An evolution modeling and prediction method and system for extreme risk coupling of an aviation hub network

Abstract

The application discloses an evolution modeling and prediction method and system for extreme risk coupling of an aviation hub network, and the method comprises the following steps: acquiring static section data of an aviation transportation system; constructing a space coupling dynamics model according to the static section data, wherein each airport node in the space coupling dynamics model corresponds to an attraction weight, and the attraction weight represents the co-mobility characteristics of each airport node under an extreme load state; using a maximization minimization algorithm to inverse the attraction weight; based on the inverted attraction weight, acquiring a space core parameter group through a Markov chain Monte Carlo algorithm, and obtaining regional congestion co-mobility early warning information based on the space core parameter group. The application can accurately simulate the dynamic growth logic of a multi-center airport group, and can more quantitatively identify the cascade load pressure distribution within the space dependence under the extreme scenario of the failure of a specific core hub.

Description

Technical Field

[0001] This invention relates to the field of air transport planning technology, specifically to an evolutionary modeling and prediction method and system for extreme risk coupling in air hub networks. Background Technology

[0002] As a typical spatially embedded directed network, the evolutionary dynamics of aviation networks are driven by both geospatial constraints and the need for topological growth. For a long time, academia and engineering have primarily used models based on the preferential attachment (PA) mechanism to explain the formation of aviation hubs and their scale-free characteristics. The core of this theory lies in the "rich get richer" mechanism, which can well fit the power-law distribution characteristics of airport in-degree and out-degree.

[0003] However, when addressing systemic risk assessments in civil aviation, traditional PA models and their simple geographical extension schemes (such as simply introducing a distance decay function) have the following technical limitations:

[0004] First, it ignores the physical connections of extreme comovements. In actual civil aviation operations, geographically proximate airport clusters (such as JFK, EWR, and LGA in the New York metropolitan area, or the Beijing-Tianjin-Hebei and Yangtze River Delta airport clusters in my country) often exhibit strong spatial coupling characteristics. This coupling is particularly pronounced during tail events such as extreme weather, airspace control, or major public health emergencies, leading to simultaneous significant reductions in routes, traffic saturation, or systemic paralysis at multiple surrounding hub airports. Traditional independent incremental models assume that node growth is probabilistically relatively isolated, failing to characterize this phenomenon of extreme spatial dependence.

[0005] Second, there is a challenge in decoupling static snapshots from dynamic evolution parameters. Existing network dynamics analysis heavily relies on complete historical evolutionary sequence data, but in practice, only a static topological snapshot at a specific time dimension can often be obtained. In static data, the implicit attraction weights caused by geospatial factors are deeply coupled with the explicit degree weights derived from topological structure. Existing techniques struggle to accurately separate the hidden spatially relevant parameters—spatial dependence range ξ and spatial association strength γ—from a single snapshot in the absence of an evolutionary path, resulting in extremely vague definitions of risk transmission boundaries in the model.

[0006] Third, the prediction of regional cascading failures suffers from significant accuracy bottlenecks. Due to the lack of effective modeling of spatial stochastic weight fields, existing prediction schemes often simplify regional cascading failures to random fluctuations or simple linear propagation. This approach ignores the strong correlation at the tails of the distribution, resulting in overly optimistic model outputs when predicting the synchronous stress-bearing capacity of surrounding nodes after the failure of a hub. This fails to provide a scientific and reliable basis for decision-making by management departments such as the Civil Aviation Administration of China when formulating cross-regional coordinated scheduling, alternate airport capacity planning, and dynamic allocation of air traffic rights. Summary of the Invention

[0007] This invention provides an evolutionary modeling and prediction method and system for extreme risk coupling in aviation hub networks, which can accurately simulate the dynamic growth logic of multi-center airport clusters and, moreover, quantitatively identify the cascading load pressure distribution within the spatial dependency zone under extreme scenarios of failure of specific core hubs.

[0008] In a first aspect, the present invention provides an evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks, the method comprising:

[0009] The data acquisition steps involve acquiring static cross-sectional data of the air transport system.

[0010] The model construction steps are as follows: Based on the static cross-sectional data, a spatial coupling dynamics model is constructed. In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, and the attraction weight represents the co-movement characteristics of each airport node under extreme load conditions.

