Infrastructure cascade failure prediction method and system constructed in flood disaster scene

Abstract

The invention provides an infrastructure cascade failure prediction method and system for flood disaster scene construction, and the method comprises the steps: constructing an extreme rainstorm flood disaster scene through a numerical simulation method; constructing an independent network topology structure of the traffic infrastructure, the electric power infrastructure and the communication infrastructure; the method comprises the following steps: under the impact of extreme rainstorm, analyzing a cascade relationship and a disaster propagation mechanism of power-communication-traffic infrastructure, and constructing a cascade failure three-order network model of urban key infrastructure, the three-order network model comprising a three-order infrastructure network topology model, an extreme rainstorm flood model, a disaster propagation model and a cascade failure model; according to the submerging water depth of the urban infrastructure influenced by the extreme rainstorm, infrastructure failure conditions are set, infrastructure node failure prediction is carried out, and then independent infrastructure network failure prediction and urban key infrastructure cascade failure prediction are carried out. According to the invention, infrastructure cascade failure prediction can be carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of disaster chain prediction, and in particular to a method and system for predicting infrastructure cascading failures based on flood disaster scenarios. Background Art

[0002] Floods and waterlogging pose a systemic threat to urban infrastructure (electricity, communications, and transportation). As the lifeline of the city, the interdependence and coupling of these infrastructures mean that the failure of a single facility may trigger a cascading collapse across the system, such as power outages leading to communication and transportation paralysis.

[0003] Current research on disaster chains caused by floods and waterlogging has the following limitations: First, existing models often focus on a single infrastructure network (e.g., analyzing power grids or road networks independently), failing to fully model the complex dependencies between multiple types of infrastructure. Second, simulations of the spatiotemporal evolution of disasters lack refinement, making it difficult for traditional linear models to capture high-order dynamic propagation characteristics. Third, data-driven approaches suffer from oversmoothing, meaning that the local structural dependencies of the initial fault node are easily overwhelmed in global predictions, resulting in inaccurate predictions of cascading failures. Furthermore, existing technologies often employ static rule-based modeling, failing to balance control costs with predictive reliability. Summary of the Invention

[0004] In order to solve the technical problems existing in the above-mentioned prior art, the present invention provides a method and system for predicting infrastructure cascading failures based on flood disaster scenarios. The technical solution is as follows:

[0005] On the one hand, a method for predicting infrastructure cascading failures based on flood disaster scenarios is provided, which includes:

[0006] S1. Based on territorial characteristics analysis and historical extreme rainstorm case studies, numerical simulation methods are used to construct extreme rainstorm flood disaster scenarios, including the inundation depth of urban infrastructure affected by extreme rainstorms;

[0007] S2. Build independent network topologies for transportation infrastructure, power infrastructure, and communication infrastructure;

[0008] S3. Under the impact of extreme rainstorms, analyze the cascading relationship and disaster propagation mechanism of power, communication, and transportation infrastructure. Based on the independent network topology structures of the three infrastructures, construct a three-order network model of cascading failure of urban critical infrastructure. The three-order network model of cascading failure of urban critical infrastructure includes a third-order infrastructure network topology model, an extreme rainstorm and flooding model, a disaster propagation model, and a cascading failure model.

[0009] S4. Based on the flooding depth of the urban infrastructure affected by extreme rainstorms, infrastructure failure conditions are set to predict infrastructure node failures, and then independent infrastructure network failures are predicted. In addition, based on the third-order network model of cascading failures of urban critical infrastructure, cascading failures of urban critical infrastructure are predicted.

[0010] Optionally, the step S1 utilizes a numerical simulation method to construct an extreme rainstorm and flood disaster scenario, specifically including:

[0011] Collect and compile typical historical rainstorm and flood data in cities across the country, and conduct rainfall regime design based on extreme rainstorm and flood disaster scenarios;

[0012] Collect and statistically study the city's topographical characteristics and river design data, and conduct water regime design based on extreme rainstorm and flood disaster scenarios;

[0013] Collect and compile geographic information system data and infrastructure system data of the study city, import the geographic information system data, infrastructure system data, designed rainfall and water conditions into the hydrodynamic model, use distributed heterogeneous parallel computing technology to divide the study city into grids, solve the two-dimensional shallow water equation at the watershed scale, obtain the submerged water depth of the infrastructure, achieve high-precision hourly simulation of the flood evolution process at the watershed scale, and complete the construction of extreme rainstorm and flood disaster scenarios.

[0014] Optionally, solving the basin-scale two-dimensional shallow water equation specifically includes:

[0015] The basin-scale two-dimensional shallow water equation is a set of governing equations consisting of the mass conservation equation and the momentum conservation equation, where the mass conservation equation is expressed as:

[0016]

[0017] Where h is the flood depth; t is time; x and y are Cartesian coordinates; u and v are the flow velocities in the coordinate axis directions; i0 is the net rainfall; and the rate of change of flood depth with time is Plus the spatial variation of the water mass flux in the xy direction: Balance with the net rainfall i0 to maintain water balance during the calculation process;

[0018] The momentum conservation equation is expressed as:

[0019]

[0020] Formulas (2) and (3) represent the conservation of momentum in the x and y directions, respectively, and g is the acceleration due to gravity; and are the slope source items in the two coordinate axis directions; b is the ground elevation; C f=gn 2 h -1 / 3 is the ground friction coefficient; n is the Manning coefficient; u=(uv) T is the velocity vector; is the modulus of the velocity vector; and are the rates of change of momentum in the xy direction with time, and are the convection term and pressure term in the xy direction, The left side of the equation is the calculation of momentum exchange between adjacent grids, which is the core of water flow transmission between grids. 0x Used to calculate the gravity component caused by terrain slope, C f ||u||u calculates the bottom friction resistance. The right side of the equation represents the driving effect of the terrain slope on the water flow and the hindering effect of the ground friction on the water flow. Formulas (2) and (3) are used to calculate the velocity distribution of the water flow in the xy direction, considering the effects of pressure gradient, gravity, and friction on the water flow, and are used to determine the direction and intensity of the water flow.

