence of disturbance events on network resilience, but also accurately quantify the dynamic changes of resilience at the urban rail transit network level and station level at different times, and finely analyze the resilience performance of each station under disturbance events, helping management departments to accurately evaluate the spatiotemporal process of resilience response under disturbance.

[0033] 2. The resilience index in the present invention is the standard score between the travel time of each OD pair in the data set of the tested scenario and the normal base value. By considering and introducing the normal base value of each OD pair in terms of travel time, an exemplary function is introduced, and the resilience index is applied to calculate the network resilience value of the tested scenario; the resilience evaluation method of the present invention can weaken and eliminate the noise influence in the real data, better distinguish the resilience changes caused by disturbance events and the resilience changes caused by data noise, so that the quantitative results of resilience are more scientific and accurate, which helps the management department to more accurately evaluate the resilience characteristics of the urban rail transit network under disturbance events, and provide more reliable support for the emergency evaluation and response of the urban rail transit system under disturbance. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the steps of the transportation network resilience assessment method in the present invention;

[0035] Figure 2 It is a schematic diagram of the resilience change of a certain city's rail transit network evaluated based on the method of the present invention in the embodiment;

[0036] Figure 3 It is a schematic diagram of the resilience change of a certain city's rail transit network evaluated based on the average travel time in the embodiment;

[0037] Figure 4 It is a schematic diagram of the resilience change of a certain city's rail transit network evaluated based on the average delay time in the embodiment;

[0038] Figure 5It is a schematic diagram of the resilience change of a certain city's rail transit network evaluated based on the Mahalanobis distance in the embodiment;

[0039] Figure 6 This is a schematic diagram of the resilience values ​​of each station in a certain city's rail transit network within 1 hour before and after the occurrence of event 1 in the embodiment;

[0040] Figure 7 This is a schematic diagram of the resilience values ​​of each station in a certain city's rail transit network within 1 hour before and after the occurrence of event 4 in the embodiment;

[0041] Figure 8 This is a schematic diagram of the resilience values ​​of each station in a certain city's rail transit network within 1 hour before and after the occurrence of event 7 in the embodiment;

[0042] Fig. 9 This is a schematic diagram of the resilience values ​​of each station in a certain city's rail transit network within 1 hour before and after the occurrence of event 8 in the embodiment. DETAILED DESCRIPTION

[0043] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0044] As the main component of urban public transportation, urban rail transit has become a key pillar for the sustainable development of urban transportation with its high capacity, high speed and high efficiency. However, with the continuous expansion of the scale of urban rail transit network, the passenger volume has increased sharply, and the large passenger flow of tens of millions has posed a huge challenge to the safety and reliability of the operation organization. Urban rail transit is extremely vulnerable to disturbance events such as train accidents, equipment failures, extreme weather, major events, etc., which will cause the risk of sudden large passenger flow and impact the normal operation of the network system. In the above context, the resilience assessment of urban rail transit systems with high travel demand intensity, high frequency of disturbance impact, multiple types of risk factors, and deep degree of closure is the key to ensuring urban traffic order and urban development.

[0045] Resilience refers to the ability to effectively resist, absorb, adapt and recover in the face of disturbances. It is an emerging indicator for evaluating the ability of public transportation systems to provide reliable services. Urban rail transit network resilience assessment usually uses two types of indicators: topology-based indicators and performance-based indicators. Topology-based indicators are based on complex network theory, including average node degree, network efficiency, betweenness centrality, relative size of the largest connected graph, and average shortest path length. However, topology-based indicators can only reflect the damage to the network structure, while ignoring the impact of disturbance events on real operational characteristics such as passenger flow and travel time delays. Performance-based indicators can evaluate the impact of disturbance events on real network operational events, including the frequency of disturbance events, average passenger travel time, and total travel time delays.

[0046] However, data noise is inevitable in actual data sets. Current traffic resilience assessment indicators cannot distinguish and weaken the impact of data noise. Their quantitative accuracy can easily be reduced due to the influence of noise in the data set, resulting in a significant deviation between the quantitative results of the indicators and the actual resilience value of the network. For example, in the absence of disturbance events, due to the influence of data noise, the network performance values ​​calculated using the current traffic resilience indicators may still show significant fluctuations, resulting in unreliable network resilience results obtained based on the existing indicators.

