A dynamic management and control method and system for traffic infrastructure based on non-stationary climate disaster factors

Abstract

This invention discloses a dynamic management and control method and system for transportation infrastructure based on non-stationary climate disaster-causing factors. The method includes the following steps: Step 1, acquiring basic data; Step 2, constructing a non-stationary extreme value statistical model; Step 3, calculating the engineering adaptability gap index; Step 4, constructing a connected topology model and performing cascade failure simulation; Step 5, generating and executing traffic control instructions. This invention, by establishing a non-stationary extreme value statistical model, can capture the drift trend of climate extreme values ​​over time, breaking through the limitations of static assessment and achieving dynamic and accurate quantification of climate risk. Furthermore, it realizes a paradigm shift from "passive response" to "proactive resilience management." Based on the management mechanism of this invention, "flexible degraded operation" effectively avoids the impact of traffic flow on vulnerable road networks, thereby preventing structural damage and global paralysis of regional road networks, and significantly improving the system resilience of transportation infrastructure throughout its entire life cycle.

Description

Technical Field

[0001] This invention relates to the field of traffic control and disaster prevention and mitigation technology, and more specifically, to a method and system for dynamic control of traffic infrastructure based on non-stationary climate disaster-causing factors. Background Technology

[0002] Transportation infrastructure (especially the highway network) is the lifeblood of a country's economic operation. According to the current "Technical Standards for Highway Engineering", the design of highway subgrades and bridges and culverts is usually based on historical hydrological and meteorological records at the time of construction, setting a specific design flood frequency (such as "once in a century") as a safety defense standard, and its design service life is usually as long as several decades.

[0003] However, with the intensification of global climate change, the frequency and intensity of extreme weather events (such as torrential rains and prolonged high temperatures) exhibit significant non-stationary variations. Existing technologies for risk assessment and management of transportation networks generally suffer from the following technical deficiencies:

[0004] First, existing assessment methods are mostly based on the climate stability assumption, which assumes that the distribution patterns of future extreme weather events will remain consistent with historical records. This static assumption cannot reflect the dynamic changes in extreme values ​​of disaster-causing factors over time. This means that design defense parameters determined based on early historical data may no longer be applicable under current changing weather conditions, exposing road networks to disaster risks exceeding the original design standards.

[0005] Second, existing assessment systems typically treat the entire transportation network as a homogeneous whole for vulnerability analysis, ignoring the differences in the construction periods of individual road segments. In reality, road segments built in different eras adhere to different design specifications, construction techniques, and referenced meteorological baseline data. A holistic assessment cannot accurately identify vulnerable road segments with "climate adaptability gaps" due to earlier construction dates and lower initial design standards, easily leading to a misallocation of road maintenance and management resources.

[0006] Third, existing traffic control measures mostly employ a passive response mechanism. Typically, guidance systems are only used to direct vehicles to alternative routes after substantial damage or blockage of roads. In complex traffic networks, passive, rigid blocking of a single node or road segment can cause a large-scale shift of traffic flow to surrounding lower-grade or less capable road networks within a short period. This sudden surge in traffic can easily trigger secondary congestion, even overloading surrounding road networks, leading to cascading failures in the regional traffic network and a significant decline in overall network connectivity.

[0007] To address the aforementioned issues, there is an urgent need in this field for a dynamic management and control scheme for transportation infrastructure that can dynamically quantify non-stationary climate risks, accurately identify climate adaptability gaps in road sections from different eras, and proactively generate physical control signals to prevent cascading failures. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for dynamic management and control of transportation infrastructure based on non-stationary climate disaster-causing factors, so as to solve the above-mentioned technical problems.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] This invention discloses a dynamic management and control method for transportation infrastructure based on non-stationary climate disaster-causing factors, the method comprising the following steps:

[0011] Step 1: Obtain basic data: Obtain vector data of the traffic network in the target area, including the construction year attributes and design level information of each road segment in the traffic network; at the same time, obtain historical meteorological grid data of the target area in a continuous time period.

[0012] Step 2: Construct a non-stationary extreme value statistical model: Based on the acquired historical meteorological grid data, extract the meteorological time series with the maximum daily precipitation as a specific climate disaster factor; establish a non-stationary extreme value statistical model, and use the non-stationary extreme value statistical model to fit the evolution relationship of the distribution parameters of the meteorological time series over time, and calculate the spatial distribution data of precipitation extreme values ​​in the target area at the current moment.

[0013] Step 3: Calculate the Engineering Adaptability Gap Index: Based on the construction year of each road segment, the transportation network is divided into multiple road network subsets, each corresponding to a specific construction period. For any target road segment within a road network subset, historical design baseline precipitation corresponding to its construction period and design grade is retrieved. Simultaneously, based on the spatial coordinates of the target road segment, its spatial location is matched with the spatial distribution data of current precipitation extremes to calculate the Engineering Adaptability Gap Index, which characterizes the current degree of climate adaptability degradation. The specific calculation process includes:

[0014] 1) Calculate the actual probability of occurrence: Obtain the historical design baseline precipitation of the target road section during the construction period, denoted as Z. design , will Z design Substituting this as an input threshold into the non-stationary extreme value statistical model, the actual probability P of the historical design baseline precipitation occurring at the current time T is calculated. actual The calculation formula is as follows:

[0015] P actual =1-G(Z) design ,T) (3)

[0016] In the formula, G(Z) design (T) represents the cumulative distribution function value of a non-stationary extreme value statistical model at a given current time T;

[0017] 2) Calculate the effective return period: Based on the actual occurrence probability mentioned above, define and calculate the effective return period (ERP) of the target road segment under the current climate conditions:

[0018] ERP = 1 / P actual (4)

[0019] 3) Calculate the engineering adaptability gap index: Extract the historical design standard return period specified in the highway engineering technical standards during the construction period of the target road section, calculate the attenuation of the effective return period (ERP) relative to the historical design standard return period, and use the attenuation as the engineering adaptability gap index;

[0020] Step 4: Construct a connected topology model and perform cascading failure simulation: Using a grid-based dual topology construction method, a connected topology model reflecting the physical connection relationships of the road network is constructed, and cascading failure simulation is performed. The specific process includes:

[0021] 1) Establish a physical topology model: Define a grid system with a preset resolution covering the target area; map the structured vector segments of the traffic network to the grid system. If an actual physical road segment exists within a grid, then the grid is marked as an accessible grid; abstract each accessible grid as a node in the topology network, and establish connections between adjacent accessible grids based on the physical adjacency relationships of the vector road network in real space, thereby constructing a grid connectivity graph; the initial weight of each connection is set based on the physical distance of the corresponding grid road segment or the travel time under free-flow conditions.

