Urban traffic system toughness enhancement and recovery method based on safety linear interval

Abstract

The invention discloses an urban traffic system toughness enhancement and recovery method based on a safety linear interval. The method comprises the following steps: firstly, constructing a time-varying traffic network diagram of an urban road network; identifying a bottleneck link of each OD pair, defining a limited link, calculating a demand weighting coefficient of the limited link, and proposing a safety linear interval for describing a change rule of contribution of link quality improvement to network toughness in different intervals; establishing a linear relation between the link quality improvement quantity and the toughness gain in the safe linear interval; the demand weighting coefficient and the static connectivity are combined, and a dual-objective optimization method is combined to optimize weight parameters of a matrix, so that a traffic network weight matrix is constructed, node competition interaction is simulated, and key nodes are identified; and finally simulating a node removal and recovery process. According to the method, the anti-disturbance and recovery capability of the urban road traffic system is quantitatively improved, and a unified and efficient toughness enhancement and recovery scheme is provided for daily disturbance and post-disaster interruption.

Description

Technical Field

[0001] This invention belongs to the technical field of urban traffic management and intelligent transportation systems, and particularly relates to a method for enhancing and restoring the resilience of urban traffic systems based on safe linear intervals. Background Technology

[0002] In modern cities, reliable and efficient urban road transportation systems are a core component of urban infrastructure, supporting daily travel, emergency response, and resource allocation. However, with the acceleration of urbanization, the impact of climate change, and the continuous increase in travel demand, urban road transportation systems face increasingly severe challenges. In particular, the vulnerability of transportation systems is gradually increasing under the influence of various disturbances, mainly manifested in two types of disturbances: major disturbances (such as natural disasters and man-made disasters) and frequent daily disturbances (such as traffic accidents, equipment failures, and traffic light malfunctions). Although their impacts differ, both negatively affect the normal operation of the transportation system.

[0003] Current research on urban road traffic systems primarily focuses on resilience enhancement and recovery in response to major disasters and long-term disturbances, with insufficient research addressing frequent daily disturbances. While these daily disturbances may have smaller impacts, their high frequency gradually affects system efficiency, leading to traffic congestion, delays, and ultimately impacting citizens' travel experience. Therefore, existing resilience enhancement and recovery frameworks largely neglect this issue, making it difficult to effectively apply resilience theory to the daily operation and micro-level decision-making of urban road traffic systems.

[0004] Existing recovery strategies primarily rely on complex optimization models. These methods suffer from high computational complexity and low efficiency in large-scale transportation networks, making them difficult to apply effectively in real-time decision-making at traffic control centers. While index-based recovery strategies can address computational efficiency issues to some extent, these methods often neglect the dynamic role of network vulnerability and importance, resulting in insufficient real-time performance and efficiency in recovery decisions. Furthermore, most existing research focuses on link optimization, neglecting the criticality at the node level. This makes balancing network optimization and recovery strategies in daily operations and post-disaster recovery a pressing issue. Summary of the Invention

[0005] Purpose of the invention: To address the shortcomings of the aforementioned background technologies, this invention provides a method, apparatus, computer equipment, and storage medium for enhancing and restoring the resilience of urban road traffic systems within safe linear intervals. This addresses the technical problems in existing research, such as the lack of effective responses to daily disturbances, high computational complexity, insufficient real-time performance, and insufficient consideration of node criticality. It enables quantitative resilience enhancement and efficient restoration of traffic systems under limited resource conditions, thereby achieving a unified optimization goal that balances daily operation and post-disaster disruption scenarios.

[0006] Technical solution: The present invention provides a method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals, comprising the following steps:

[0007] Step S1: Obtain hourly link quality data and corresponding demand data using multiple data sources, embed the link quality data and demand data into the static road network topology, and construct a time-varying traffic network diagram of the urban road network.

[0008] Step S2: In the hourly time-varying traffic network graph, identify the bottleneck links for each origin-endpoint (OD) pair, define restricted links, calculate the demand weighting coefficients for these restricted links, and propose a safe linear interval to characterize the changing contribution of link quality improvements to network resilience across different intervals; within a given time interval... [ t 1, t 2 ] Internally, based on the link demand weighting coefficient and link quality, the resilience index R is obtained by integration or hourly discretization to measure the overall network's ability to resist disturbances.

[0009] Step S3: Within the safe linear interval, establish a linear relationship between the link quality improvement and resilience gain, and use the resilience gain formula to quantitatively calculate the resilience improvement brought about by the link improvement.