[0047] Based on this, the present invention proposes a big data-based urban rail transit network resilience assessment method. The assessment method is based on real AFC data and proposes a new idea for constructing a resilience assessment indicator, which can better distinguish between network resilience changes caused by disturbance events and resilience changes caused by data noise, ensure the accuracy and credibility of resilience quantification results, and use the calculated results to perform network resilience analysis.

[0048] Example 1

[0049] In this embodiment, a method for evaluating the resilience of an urban rail transit network is applied. Figure 1 As shown, the following steps are included:

[0050] S1, extracting the AFC data of the urban rail transit network, processing the AFC data to obtain OD pair data, then cleaning the OD pair data, and finally integrating them to form an overall data set;

[0051] S2, calculate the normal base value of each OD pair in the overall data set;

[0052] S3, determine the scope of the scene to be tested, and extract the scene data set to be tested from the overall data set;

[0053] S4. Calculate and obtain the standard score using the normal base value and the measured scene data set;

[0054] S5. Construct a resilience index based on the standard score and evaluate the resilience change of the network based on the resilience index.

[0055] In this method, the data processing method of step S1 includes the following steps:

[0056] S11. Obtain AFC data of the urban rail transit network, where the AFC data includes fields such as “card number ID, card swiping time, card swiping station, and entry and exit type”;

[0057] S12, sorting the AFC data according to the card swiping time, matching the entry card swiping data of the same card number with the adjacent exit card swiping data in sequence, forming OD pair data, each OD pair data of which contains the fields of "card number ID, entry card swiping time, entry card swiping station, exit card swiping time, exit card swiping station"; calculating the travel time of the OD pair according to the exit card swiping time and entry card swiping time of the OD pair;

[0058] S13, data cleaning, delete the data items with the travel time less than or equal to 0 in the OD pair; delete the data with the same card number ID and entry card swiping time, and only keep the OD pairs formed by the entry and exit connections with the closest time; delete the data with the same entry and exit stations.

[0059] In this embodiment, the AFC data of the urban rail transit network is first obtained. The AFC data of a certain city's rail transit is obtained, with a time span of three full months and a total of 1.451 billion records, including the fields of "card number ID, card swiping time, card swiping station, and entry and exit type".

[0060] Table 1 Processed OD pair dataset of a certain city's urban rail transit network

[0061] Card ID Entry card swiping station Entry card swiping time Exit card swiping station Exit card swiping time Travel time (min) 7791 Line 4 Station 1 19:14:08 Line 10 Station 1 19:46:37 1949 52896 Line 8 Station 1 17:09:01 Line 10 Station 2 17:53:20 2659 55493 Line 10 Station 3 14:05:25 Line 2 Station 1 14:29:21 1436 55493 Line 2 Station 1 15:26:49 Line 10 Station 3 15:47:59 1270 55966 Line 1 Station 1 8:12:33 Line 8 Station 2 8:51:12 2319 55966 Line 1 Station 2 17:36:49 Line 1 Station 3 18:07:33 1844 56769 Line 6 Station 1 17:26:16 Line 6 Station 2 17:31:38 322 56936 Line 8 Station 3 12:37:53 Line 8 Station 2 12:47:33 580

[0062] In this embodiment, the AFC data is sorted by the card swiping time, and the entry card swiping data of the same card number and the adjacent exit card swiping data are matched in sequence to form OD pair data and calculate the travel time of the OD pair. Data cleaning is performed to delete the data strips with travel time less than or equal to 0 in the OD pair; delete the data with repeated card number ID and progress card swiping time, and only retain the OD pairs formed by the entry and exit connections with the closest time; delete the data with the same entry and exit stations. The processed OD pair data set contains 455 million data. Table 1 is the processed OD pair data set of part of the urban rail transit network of a certain city.