[0022] 2) Setting thresholds and dynamic failure rules: The calculated effective return period is used as the failure judgment index for each accessible grid node, and cascade failure simulation calculation is performed on the grid connectivity graph; a first critical threshold Th1 and a second early warning threshold Th2 are set, where Th1 < Th2; for each node in the network, its dynamic change rules in the topology graph are determined by numerical comparison.

[0023] 3) Cascaded Failure Simulation and Vulnerable Node Identification: After adjusting one or more initial test nodes according to the above rules, the original topological balance of the road network is broken. The system will simulate the redistribution of traffic flow to the surrounding road network due to local obstruction. The shortest path and connectivity efficiency between all network nodes are recalculated using graph theory algorithms, thereby obtaining the decline rate of the overall connectivity performance index of the remaining network under the cascade effect, i.e., the performance degradation rate. If the failure test of an initial test node causes the connectivity performance degradation rate of the entire network to be greater than or equal to the preset sensitivity threshold, then the node is marked as a critical vulnerable node.

[0024] Step 5: Generate and execute traffic control instructions: Based on the identified critical vulnerable nodes and their corresponding engineering adaptability gap indices, generate corresponding physical control signals, i.e., control instructions; send the control instructions to the on-site traffic control equipment of the corresponding road section to drive the traffic control facilities to perform physical actions.

[0025] Furthermore, the vector data of the transportation network mentioned in step 1 comes from a transportation geographic information system or a transportation network database; the historical meteorological grid data comes from public meteorological data platforms, including the National Meteorological Science Data Center, the ERA5 reanalysis dataset issued by the European Centre for Medium-Range Weather Forecasts, or the MSWEP multi-source fusion precipitation dataset; the historical meteorological data spans more than or equal to 30 years, the grid size corresponding to the spatial resolution is no greater than 0.25°×0.25°, and the temporal resolution is daily.

[0026] Furthermore, the non-stationary extreme value statistical model described in step 2 adopts a generalized extreme value distribution model, and its cumulative distribution function is constructed as follows:

[0027] (1)

[0028] In the formula: z is the extreme value of precipitation; σ is the scale parameter; For shape parameters; For position parameters;

[0029] Among them, position parameters Set as a linear function of time t:

[0030] (2)

[0031] Where α is the intercept parameter, which characterizes the baseline intensity of the disaster-causing factor; β is the slope parameter, which characterizes the linear trend of the intensity of the disaster-causing factor changing with time; σ, α, and β are all parameters to be estimated;

[0032] The specific process for calculating the spatial distribution data of precipitation extreme values ​​in the target area at the current time is as follows: the above-mentioned parameters to be estimated are solved by the maximum likelihood estimation method, the solved parameters are substituted into the generalized extreme value distribution model, the theoretical intensity threshold of each spatial grid in the target area at the current time for any return period is calculated, and the spatial distribution data of precipitation extreme values ​​at the current time is generated.

[0033] Furthermore, step 4, which describes determining the dynamic change rules of each node in the network through numerical comparison, specifically includes:

[0034] Blockage failure simulation: When ERP≤Th1, it is determined that the road segment corresponding to the node has been physically blocked or completely paralyzed due to climate disaster factors; in the algorithm implementation of the grid connected graph, this is specifically manifested as removing the node and all its connected edges.

[0035] Damage-related traffic restriction simulation: When Th1 < ERP ≤ Th2, it is determined that the traffic capacity of the road segment corresponding to the node is damaged; in the algorithm implementation, this is specifically manifested by increasing the weight of the edges connected to the node.

[0036] Safe passage simulation: When ERP > Th2, it is determined that the effective defense capability of the road segment is within the safe range, and the initial weights of the node and its connected edges remain unchanged.

[0037] Furthermore, the feature is that the specific process of generating corresponding physical control signals, i.e., control instructions, based on the identified key vulnerable nodes and their corresponding engineering adaptability gap indices in step 5 is as follows:

[0038] The physical control signals include blocking signals and flow limiting signals; for any critical vulnerable node, a corresponding physical control signal is generated based on the comparison between its calculated effective return period (ERP) and the first critical threshold Th1 and the second early warning threshold Th2.

[0039] 1) Extremely high risk blocking control: When ERP≤Th1, it indicates that the effective disaster prevention and defense capability of the road section corresponding to the critical vulnerable node has dropped to the first critical threshold and below. The system determines that the road section faces an extremely high risk of water damage or physical interruption. At this time, the system generates a blocking signal.

[0040] 2) Medium- and high-risk traffic restriction and control: When Th1 < ERP ≤ Th2, it is determined that the traffic capacity of the road segment corresponding to the vulnerable node is damaged, but has not yet reached the limit state of complete blockage. At this time, the system generates the traffic restriction signal.

[0041] 3) Safety status maintenance: When ERP > Th2, it is determined that the effective defense capability of the road section under the current weather conditions is still within the safety threshold range, and the system does not generate physical control signals.

[0042] The present invention also discloses a dynamic management and control system for transportation infrastructure based on non-stationary climate disaster-causing factors. The system includes interconnected data acquisition modules, dynamic model construction modules, gap index calculation modules, cascade failure simulation modules, and traffic control execution modules.

[0043] The data acquisition module is used to acquire structured vector data of the traffic network in the target area and historical meteorological grid data of the target area over a continuous time period.

[0044] The dynamic model construction module is communicatively connected to the data acquisition module, and is used to receive the historical meteorological grid data and extract the meteorological time series, establish and use a non-stationary extreme value statistical model to fit the evolution relationship of the distribution parameters of the meteorological time series over time, and calculate the spatial distribution number of precipitation extreme values ​​in the target area at the current moment.

[0045] The gap index calculation module is communicatively connected to the dynamic model construction module. It is used to obtain the corresponding historical design benchmark precipitation based on the construction year attribute of each road segment, calculate the effective return period of the historical design benchmark at the current moment using the aforementioned non-stationary extreme value statistical model and precipitation extreme value spatial distribution data, and compare the effective return period with the legally mandated historical design standard return period to output the engineering adaptability gap index of the target road segment.