[0010] Step S4: Combine the demand weighting coefficient with static connectivity, and optimize the weight parameters of the matrix using a bi-objective optimization method to construct the traffic network weight matrix. Then, use the CSDK algorithm to simulate node competition and interaction to identify key nodes.

[0011] Step S5: Simulate the node removal and recovery process. Under two scenarios, namely daily disturbance and post-disaster recovery, compare five recovery strategies based on degree centrality, betweenness centrality, PageRank priority, demand priority, and CSDK to verify the superiority of the CSDK method under limited resource conditions.

[0012] Further, step 1 specifically involves: collecting multi-source basic data, including: road segment speed data, open street map data, population distribution data, land use data, and traffic analysis area data; then processing the data: on the one hand, extracting hourly link operation status to form link quality data, reflecting the speed and traffic conditions of different roads at different times; on the other hand, based on population and land use factors, and combined with a gravity model, constructing an hourly origin-destination OD demand matrix. F t =[ f of t ] ,in This represents the travel demand from origin o to destination d within time period t. O-D demand data characterizes the strength of travel connections and travel distribution characteristics between regions.

[0013] To obtain the operational characteristics of road links at different time periods, traffic data is collected periodically. The average speed of each origin-destination (OD) pair within its corresponding area is acquired at preset time intervals and summarized on an hourly scale to obtain hourly average link speed information. Based on the speed data and travel time parameters, a link operation quality index is defined to measure the link's performance at time t. The passage performance is expressed as follows:

[0014]

[0015] in, This indicates that throughout all time periods of the day, via the link The shortest required travel time; Indicates the time period Within, the travel time through this link;

[0016] Constructing a network link quality matrix Λ t =[ λ ij t ] ; in each non-overlapping hour interval Construct urban road traffic network maps for different time periods. ,in Represents the set of road intersection nodes. Let A represent the set of roads, and let A represent the adjacency matrix. Representing the demand matrix, This represents the link quality matrix.

[0017] Furthermore, in step 2, identifying bottleneck links for each origin-endpoint (OD) pair in the hourly time-varying traffic network graph, defining restricted links, calculating demand weighting coefficients for restricted links, and proposing safe linear intervals specifically involves: in the hourly time-varying traffic network... In this process, for each origin-endpoint pair (o, d), where o and d represent the origin and endpoint in the transportation network, all undirected paths are extracted. The bottleneck link is identified on each path, and its quality is defined as the minimum quality among all links on the path. Then, the link with the highest quality among all bottleneck links is selected as the limiting link for that OD pair. Based on this, the link quality is calculated. The weighting coefficients, where, when When dealing with a single OD pair, the weighting coefficients are:

[0018]

[0019] when When used as a limiting link for multiple OD pairs, its weighting coefficient is:

[0020]

[0021] in, This represents the traffic demand of the OD pair within time period t. Indicates the total network traffic demand; Represents the set of road intersection nodes; for each restricted link Define a safe linear interval [ , ],in, This indicates that OD (Original Design Inventory) restricts links in the network. Maximum link quality, This represents the maximum quality the link can achieve within a safe linear range. If the link quality improves within this range, its demand weighting coefficient remains unchanged; if the link quality exceeds... Traffic demand will be redistributed, and the weighting coefficients will change, thereby enabling the assessment of the contribution of links to network resilience under different quality conditions.

[0022] Furthermore, in step 2, the given time interval [ t 1, t 2 ] Within a given time interval, based on the link's demand-weighted coefficients and link quality, the resilience index R is obtained through integration or hourly discretization. [ t 1, t 2 ] Within this framework, based on demand-weighted coefficients and link quality, the average resilience level of the network under disturbances is calculated; the resilience index R is defined as:

[0023]

[0024] in, This represents the demand weighting coefficient for the limiting link at time t. This indicates the quality of the link at time t. Let R represent the set of roads; the resilience index R is a strictly linear function that limits link quality, with the slope as: ;

[0025] Since the analysis is performed on an hourly timescale, the above equation can be discretized as follows:

[0026]

[0027] Furthermore, step 3 specifically involves: when restricting links The mass increases within the safe linear range At that time, the corresponding network resilience gain Defined as:

[0028]

[0029] For any critical link At any time t, the instantaneous toughness gain resulting from a unit mass improvement is:

[0030] =

[0031] Furthermore, step 4 specifically involves: calculating the weighting coefficients of each link based on link traffic demand to reflect its traffic load and potential vulnerability, and constructing a link weighting matrix. Simultaneously, the topological connectivity information of the static network is extracted to construct the network connectivity matrix. A time-varying traffic network weight matrix is ​​proposed. Its definition is:

[0032]

[0033] in: W t =[ w of t ] The link weighting matrix based on critical links is defined as follows:

[0034]

[0035] Indicates the time period Internal link quality A=[ a of ] The network connectivity matrix is ​​defined as follows:

[0036]

[0037] Parameters a and b are used to balance the importance of traffic demand and network topology, and satisfy... ;

[0038] After the weight matrix is ​​constructed, in order to achieve a balance between local node dominance and global network structure effects, a parameter solution method based on multi-objective optimization is proposed to maximize the area of ​​the unaffected demand curve. And maximizing the relative size of the largest connected subgraph To achieve the dual optimization objectives, the Pareto optimality method is used to optimize and solve for parameters a and b.