[0011] The inversion step involves using a maximization-minimization algorithm to invert the attraction weights;

[0012] The solution process involves obtaining a spatial core parameter group based on the inverted attraction weights using the Markov chain Monte Carlo algorithm, and then obtaining regional congestion coordination early warning information based on the spatial core parameter group.

[0013] In some embodiments of the present invention, the spatial coupling dynamics model is characterized as follows:

[0014]

[0015] In the formula, W i W j Let D be the attraction weights of airport node i and airport node j, respectively. i D j Let be the observed degree of airport node i and airport node j, respectively; δ be the initial attraction constant; α be the nonlinear exponent controlling the power-law characteristic of the network; and V be...

[0016] Wherein, the attraction weight W of airport node i i The attraction weight W of airport node j j The logarithmic transformation of follows a multivariate normal distribution, and the corresponding covariance matrix Σ follows a spatial exponential decay law:

[0017]

[0018] In the formula, ξ represents the spatially dependent correlation scale, γ represents the intensity of spatial fluctuations, and xi and xj are the location information of airport nodes i and j, respectively.

[0019] In some embodiments of the present invention, the step of inverting the attraction weights using a maximization-minimization algorithm includes:

[0020] Initialize the attraction weight ;

[0021] Construct proxy function ,

[0022] Update attraction weight Until it converges.

[0023] In some embodiments of the present invention, the spatial core parameter set includes a spatial scale parameter γ and a distance attenuation parameter ξ. The acquisition of the spatial core parameter set based on the inverted attraction weights using a Markov chain Monte Carlo algorithm includes:

[0024] A hierarchical Bayesian model is established, and non-information priors are set for the distance decay parameter ξ.

[0025] The spatial scale parameter γ is updated using Gibbs sampling, and the distance decay parameter ξ is updated using random walk MH sampling.

[0026] After the warm-up period, the posterior mean of the parameters is obtained and used as the final spatial core parameter group.

[0027] In some embodiments of the present invention, after obtaining the space core parameter group, the method further includes:

[0028] The space core parameter group was verified using the F-madogram algorithm;

[0029] If the verification results indicate that the prediction accuracy meets the preset conditions, regional congestion coordination early warning information is obtained based on the spatial core parameter group.

[0030] If the verification result indicates that the prediction accuracy does not meet the preset conditions, the inversion step and the solution step are repeated.

[0031] In some embodiments of the present invention, the verification of the spatial core parameter group using the F-madogram algorithm includes:

[0032] A synthetic network is generated using the aforementioned space core parameter group;

[0033] Calculate the F-madogram curve between the computational synthesis network and the original snapshot;

[0034] If the tail fit in the F-madogram curve exceeds a threshold, the prediction accuracy is determined to meet a preset condition.

[0035] In some embodiments of the present invention, after acquiring static cross-sectional data of the air transport system, the method further includes:

[0036] Based on the static cross-sectional data, a spatial distance matrix is ​​constructed, and the distances between each airport node are obtained according to the spatial distance matrix.

[0037] The in-degree and out-degree distributions of each airport node are statistically analyzed, and the logarithmic binning method is used to process the in-degree and out-degree distributions.

[0038] Secondly, the present invention also provides an evolutionary modeling and prediction system for extreme risk coupling in aviation hub networks, the system comprising:

[0039] The data acquisition module is used to acquire static cross-sectional data of the air transport system;

[0040] The model building module is used to construct a spatial coupling dynamics model based on the static cross-sectional data. In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, and the attraction weight represents the co-movement characteristics of each airport node under extreme load conditions.

[0041] The inversion module is used to invert the attraction weights using a maximization-minimization algorithm;

[0042] The solution module is used to obtain the spatial core parameter group based on the inverted attraction weights using the Markov chain Monte Carlo algorithm, and to obtain regional congestion coordination early warning information based on the spatial core parameter group.