[0046] The cascaded failure simulation module is communicatively connected to the gap index calculation module. It is used to construct a grid connectivity graph reflecting physical connection relationships based on the vector data of the traffic network. The effective return period is used as the failure judgment parameter for the reachable grid nodes. Combined with the preset first critical threshold and second early warning threshold, the module simulates network paralysis and traffic restriction by dynamically removing nodes or increasing the edge weights in the grid connectivity graph. Then, it calculates the decline of the overall network connectivity performance index and identifies key vulnerable nodes.

[0047] The traffic control execution module is communicatively connected to the cascaded failure simulation module. It is used to generate corresponding blocking signals or flow restriction signals and other control instructions based on the risk level of the critical vulnerable nodes, and send the control instructions to the traffic control equipment on the corresponding road section via the communication network to drive the relevant facilities to perform physical actions such as lane closure or speed limit adjustment.

[0048] The beneficial effects of this invention are as follows:

[0049] 1) This invention, by establishing a non-stationary extreme value statistical model that includes time covariates, can capture the drift trend of climate extremes over time, breaking through the limitations of static assessment and achieving dynamic and accurate quantification of climate risk. Compared with traditional static assessments based on historical averages, this invention can calculate the "effective return period" of design standards under the current climate, thereby accurately revealing the risk of substantial failure of design parameters due to climate change.

[0050] 2) This invention abandons the traditional approach of treating the road network as a homogeneous whole. It innovatively utilizes the "construction era attribute" to deconstruct the road network into different incremental subsets and employs a "retrospective matching" logic to quantify the adaptability gaps of road sections built in different eras under the current climate. This method can accurately identify those hidden vulnerable nodes that, although meeting the construction standards of their time, have become "weak links" under the current climate, solving the problem of assessing intergenerational differences in the road network and providing a scientific basis for differentiated maintenance.

[0051] 3) Addressing the nonlinear traffic flow transfer and cascading failures caused by traditional passive road closures, this invention achieves a paradigm shift from "passive response" to "active resilience management," generating proactive physical control signals (such as speed limits and road closures). Based on the management mechanism of this invention, although this strategy sacrifices some traffic efficiency locally, it effectively avoids the impact of traffic flow on the vulnerable road network through "flexible degraded operation," thereby preventing structural damage and global paralysis of the regional road network and significantly improving the system resilience of transportation infrastructure throughout its entire life cycle.

[0052] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the method flow described in this invention;

[0054] Figure 2 This is a schematic diagram illustrating the adaptive gap between the physical road network and dynamic weather in Example 1;

[0055] Figure 3 This is a schematic diagram illustrating the construction of the physical connectivity topology model and the rasterization process in Example 1;

[0056] Figure 4 This is a schematic diagram comparing the network connectivity status based on MATLAB simulation in Example 1. Detailed Implementation

[0057] This invention discloses a dynamic management and control method for transportation infrastructure based on non-stationary climate disaster-causing factors, such as... Figure 1 As shown, the method includes the following steps:

[0058] Step 1: Obtain Basic Data. Obtain vector data of the target area's transportation network. This vector data includes structured data such as the construction year and design level information of each road segment. In practical engineering applications, this data source is generally a Geographic Information System (GIS) or a transportation network database. Simultaneously, obtain historical meteorological grid data of the target area over a continuous time period. To ensure the reliability of extreme value statistics, the span of the historical meteorological data is preferably greater than or equal to 30 years, the grid size corresponding to the spatial resolution is no greater than 0.25° × 0.25°, and the temporal resolution is daily. In specific implementation, the aforementioned historical meteorological grid data can be obtained from publicly available meteorological data platforms, such as the National Meteorological Science Data Center, the ERA5 reanalysis dataset issued by the European Centre for Medium-Range Weather Forecasts (ECMWF), or the MSWEP multi-source fusion precipitation dataset.

[0059] Step 2: Constructing a Non-Stationary Extreme Value Statistical Model: Based on the acquired historical meteorological grid data, meteorological time series are extracted. In this invention, the maximum daily precipitation is used as a specific climate-causing factor for extraction. A non-stationary extreme value statistical model is established to fit the evolution of the distribution parameters of the meteorological time series over time.

[0060] Specifically, the non-stationary extreme value statistical model preferably adopts the generalized extreme value distribution (GEV) model, whose cumulative distribution function is constructed as follows:

[0061] (1)

[0062] In the formula: z is the extreme value of precipitation; σ is the scale parameter; For shape parameters; For position parameters.

[0063] Among them, position parameters Set as a linear function of time t:

[0064] (2)

[0065] Where α is the intercept parameter, which characterizes the baseline intensity of the disaster-causing factor; β is the slope parameter, which characterizes the linear trend of the intensity of the disaster-causing factor changing with time; σ, α, and β are all parameters to be estimated. By setting these parameters, the model can intuitively reflect the trend of climate evolution. For example, if the calculated value of β > 0, it indicates that the intensity of the disaster is increasing over time.

[0066] The parameters to be estimated (i.e., scale, shape, intercept, and slope parameters) are solved using the maximum likelihood estimation (MLE) method. These solved parameters are then substituted into the GEV model to calculate the theoretical intensity threshold for any return period for each spatial grid within the target area at the current time, generating spatial distribution data of precipitation extremes at the current moment. This data will serve as the baseline input for subsequent matching with historical design baseline intensities.

[0067] Step 3: Calculate the Engineering Adaptability Gap Index: Based on the construction year of each road segment, the transportation network is divided into multiple road network subsets, each corresponding to a specific construction period. For any target road segment within a road network subset, historical design baseline precipitation corresponding to its construction period and design grade is retrieved. Simultaneously, based on the spatial coordinates of the target road segment, it is spatially matched with the spatial distribution data of current precipitation extremes to calculate the Engineering Adaptability Gap Index, which characterizes the current degree of climate adaptability degradation. The specific calculation process is as follows:

[0068] 1) Calculate the actual probability of occurrence: Obtain the historical design baseline precipitation of the target road section during the construction period, denoted as Z. design , will Z design As the input threshold, it is substituted into the non-stationary extreme value statistical model established in step 2 to calculate the actual probability P of the historical design baseline precipitation at the current time T. actual The calculation formula is as follows:

[0069] P actual =1-G(Z) design ,T) (3)

[0070] In the formula, G(Z) design (T) represents the cumulative distribution function value of a non-stationary extreme value statistical model at a given current time T.