[0039] in, The definition of is:

[0040]

[0041] Indicates at the threshold The proportion of OD flow that can still be satisfied; among which, Represents the total traffic of all OD pairs in the network. This indicates that during the penetration process, the total traffic of the service's OD pairs can be maintained; The definition of is:

[0042]

[0043] Represents the relative size of the largest connected subgraph; represents the number of nodes in the maximally connected subgraph, and is the total number of nodes in the URTS network. Through Pareto optimization, a trade-off solution can be obtained in the solution space of UD and LCC, thus determining the optimal parameters a and b, ensuring that the constructed weight matrix can balance traffic vulnerability and network topology characteristics. The optimized parameters are then substituted into the time-varying traffic network weight matrix, and the CSDK algorithm is used to simulate the competitive interaction process between nodes. Its dynamic evolution equation is:

[0044]

[0045] in, For node score vectors, Control the intensity of competition. , All represent the optimal parameters; through iterative solution, the stable solution of the system in the evolution process is obtained, and the set of key nodes is identified;

[0046] To avoid the computational and storage overhead of large-scale sparse matrix inversion, a recursive approximation method is proposed, simplifying the dynamic process as follows:

[0047]

[0048] in, The traffic network weight matrix at time t is composed of the link importance matrix and the network connectivity matrix. , and when At ∞, Converging to steady-state solution By adjusting The value of is used to balance the dominance at the local node level, i.e., low. With the overall network structure effect, i.e., high This allows us to obtain the key node identification results.

[0049] Further, step 5 specifically involves: simulating network outage and recovery scenarios through node removal and recovery processes, and evaluating the network performance recovery effect based on different node recovery sorting strategies; firstly, assuming that when a node fails, all its associated links fail simultaneously, and passenger flow cannot pass through that node; during the recovery process, traffic is restored by reactivating the node and its links. Under this framework, five recovery strategies based on degree centrality, betweenness centrality, PageRank priority, demand priority, and CSDK are selected for simulation, and the performance of different strategies in short-term and long-term scenarios is evaluated using the overall network resilience index; in short-term daily recovery scenarios, for minor disturbances such as traffic accidents, local congestion, and severe weather, the recovery effect is calculated using the linear gain formula after identifying key nodes; in long-term large-scale post-disaster recovery scenarios, for outages caused by natural disasters or extreme weather, the system resilience recovery level is evaluated through multi-day or long-term recovery processes.

[0050] The present invention also discloses a device for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals, the device comprising:

[0051] The time-varying traffic network construction module is used to acquire road link quality data and travel demand data to construct a time-varying network diagram of the urban road traffic system.

[0052] The resilience quantification calculation module is used to establish a network resilience measurement model based on demand weighting coefficients and link quality, and obtain network resilience indicators under different operating conditions.

[0053] The resilience enhancement calculation module is used to calculate the linear relationship between the link quality increment and resilience improvement within the safety linear range, obtain the quantitative resilience gain brought about by the link improvement, and verify the law of diminishing marginal benefits.

[0054] The key node identification module is used to construct a traffic network weighted matrix based on the link importance matrix and the network connectivity matrix, and to identify key nodes in the network by combining a non-dominated ranking algorithm.

[0055] The resilience recovery optimization module is used to determine the priority of link repair and node recovery based on the key node identification results in short-term disturbance and long-term disaster recovery scenarios, generate rapid recovery efficiency estimates and optimization schemes, and achieve efficient recovery of system resilience.

[0056] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method of the present invention.

[0057] The present invention also discloses a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the method of the present invention.