[0063] In this method, the calculation steps of the normal base value of the OD pair in terms of travel time in step S2 are as follows:

[0064] First, count the days covered by the overall data set of the urban rail transit network N，According to the entry and exit card swiping time, each OD pair is grouped according to a certain time interval. For example, if the grouping time interval is 1 hour, the OD pair data of hours 0-1, 1-2, ..., 22-23, and 23-24 are grouped respectively; according to the entry and exit card swiping sites, the OD pairs are grouped. In each time interval, the OD pairs with the same entry and exit card swiping sites are grouped together; the statistics of the i Tiandi t In the time interval, p Traffic of group OD pair w p,t,i and travel time v p,t,i ;

[0065] Secondly, calculate the t In the time interval p Average travel time of OD pairs and the standard deviation of travel time σ p,t , the calculation formula is as follows:

[0066]

[0067]

[0068] in, The overall data set of urban rail transit network i Tiandi t In the time interval p Mean travel time of group OD pairs; w p,t,i For the overall data set i Tiandi t In the time interval p Traffic of group OD pairs; N is the number of days covered by the overall data set; v p,t,i For the overall data set t In the time interval p The travel time of the OD pair of the group; σ p,t For the overall data set t In the time interval p The standard deviation of the travel time for the OD pairs in the group.

[0069] In the present method, the spatiotemporal scope of the measured scene in step S3 includes: the scope of the time dimension and the scope of the space dimension, the scope of the time dimension refers to the time period covered by the scene to be studied, and the scope of the space dimension refers to the sites covered by the scene to be studied; the method for extracting the measured scene data set in step S3 is to extract the OD pairs whose entry card swiping time is within the time dimension range of the measured scene and whose entry card swiping sites are within the space dimension range of the measured scene from the OD pair data set of the entire urban rail transit network, thereby forming a measured scene data set.

[0070] In this embodiment, the time range of the tested scenario is determined as the five days from July 4 to July 8, and the spatial range is determined as the entire urban rail transit network of a certain city. From the OD pair data set of the entire urban rail transit network of a certain city, the OD pairs with the entry card swiping time within the five days from July 4 to July 8 (regardless of the entry card swiping station) are extracted to form the tested scenario data set.

[0071] In this method, the specific process of calculating the standard score using the normal base value and the measured scene data set in step S4 is as follows:

[0072] First, count the tested scene data sets, t In the time interval, p Traffic of group OD pair q p,t and average travel time u p,t ;

[0073] Then calculate the standard score between the travel time of each OD pair and the normal base value in the tested scenario data set z p,t :

[0074]

[0075] in, For the overall data set t In the time interval p The average travel time of the group OD pairs; σ p,t For the overall data set t In the time interval p Standard deviation of travel time for group OD pairs; u p,t The first t In the time interval p The average travel time of the OD pairs in the group.

[0076] In this method, in step S5, the resilience value of the urban rail transit network in the measured scenario is calculated according to the resilience index, and the resilience index calculation formula is as follows:

[0077]

[0078] in, Q(t) is an indicator of resilience; p is the ordinal number of the OD pair; z p,t It is the standard score between the travel time of each OD pair in the tested scenario data set and the normal base value; I is an indicative function, β is the threshold value, if z p,t >β ,but I is equal to 1, otherwise it is equal to 0; q p,t The first t In the time interval p Traffic of the OD pair of the group.

[0079] β is a threshold, that is, only the standard score z p,t Exceeding the threshold β Only the OD pairs will be included in the resilience index Q(t) The standard score is often used in statistics to identify outliers. It can measure the degree to which a data point deviates from the average. Data points with a standard score greater than 3 are usually considered outliers.

[0080] In this embodiment β Take 3.

[0081] In this embodiment, the data is substituted into the above toughness index to calculate the toughness value.

[0082] In this embodiment, a toughness triangle is drawn according to the toughness value calculation results of each time period in the tested scene. The drawing result is as follows: Figure 2 As shown, Figure 2 This is the change in resilience of a city’s urban rail transit network under the measured scenario in this embodiment.

[0083] In this example, a total of 8 disturbance events occurred in a certain city's urban rail transit network, such as Figures 2 to 5 As shown, the gray areas from left to right are events 1 to 8, which are 3 equipment failures, 3 vehicle failures, 1 door failure and 1 foreign object intrusion disturbance event. Specifically, event 1, event 3 and event 4 are equipment failures; event 6 is a door failure; event 2, event 5 and event 7 are vehicle failures; event 8 is a foreign object intrusion.

[0084] In this example, in order to compare and analyze the superiority of this resilience quantification method, three other commonly used urban rail transit resilience indicators are selected: average travel time, average delay time and Mahalanobis distance. The resilience changes of these indicators are as follows: Figure 3 , Figure 4 and Figure 5 As shown, Figure 3 is the resilience result based on the average travel time in common indicators, Figure 4 is the resilience result based on the average delay time in common indicators, Figure 5 It is the toughness result based on the Mahalanobis distance in common indicators.