[0071] 2) Calculate the Effective Return Period (ERP): Based on the actual occurrence probability mentioned above, define and calculate the effective return period (ERP) of the target road segment under the current climate conditions:

[0072] ERP = 1 / P actual (4)

[0073] The physical significance of Effective Return Period (ERP) lies in reflecting the drift of defense capabilities under non-stationary climate conditions. For example, when a certain road section was constructed, based on the then-current stable climate data, its design reference precipitation return period was the legally mandated "once-in-a-century" (i.e., the historical design standard return period was 100 years, and the probability of occurrence at that time was 1%). However, under current constantly changing meteorological conditions, if the actual probability P of that precipitation occurs increases due to increased precipitation extremes... actualIf it rises to 2%, then its effective return period (ERP) at the current moment has actually decreased to 50 years.

[0074] 3) Calculate the Engineering Adaptability Gap Index: Extract the historical design standard return period (e.g., the aforementioned "once-in-a-century" statutory benchmark) specified in the highway engineering technical standards during the construction period of the target road section, calculate the attenuation of the effective return period (ERP) relative to the historical design standard return period, and use this attenuation as the engineering adaptability gap index. When the calculated ERP is less than the historical design standard return period, the system determines that the target road section has a climate adaptability gap, indicating that its current actual disaster resistance and defense capabilities are lower than the initial design safety standards.

[0075] Step 4: Construct a connected topology model and perform cascading failure simulation: Using a grid-based dual topology construction method, a connected topology model reflecting the physical connection relationships of the road network is constructed, and cascading failure simulation is performed. The specific process includes:

[0076] 1) Establishing a physical topology model: First, define a grid system with a preset resolution covering the entire target area. Second, map the structured vector segments of the traffic network onto this grid system. If an actual physical road segment exists within a grid, mark it as an accessible grid. Subsequently, abstract each accessible grid as an independent node in the topology network, and establish edges between adjacent accessible grids based on the physical adjacency relationships of the vector road network in real space (e.g., using the 8-neighborhood connectivity rule). This constructs a grid connectivity graph based on grid nodes that reflects the physical connectivity relationships of the real road network. In this connectivity graph, the initial weight of each edge is set based on the actual physical distance of the corresponding grid road segment or the travel time under free-flow conditions.

[0077] 2) Setting Thresholds and Dynamic Failure Rules: Using the calculated Effective Return Period (ERP) as the failure criterion for each accessible grid node, cascade failure simulation calculations are performed on the grid connectivity graph. To achieve accurate simulation of the impact of dynamic disaster-causing factors on the physical road network, a first critical threshold Th1 (e.g., a 50-year return period) and a second warning threshold Th2 (e.g., an 80-year return period) are set, where Th1 < Th2. For each node in the network, its dynamic change rules in the topology graph are determined through numerical comparison:

[0078] Blockage failure simulation: When ERP≤Th1, it is determined that the road segment corresponding to the node has been physically blocked or completely paralyzed due to climate-related disaster factors (such as deep water accumulation caused by severe rainstorms). In the algorithm implementation of the grid connected graph, this is specifically manifested by removing the node and all its connected edges.

[0079] Damage-related traffic restriction simulation: When Th1 < ERP ≤ Th2, the traffic capacity of the road segment corresponding to the node is determined to be impaired (e.g., vehicles must slowly wade through shallow water due to accumulation). In the algorithm implementation, this is specifically manifested by increasing the weight of the edges connected to the node (i.e., simulating a significant increase in the actual travel time).

[0080] Safe passage simulation: When ERP > Th2, it is determined that the effective defense capability of the road segment is still within the safe range, and the initial weights of the node and its connected edges remain unchanged.

[0081] 3) Cascaded Failure Simulation and Vulnerable Node Identification: After triggering "removal" or "weight increase" operations on one or more initial test nodes according to the above rules, the original topological balance of the road network is disrupted. The system will simulate the redistribution of traffic flow to surrounding road networks due to local obstruction. Graph theory algorithms are used to recalculate the shortest paths and connectivity efficiency between all network nodes, thereby deriving the decline in the overall connectivity performance indicators (such as global efficiency) of the remaining network under the cascade effect (i.e., performance degradation rate). If the failure test of a single initial test node causes the connectivity performance degradation rate of the entire network to be greater than or equal to a preset sensitivity threshold, then that initial test node is explicitly marked as a critical vulnerable node. This judgment logic effectively identifies core bottleneck road sections that play an irreplaceable role in regional disaster prevention connectivity.

[0082] Step 5: Generate and execute traffic control instructions: Based on the identified critical vulnerable nodes and their corresponding engineering adaptability gap indices, determine the specific risk level, and then generate corresponding physical control signals (control instructions). The generated physical control signals are sent to the on-site traffic control equipment on the corresponding road sections via the communication network to drive the traffic control facilities to perform physical actions. By changing the physical traffic state of locally vulnerable road sections, the nonlinear transfer of traffic flow and the risk of road network cascading failure are proactively prevented.

[0083] Specifically, the physical control signals mainly include blocking signals and current limiting signals. For any critical vulnerable node, based on the comparison between its calculated effective return period (ERP) and preset thresholds (Th1, Th2), the system executes the following differentiated dynamic control strategies:

[0084] 1) Extremely High Risk Blocking and Control: When ERP ≤ Th1, it indicates that the effective disaster prevention and mitigation capability of the road segment corresponding to the vulnerable node has substantially decreased to the first critical threshold or below, and the system determines that the road segment faces an extremely high risk of water damage or physical interruption. At this time, the system generates the blocking signal and sends the signal to the traffic control equipment on site. For example, it drives the intelligent barrier gate (PLC) based on programmable logic controller to automatically execute the lowering action, or controls the lane indicator matrix to display a red cross (no entry status) to execute a forced physical closure of the lane or the entire road segment.