[0058] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: It can effectively improve the resilience of Urban Road Traffic Systems (URTS) under frequent daily disturbances. By introducing the theory of safe linear intervals, this method establishes a precise linear relationship between incremental link improvement and resilience enhancement, transforming the complex high-dimensional control problem into an operable linear scoring task, thereby significantly reducing computational complexity and supporting near real-time applications. Simultaneously, combined with a CSDK-based key node identification method, it can simultaneously consider dynamic traffic demand and network topology characteristics, achieving a unified recovery strategy in both short-term daily disturbances and long-term post-disaster recovery. This technology not only verifies the diminishing marginal resilience law brought about by link improvement but also provides traffic management departments with clear quantitative basis for "when, where, and how" to intervene, scientifically guiding priority link repair and node recovery, and improving resilience recovery efficiency. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating a method for enhancing and restoring the resilience of urban road traffic systems based on safe linear intervals in one embodiment.

[0060] Figure 2 This is a schematic diagram of a key node identification framework based on CSDK. Detailed Implementation

[0061] To overcome the shortcomings of existing urban road traffic system (URTS) resilience studies, such as insufficient attention to daily high-frequency disturbances, high computational complexity, difficulty in real-time application, and lack of node-level recovery analysis, this invention proposes a method for enhancing and restoring the resilience of urban road traffic systems oriented towards safe linear intervals. This method introduces the theory of safe linear intervals to establish a linear relationship between incremental link improvement and resilience enhancement, simplifying and quantifying complex high-dimensional control problems. Furthermore, it combines a key node identification method based on CSDK, considering both dynamic traffic demand and network topology, to identify the links and nodes most critical to system resilience. Based on this, the method can uniformly conduct resilience assessment and recovery optimization under both daily small disturbances and large-scale post-disaster disruptions, providing a quantitative basis for prioritizing link repair and node recovery, and improving the overall resilience level of the traffic system under different disturbance conditions.

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0063] In one exemplary embodiment, such as Figure 1 As shown, a method includes the following steps: Step S1, using multiple data sources to obtain hourly link quality data and corresponding demand data, embedding the link quality data and demand data into a static road network topology to construct a time-varying network diagram of the urban road network;

[0064] Step S2 involves identifying bottleneck links for each OD pair in the hourly time-varying traffic network, defining restricted links, calculating demand weighting coefficients for these restricted links, and further proposing a safe linear interval to characterize the changing contribution of link quality improvements to network resilience across different intervals; within a given time interval. [ t 1, t 2 ] Within the network, based on the link demand weighting coefficient and link quality, the resilience index R is obtained through integration (or hourly discretization) to measure the overall network's ability to withstand disturbances.

[0065] Step S3: Within the safe linear interval, establish a linear relationship between the link quality improvement and resilience gain, and use the resilience gain formula to quantitatively calculate the resilience improvement brought about by the link improvement.

[0066] Step S4: Combine the demand weighting coefficient with static connectivity, and optimize the weight parameters of the matrix using a bi-objective optimization method to construct the traffic network weight matrix. Finally, use the CSDK algorithm to simulate node competition and interaction, identify key nodes, and improve the scientific accuracy of key node identification.

[0067] Step S5 simulates the node removal and recovery process. Under two scenarios, namely daily disturbances (traffic accidents, local congestion) and post-disaster recovery (natural disasters, extreme weather), five recovery strategies based on degree centrality, betweenness centrality, PageRank priority, demand priority, and CSDK are compared to verify the superiority of the CSDK method under limited resource conditions.

[0068] In one embodiment, step S1 includes:

[0069] Multi-source basic data was collected, including: road segment speed data, open street map data, population distribution data, land use data, and traffic analysis zone data. This data was then comprehensively processed: firstly, hourly link operation status was extracted to form link quality data, reflecting the speed and traffic conditions of different roads at different times; secondly, based on factors such as population and land use, an hourly origin-destination (OD) demand matrix was constructed using a gravity model. F t =[ f of t ] ,in This represents the travel demand from origin o to destination d within time period t. The OD demand data can characterize the strength of travel connections and travel distribution characteristics between regions.

[0070] In this method, to obtain the operational characteristics of road links at different time periods, traffic data is collected periodically using the Baidu Maps API. Specifically, the average travel speed of each origin-destination (OD) pair within its corresponding area is obtained at 5-minute intervals and summarized on an hourly scale to obtain hourly average link speed information. Based on the speed data and travel time parameters, a link operation quality index is further defined to measure the link at time t. The passage performance is expressed as follows:

[0071]

[0072] in, This indicates that throughout all time periods of the day, via the link The shortest required travel time; Indicates the time period Within, the travel time through this link.

[0073] Based on this, a network link quality matrix is ​​constructed. Λ t =[ λ ij t ] In each non-overlapping hour interval Construct urban road traffic network maps for different time periods. ,in Represents the set of road intersection nodes. Let A represent the set of roads, and let A represent the adjacency matrix. Representing the demand matrix, This represents the link quality matrix.