[0021] A numerical method is used to solve the governing equations, converting the differential form of the governing equations into an integral form and decomposing the continuous boundary into a finite number of edges to facilitate computer solution. The numerical method updates the submerged water depth of the discrete grid under the time step and ensures that physical conservation laws are still satisfied at the structural discrete level. The numerical method includes:

[0022] The Godunov finite volume method is used to discretize the control equations into a structured grid. The entire calculation area is discretized into a continuous mathematical theoretical expression. The integral form of a single calculation unit is expressed as:

[0023]

[0024] Where Ω is a single computational unit; dΩ is the boundary of the computational unit; is the flow vector; and is the transport flow in the xy axis direction, is the unit normal vector; n x and n y is the normal vector component along the coordinate axis, is a conserved variable, are the ground slope source term and the friction source term, S b Representing the source term affected by the riverbed, formula (4) transforms the differential equation into an integral equation, ensuring the conservation of mass momentum of the numerical solution of each grid even in the case of discontinuous solutions, and supporting computer numerical calculations;

[0025] The structured grid is discretized into a structured Cartesian grid using the first-order Godunov finite volume method, and the boundary integral term of formula (4) is numerically discretized. For the Cartesian grid, the flux passing through the unit boundary is the sum of the fluxes on each boundary surface, which is calculated by the following formula:

[0026]

[0027] Wherein, the subscript k is the serial number of the unit boundary, where each unit grid has 4 edges; the subscripts “+” and “-” represent the positive and negative sides of the coordinate axis of the edge, respectively; F k (Q - ,Q + ) is the Riemann flow of edge k; Q - ,Q + are the conserved variables on both sides of the boundary; n k is the normal vector of the edge; l k is the length of the edge, and formula (5) is used to calculate the mass and momentum exchange between each grid and the adjacent grids, and to update the submerged water depth to obtain the submerged water depth of each grid at the next moment.

[0028] Optionally, the S2 specifically includes:

[0029] The transportation network topology adopts the existing mature subway network;

[0030] When constructing the power and communication network topology, based on the influence range and density of the ring main unit and the communication base station, a network construction technology N combining K-nearest neighbor and ε-radius is adopted. When the number of connected nodes within the ε-radius of a node exceeds k, the ε-radius network construction technology is selected. When it is less than k, the k-nearest neighbor network construction technology is selected, as shown in the following formula:

[0031]

[0032] where v i represents the i-th node in the network, k represents the minimum number of connections of this node, and ε represents the connection radius.

[0033] Optionally, the three-order infrastructure network topology model integrates three independent infrastructure network structures and is represented as follows:

[0034] G(t)={V,E(t),H(t)} (7)

[0035] where V = {v1,...,v i} is the infrastructure node set; is an edge set, which is affected by time t and changes at different times. R(B) is the node influence range. H(t) is an extreme rainstorm flood model, including the flooding depth and failure state of the node at time t.

[0036] Definition of critical infrastructure nodes:

[0037] v i ={Β i ,pos i ,h i (t),φ i (t),N i (t)} (8)

[0038] B i ∈{E,C,T} represents the node type, including: E ​​is the power ring network cabinet node, C is the communication base station node, and T is the transportation subway node;

[0039] POS i =(lat i ,lon i ,elev i ) are the latitude, longitude and elevation coordinates of the node;

[0040] h i (t) represents infrastructure The depth of water inundation at a node under a rainfall attack at time t;

[0041] φ i (t)∈{0,1} is the failure state of the node. When a node is under a rainfall attack at time t, if the node fails, it is represented as "0", and if the node does not fail, it is represented as "1";

[0042] N i (t) is The dynamic association list of node associations determines the number of nodes under the rainfall attack at time t according to the coupling mechanism of power-communication-transportation infrastructure. The node's associated node, the edge relationship between the two is represented by the adjacency matrix A at time t ij express.

[0043] Optionally, the extreme rainstorm flood model, based on the construction of an extreme rainstorm scenario, obtains the submerged water depth h(t) of urban infrastructure over rainfall time as an input parameter for analyzing the impact of extreme rainstorms on the network. The constructed flood attack model is as follows:

[0044]

[0045] Among them, the extreme rainstorm flood model H(t): includes the flooding depth and failure state of the node, posi is the coordinate of node i. Under the extreme rainfall attack at time t, the water depth function h(pos i ,t) represents the submerged water depth of node i at time t. If the submerged water depth of node i is greater than or equal to 0.5 meters, node i fails;

[0046] Failure state update φ i (t): For all nodes, the state φ at each time point t i (t) According to h(pos i ,t)Update.