[0085] 2) Medium- and High-Risk Traffic Restriction Management: When Th1 < ERP ≤ Th2, it is determined that the traffic capacity of the road segment corresponding to the vulnerable node is impaired, but has not yet reached the extreme state of complete blockage. At this time, the system generates the aforementioned traffic restriction signal. For example, the variable message signs (VMS) or intelligent speed limit signs at the scene display the reduced speed limit adjustment value to the driver (such as dynamically lowering the legal speed limit from 120km / h to 80km / h or lower), and can simultaneously issue relevant weather warning information to delay the formation of cascading congestion by reducing traffic density and speed.

[0086] 3) Maintaining a safe state: When ERP > Th2, it is determined that the effective defense capability of this road segment under the current weather conditions is still within the safe threshold range. The system does not generate additional physical intervention signals and controls the relevant traffic facilities on site to maintain the normal traffic status of this road segment.

[0087] This invention also provides a dynamic management and control system for transportation infrastructure based on non-stationary climate disaster-causing factors. Specifically, the system includes interconnected data acquisition modules, dynamic model construction modules, gap index calculation modules, cascaded failure simulation modules, and traffic control execution modules.

[0088] Data Acquisition Module: Used to perform the basic data acquisition operations in step 1 above. Specifically, it is configured to acquire structured vector data of the target area's transportation network (including construction year and design level attributes), and acquire historical meteorological grid data of the target area over a continuous time period.

[0089] Dynamic model construction module: Communicatively connected to the data acquisition module, used to perform the calculation operation in step 2 above. Specifically, it is configured to receive the historical meteorological grid data and extract the meteorological time series, establish and utilize a non-stationary extreme value statistical model, fit the evolution relationship of the distribution parameters of the series over time, and finally calculate the spatial distribution data of precipitation extreme values ​​of the target area at the current moment.

[0090] The gap index calculation module is communicatively connected to the dynamic model construction module and is used to perform the comparison calculation operation in step 3 above. Specifically, it is configured to obtain the corresponding historical design baseline precipitation based on the construction year attribute of each road segment, calculate the effective return period (ERP) of the historical design baseline at the current moment using the aforementioned non-stationary extreme value statistical model and spatial distribution data of precipitation extreme values, and compare the effective return period with the legally mandated historical design standard return period to output the engineering adaptability gap index of the target road segment.

[0091] Cascaded Failure Simulation Module: Communicatively connected to the Gap Index Calculation Module, it is used to perform the network simulation operation in step 4 above. Specifically, it is configured to construct a grid connectivity graph reflecting physical connections based on vector data of the traffic network; using the Effective Return Period (ERP) as the failure judgment parameter for accessible grid nodes, and combining it with a preset first critical threshold and second early warning threshold, it simulates network paralysis and traffic restriction by dynamically removing nodes or increasing edge weights in the grid connectivity graph, thereby calculating the decline in the overall network connectivity performance index and identifying key vulnerable nodes.

[0092] Traffic control execution module: Communicatively connected to the cascaded failure simulation module, used to execute the control command issuance operation in step 5 above. Specifically, it is configured to generate corresponding blocking signals or flow restriction signals and other control commands based on the risk level of the critical vulnerable nodes, and send the control commands to the traffic control equipment (such as intelligent gate PLC, VMS variable message sign, etc.) on the corresponding road section via the communication network to drive the relevant facilities to perform physical actions such as lane closure or speed limit adjustment.

[0093] Example 1

[0094] This embodiment discloses a dynamic management and control method for transportation infrastructure based on non-stationary climate disaster-causing factors, including the following steps:

[0095] Step 1: Obtain basic data: In this embodiment, vector data (Shapefile format) of the highway network in the target area is obtained from a Geographic Information System (GIS). The attribute table of the data contains the "year of opening to traffic" of each road segment (e.g., a road segment was built in 1995 and opened to traffic in 1998) and the design grade information of the road segment; at the same time, CN05.1 meteorological grid data of the region from 1980 to 2024 is obtained, with a spatial resolution of not less than 0.25°×0.25° and a temporal resolution of daily.

[0096] Step 2: Constructing a Non-Stationary Extreme Value Statistical Model: For each meteorological grid point, extract the time series of its specific disaster-causing factors. In this example, it is the annual maximum daily precipitation Rx1day series. Considering the non-stationarity of climate change, this example uses a generalized extreme value distribution (GEV) model that includes time covariates. The cumulative distribution function of the model is constructed as follows:

[0097] (1)

[0098] In the formula: z is the extreme value of precipitation; σ is the scale parameter; For shape parameters; For position parameters.

[0099] Among them, position parameters Set as a linear function of time t:

[0100] (2)

[0101] Where α is the intercept parameter, which characterizes the baseline intensity of the disaster-causing factor; β is the slope parameter, which characterizes the linear trend of the intensity of the disaster-causing factor changing with time; σ, α, and β are all parameters to be estimated.

[0102] To ensure the numerical stability and convergence of the parameter estimates, this embodiment normalizes the time variable t, linearly mapping the original year sequence to the [0,1] interval, instead of directly using absolute year values. After solving the above parameters using the maximum likelihood estimation (MLE) method, if β > 0, it objectively indicates that the extreme precipitation intensity in this region is increasing over time. Based on this solved model, the theoretical intensity threshold for any recurrence period event occurring at this grid point at the current time is calculated, thereby generating the spatial distribution data of precipitation extremes for the current year.

[0103] Step 3: Calculate the engineering adaptability gap index: This step aims to accurately quantify the adaptability gap between the physical road network and dynamic weather, such as... Figure 2 As shown, the specific process is as follows:

[0104] First, based on the "year of opening to traffic" attribute, the entire network is deconstructed into different incremental sets (e.g., set A consists of roads built in the 1990s, set B consists of roads built in the 2010s); second, for road segments in set A, the historical design benchmarks corresponding to their construction period are retrieved (i.e., Figure 2 (Historical baseline A); then, using the GEV model, the actual probability of the precipitation corresponding to this design flood frequency occurring in the current year (T) is calculated. For example, a certain road section is designed to withstand a once-in-a-century flood event. However, given climate drift, the probability of the same intensity of precipitation occurring in 2024 has changed. (i.e., a 40-year return period); finally, the engineering adaptability gap index is calculated, and the effective return period is defined in this embodiment. If ERP < 100, it indicates an adaptation gap. Figure 2 The results are clearly illustrated in the text. The old road network built in the 1990s faces a huge adaptation gap (Gap 1) at present due to its early construction and low baseline. In contrast, the road network built in the 2010s, although also affected by climate non-stability, has a relatively small adaptation gap (Gap 2) due to its higher historical baseline B and shorter service life.