[0074] In one embodiment, step S2 involves identifying bottleneck links for each OD pair in the hourly time-varying traffic network, defining restricted links, calculating demand weighting coefficients for the restricted links, and further proposing a safe linear interval to characterize the changing contribution of link quality improvement to network resilience in different intervals, including:

[0075] Hourly time-varying traffic network In the middle, firstly for each start-end pair Extract all of them The bottleneck link is identified on each path, and its quality is defined as the minimum quality among all links on that path. Then, the link with the highest quality among all bottleneck links is selected as the limiting link for that OD pair. Based on this, link calculations are performed. The weighting coefficients, where, when For a single OD pair, the weighting coefficient is:

[0076]

[0077] when When used as a limiting link for multiple OD pairs, its weighting coefficient is:

[0078]

[0079] in, This represents the traffic demand of the OD pair within time period t. This represents the total network traffic demand. In this process, for each restricted link... Further define a safe linear interval [ , If the link quality improves within this range, its demand weighting coefficient remains unchanged; if the link quality exceeds... Traffic demand will be redistributed, and the weighting coefficients will change, thereby enabling the assessment of the contribution of links to network resilience under different quality conditions.

[0080] In step S2, within the given time interval [ t 1, t 2 ] Within the network, based on the demand-weighted coefficients and link quality, a resilience index R is obtained through integration (or hourly discretization), which measures the overall network's ability to withstand disturbances, including:

[0081] Within a given time interval [ t 1, t 2 ] Within this framework, based on demand-weighted coefficients and link quality, the average resilience level of the network under disturbances is calculated. Specifically, this includes:

[0082] In time interval [ t 1, t 2 ] Internally, the resilience index R is defined as:

[0083]

[0084] in, This represents the demand weighting coefficient for the limiting link at time t. This indicates the quality of the link at time t.

[0085] Since the analysis is performed on an hourly timescale, the above equation can be discretized as follows:

[0086]

[0087] In one embodiment, such as Figure 2 As shown, step S3 includes:

[0088] Determine restricted links And determine its maximum mass. Is it within the safe linear range? , Under these conditions, the network resilience index R can be expressed as:

[0089]

[0090] This metric is a strictly linear function that limits link quality, and its slope is:

[0091] When the link is restricted The mass increases within the safe linear range At that time, the corresponding network resilience gain Defined as:

[0092]

[0093] For any critical link At any time t, the instantaneous toughness gain resulting from a unit mass improvement is:

[0094] =

[0095] This demonstrates that within the safe linear range, there is a strict linear relationship between resilience gain and limiting link quality improvement, thus providing traffic managers with precise quantitative data to determine which links to prioritize for improvement under limited resource conditions in order to maximize network resilience.

[0096] In one embodiment, such as Figure 2 As shown, step S4 includes:

[0097] Based on link traffic demand, weighting coefficients are calculated for each link to reflect its traffic load and potential vulnerability, and a link weighting matrix is ​​constructed. Simultaneously, the topological connectivity information of the static network is extracted to construct a network connectivity matrix. Based on this, a time-varying traffic network weight matrix is ​​proposed. Its definition is:

[0098]

[0099] in: W t =[ w of t ] The link weighting matrix based on critical links is defined as follows:

[0100]

[0101] A=[ a of ] The network connectivity matrix is ​​defined as follows:

[0102]

[0103] Parameters a and b are used to balance the importance of traffic demand and network topology, and satisfy... .

[0104] After the weight matrix is ​​constructed, to achieve a balance between local node dominance and global network structure effects, this invention proposes a parameter solution method based on multi-objective optimization. Specifically, this method aims to maximize the area under the unaffected demand curve (AUC). ) and maximizing the relative size of the largest connected subgraph ( With dual optimization objectives, the Pareto optimality method is used to optimize and solve for parameters a and b.

[0105] in, The definition of is:

[0106]

[0107] Indicates at the threshold The proportion of OD flow that can still be satisfied;

[0108] The definition of is:

[0109]

[0110] This represents the relative size of the largest connected subgraph.

[0111] Pareto optimization can yield a trade-off solution in the solution space of UD and LCC, thereby determining the optimal parameters a and b, ensuring that the constructed weight matrix can take into account both traffic vulnerability and network topology characteristics.

[0112] The optimized parameters are substituted into the time-varying traffic network weight matrix, and the CSDK algorithm is used to simulate the competitive interaction process between nodes. Its dynamic evolution equation is:

[0113]

[0114] in, For node score vectors, Controlling the intensity of competition. Through iterative solutions, stable solutions to the system during its evolution can be obtained, and the set of key nodes can be identified.