[0043] In the evolutionary modeling and prediction method and system for extreme risk coupling in aviation hub networks provided by this invention, the Minimize-Maximize (MM) algorithm is used to efficiently reconstruct the implicit spatial attraction weights of airport nodes modulated by the geographical environment from static flight path cross-section data. Subsequently, the Geographical Distance Attenuation Parameter and Spatial Correlation Scale are inferred through the Markov Chain Monte Carlo (MCMC) algorithm. This invention can not only accurately simulate the dynamic growth logic of multi-center airport clusters, but also quantitatively identify the cascading load pressure distribution within the spatial dependency zone under extreme scenarios of failure of specific core hubs. This provides scientific quantitative decision support for the resilience assessment of civil aviation systems, air rights resource allocation strategies, and cross-regional emergency dispatch under sudden disasters. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating the evolution modeling and prediction method for extreme risk coupling in aviation hub networks provided in this embodiment of the invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0049] The use of "applies to" or "configured to" in this invention implies an open and inclusive language, which does not exclude the applicability to or configuration to devices performing additional tasks or steps. Additionally, the use of "based on" implies openness and inclusivity, because processes, steps, calculations, or other actions "based on" one or more conditions or values ​​may in practice be based on additional conditions or values ​​beyond those conditions.

[0050] In this invention, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0051] The following describes, with reference to the accompanying drawings, an evolutionary modeling and prediction method and system for extreme risk coupling in aviation hub networks provided by embodiments of the present invention.

[0052] like Figure 1 As shown in the figure, this invention provides an evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks. The method includes the following steps:

[0053] S101 Data Acquisition Steps: Acquire static cross-sectional data of the air transport system.

[0054] The static cross-sectional data includes the airport set V, the existing route connection edges E, and the geographic coordinates (x, y) of each airport.

[0055] The S102 model construction steps involve constructing a spatially coupled dynamic model based on the static cross-sectional data.

[0056] In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, which represents the co-movement characteristics of each airport node under extreme load conditions.

[0057] In step S103, the attraction weights are inverted using a maximization-minimization algorithm.

[0058] In the S104 solution step, based on the inverted attraction weights, the spatial core parameter group is obtained through the Markov chain Monte Carlo algorithm, and regional congestion coordination early warning information is obtained based on the spatial core parameter group.

[0059] The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks provided in this invention utilizes the Minimize-Maximize (MM) algorithm to efficiently reconstruct the implicit spatial attraction weights of airport nodes modulated by the geographical environment from static flight path cross-section data. Subsequently, the Markov Chain Monte Carlo (MCMC) algorithm is used to infer the geographical distance decay parameter and spatial correlation scale. This invention not only accurately simulates the dynamic growth logic of multi-center airport clusters but also quantifies and identifies the cascading load pressure distribution within spatial dependence zones under extreme scenarios of specific core hub failures. This provides scientific quantitative decision support for resilience assessment of civil aviation systems, air rights allocation strategies, and cross-regional emergency dispatching under sudden disasters.

[0060] In some embodiments of the present invention, each airport node in the spatial coupling dynamics model is represented by a node growth probability model. This node growth probability model not only considers the explicit node degree but also introduces an implicit spatial weight field. For the route network, the probability of node i acquiring a new connection is:

[0061] ;

[0062] In the formula, W i W j Let D be the attraction weights of airport node i and airport node j, respectively. i D j Let be the observed degree of airport node i and airport node j, respectively; δ be the initial attraction constant; α be the nonlinear exponent controlling the power-law characteristic of the network; and V be the set of airport nodes.

[0063] Wherein, the attraction weight W of airport node i i The attraction weight W of airport node j j The logarithmic transformation of follows a multivariate normal distribution, thus forming a log-normal random field. The covariance matrix Σ of this log-normal random field follows a spatial exponential decay law.

[0064] ;

[0065] In the formula, ξ represents the distance attenuation parameter, γ represents the spatial scale parameter, and x i and x j These represent the location information of airport nodes i and j, respectively. This model quantifies the transmission efficiency of geospatial risks to extreme risks through spatial covariance structure, and characterizes the co-movement risk coupling characteristics of geographically adjacent hubs under extreme disturbances.

[0066] The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks provided in this invention, by introducing the above-mentioned log-normal random field assumption and covariance matrix space construction rules, for the first time realizes the deconstruction of the explicit degree weight and implicit spatial attraction weight of nodes under static snapshot data.