[0105] Step 4: Construct a connected topology model and perform cascading failure simulation: This embodiment uses a grid-based dual topology construction method to transform the physical road network into a computable graph theory model and perform cascading failure simulation, such as... Figure 3 As shown, the specific process includes:

[0106] 1) Establish a physical topology model: Define a grid system with a preset resolution covering the target area, and map the structured vector segments of the traffic network onto this grid system. If an actual physical road segment exists within a grid, the grid is marked as an accessible grid. Subsequently, each accessible grid is abstracted as an independent node in the topology network, and edges are established between adjacent accessible grids based on the physical adjacency relationships of the vector road network in real space (e.g., using the 8-neighborhood connectivity rule), thereby constructing a grid connectivity graph that reflects the physical connectivity relationships of the real road network. The initial weight of each edge is set based on the actual physical distance of the corresponding grid road segment or the travel time under free-flow conditions.

[0107] 2) Setting Thresholds and Dynamic Failure Rules: The Effective Return Period (ERP) calculated in step 3 is used as the key indicator for failure determination of each accessible grid node. Specifically, this embodiment sets the first critical threshold Th1 to 50 years (representing the critical state of a road segment facing high-risk paralysis due to extreme disaster factors) and the second warning threshold Th2 to 80 years (representing the warning state of a road segment facing medium-risk damage and reduced traffic capacity). For each node in the grid connectivity graph, a corresponding topology change operation is forcibly triggered through numerical comparison:

[0108] Blockage failure simulation: When the ERP of a certain test node is ≤ 50, simulate the functional paralysis of the road section due to deep water accumulation or physical water damage, and remove the node and all its connected edges in the grid connectivity graph;

[0109] Damage-related traffic restriction simulation: When 50 < ERP ≤ 80 for a certain test node, the simulation shows that vehicles must slowly wade through the shallow water in this section of the road. The weight of the edges connected to this node is increased in the grid connectivity graph (i.e., the actual travel time is significantly longer).

[0110] Safe passage simulation: When the ERP of a certain test node is greater than 80, the effective defense capability of the simulated road segment is still within the safe range, and the initial weights of the node and its connected edges remain unchanged.

[0111] 3) Cascaded Failure Simulation and Vulnerable Node Identification: After triggering "removal" or "weight increase" operations on one or more initial test nodes according to the above rules, the original topological balance of the road network is disrupted. The system will simulate the redistribution of traffic flow to surrounding road networks due to local obstruction. Graph theory algorithms are used to recalculate the shortest paths and connectivity efficiency between all network nodes, thereby obtaining the decline rate of the overall connectivity performance index of the remaining network under the cascade effect (i.e., performance degradation rate). If the failure test of a single initial test node causes the connectivity performance degradation rate of the entire network to be greater than or equal to the preset sensitivity threshold, then that node is explicitly marked as a critical vulnerable node.

[0112] Step 5: Generate and execute traffic control instructions: Based on the identified critical vulnerable nodes and their corresponding engineering adaptability gap indices, the specific risk level is determined, and corresponding physical control signals are generated, including blocking signals and flow restriction signals. For any critical vulnerable node, based on the comparison between its calculated effective return period (ERP) and preset thresholds (Th1, Th2), the system executes the following differentiated dynamic control strategies:

[0113] When ERP≤50, that is, when the effective defense capability of the road section indicated by the Engineering Adaptability Gap Index drops below the first critical threshold, the system generates a blocking signal to drive the lane indicator lights or intelligent barrier gate (PLC) to perform lane closure actions.

[0114] When 50 < ERP ≤ 80, the system generates a current limiting signal to drive the variable message sign (VMS) to display the speed limit adjustment value;

[0115] When ERP > 80, the effective defense capability is determined to be within a safe range, the system does not generate an intervention signal, and maintains the normal traffic status of the road segment.

[0116] To verify the effectiveness of the control strategy of this invention, this embodiment constructs a highway network model with typical topological characteristics (containing 30 key nodes and 85 road segments) in the MATLAB simulation environment to simulate the operation of a regional traffic network. Regarding the experimental setup, an extreme precipitation scenario during the 2024 flood season is simulated. The original design standard for the road network is set to a 100-year return period; however, due to climate drift and differences in construction age, the effective return period (ERP) of some older road segments has decreased to 30-40 years. Regarding the evaluation index, this embodiment introduces the network global efficiency E as an evaluation index, and its calculation formula is as follows:

[0117] (5)

[0118] In the formula: This represents the total number of nodes in the grid. For nodes To the node The shortest path length between two points when they cannot be connected due to a road network break. Its reciprocal term is 0.

[0119] like Figure 4 As shown, this embodiment compares the following three network states:

[0120] The first type ( Figure 4 The left figure shows the original network (baseline state), where the road network is intact and traffic is smooth. The calculations show... .

[0121] The second type ( Figure 4 The middle image shows the control group (existing technology—passive response), which uses the traditional "post-disaster repair" model. When a road section encounters an obstacle exceeding its actual defense capacity (…), the road is then repaired. During rainfall, assuming physical water damage (such as landslides) occurs on a road segment, the edge is directly removed from the topology graph. Results show that removing critical nodes leads to topological breaks in the road network, requiring vehicles to detour extremely long distances or making the network inaccessible. Calculations show... This represents a 14.8% decrease compared to the benchmark.

[0122] The third type ( Figure 4 The right figure shows the experimental group (this invention—active control), which adopts the "pre-emptive active control" model. It identifies... For road sections where the edge is not removed, the traffic impedance weight is increased by 4 times (simulating road closure); identify For certain road sections, the weight is increased by 1.5 times (simulating speed limits). The results show that the road network remains fully connected (no nodes are disconnected), but the travel time on some paths increases. The calculated... This represents a 19.0% decrease compared to the benchmark.

[0123] The results show that although the numerical efficiency of the experimental group (0.0315) was slightly lower than that of the control group (0.0332), this precisely reflects the core technical value of this invention. The "high efficiency" of the control group is based on structural damage and physical fracture of the road network, which has a long recovery period and does not include implicit secondary road network congestion. In contrast, this invention, by actively sacrificing approximately 4.2% of marginal efficiency (19.0%-14.8%), successfully avoids the physical removal of nodes and preserves 100% of the physical integrity of the road network. This means that after the disaster ends, the road network managed by this invention can immediately resume normal traffic, exhibiting extremely high life-cycle resilience.