[0085] Depend on Figure 2 and Figures 3 to 5 By comparison, it can be found that the results of urban rail transit resilience quantified based on commonly used resilience indicators still show significant fluctuations under normal conditions (i.e., when no disturbance events occur). In particular, these indicators lack the ability to quantify small-scale disturbance events (such as Event 2, Event 5, and Event 6), which leads to inaccurate calculation and analysis results of network resilience. The network resilience performance quantified by this resilience assessment method remains stable under normal conditions, and only shows resilience fluctuations when 8 disturbance events occur.

[0086] In summary, the resilience indicators and evaluation methods in this scheme demonstrate superior capabilities, which can weaken the impact of noisy noise in real data sets, better distinguish between changes in network resilience caused by disturbance events and changes in resilience caused by data noise, and ensure the accuracy and credibility of resilience quantification results.

[0087] Example 2

[0088] In this embodiment, an urban rail transit network resilience assessment system is applied, which includes a data processing module, a normal base value calculation module, a measured scenario data set extraction module, a standard score calculation module and a resilience assessment module;

[0089] Among them, the data processing module is used to extract and process the AFC data of the urban rail transit network, obtain the passenger OD pair data, perform data cleaning on it, and integrate it to obtain the overall data set of the urban rail transit network; the normal base value calculation module is used to calculate the normal base value of each OD pair in the overall data set; the measured scenario data set extraction module is used to determine the scope of the measured scenario, and extract the measured scenario data set from the overall data set to form a measured scenario data set; the standard score calculation module is used to use the normal base value and the measured scenario data set to calculate the standard score; the resilience assessment module is used to construct a resilience index based on the standard score, and evaluate the resilience changes of the network based on the resilience index.

[0090] In this embodiment, the following steps are first performed using a data processing module:

[0091] S11. Obtain AFC data of the urban rail transit network, where the AFC data includes fields such as “card number ID, card swiping time, card swiping station, and entry and exit type”;

[0092] S12, sorting the AFC data according to the card swiping time, matching the entry card swiping data of the same card number with the adjacent exit card swiping data in sequence, forming OD pair data, each OD pair data of which contains the fields of "card number ID, entry card swiping time, entry card swiping station, exit card swiping time, exit card swiping station"; calculating the travel time of the OD pair according to the exit card swiping time and entry card swiping time of the OD pair;

[0093] S13, data cleaning, delete the data items with the travel time less than or equal to 0 in the OD pair; delete the data with the same card number ID and entry card swiping time, and only keep the OD pairs formed by the entry and exit connections with the closest time; delete the data with the same entry and exit stations.

[0094] In this embodiment, the AFC data of the urban rail transit network is first obtained. The AFC data of a certain city's rail transit is obtained, with a time span of three full months and a total of 1.451 billion records, including the fields of "card number ID, card swiping time, card swiping station, and entry and exit type".

[0095] In this embodiment, the AFC data is sorted by the card swiping time, and the entry card swiping data of the same card number and the adjacent exit card swiping data are matched in sequence to form OD pair data and calculate the travel time of the OD pair. Data cleaning is performed to delete the data strips with travel time less than or equal to 0 in the OD pair; delete the data with repeated card number ID and progress card swiping time, and only retain the OD pairs formed by the entry and exit connections with the closest time; delete the data with the same entry and exit stations. The processed OD pair data set contains 455 million data.

[0096] In this embodiment, after processing the data, the normal base value calculation module is used to perform the following steps:

[0097] First, count the days covered by the overall data set of the urban rail transit network N , according to the entry and exit card swiping time, each OD pair is grouped according to a certain time interval. For example, if the grouping time interval is 1 hour, the OD pair data of hours 0-1, 1-2, ..., 22-23, and 23-24 are grouped respectively; according to the entry and exit card swiping sites, the OD pairs are grouped. In each time interval, the OD pairs with the same entry and exit card swiping sites are grouped together; the statistical calculation of the i Tiandi t In the time interval, p Traffic of group OD pair w p,t,i and travel timev p,t,i ;

[0098] Secondly, calculate the t In the time interval p Average travel time of OD pairs and the standard deviation of travel time σ p,t , the calculation formula is as follows:

[0099]

[0100]

[0101] in, The overall data set of urban rail transit network i Tiandi t In the time interval p Mean travel time of group OD pairs; w p,t,i For the overall data set i Tiandi t In the time interval p Traffic of group OD pairs; N is the number of days covered by the overall data set; v p,t,i For the overall data set t In the time interval p The average travel time of the group OD pairs; σ p,t For the overall data set t In the time interval p The standard deviation of the travel time for the OD pairs in the group.