[0067] In some embodiments of the present invention, addressing the technical challenge of highly coupled node parameters in the likelihood function of aviation network evolution, which is difficult to solve directly, the present invention constructs a surrogate function and utilizes iterative lower bound maximization to achieve parameter inversion, specifically including the following steps:

[0068] Initialize the attraction weight .

[0069] Let L(W) be the log-likelihood function of the observed flight route network under a static snapshot. For a network with n airport nodes, the core challenge of its likelihood function lies in the logarithmic summation term:

[0070] ;

[0071] In the formula, k i Let i be the total number of observed edges. This is the topological attraction term. Since W... j The coupling inside the logarithmic summation sign of the second term results in a non-sparse Hessian matrix, making direct differentiation computationally extremely complex.

[0072] This invention utilizes the convexity of the negative logarithmic function and Jensen's inequality to estimate the value W in the m-th iteration. (m) Construct a proxy function at the location.

[0073] Introducing auxiliary weight variables Defined as the contribution percentage of the attractiveness of each node at the current iteration point:

[0074] .

[0075] Obviously Based on the concavity of the logarithmic function, for any W > 0, the following inequality holds:

[0076] .

[0077] Therefore, this invention constructs a surrogate function Q(W|W) for the original likelihood function L(W). (m) ):

[0078] ;

[0079] In the formula, C is a constant term that is independent of the parameter W to be determined.

[0080] Furthermore, by using Jensen's inequality to bound the logarithmic summation term, we define the closed-form expression of the lower bound function as:

[0081]

[0082] Among them, auxiliary coefficient This characterizes the proportion of node i's contribution to the overall network attractiveness under the current weight estimate:

[0083] .

[0084] The aforementioned lower bound function possesses the following strict mathematical properties, which constitute the technical support for the stability of the algorithm in this system:

[0085] Tangency: To ensure consistency of the iteration start point;

[0086] Global: This ensures that the algorithm remains within the lower envelope of the original likelihood surface throughout the optimization process.

[0087] Monotonic convergence: By maximizing this lower bound, the original likelihood function can be induced to satisfy... .

[0088] By taking the first derivative of the lower bound function, this invention realizes the parameter W i The decoupling update formula:

[0089]

[0090] The above decoupling update formula directly provides a closed-form solution for maximizing the lower bound, which significantly reduces the computational cost of large-scale aviation hub network inversion.

[0091] By updating attraction weights This process continues until convergence, reflecting the distribution of the pure implicit attraction derived from geographical location after removing the influence of degree.

[0092] The maximization-minimization algorithm provided in this invention can quickly invert the attraction weight vector W using only static flight path snapshots without relying on historical evolution sequences. Through iteration, the model estimates achieve statistical consistency with the observation degree distribution, effectively separating the implicit contribution of geographical location to node growth. Furthermore, the MM algorithm utilizes a surrogate function to construct the iterative lower bound, avoiding complex matrix inversion operations.

[0093] In some embodiments of the present invention, the spatial core parameter set includes a spatial scale parameter γ and a distance attenuation parameter ξ. The acquisition of the spatial core parameter set based on the inverted attraction weights using a Markov chain Monte Carlo algorithm includes:

[0094] A hierarchical Bayesian model is established, and non-information priors are set for the distance decay parameter ξ.

[0095] The spatial scale parameter γ is updated using Gibbs sampling, and the distance decay parameter ξ is updated using random walk MH sampling. Here, γ reflects the overall regional volatility, and ξ characterizes how quickly congestion risk disappears with geographical distance.

[0096] After the warm-up period, the posterior mean of the parameters is obtained and used as the final spatial core parameter group.

[0097] Schematic, based on the inverted weights W, the spatial core parameter group {ξ, γ} is inferred using the Markov Chain Monte Carlo (MCMC) method under a hierarchical Bayesian architecture.

[0098] By utilizing the characteristics of conjugate prior distribution, posterior samples of the spatial scale parameter γ are efficiently extracted to serve as a benchmark for characterizing the volatility of overall regional risk.

[0099] For the distance decay parameter ξ, which has nonlinear characteristics, rejection sampling is performed using a random walk scheme to accurately characterize the physical boundary of risk decay with geographical distance.