[0115] To avoid the computational and storage overhead of large-scale sparse matrix inversion, this invention further proposes a recursive approximation method, simplifying the above dynamic process as follows:

[0116]

[0117] in, The traffic network weight matrix at time t is composed of the link importance matrix and the network connectivity matrix. , and when At ∞, Converging to steady-state solution During this process, adjustments can be made. The value of is used to balance the dominance at the local node level (low). ) and overall network structure effect (high This allows for the acquisition of more reasonable key node identification results.

[0118] In one embodiment, step S5 includes:

[0119] This study simulates network outage and recovery scenarios through node removal and restoration, evaluating the network performance recovery effectiveness based on different node restoration ranking strategies. Specifically, it first assumes that when a node fails, all its associated links fail simultaneously, preventing passenger flow through that node. During the restoration process, traffic is restored by reactivating the node and its links. Within this framework, five restoration strategies—degree centrality-based, betweenness centrality-based, PageRank-based, demand-based, and CSDK-based—are simulated, and the performance of different strategies in short-term and long-term scenarios is evaluated using the overall network resilience index. In the short-term daily recovery scenario, focusing on minor disturbances such as traffic accidents, localized congestion, and severe weather, the restoration effect is directly calculated using the linear gain formula after identifying key nodes. In the long-term large-scale post-disaster recovery scenario, addressing widespread outages caused by natural disasters or extreme weather, the system resilience recovery level is evaluated through multi-day or long-term restoration processes. Comparative results show that the CSDK-based restoration strategy can achieve more efficient resilience recovery under limited resource conditions.

[0120] This invention proposes a method for enhancing and restoring the resilience of urban road traffic systems within safe linear intervals. This method overcomes the shortcomings of existing technologies, such as insufficient resilience assessment under daily high-frequency minor disturbances, high computational complexity, lack of node-level analysis, and real-time application capabilities. By introducing the theory of safe linear intervals, this method establishes a linear gain relationship between incremental link improvement and system resilience enhancement, thereby transforming a high-dimensional control problem into an operable linear scoring task, reducing algorithm complexity and meeting near-real-time application requirements. Simultaneously, by combining the CSDK key node identification method, it considers both dynamic traffic demand and network topology characteristics, achieving unified identification and prioritized restoration of key nodes and links. Based on this, this invention can optimize resilience enhancement and restoration under both daily disturbances and post-disaster interruptions, providing quantitative basis for link repair and node recovery, and significantly improving the overall resilience level of urban road traffic systems under different disturbance conditions.

[0121] Based on the same inventive concept, this application also provides an apparatus for implementing the above-mentioned method for enhancing and restoring the resilience of urban road traffic systems based on safe linear intervals. The solution provided by this apparatus is similar to the implementation described in the above method. Therefore, the specific limitations of one or more embodiments of the method for enhancing and restoring the resilience of urban road traffic systems based on safe linear intervals provided below can be found in the limitations of the traffic flow detector data super-resolution method described above, and will not be repeated here.

[0122] In one exemplary embodiment, a device for enhancing and restoring the resilience of urban road traffic systems based on safe linear intervals is provided, comprising:

[0123] The time-varying traffic network construction module is used to acquire road link quality data and travel demand data to construct a time-varying network diagram of the urban road traffic system.

[0124] The resilience quantification calculation module is used to establish a network resilience measurement model based on demand weighting coefficients and link quality, and obtain network resilience indicators under different operating conditions.

[0125] The resilience enhancement calculation module is used to calculate the linear relationship between the link quality increment and resilience improvement within the safety linear range, obtain the quantitative resilience gain brought about by the link improvement, and verify the law of diminishing marginal benefits.

[0126] The key node identification module is used to construct a traffic network weighted matrix based on the link importance matrix and the network connectivity matrix, and to identify key nodes in the network by combining a non-dominated ranking algorithm.

[0127] The resilience recovery optimization module is used to determine the priority of link repair and node recovery based on the key node identification results in short-term disturbance and long-term disaster recovery scenarios, generate rapid recovery efficiency estimates and optimization schemes, and achieve efficient recovery of system resilience.

[0128] The modules in the aforementioned urban road traffic system resilience enhancement and recovery device based on safe linear intervals can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0129] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0130] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0131] For those skilled in the art, various modifications and improvements can be made without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application shall be determined by the appended claims.