[0047] Optionally, the disaster propagation model includes a flood disaster propagation path and a set of dynamically associated nodes under the path, and is used to analyze the propagation mechanism of rainstorm disasters on three types of infrastructure: if a power node fails, it will further affect communication and transportation nodes within its geographically associated influence radius; if a communication node fails, it will further affect transportation nodes within its geographically associated influence radius; if a transportation node fails, it will only affect the links connected to this node and will not affect other nodes;

[0048] Set the distance function d(v i ,v j ): Calculate the Euclidean distance between node i and node j. The formula is as follows:

[0049]

[0050] Among them, x i and x j Represents two infrastructure nodes, x i (k) represents the k-dimensional coordinate value of the i-th node, P represents the number of dimensions, and is a two-dimensional or three-dimensional coordinate according to the actual situation. According to the distance function and the different node types, the influence range and associated nodes of each node are also different;

[0051] Set the distance parameter R to further affect the edge E(t), and construct the adjacency matrix A based on whether there is an edge ij , which is expressed as follows:

[0052]

[0053] Among them, R(B i ) represents the influence radius of the infrastructure node, Bi represents the type of node i, the power infrastructure node will affect a circular area with a radius of 500 meters, the communication base station coverage area has an influence radius of 1 km, and the adjacency matrix A ij It includes both the inter-layer connections of single-order networks and the associated edges of high-order networks. The edge priority is power E>communication C>transport T;

[0054] The disaster propagation path of the disaster propagation model is fixed, the propagation of the disaster is immediate, and the associated node set is updated as the spatiotemporal flood field is updated. The associated node list of node i at time t is as follows:

[0055]

[0056] in, is the set of associated nodes, is the actual geographical distance between node i and communication node j, R(B) is the influence distance of the infrastructure node. When the node attribute is a power node, it will be associated with the communication nodes and transportation nodes within the distance range. When the node attribute is a communication node, it will be associated with the transportation nodes within the distance range.

[0057] Optionally, according to the disaster propagation model, for each infrastructure node's associated node set, when the power node fails at time t, the communication and transportation nodes at time t will also be marked as failed; when the communication node fails at time t, the transportation node at time t will also be marked as failed; when transportation node i fails at time t, all edges associated with i are marked as failed at time t. Then, based on the associated node set, the dynamic failure state and adjacency matrix of the node are constructed to obtain the cascading failure model, which is expressed as:

[0058]

[0059] V F (t)={v i ∈V|φ i (t)=0} (15)

[0060] Among them, the node state φ i When (t) is 1, it indicates that the submerged depth of the node is less than 0.5, and the node is not invalid. At this time, the associated nodes with higher priority around the node v j If the state of is also 1, the node remains in the non-failed state, otherwise it is updated to failed;

[0061] N i (t) is the associated node set of node i, and the network topology state is represented by A ij (t) indicates that f(B j ,B i ) is expressed as the priority of the node: the priority relationship of power E is greater than communication C and greater than transportation T. The node with a higher priority will affect the failure status of other nodes; V F (t) represents all failed nodes;

[0062] The cascading failure model comprehensively considers the direct destructive effects of the spatiotemporal flood field on nodes under flood disasters and the cross-system cascading failure process caused by it. On the basis of the three-order infrastructure network topology model, the extreme rainstorm and flood model is superimposed, and the disaster propagation mechanism is considered to construct a cascading failure model of power E>communication C>transportation T. The model analyzes the dynamic failure set of direct failure nodes and cascading failure nodes, systematically characterizes the failure characteristics of multiple types of infrastructure nodes, network coupling relationships and failure dynamics under the action of floods, quantitatively analyzes the response process of key infrastructure to flood disasters in the continuous time domain, and realizes the dynamic propagation prediction of disasters in the power-communication-transportation coupling network.

[0063] Optionally, the S4 specifically includes:

[0064] S41, infrastructure node failure prediction;

[0065] Based on the results of numerical simulations, the flooded water depth of all infrastructure nodes at each time point was extracted, and a node failure threshold h was set. When the flooded water depth was greater than h, the node failed. The number of all failed nodes and the number of failed nodes of different types of infrastructure were counted, and the dynamic changes in the failure of urban infrastructure nodes over time under urban flood disasters were analyzed.

[0066] S42, independent infrastructure network failure prediction;

[0067] Based on the extreme rainstorm flood model and the independent critical infrastructure network, failure nodes at time nodes are extracted. When a node fails, if the node is a transportation node, then the associated links of the node will fail. If the node is a communication node or a power node, then the associated links of the node and the area within the affected radius will also fail. ArcGIS or Python is used to visualize the time evolution state and failure status of the two-dimensional independent network.

[0068] S43, prediction of cascading failure of urban critical infrastructure;

[0069] Based on the third-order infrastructure network topology model, considering the disaster propagation model and cascading failure model, predict other nodes affected by the failed node and count the number of infrastructure nodes under cascading failure;

[0070] A three-dimensional network structure was constructed using Python, and the extreme rainfall inundation depth was superimposed. Based on the cascading failure model, a node association adjacency matrix was constructed to predict all failed nodes, failed links, and failed areas, and to visualize the dynamic evolution of cascading failures of critical infrastructure.

[0071] In another aspect, a system for predicting infrastructure cascading failures based on flood disaster scenarios is provided, the system comprising:

[0072] A scenario construction module is used to construct extreme rainstorm and flood disaster scenarios based on territorial characteristics analysis and historical extreme rainstorm case studies using numerical simulation methods. The extreme rainstorm and flood disaster scenarios include the inundation depth of urban infrastructure affected by extreme rainstorms;

[0073] Independent network building modules for building independent network topologies for transportation infrastructure, power infrastructure, and communication infrastructure;

[0074] A third-order network construction module is used to analyze the cascading relationship and disaster propagation mechanism of power, communication, and transportation infrastructure under the impact of extreme rainstorms. Based on the independent network topology structures of the three infrastructures, a third-order network model of cascading failure of urban critical infrastructure is constructed. The third-order network model of cascading failure of urban critical infrastructure includes a third-order infrastructure network topology model, an extreme rainstorm and flooding model, a disaster propagation model, and a cascading failure model.