[0100] Through this hierarchical inference technology, the present invention can pinpoint the sensitivity radius of the aviation network affected by the geographical environment, providing a core threshold reference for cross-regional scheduling.

[0101] In some embodiments of the present invention, after obtaining the space core parameter group, the method further includes:

[0102] The space core parameter group was verified using the F-madogram algorithm.

[0103] If the verification results indicate that the prediction accuracy meets the preset conditions, regional congestion coordination early warning information is obtained based on the spatial core parameter group.

[0104] If the verification result indicates that the prediction accuracy does not meet the preset conditions, the inversion step and the solution step are repeated.

[0105] In some embodiments of the present invention, the verification of the spatial core parameter group using the F-madogram algorithm includes:

[0106] A synthetic network is generated using the aforementioned space core parameter group.

[0107] Calculate the F-madogram curve between the computational synthesis network and the original snapshot.

[0108] It is understandable that for two random variables Z(s) and Z(s+h) defined at spatial locations s and s+h, the formula for their F-madogram curves is:

[0109] ;

[0110] In the formula, ν(h) represents the F-madogram value at a distance of h, reflecting the difference in cumulative probability between variables at two locations at a distance of h. E[·] is the mathematical expectation function, which represents the statistical average of all sample point pairs satisfying a distance of h. F(·) is the marginal cumulative distribution function; the range of F(Z(s)) is between [0,1]. Through the F(·) function, the original observed data (such as airport delay time, traffic, etc.) are mapped to the probability space. This rank transformation eliminates the influence of the original data dimensions, focusing only on their relative positions in the distribution. Z(s) and Z(s+h) represent the random variables observed at spatial locations s and s+h, respectively. For example, in an aviation network, they can represent the saturation of the airport at location s and the airport at location s+h at a certain moment.

[0111] If the tail fit of the F-madogram curve exceeds a threshold, the prediction accuracy is determined to meet a preset condition. That is, if the two curves are highly fitted at the tail, it proves that the system has accurately captured the extreme risk coupling caused by geographical location. It can be understood that the value of v(h) is used to determine whether the two curves are highly fitted at the tail.

[0112] This invention verifies whether the system accurately reproduces the extreme co-movement characteristics of geographically proximate airports reaching their load peak simultaneously by comparing the consistency of real network snapshots and simulated synthetic networks on F-madogram curves (especially in the close-range interval), thereby providing statistical confidence support for subsequent risk warnings.

[0113] In some embodiments of the present invention, after acquiring static cross-sectional data of the air transport system, the method further includes:

[0114] Based on the static cross-sectional data, a spatial distance matrix H is constructed, and the distances between each airport node are obtained according to the spatial distance matrix. The static cross-sectional data includes the airport set V, existing flight route edges E, and the precise geographic coordinates (x, y) of each airport.

[0115] In some examples, the Euclidean distance between node pairs is calculated based on the spatial distance matrix H.

[0116] The in-degree and out-degree distributions of each airport node are statistically analyzed, and logarithmic binning is used to process the in-degree and out-degree distributions, thereby reducing the impact of statistical fluctuations on tail estimation.

[0117] The data results produced in the preprocessing stage in this embodiment of the invention support the execution of the subsequent core algorithm according to the following logic:

[0118] Spatial feature mapping: The spatial distance matrix obtained from preprocessing; it forms the underlying mapping of the subsequent spatial covariance matrix and directly constrains the posterior inference range of the spatial attenuation parameter in the MCMC process.

[0119] The observation-driven approach of MM inversion: The statistically obtained distribution of nodal degree of each airport is used as the explicit observation input of the maximization-minimization (MM) algorithm, which drives the surrogate function to achieve mathematical decoupling between topological contribution and implicit geographic attraction during the iteration process.

[0120] In some embodiments of the present invention, obtaining regional congestion coordination early warning information based on the spatial core parameter group includes:

[0121] When a simulated extreme load is input (such as simulating the shutdown of a core hub), the system automatically identifies a list of nodes in the surrounding high-correlation zone and their synchronization failure probability based on the inverted spatial covariance Σ.