[0124] Example 2

[0125] This embodiment discloses a dynamic management and control system for transportation infrastructure based on non-stationary climate disaster-causing factors. The system includes interconnected data acquisition modules, dynamic model construction modules, gap index calculation modules, cascade failure simulation modules, and traffic control execution modules.

[0126] The data acquisition module is used to acquire historical meteorological grid data issued by publicly available meteorological data platforms in real time through API interfaces, and to acquire structured vector data of road networks in real time from transportation geographic information systems.

[0127] The dynamic model building module, which communicates with the data acquisition module, is mounted on a high-performance server node. This module has a built-in non-stationary GEV computing unit for high-speed extraction of meteorological time series, establishing a non-stationary extreme value statistical model and solving for relevant parameters, thereby generating spatial distribution data of precipitation extreme values ​​at the current moment.

[0128] The gap index calculation module is communicatively connected to the dynamic model construction module. This module has a built-in spatiotemporal matching engine, which is used to obtain the corresponding historical design benchmark precipitation based on the construction year attribute of each road segment, calculate the effective return period (ERP) of the historical design benchmark at the current moment, compare the effective return period with the historical design standard return period, and output the engineering adaptability gap index of the target road segment.

[0129] The cascading failure simulation module is communicatively connected to the gap index calculation module. This module is used to construct a digital twin of the road network in the server memory, that is, to map the vector road network into a grid connectivity dual topology model that reflects the physical connection relationship; and based on the input ERP as the node failure judgment index, it performs millisecond-level cascading failure inference simulation calculations in memory (including dynamically removing nodes or increasing the impedance weight of the connection edges), thereby quickly identifying the key vulnerable nodes that cause the overall network connectivity performance degradation.

[0130] The traffic control execution module is communicatively connected to the cascaded failure simulation module. This module is equipped with an industrial Ethernet communication interface and connects to the variable message sign (VMS) controller and the programmable logic controller (PLC) of the smart barrier gate at the corresponding road section via standard industrial control protocols such as TCP / IP. It is used to generate and issue physical control commands (including blocking signals and flow restriction signals) in real time based on the risk level of the critical vulnerable nodes, thereby driving the on-site traffic control equipment to change its physical operating state.

[0131] Finally, it should be noted that the above description is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred arrangement, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A dynamic management and control method for transportation infrastructure based on non-stationary climate disaster-causing factors, characterized in that, The method includes the following steps: Step 1: Obtain basic data: Obtain vector data of the traffic network in the target area, including the construction year attributes and design level information of each road segment in the traffic network; at the same time, obtain historical meteorological grid data of the target area in a continuous time period. Step 2: Construct a non-stationary extreme value statistical model: Based on the acquired historical meteorological grid data, extract the meteorological time series with the maximum daily precipitation as a specific climate disaster factor; establish a non-stationary extreme value statistical model, and use the non-stationary extreme value statistical model to fit the evolution relationship of the distribution parameters of the meteorological time series over time, and calculate the spatial distribution data of precipitation extreme values ​​in the target area at the current moment. Step 3: Calculate the Engineering Adaptability Gap Index: Based on the construction year of each road segment, the transportation network is divided into multiple road network subsets, each corresponding to a specific construction period. For any target road segment within a road network subset, historical design baseline precipitation corresponding to its construction period and design grade is retrieved. Simultaneously, based on the spatial coordinates of the target road segment, its spatial location is matched with the spatial distribution data of current precipitation extremes to calculate the Engineering Adaptability Gap Index, which characterizes the current degree of climate adaptability degradation. The specific calculation process includes: 1) Calculate the actual occurrence probability: Obtain the historical design reference precipitation of the target section in the construction period, denoted as Z design , and substitute it into the non-stationary extreme value statistical model as the input threshold value, calculate the actual occurrence probability P design of the historical design reference precipitation at the current time T actual , and the calculation formula is: P actual =1-G(Z design ,T) (3) In the formula, G(Z) design (T) represents the cumulative distribution function value of a non-stationary extreme value statistical model at a given current time T; 2) Calculate the effective return period: Based on the actual occurrence probability mentioned above, define and calculate the effective return period (ERP) of the target road segment under the current climate conditions: ERP = 1 / P actual (4) 3) Calculate the engineering adaptability gap index: Extract the historical design standard return period specified in the highway engineering technical standards during the construction period of the target road section, calculate the attenuation of the effective return period (ERP) relative to the historical design standard return period, and use the attenuation as the engineering adaptability gap index; Step 4: Construct a connected topology model and perform cascading failure simulation: Using a grid-based dual topology construction method, a connected topology model reflecting the physical connection relationships of the road network is constructed, and cascading failure simulation is performed. The specific process includes: 1) Establish a physical topology model: Define a grid system with a preset resolution covering the target area; map the structured vector segments of the traffic network to the grid system. If an actual physical road segment exists within a grid, then the grid is marked as an accessible grid; abstract each accessible grid as a node in the topology network, and establish connections between adjacent accessible grids based on the physical adjacency relationships of the vector road network in real space, thereby constructing a grid connectivity graph; the initial weight of each connection is set based on the physical distance of the corresponding grid road segment or the travel time under free-flow conditions. 2) Setting thresholds and dynamic failure rules: The calculated effective return period is used as the failure judgment index for each accessible grid node, and cascade failure simulation calculation is performed on the grid connectivity graph; a first critical threshold Th1 and a second early warning threshold Th2 are set, where Th1 < Th2; for each node in the network, its dynamic change rules in the topology graph are determined by numerical comparison. 3) Cascaded Failure Simulation and Vulnerable Node Identification: After adjusting one or more initial test nodes according to the above rules, the original topological balance of the road network is broken. The system will simulate the redistribution of traffic flow to the surrounding road network due to local obstruction. The shortest path and connectivity efficiency between all network nodes are recalculated using graph theory algorithms, thereby obtaining the decline rate of the overall connectivity performance index of the remaining network under the cascade effect, i.e., the performance degradation rate. If the failure test of an initial test node causes the connectivity performance degradation rate of the entire network to be greater than or equal to the preset sensitivity threshold, then the node is marked as a critical vulnerable node. Step 5: Generate and execute traffic control instructions: Based on the identified critical vulnerable nodes and their corresponding engineering adaptability gap indices, generate corresponding physical control signals, i.e., control instructions; send the control instructions to the on-site traffic control equipment of the corresponding road section to drive the traffic control facilities to perform physical actions.