[0075] The prediction module is used to set infrastructure failure conditions based on the flooding depth of the urban infrastructure affected by extreme rainstorms, predict infrastructure node failures, and then predict independent infrastructure network failures, as well as predict cascading failures of urban key infrastructure based on the third-order network model of cascading failures of urban key infrastructure.

[0076] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the above-mentioned infrastructure cascading failure prediction method constructed based on the flood disaster scenario.

[0077] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the infrastructure cascading failure prediction method constructed based on the above-mentioned flood disaster scenario.

[0078] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0079] The present invention proposes a method and system for predicting cascading failures of infrastructure based on flood disaster scenarios. The method and system have wide applicability, strong popularity, moderate accuracy, and require less data. It helps to analyze the impact of flood disasters on the coupled systems of key urban infrastructure, thereby formulating scientific emergency response strategies, reducing urban economic losses, and improving urban resilience. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0081] Figure 1 This is a flow chart of a method for predicting infrastructure cascading failures based on flood disaster scenarios provided by an embodiment of the present invention;

[0082] Figure 2 This is a schematic diagram of the relationship between power, communication, and transportation infrastructure under the impact of extreme rainstorms provided by an embodiment of the present invention;

[0083] Figure 3 This is a block diagram of an infrastructure cascading failure prediction system based on flood disaster scenarios provided by an embodiment of the present invention;

[0084] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0085] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0086] An embodiment of the present invention provides a method for predicting infrastructure cascading failures based on flood disaster scenarios. The method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of the method is shown, and the processing flow may include the following steps:

[0087] S1. Based on territorial characteristics analysis and historical extreme rainstorm case studies, numerical simulation methods are used to construct extreme rainstorm flood disaster scenarios, including the inundation depth of urban infrastructure affected by extreme rainstorms;

[0088] Optionally, the step S1 utilizes a numerical simulation method to construct an extreme rainstorm and flood disaster scenario, specifically including:

[0089] Collect and compile typical historical rainstorm and flood data in cities across the country, and conduct rainfall regime design based on extreme rainstorm and flood disaster scenarios;

[0090] Collect and statistically study the city's topographical characteristics and river design data, and conduct water regime design based on extreme rainstorm and flood disaster scenarios;

[0091] Collect and count the geographic information system data of the study city (mainly including the latitude and longitude range of the study city, elevation DEM data) and infrastructure system data (including the latitude and longitude of key infrastructure points, including transportation, electricity, communication and other infrastructure, and decrypt the encrypted data), import the geographic information system data, infrastructure system data, designed rainfall and water conditions into the hydrodynamic model, adopt distributed heterogeneous parallel computing technology to divide the study city into grids, solve the two-dimensional shallow water equation at the watershed scale, obtain the submerged water depth of the infrastructure, realize the hourly simulation of the high-precision flood evolution process at the watershed scale, and complete the construction of extreme rainstorm and flood disaster scenarios.

[0092] Optionally, solving the basin-scale two-dimensional shallow water equation specifically includes:

[0093] The basin-scale two-dimensional shallow water equation is a set of governing equations consisting of the mass conservation equation and the momentum conservation equation, where the mass conservation equation is expressed as:

[0094]

[0095] Where h is the flood depth; t is time; x and y are Cartesian coordinates; u and v are the flow velocities in the coordinate axis directions; i0 is the net rainfall; and the rate of change of flood depth with time is Plus the spatial variation of the water mass flux in the xy direction: Balance with the net rainfall i0 to maintain water balance during the calculation process;

[0096] The momentum conservation equation is expressed as:

[0097]

[0098] Formulas (2) and (3) represent the conservation of momentum in the x and y directions, respectively, and g is the acceleration due to gravity; and are the slope source items in the two coordinate axis directions; b is the ground elevation; C f =gn 2 h -1 / 3 is the ground friction coefficient; n is the Manning coefficient; u=(uv) T is the velocity vector; is the modulus of the velocity vector; and are the rates of change of momentum in the xy direction with time, and are the convection term and pressure term in the xy direction, The left side of the equation is the calculation of momentum exchange between adjacent grids, which is the core of water flow transmission between grids. 0xGravity component generated by terrain slope, C f ||u||uCalculate the bottom friction resistance, the right side of the equation represents the driving effect of terrain slope on water flow and the hindering effect of ground friction on water flow, equations (2) and (3) are used to calculate the velocity distribution of water flow in the xy direction, considering the influence of pressure gradient, gravity and friction on water flow movement, for determining the direction and intensity of water flow movement;

[0099] The control equation group is solved by a numerical method, the differential form of the control equation group is converted into an integral form, and the continuous boundary is divided into a finite strip edge, so as to facilitate computer solving, the numerical method updates the submerged water depth of the discrete grid at the time step, and ensures that the physical conservation law is still satisfied at the structure discrete level, the numerical method comprises:

[0100] The control equation group is discretized into a structured grid by using a Godunov finite volume method, the entire calculation area is discretized into a continuous mathematical theoretical expression, and the integral form expression of a single calculation unit is:

[0101]