Claims

1. A method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals, characterized in that, Includes the following steps: Step S1: Obtain hourly link quality data and corresponding demand data using multiple data sources, embed the link quality data and demand data into the static road network topology, and construct a time-varying traffic network diagram of the urban road network. Step S2: In the hourly time-varying traffic network graph, identify the bottleneck links for each origin-endpoint (OD) pair, define restricted links, calculate the demand weighting coefficients for these restricted links, and propose a safe linear interval to characterize the changing contribution of link quality improvements to network resilience across different intervals; within a given time interval... Internally, based on the link demand weighting coefficient and link quality, the resilience index R is obtained by integration or hourly discretization to measure the overall network's ability to resist disturbances. Step S3: Within the safe linear interval, establish a linear relationship between the link quality improvement and resilience gain, and use the resilience gain formula to quantitatively calculate the resilience improvement brought about by the link improvement. Step S4: Combine the demand weighting coefficient with static connectivity, and optimize the weight parameters of the matrix using a bi-objective optimization method to construct the traffic network weight matrix. Then, use the CSDK algorithm to simulate node competition and interaction to identify key nodes. Step S5: Simulate the node removal and recovery process. Under two scenarios, namely daily disturbance and post-disaster recovery, compare five recovery strategies based on degree centrality, betweenness centrality, PageRank priority, demand priority, and CSDK to verify the superiority of the CSDK method under limited resource conditions.

2. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 1, characterized in that, Step 1 specifically involves collecting multi-source basic data, including: road segment speed data, open street map data, population distribution data, land use data, and traffic analysis area data. The data is then processed: on the one hand, hourly link operation status is extracted to form link quality data, reflecting the speed and traffic conditions of different roads at different times; on the other hand, based on population and land use factors, an hourly origin-destination (OD) demand matrix is ​​constructed using a gravity model. ,in This represents the travel demand from origin o to destination d within time period t. O-D demand data characterizes the strength of travel connections and travel distribution characteristics between regions. To obtain the operational characteristics of road links at different time periods, traffic data is collected periodically. The average speed of each origin-destination (OD) pair within its corresponding area is acquired at preset time intervals and summarized on an hourly scale to obtain hourly average link speed information. Based on the speed data and travel time parameters, a link operation quality index is defined to measure the link's performance at time t. The passage performance is expressed as follows: ； in, This indicates that throughout all time periods of the day, via the link The shortest required travel time; Indicates the time period Within, the travel time through this link; Constructing a network link quality matrix ; in each non-overlapping hour interval Construct urban road traffic network maps for different time periods. ,in Represents the set of road intersection nodes. Let A represent the set of roads, and let A represent the adjacency matrix. Representing the demand matrix, This represents the link quality matrix.

3. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 1, characterized in that, In step 2, identifying bottleneck links for each origin-destination (OD) pair in the hourly time-varying traffic network graph, defining restricted links, calculating demand weighting coefficients for restricted links, and proposing safe linear intervals specifically involves: in the hourly time-varying traffic network... In this process, for each origin-endpoint pair (o, d), where o and d represent the origin and endpoint in the transportation network, all undirected paths are extracted. The bottleneck link is identified on each path, and its quality is defined as the minimum quality among all links on the path. Then, the link with the highest quality among all bottleneck links is selected as the limiting link for that OD pair. Based on this, the link quality is calculated. The weighting coefficients, where, when When dealing with a single OD pair, the weighting coefficients are: ； when When used as a limiting link for multiple OD pairs, its weighting coefficient is: ； in, This represents the traffic demand of the OD pair within time period t. Indicates the total network traffic demand; Represents the set of road intersection nodes; for each restricted link Define a safe linear interval [ , ],in, This indicates that OD (Original Design Inventory) restricts links in the network. Maximum link quality, This represents the maximum quality the link can achieve within a safe linear range. If the link quality improves within this range, its demand weighting coefficient remains unchanged; if the link quality exceeds... Traffic demand will be redistributed, and the weighting coefficients will change, thereby enabling the assessment of the contribution of links to network resilience under different quality conditions.

4. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 1, characterized in that, In step 2, the given time interval Within a given time interval, based on the link's demand-weighted coefficients and link quality, the resilience index R is obtained through integration or hourly discretization. Within this framework, based on demand-weighted coefficients and link quality, the average resilience level of the network under disturbances is calculated; the resilience index R is defined as: ； in, This represents the demand weighting coefficient for the limiting link at time t. This indicates the quality of the link at time t. Let R represent the set of roads; the resilience index R is a strictly linear function that limits link quality, with the slope as: ; Since the analysis is performed on an hourly timescale, the above equation can be discretized as follows: 。 5. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 4, characterized in that, Step 3 specifically involves: when restricting links The mass increases within the safe linear range At that time, the corresponding network resilience gain Defined as: ； For any critical link At any time t, the instantaneous toughness gain resulting from a unit mass improvement is: = 。 6. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 1, characterized in that, Step 4 specifically involves: calculating the weighting coefficients for each link based on link traffic demand to reflect its traffic load and potential vulnerability, and constructing a link weighting matrix. Simultaneously, the topological connectivity information of the static network is extracted to construct the network connectivity matrix. A time-varying traffic network weight matrix is ​​proposed. Its definition is: ； in: The link weighting matrix based on critical links is defined as follows: ； Indicates the time period Internal link quality The network connectivity matrix is ​​defined as follows: ； Parameters a and b are used to balance the importance of traffic demand and network topology, and satisfy... ; After the weight matrix is ​​constructed, in order to achieve a balance between local node dominance and global network structure effects, a parameter solution method based on multi-objective optimization is proposed to maximize the area of ​​the unaffected demand curve. And maximizing the relative size of the largest connected subgraph To achieve the dual optimization objectives, the Pareto optimality method is used to optimize and solve for parameters a and b. in, The definition of is: ； Indicates at the threshold The proportion of OD flow that can still be satisfied; among which, Represents the total traffic of all OD pairs in the network. This indicates that during the penetration process, the total traffic of the service's OD pairs is maintained; The definition of is: ； This represents the relative size of the largest connected subgraph; where, This represents the number of nodes in the largest connected subgraph. It is the total number of nodes in the URTS network; Pareto optimization can yield a trade-off solution in the solution space of UD and LCC, thereby determining the optimal parameters a and b, ensuring that the constructed weight matrix can take into account both traffic vulnerability and network topology characteristics. Substituting the optimized parameters into the time-varying traffic network weight matrix, and using the CSDK algorithm to simulate the competitive interaction process between nodes, the dynamic evolution equation is as follows: ； in, For node score vectors, Control the intensity of competition. , All represent the optimal parameters; through iterative solution, the stable solution of the system in the evolution process is obtained, and the set of key nodes is identified; To avoid the computational and storage overhead of large-scale sparse matrix inversion, a recursive approximation method is proposed, simplifying the dynamic process as follows: ； in, The traffic network weight matrix at time t is composed of the link importance matrix and the network connectivity matrix. , and when At ∞, Converging to steady-state solution By adjusting The value of is used to balance the dominance at the local node level, i.e., low. With the overall network structure effect, i.e., high This allows us to obtain the key node identification results.

7. The method for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals according to claim 1, characterized in that, Step 5 specifically involves simulating network outage and recovery scenarios through node removal and recovery processes, and evaluating the recovery effect of network performance based on different node recovery sorting strategies. First, it is assumed that when a node fails, all its associated links fail simultaneously, preventing passenger flow from passing through that node. During the recovery process, traffic is restored by reactivating the node and its links. Within this framework, five recovery strategies—degree centrality-based, betweenness centrality-based, PageRank-based, demand-based, and CSDK-based—are simulated, and the performance of different strategies in short-term and long-term scenarios is evaluated using the overall network resilience index. In short-term daily recovery scenarios, for minor disturbances such as traffic accidents, localized congestion, and severe weather, the recovery effect is calculated using the linear gain formula after identifying key nodes. In long-term large-scale post-disaster recovery scenarios, for outages caused by natural disasters or extreme weather, the system resilience recovery level is evaluated through multi-day or long-term recovery processes.

8. A device for enhancing and restoring the resilience of urban transportation systems based on safe linear intervals, used to implement the method as described in claim 1, characterized in that, The device includes: The time-varying traffic network construction module is used to acquire road link quality data and travel demand data to construct a time-varying network diagram of the urban road traffic system. The resilience quantification calculation module is used to establish a network resilience measurement model based on demand weighting coefficients and link quality, and obtain network resilience indicators under different operating conditions. The resilience enhancement calculation module is used to calculate the linear relationship between the link quality increment and resilience improvement within the safety linear range, obtain the quantitative resilience gain brought by the link improvement, and verify the law of diminishing marginal benefits. The key node identification module is used to construct a traffic network weighted matrix based on the link importance matrix and the network connectivity matrix, and then identify key nodes in the network by combining a non-dominated ranking algorithm. The resilience recovery optimization module is used to determine the priority of link repair and node recovery based on the key node identification results in short-term disturbance and long-term disaster recovery scenarios, generate rapid recovery efficiency estimates and optimization schemes, and achieve efficient recovery of system resilience.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method of claim 1.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method of claim 1.

Images