[0102] In this embodiment, a measured scene data set extraction module is used to determine the scope of the measured scene, and a measured scene data set is extracted from the overall data set to form a measured scene data set; wherein, the spatiotemporal scope of the measured scene includes: the scope of the time dimension and the scope of the space dimension, the scope of the time dimension refers to the time period covered by the scene to be studied, and the scope of the space dimension refers to the sites covered by the scene to be studied.

[0103] In this embodiment, the method for extracting the measured scene data set is: from the OD pair data set of the entire urban rail transit network, extract the OD pairs whose entry card swiping time is within the time dimension of the measured scene and whose entry card swiping stations are within the space dimension of the measured scene, thereby forming a measured scene data set.

[0104] In this embodiment, the time range of the tested scene is determined as within 1 hour before and after the occurrence of event 1, event 4, event 7 and event 8, and the spatial range is determined as each station in the urban rail transit network of a certain city. OD pairs within the range of the tested scene are extracted to form a tested scene data set.

[0105] In this method, the following steps are performed using the standard score calculation module:

[0106] First, count the tested scene data sets, t In the time interval, p Traffic of group OD pair q p,t and average travel time u p,t ;

[0107] Then calculate the standard score between the travel time of each OD pair and the normal base value in the measured scenario data set z p,t :

[0108]

[0109] in, For the overall data set t In the time interval p The average travel time of the group OD pairs; σ p,t For the overall data set t In the time interval p Standard deviation of travel time for group OD pairs; u p,t The first t In the time interval p The average travel time of the OD pairs in the group.

[0110] S43. Calculate the resilience value of the urban rail transit network in the tested scenario based on the resilience index. The resilience index calculation formula is as follows:

[0111]

[0112] in, Q(t) is an indicator of resilience; p is the ordinal number of the OD pair; z p,t It is the standard score between the travel time of each OD pair in the tested scenario data set and the normal base value; I is an indicative function, β is the threshold value, if z p,t >β ,but I is equal to 1, otherwise it is equal to 0;q p,t The first t In the time interval p Traffic of the OD pair of the group.

[0113] In this embodiment, β is a threshold, that is, only the standard score z p,t Exceeding the threshold β Only the OD pairs will be included in the resilience index Q(t) The standard score is often used in statistics to identify outliers. It can measure the degree to which a data point deviates from the average. Data points with a standard score greater than 3 are usually considered outliers.

[0114] In this embodiment, β Take 3.

[0115] In this embodiment, the data is brought into the toughness index to calculate the toughness value.

[0116] In this embodiment, the time range of the measured scene is determined to be within 1 hour before and after the occurrence of event 1, event 4, event 7 and event 8, and the spatial range is determined to be each station in the urban rail transit network of a certain city. Figure 6 , Figure 7 , Figure 8 and Fig. 9 As shown, Figure 6 This is a schematic diagram of the resilience values ​​of each station in a city's rail transit network within 1 hour before and after event 1 occurs; Figure 7 This is a schematic diagram of the resilience values ​​of each station in a city's rail transit network within 1 hour before and after event 4 occurs; Figure 8 This is a schematic diagram of the resilience values ​​of each station in a city's rail transit network within 1 hour before and after event 7; Fig. 9 This is a schematic diagram of the resilience values ​​of each station in a city’s rail transit network within 1 hour before and after event 8. Figure 6 and Figure 7 It can be seen that events 1 and 4 have had a profound impact on the resilience of most stations on Line 3 and Line 4 of a certain city’s urban rail transit. Figure 8 and Fig. 9 It can be seen that event 7 mainly affects Line 1 of a certain city's urban rail transit, and has little impact on the resilience of each station on Line 1; while event 8 mainly affects the branch line of Line 11 of a certain city's urban rail transit. Although the impact range of this event is small, that is, the number of stations affected by the event is small, its impact lasts longer, and event 8 lasts for more than 2 hours.