[0122] By combining the α index with the spatial implicit weight W derived from the inversion, we can predict future new route pairs with high spatial correlation.

[0123] In some examples, after obtaining the spatial core parameter group {ξ,γ,Σ} and the implicit weights W, the system performs evaluation and prediction according to the following mathematical path:

[0124] 1. Calculation of risk synchronization probability under extreme load:

[0125] When a specific simulated extreme load is input (let the set of affected nodes be S), a conditional random field is constructed using the inverted spatial covariance matrix Σ. (The surrounding node set is also mentioned.) Risk response vector The posterior mean is given by the following formula:

[0126]

[0127] in, The elements of the matrix depend on the spatial correlation parameter ξ. If If the threshold for business security is exceeded, the node is determined to be in a high-correlation failure risk zone.

[0128] II. Prediction of new flight routes based on spatial evolution kernel:

[0129] The system constructs a comprehensive evolution kernel that includes spatial weights and topological features, and calculates the growth potential of any unconnected node pair (i,j):

[0130]

[0131] This formula corrects the constraint of physical distance on the probability of connection by the spatial core parameter ξ, and adjusts the nonlinear expansion capability of the hub node by the inverse α exponent, thereby achieving accurate prediction of highly spatially correlated route pairs.

[0132] In addition, this embodiment establishes the dynamic relationship between the core hub and surrounding airports through the following derivation logic:

[0133] Let the spatial core parameter group derived from the system be {ξ, γ, Σ}. When the implicit weights of the core hub s are abnormally perturbed by ∆ϵ... s At that time, based on the spatial random field covariance matrix Σ constructed according to the present invention, the state offset ∆ϵ of the surrounding satellite airport j facing the risk of synchronization failure. j It can be derived from the conditional expectation formula:

[0134] ;

[0135] Among them, the distance parameter ξ determines the effective radius of risk transmission across physical space.

[0136] Specifically, the steps include the following:

[0137] 1. Select the failed core node: Select the core airport node s to be simulated in the data perception layer;

[0138] 2. Constructing the perturbation operator: Based on the damage intensity of the core airport, define the offset perturbation amount ∆ϵ of its random field components. s ;

[0139] 3. Perform spatial correlation inference: Based on the covariance matrix Σ obtained by inversion, calculate the conditional posterior mean distribution of all network nodes under conditional perturbation;

[0140] 4. Output risk transmission results: Calculate the state offset ∆ϵ for each airport node j. j The variance confidence interval is used to identify a "high-risk synchronization failure list" that exceeds the threshold.

[0141] The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks provided in this invention, through the inversion of the spatial attenuation parameter ξ and the correlation intensity γ, can simulate how the risk, when a core hub (such as Beijing Capital International Airport) is damaged at a specific geographical coordinate, crosses physical flight routes and propagates to surrounding satellite airports through geospatial dependence. Under extreme weather conditions, major public health events, or large-scale airspace control, it assesses the probability of geographically adjacent airports simultaneously reaching their load extremes and their impact on the overall network's operational efficiency. This provides the Civil Aviation Administration of China with unprecedented quantitative scientific basis for formulating alternate airport planning, dynamically allocating cross-regional air rights resources, and coordinating traffic flow management, significantly improving the operational resilience of the air transport system.

[0142] On the other hand, embodiments of the present invention also provide an evolutionary modeling and prediction system for extreme risk coupling in aviation hub networks, the system comprising:

[0143] The data acquisition module is used to acquire static cross-sectional data of the air transport system;

[0144] The model building module is used to construct a spatial coupling dynamics model based on the static cross-sectional data. In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, and the attraction weight represents the co-movement characteristics of each airport node under extreme load conditions.

[0145] The inversion module is used to invert the attraction weights using a maximization-minimization algorithm;

[0146] The solution module is used to obtain the spatial core parameter group based on the inverted attraction weights using the Markov chain Monte Carlo algorithm, and to obtain regional congestion coordination early warning information based on the spatial core parameter group.

[0147] The evolution modeling and prediction system for extreme risk coupling in aviation hub networks provided in this embodiment corresponds to the evolution modeling and prediction method for extreme risk coupling in aviation hub networks provided in any of the above embodiments, and will not be described again here.