[0102] Wherein, Ω is a single calculation unit; dΩ is the boundary of the calculation unit; is a flow vector; and is the transport flow in the xy axis direction, is a unit normal vector; n x and n y are the normal vector components along the coordinate axis direction, is a conservation variable, S b respectively, the ground slope source term and the friction source term, formula (4) converts the differential equation into an integral equation, which ensures the mass and momentum conservation of the numerical solution of each grid even in the case of discontinuous solution, and supports computer numerical calculation;

[0103] A first-order Godunov finite volume method is further used to discretize the structured grid into a structured Cartesian grid, the boundary integral term of formula (4) is numerically discretized, and for the Cartesian grid, the flow passing through the unit boundary is the sum of the fluxes on each boundary surface, which is calculated by the following formula:

[0104]

[0105] Wherein, the subscript k is the serial number of the unit boundary, here each unit grid has four edges; the subscripts "+" and "-" respectively represent the positive side and the negative side of the edge coordinate axis; F k (Q - ,Q+ ) is the Riemann flow of edge k; Q - ,Q + are the conserved variables on both sides of the boundary; n k is the normal vector of the edge; l k is the length of the edge, and formula (5) is used to calculate the mass and momentum exchange between each grid and the adjacent grids, and to update the submerged water depth to obtain the submerged water depth of each grid at the next moment.

[0106] The hydrodynamic model of the embodiment of the present invention can accurately simulate the hydrodynamic process under complex conditions and realize hourly calculation of the evolution process of floods in a large number of grids. The scenario construction method based on numerical simulation can intuitively, in real time and dynamically reflect the evolution of disasters. According to the simulation results of the scenario construction, the changes in the flood depth of infrastructure nodes with the rainfall time are obtained, realizing the dynamic prediction of the flood depth of infrastructure.

[0107] S2. Build independent network topologies for transportation infrastructure, power infrastructure, and communication infrastructure;

[0108] Optionally, the S2 specifically includes:

[0109] The transportation network topology adopts the existing mature subway network;

[0110] When constructing the topology of the power and communication networks, based on the influence range and density of the ring main unit and the communication base station, a network construction technology N combining K-nearest neighbor and ε-radius is adopted. When the number of connected nodes within the ε-radius of a node exceeds k, the ε-radius network construction technology is selected (a network construction method based on a distance threshold. Its core concept is: in the metric space, if the distance between two nodes does not exceed a given threshold ε, then a connection is established; otherwise, no connection is established, thereby constructing a network). When the distance is less than k, the k-nearest neighbor (KNN) network construction technology is selected (which refers to building a network structure with nodes and their nearest neighbor relationships as the core by calculating the distance between nodes), as shown in the following formula:

[0111]

[0112] where v i represents the i-th node in the network, k represents the minimum number of connections of this node, and ε represents the connection radius.

[0113] S3. Under the impact of extreme rainstorms, analyze the cascading relationship and disaster propagation mechanism of power-communication-transportation infrastructure (there are physical, information, geographical and logical connections between power-communication-transportation infrastructure. The power system provides energy for traffic signals and is also the driving force for the operation of the communication system. The communication system provides a key information flow channel for the power and transportation systems). Based on the independent network topology of the three infrastructures, construct a three-order network model of cascading failure of urban key infrastructure, such as Figure 2 As shown, the three-order network model of cascading failure of urban critical infrastructure includes a three-order infrastructure network topology model, an extreme rainstorm and flood model, a disaster propagation model and a cascading failure model;

[0114] Optionally, the three-order infrastructure network topology model integrates three independent infrastructure network structures and is represented as follows:

[0115] G(t)={V,E(t),H(t)} (7)

[0116] where V = {v1,...,v i} is the infrastructure node set; is an edge set, which is affected by time t and changes at different times. R(B) is the node influence range. H(t) is an extreme rainstorm flood model, including the flooding depth and failure state of the node at time t.

[0117] Definition of critical infrastructure nodes:

[0118] v i ={Β i ,pos i ,h i (t),φ i (t),N i (t)} (8)

[0119] B i ∈{E,C,T} represents the node type, including: E ​​is the power ring network cabinet node, C is the communication base station node, and T is the transportation subway node;

[0120] POS i =(lat i ,lon i ,elev i ) are the latitude, longitude and elevation coordinates of the node;

[0121] h i (t) represents infrastructure The depth of water inundation at a node under a rainfall attack at time t;

[0122] φ i(t)∈{0,1} is the failure state of the node. When a node is under a rainfall attack at time t, if the node fails, it is represented as "0", and if the node does not fail, it is represented as "1";

[0123] N i (t) is The dynamic association list of node associations determines the number of nodes under the rainfall attack at time t according to the coupling mechanism of power-communication-transportation infrastructure. The node's associated node, the edge relationship between the two is represented by the adjacency matrix A at time t ij express.

[0124] Optionally, the extreme rainstorm flood model, based on the construction of an extreme rainstorm scenario, obtains the submerged water depth h(t) of urban infrastructure over rainfall time as an input parameter for analyzing the impact of extreme rainstorms on the network. The constructed flood attack model is as follows:

[0125]

[0126] Among them, the extreme rainstorm flood model H(t): includes the flooding depth and failure state of the node, pos i is the coordinate of node i. Under the extreme rainfall attack at time t, the water depth function h(pos i ,t) represents the submerged water depth of node i at time t. If the submerged water depth of node i is greater than or equal to 0.5 meters, node i fails;

[0127] Failure state update φ i (t): For all nodes, the state φ at each time point t i (t) According to h(pos i ,t)Update.