[0117] In summary, this resilience assessment system takes into account the changes in resilience in time and space dimensions, which helps management departments to more accurately assess the resilience characteristics of urban rail transit networks under disturbance events and provide more reliable support for emergency assessment and response of urban rail transit systems under disturbances.

Claims

1. A method for evaluating the resilience of an urban rail transit network, characterized in that: The method comprises the following steps: S1, extracting the AFC data of the urban rail transit network, processing the AFC data to obtain OD pair data, then cleaning the OD pair data, and finally integrating them to form an overall data set; S2, calculate the normal base value of each OD pair in the overall data set; S3, determine the scope of the scene to be tested, and extract the scene data set to be tested from the overall data set; S4. Calculate and obtain the standard score using the normal base value and the measured scene data set; S5. Construct a resilience index based on the standard score and evaluate the resilience change of the network based on the resilience index; The normal base value in S2 includes the average travel time of the OD pair and the standard deviation of the travel time, and the specific calculation formula is: in, is the i-th day in the urban rail transit network overall data set t In the time interval p The average travel time of the group OD pairs; w p,t,i is the number of the i-th day in the entire data set t In the time interval p The flow of group OD pairs; N is the number of days covered by the overall data set; v p,t,i For the overall data set t In the time interval p The travel time of the OD pair of the group; σ p,t For the overall data set t In the time interval p The standard deviation of the travel time of the group OD pairs; The standard score in S4 is the standard score between the travel time of each OD pair in the measured scene data set and the normal base value, and its specific calculation formula is: in, For the overall data set t In the time interval p The average travel time of the group OD pairs; σ p,t For the overall data set t In the time interval p Standard deviation of travel time for group OD pairs; u p,t The first t In the time interval p The average travel time of the OD pairs in the group.

2. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The urban rail transit network AFC data field in S1 includes card number ID, card swiping time, card swiping site and entry and exit type; the OD pair data field includes card number ID, entry card swiping time, entry card swiping site, exit card swiping time and exit card swiping site.

3. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The specific process of processing the AFC data in S1 to obtain the OD pair data is: sorting the AFC data according to the card swiping time, matching the inbound card swiping data of the same card number with its adjacent outbound card swiping data to form the OD pair data.

4. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The process of cleaning the OD pair data in S1 is as follows: deleting data records in the OD pair where the travel time is less than or equal to 0; deleting data with duplicate card number ID and entry card swiping time, and only retaining the OD pairs formed by the entry and exit connections with the closest time; Delete the data that is the same for inbound and outbound sites.

5. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The scope of the measured scene in S3 includes the scope of the time dimension and the scope of the space dimension, wherein the scope of the time dimension includes the time period covered by the measured scene, and the scope of the space dimension includes the sites covered by the measured scene.

6. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The method for extracting the measured scene data set in S3 is: from the OD pair data set of the entire urban rail transit network, extract the OD pairs whose entry card swiping time is within the time dimension of the measured scene, and the entry card swiping stations are within the spatial dimension of the measured scene, thereby forming a measured scene data set.

7. The urban rail transit network resilience assessment method according to claim 1, characterized in that: The specific calculation formula of the traffic network resilience index of the measured scenario in S5 is: in, Q ( t ) is the toughness index; p is the ordinal number of the OD pair; z p,t It is the standard score between the travel time of each OD pair in the tested scenario data set and the normal base value; I is an indicative function, β is the threshold, if z p,t > β ,but I is equal to 1, otherwise it is equal to 0; q p,t The first t In the time interval p Traffic of the OD pair of the group.

8. An urban rail transit network resilience assessment system, characterized in that: The system applies an urban rail transit network resilience assessment method as described in any one of claims 1 to 7, and the system includes a data processing module, a normal base value calculation module, a measured scenario data set extraction module, a standard score calculation module and a resilience assessment module; The data processing module is used to extract and process the AFC data of the urban rail transit network, obtain the passenger OD pair data, perform data cleaning on it, and integrate it to obtain the overall data set of the urban rail transit network; The normal base value calculation module is used to calculate the normal base value of each OD pair in the overall data set; The measured scene data set extraction module is used to determine the range of the measured scene and extract the measured scene data set from the overall data set; The standard score calculation module is used to calculate and obtain the standard score using the normal base value and the measured scene data set; The resilience assessment module is used to construct a resilience index from a standard score, and to assess the resilience change of the network based on the resilience index.

Images