2. The method for dynamic management and control of transportation infrastructure based on non-stationary climate disaster-causing factors according to claim 1, characterized in that, The vector data of the transportation network mentioned in step 1 comes from a transportation geographic information system or a transportation network database; the historical meteorological grid data comes from public meteorological data platforms, including the National Meteorological Science Data Center, the ERA5 reanalysis dataset issued by the European Centre for Medium-Range Weather Forecasts, or the MSWEP multi-source fusion precipitation dataset; the historical meteorological data spans more than or equal to 30 years, the grid size corresponding to the spatial resolution is no greater than 0.25°×0.25°, and the temporal resolution is daily.

3. The method for dynamic management and control of transportation infrastructure based on non-stationary climate disaster-causing factors according to claim 1, characterized in that, The non-stationary extreme value statistical model described in step 2 adopts the generalized extreme value distribution model, and its cumulative distribution function is constructed as follows: （1） In the formula: z is the extreme value of precipitation; σ is the scale parameter; For shape parameters; For position parameters; Among them, position parameters Set as a linear function of time t: （2） Where α is the intercept parameter, which characterizes the baseline intensity of the disaster-causing factor; β is the slope parameter, which characterizes the linear trend of the intensity of the disaster-causing factor changing with time; σ, α, and β are all parameters to be estimated; The specific process for calculating the spatial distribution data of precipitation extreme values ​​in the target area at the current time is as follows: the above-mentioned parameters to be estimated are solved by the maximum likelihood estimation method, the solved parameters are substituted into the generalized extreme value distribution model, the theoretical intensity threshold of each spatial grid in the target area at the current time for any return period is calculated, and the spatial distribution data of precipitation extreme values ​​at the current time is generated.

4. The method for dynamic management and control of transportation infrastructure based on non-stationary climate disaster-causing factors according to claim 1, characterized in that, Step 4, which describes determining the dynamic change rules of each node in the network through numerical comparison, specifically includes: Blockage failure simulation: When ERP≤Th1, it is determined that the road segment corresponding to the node has been physically blocked or completely paralyzed due to climate disaster factors; in the algorithm implementation of the grid connected graph, this is specifically manifested as removing the node and all its connected edges. Damage-related traffic restriction simulation: When Th1 < ERP ≤ Th2, it is determined that the traffic capacity of the road segment corresponding to the node is damaged; in the algorithm implementation, this is specifically manifested by increasing the weight of the edges connected to the node. Safe passage simulation: When ERP > Th2, it is determined that the effective defense capability of the road segment is within the safe range, and the initial weights of the node and its connected edges remain unchanged.

5. The method for dynamic management and control of transportation infrastructure based on non-stationary climate disaster-causing factors according to claim 1, characterized in that, Its features are, The specific process described in step 5, which involves generating corresponding physical control signals, i.e., control instructions, based on the identified critical vulnerable nodes and their corresponding engineering adaptability gap indices, is as follows: The physical control signals include blocking signals and flow limiting signals; for any critical vulnerable node, a corresponding physical control signal is generated based on the comparison between its calculated effective return period (ERP) and the first critical threshold Th1 and the second early warning threshold Th2. 1) Extremely high risk blocking control: When ERP≤Th1, it indicates that the effective disaster prevention and defense capability of the road section corresponding to the critical vulnerable node has dropped to the first critical threshold and below. The system determines that the road section faces an extremely high risk of water damage or physical interruption. At this time, the system generates a blocking signal. 2) Medium- and high-risk traffic restriction and control: When Th1 < ERP ≤ Th2, it is determined that the traffic capacity of the road segment corresponding to the vulnerable node is damaged, but has not yet reached the limit state of complete blockage. At this time, the system generates the traffic restriction signal. 3) Safety status maintenance: When ERP > Th2, it is determined that the effective defense capability of the road section under the current weather conditions is still within the safety threshold range, and the system does not generate physical control signals.

6. A dynamic management and control system for transportation infrastructure based on non-stationary climate disaster-causing factors, the system being used to implement the dynamic management and control method for transportation infrastructure based on non-stationary climate disaster-causing factors as described in any one of claims 1 to 5, characterized in that, The system includes interconnected data acquisition modules, dynamic model building modules, gap index calculation modules, cascade failure simulation modules, and traffic control execution modules. The data acquisition module is used to acquire structured vector data of the traffic network in the target area and historical meteorological grid data of the target area over a continuous time period. The dynamic model construction module is communicatively connected to the data acquisition module, and is used to receive the historical meteorological grid data and extract the meteorological time series, establish and use a non-stationary extreme value statistical model to fit the evolution relationship of the distribution parameters of the meteorological time series over time, and calculate the spatial distribution number of precipitation extreme values ​​in the target area at the current moment. The gap index calculation module is communicatively connected to the dynamic model construction module. It is used to obtain the corresponding historical design benchmark precipitation based on the construction year attribute of each road segment, calculate the effective return period of the historical design benchmark at the current moment using the aforementioned non-stationary extreme value statistical model and precipitation extreme value spatial distribution data, and compare the effective return period with the legally mandated historical design standard return period to output the engineering adaptability gap index of the target road segment. The cascaded failure simulation module is communicatively connected to the gap index calculation module. It is used to construct a grid connectivity graph reflecting physical connection relationships based on the vector data of the traffic network. The effective return period is used as the failure judgment parameter for the reachable grid nodes. Combined with the preset first critical threshold and second early warning threshold, the module simulates network paralysis and traffic restriction by dynamically removing nodes or increasing the edge weights in the grid connectivity graph. Then, it calculates the decline of the overall network connectivity performance index and identifies key vulnerable nodes. The traffic control execution module is communicatively connected to the cascaded failure simulation module. It is used to generate corresponding blocking signals or flow restriction signals and other control instructions based on the risk level of the critical vulnerable nodes, and send the control instructions to the traffic control equipment on the corresponding road section via the communication network to drive the relevant facilities to perform physical actions such as lane closure or speed limit adjustment.

Images