[0128] Optionally, the disaster propagation model includes a flood disaster propagation path and a set of dynamically associated nodes under the path, and is used to analyze the propagation mechanism of rainstorm disasters on three types of infrastructure: if a power node fails, it will further affect communication and transportation nodes within its geographically associated influence radius; if a communication node fails, it will further affect transportation nodes within its geographically associated influence radius; if a transportation node fails, it will only affect the links connected to this node and will not affect other nodes;

[0129] Set the distance function d(v i ,v j ): Calculate the Euclidean distance between node i and node j. The formula is as follows:

[0130]

[0131] Among them, xi and x j Represents two infrastructure nodes, x i (k) represents the k-dimensional coordinate value of the i-th node, P represents the number of dimensions, and is a two-dimensional or three-dimensional coordinate according to the actual situation. According to the distance function and the different node types, the influence range and associated nodes of each node are also different;

[0132] Set the distance parameter R to further affect the edge E(t), and construct the adjacency matrix A based on whether there is an edge ij , which is expressed as follows:

[0133]

[0134] Among them, R(B i ) represents the influence radius of the infrastructure node, B i Indicates the type of node i. The power infrastructure node will affect a circular area with a radius of 500 meters. The communication base station coverage area has an influence radius of 1 kilometer. The adjacency matrix A ij It includes both the inter-layer connections of single-order networks and the associated edges of high-order networks. The edge priority is power E>communication C>transport T;

[0135] The disaster propagation path of the disaster propagation model is fixed, the propagation of the disaster is immediate, and the associated node set is updated as the spatiotemporal flood field is updated. The associated node list of node i at time t is as follows:

[0136]

[0137] in, is the set of associated nodes, is the actual geographical distance between node i and communication node j, R(B) is the influence distance of the infrastructure node. When the node attribute is a power node, it will be associated with the communication nodes and transportation nodes within the distance range. When the node attribute is a communication node, it will be associated with the transportation nodes within the distance range.

[0138] Optionally, according to the disaster propagation model, for each infrastructure node's associated node set, when the power node fails at time t, the communication and transportation nodes at time t will also be marked as failed; when the communication node fails at time t, the transportation node at time t will also be marked as failed; when transportation node i fails at time t, all edges associated with i are marked as failed at time t. Then, based on the associated node set, the dynamic failure state and adjacency matrix of the node are constructed to obtain the cascading failure model, which is expressed as:

[0139]

[0140] V F (t)={vi ∈V|φ i (t)=0} (15)

[0141] Among them, the node state φ i When (t) is 1, it indicates that the submerged depth of the node is less than 0.5, and the node is not invalid. At this time, the associated nodes with higher priority around the node v j If the state of is also 1, the node remains in the non-failed state, otherwise it is updated to failed;

[0142] N i (t) is the associated node set of node i, and the network topology state is represented by A ij (t) indicates that f(B j ,B i ) is expressed as the priority of the node: the priority relationship of power E is greater than communication C and greater than transportation T. The node with a higher priority will affect the failure status of other nodes; V F (t) represents all failed nodes;

[0143] The cascading failure model comprehensively considers the direct destructive effects of the spatiotemporal flood field on nodes under flood disasters and the cross-system cascading failure process caused by it. On the basis of the three-order infrastructure network topology model, the extreme rainstorm and flood model is superimposed, and the disaster propagation mechanism is considered to construct a cascading failure model of power E>communication C>transportation T. The model analyzes the dynamic failure set of direct failure nodes and cascading failure nodes, systematically characterizes the failure characteristics of multiple types of infrastructure nodes, network coupling relationships and failure dynamics under the action of floods, quantitatively analyzes the response process of key infrastructure to flood disasters in the continuous time domain, and realizes the dynamic propagation prediction of disasters in the power-communication-transportation coupling network.

[0144] S4. Based on the flooding depth of the urban infrastructure affected by extreme rainstorms, infrastructure failure conditions are set to predict infrastructure node failures, and then independent infrastructure network failures are predicted. In addition, based on the third-order network model of cascading failures of urban critical infrastructure, cascading failures of urban critical infrastructure are predicted.

[0145] Optionally, the S4 specifically includes:

[0146] S41, infrastructure node failure prediction;

[0147] Based on the results of numerical simulations, the flooded water depth of all infrastructure nodes at each time point was extracted, and a node failure threshold h was set. When the flooded water depth was greater than h, the node failed. The number of all failed nodes and the number of failed nodes of different types of infrastructure were counted, and the dynamic changes in the failure of urban infrastructure nodes over time under urban flood disasters were analyzed.

[0148] S42, independent infrastructure network failure prediction;

[0149] Based on the extreme rainstorm flood model and the independent critical infrastructure network, failure nodes at time nodes are extracted. When a node fails, if the node is a transportation node, then the associated links of the node will fail. If the node is a communication node or a power node, then the associated links of the node and the area within the affected radius will also fail. ArcGIS or Python is used to visualize the time evolution state and failure status of the two-dimensional independent network.

[0150] S43, prediction of cascading failure of urban critical infrastructure;

[0151] Based on the third-order infrastructure network topology model, considering the disaster propagation model and cascading failure model, predict other nodes affected by the failed node and count the number of infrastructure nodes under cascading failure;

[0152] A three-dimensional network structure was constructed using Python, and the extreme rainfall inundation depth was superimposed. Based on the cascading failure model, a node association adjacency matrix was constructed to predict all failed nodes, failed links, and failed areas, and to visualize the dynamic evolution of cascading failures of critical infrastructure.