[0148] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0150] The above provides a detailed description of an evolutionary modeling and prediction method and system for extreme risk coupling in aviation hub networks, as provided by embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks, characterized in that, The method includes: The data acquisition steps involve acquiring static cross-sectional data of the air transport system. The model construction steps are as follows: Based on the static cross-sectional data, a spatial coupling dynamics model is constructed. In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, and the attraction weight represents the co-movement characteristics of each airport node under extreme load conditions. The inversion step involves using a maximization-minimization algorithm to invert the attraction weights; The solution process involves obtaining a spatial core parameter group based on the inverted attraction weights using the Markov chain Monte Carlo algorithm, and then obtaining regional congestion coordination early warning information based on the spatial core parameter group.

2. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to claim 1, characterized in that, The spatial coupling dynamics model is characterized as follows: ； In the formula, W i W j Let D be the attraction weights of airport node i and airport node j, respectively. i D j Let be the observed degree of airport node i and airport node j, respectively; δ be the initial attraction constant; α be the nonlinear exponent controlling the power-law characteristic of the network; and V be... Wherein, the attraction weight W of airport node i i The attraction weight W of airport node j j The logarithmic transformation of follows a multivariate normal distribution, and the corresponding covariance matrix Σ follows a spatial exponential decay law: ； In the formula, ξ represents the spatially dependent correlation scale, γ represents the intensity of spatial fluctuations, and xi and xj are the location information of airport nodes i and j, respectively.

3. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to claim 1, characterized in that, The process of inverting the attraction weights using a maximization-minimization algorithm includes: Initialize the attraction weight ; Construct proxy function , Update attraction weight Until it converges.

4. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to claim 1, characterized in that, The core spatial parameter set includes the spatial scale parameter γ and the distance decay parameter ξ. Based on the inverted attraction weights, the core spatial parameter set is obtained using a Markov chain Monte Carlo algorithm, including: A hierarchical Bayesian model is established, and non-information priors are set for the distance decay parameter ξ. The spatial scale parameter γ is updated using Gibbs sampling, and the distance decay parameter ξ is updated using random walk MH sampling. After the warm-up period, the posterior mean of the parameters is obtained and used as the final spatial core parameter group.

5. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to claim 1, characterized in that, After obtaining the core spatial parameter set, the method further includes: The space core parameter group was verified using the F-madogram algorithm; If the verification results indicate that the prediction accuracy meets the preset conditions, regional congestion coordination early warning information is obtained based on the spatial core parameter group. If the verification result indicates that the prediction accuracy does not meet the preset conditions, the inversion step and the solution step are repeated.

6. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to claim 5, characterized in that, The verification of the space core parameter group using the F-madogram algorithm includes: A synthetic network is generated using the aforementioned space core parameter group; Calculate the F-madogram curve between the computational synthesis network and the original snapshot; If the tail fit in the F-madogram curve exceeds a threshold, the prediction accuracy is determined to meet a preset condition.

7. The evolutionary modeling and prediction method for extreme risk coupling in aviation hub networks according to any one of claims 1 to 6, characterized in that, After acquiring static cross-sectional data of the air transport system, the method further includes: Based on the static cross-sectional data, a spatial distance matrix is ​​constructed, and the distances between each airport node are obtained according to the spatial distance matrix. The in-degree and out-degree distributions of each airport node are statistically analyzed, and the logarithmic binning method is used to process the in-degree and out-degree distributions.

8. An evolutionary modeling and prediction system for extreme risk coupling in aviation hub networks, characterized in that, The system includes: The data acquisition module is used to acquire static cross-sectional data of the air transport system; The model building module is used to construct a spatial coupling dynamics model based on the static cross-sectional data. In the spatial coupling dynamics model, each airport node corresponds to an attraction weight, and the attraction weight represents the co-movement characteristics of each airport node under extreme load conditions. The inversion module is used to invert the attraction weights using a maximization-minimization algorithm; The solution module is used to obtain the spatial core parameter group based on the inverted attraction weights using the Markov chain Monte Carlo algorithm, and to obtain regional congestion coordination early warning information based on the spatial core parameter group.

Images