[0153] like Figure 3 As shown, an embodiment of the present invention further provides an infrastructure cascading failure prediction system based on flood disaster scenarios, the system comprising:

[0154] Scenario construction module 310 is used to construct extreme rainstorm and flood disaster scenarios based on territorial characteristics analysis and historical extreme rainstorm case studies using numerical simulation methods. The extreme rainstorm and flood disaster scenarios include the inundation depth of urban infrastructure affected by extreme rainstorms.

[0155] An independent network construction module 320 is used to construct independent network topologies of transportation infrastructure, power infrastructure, and communication infrastructure;

[0156] A third-order network construction module 330 is used to analyze the cascading relationship and disaster propagation mechanism of power, communication, and transportation infrastructure under the impact of extreme rainstorms. Based on the independent network topologies of the three infrastructures, a third-order network model of cascading failure of urban critical infrastructure is constructed. The third-order network model of cascading failure of urban critical infrastructure includes a third-order infrastructure network topology model, an extreme rainstorm and flooding model, a disaster propagation model, and a cascading failure model.

[0157] The prediction module 340 is used to set infrastructure failure conditions based on the flooding depth of the urban infrastructure affected by extreme rainstorms, predict infrastructure node failures, and then predict independent infrastructure network failures, as well as predict cascading failures of urban key infrastructure based on the third-order network model of cascading failures of urban key infrastructure.

[0158] An infrastructure cascading failure prediction system based on flood disaster scenarios provided by an embodiment of the present invention has a functional structure corresponding to an infrastructure cascading failure prediction method based on flood disaster scenarios provided by an embodiment of the present invention, which will not be described in detail here.

[0159] Figure 4 This is a structural diagram of an electronic device 400 provided in an embodiment of the present invention. The electronic device 400 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 401 and one or more memories 402, wherein the memory 402 stores at least one instruction, and the at least one instruction is loaded and executed by the processor 401 to implement the steps of the infrastructure cascading failure prediction method constructed in the above-mentioned flood disaster scenario.

[0160] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory device including instructions. The instructions are executable by a processor in a terminal to implement the aforementioned method for predicting infrastructure cascading failures based on flood disaster scenarios. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device.

[0161] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0162] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predicting infrastructure cascading failures based on flood disaster scenarios, characterized in that: The method comprises: S1. Based on territorial characteristics analysis and historical extreme rainstorm case studies, numerical simulation methods are used to construct extreme rainstorm flood disaster scenarios, including the inundation depth of urban infrastructure affected by extreme rainstorms; S2. Build independent network topologies for transportation infrastructure, power infrastructure, and communication infrastructure; S3. Under the impact of extreme rainstorms, analyze the cascading relationship and disaster propagation mechanism of power, communication, and transportation infrastructure. Based on the independent network topology structures of the three infrastructures, construct a three-order network model of cascading failure of urban critical infrastructure. The three-order network model of cascading failure of urban critical infrastructure includes a third-order infrastructure network topology model, an extreme rainstorm and flooding model, a disaster propagation model, and a cascading failure model. S4. Based on the flooding depth of the urban infrastructure affected by extreme rainstorms, infrastructure failure conditions are set to predict infrastructure node failures, and then independent infrastructure network failures are predicted. In addition, based on the third-order network model of cascading failures of urban critical infrastructure, cascading failures of urban critical infrastructure are predicted.

2. The method according to claim 1, characterized in that In S1, a numerical simulation method is used to construct an extreme rainstorm and flood disaster scenario, specifically including: Collect and compile typical historical rainstorm and flood data in cities across the country, and conduct rainfall regime design based on extreme rainstorm and flood disaster scenarios; Collect and statistically study the city's topographical characteristics and river design data, and conduct water regime design based on extreme rainstorm and flood disaster scenarios; Collect and compile geographic information system data and infrastructure system data of the study city, import the geographic information system data, infrastructure system data, designed rainfall and water conditions into the hydrodynamic model, use distributed heterogeneous parallel computing technology to divide the study city into grids, solve the two-dimensional shallow water equation at the watershed scale, obtain the submerged water depth of the infrastructure, achieve high-precision hourly simulation of the flood evolution process at the watershed scale, and complete the construction of extreme rainstorm and flood disaster scenarios.

3. The method according to claim 2, characterized in that The solution of the basin-scale two-dimensional shallow water equation specifically includes: The basin-scale two-dimensional shallow water equation is a set of governing equations consisting of the mass conservation equation and the momentum conservation equation, where the mass conservation equation is expressed as: Where h is the flood depth; t is time; x and y are Cartesian coordinates; u and v are the flow velocities in the coordinate axis directions; i0 is the net rainfall; and the rate of change of flood depth with time is Plus the spatial variation of the water mass flux in the xy direction: Balance with the net rainfall i0 to maintain water balance during the calculation process; The momentum conservation equation is expressed as: Formulas (2) and (3) represent the conservation of momentum in the x and y directions, respectively, and g is the acceleration due to gravity; and are the slope source items in the two coordinate axis directions; b is the ground elevation; C f =gn 2 h -1 / 3 is the ground friction coefficient; n is the Manning coefficient; u=(uv) T is the velocity vector; is the modulus of the velocity vector; and are the rates of change of momentum in the xy direction with time, and are the convection term and pressure term in the xy direction, The left side of the equation is the calculation of momentum exchange between adjacent grids, which is the core of water flow transmission between grids. 0x Used to calculate the gravity component caused by terrain slope, C f ||u||u calculates the bottom friction resistance. The right side of the equation represents the driving effect of the terrain slope on the water flow and the hindering effect of the ground friction on the water flow. Formulas (2) and (3) are used to calculate the velocity distribution of the water flow in the xy direction, considering the effects of pressure gradient, gravity, and friction on the water flow, and are used to determine the direction and intensity of the water flow. A numerical method is used to solve the governing equations, converting the differential form of the governing equations into an integral form and decomposing the continuous boundary into a finite number of edges to facilitate computer solution. The numerical method updates the submerged water depth of the discrete grid under the time step and ensures that physical conservation laws are still satisfied at the structural discrete level. The numerical method includes: The Godunov finite volume method is used to discretize the control equations into a structured grid. The entire calculation area is discretized into a continuous mathematical theoretical expression. The integral form of a single calculation unit is expressed as: Where Ω is a single computational unit; dΩ is the boundary of the computational unit; is the flow vector; and is the transport flow in the xy axis direction, is the unit normal vector; n x and n y is the normal vector component along the coordinate axis, is a conserved variable, are the ground slope source term and the friction source term, S b Representing the source term affected by the riverbed, formula (4) transforms the differential equation into an integral equation, ensuring the conservation of mass momentum of the numerical solution of each grid even in the case of discontinuous solutions, and supporting computer numerical calculations; The structured grid is discretized into a structured Cartesian grid using the first-order Godunov finite volume method, and the boundary integral term of formula (4) is numerically discretized. For the Cartesian grid, the flux passing through the unit boundary is the sum of the fluxes on each boundary surface, which is calculated by the following formula: Wherein, the subscript k is the serial number of the unit boundary, where each unit grid has 4 edges; the subscripts "+" and "-" represent the positive and negative sides of the coordinate axis of the edge, respectively; F k (Q - ,Q + ) is the Riemann flow of edge k; Q - ,Q + are the conserved variables on both sides of the boundary; n k is the normal vector of the edge; l k is the length of the edge, and formula (5) is used to calculate the mass and momentum exchange between each grid and the adjacent grids, and to update the submerged water depth to obtain the submerged water depth of each grid at the next moment.

4. The method according to claim 1, wherein Said S2 specifically includes: The transportation network topology adopts the existing mature subway network; When constructing the power and communication network topology, based on the influence range and density of the ring main unit and the communication base station, a network construction technology N combining K-nearest neighbor and ε-radius is adopted. When the number of connected nodes within the ε-radius of a node exceeds k, the ε-radius network construction technology is selected. When it is less than k, the k-nearest neighbor network construction technology is selected, as shown in the following formula: where v i represents the i-th node in the network, k represents the minimum number of connections of this node, and ε represents the connection radius.

5. The method according to claim 1, wherein The three-order infrastructure network topology model integrates three independent infrastructure network structures and is represented as follows: G(t)={V,E(t),H(t)} (7) where V = {v1,...,v i } is the infrastructure node set; is an edge set, which is affected by time t and changes at different times. R(B) is the node influence range. H(t) is an extreme rainstorm flood model, including the flooding depth and failure state of the node at time t. Definition of critical infrastructure nodes: v i ={Β i ,pos i ,h i (t),φ i (t),N i (t)} (8) B i ∈{E,C,T} represents the node type, including: E ​​is the power ring network cabinet node, C is the communication base station node, and T is the transportation subway node; POS i =(lat i ,lon i ,elev i ) are the latitude, longitude and elevation coordinates of the node; h i (t) represents infrastructure The depth of water inundation at a node under a rainfall attack at time t; φ i (t)∈{0,1} is the failure state of the node. When a node is under a rainfall attack at time t, if the node fails, it is represented as "0", and if the node does not fail, it is represented as "1"; N i (t) is The dynamic association list of node associations determines the number of nodes under the rainfall attack at time t according to the coupling mechanism of power-communication-transportation infrastructure. The node's associated node, the edge relationship between the two is represented by the adjacency matrix A at time t ij express.

6. The method according to claim 1, characterized in that The extreme rainstorm flood model is constructed based on the extreme rainstorm scenario. The inundation depth h(t) of urban infrastructure over rainfall time is obtained as an input parameter to analyze the impact of extreme rainstorms on the network. The constructed flood attack model is as follows: Among them, the extreme rainstorm flood model H(t): includes the flooding depth and failure state of the node, pos i is the coordinate of node i. Under the extreme rainfall attack at time t, the water depth function h(pos i ,t) represents the submerged water depth of node i at time t. If the submerged water depth of node i is greater than or equal to 0.5 meters, node i fails; Failure state update φ i (t): For all nodes, the state φ at each time point t i (t) According to h(pos i ,t)Update.

7. The method according to claim 1, characterized in that The disaster propagation model, which includes flood disaster propagation paths and dynamically associated node sets within those paths, is used to analyze the propagation mechanism of rainstorm disasters on three types of infrastructure: The failure of a power node will further affect communication and transportation nodes within its geographically associated influence radius; the failure of a communication node will further affect transportation nodes within its geographically associated influence radius; and the failure of a transportation node will only affect the links connected to that node and not other nodes. Set the distance function d(v i ,v j ): Calculate the Euclidean distance between node i and node j. The formula is as follows: Among them, x i and x j Represents two infrastructure nodes, x i (k) represents the k-dimensional coordinate value of the i-th node, P represents the number of dimensions, and is a two-dimensional or three-dimensional coordinate according to the actual situation. According to the distance function and the different node types, the influence range and associated nodes of each node are also different; Set the distance parameter R to further affect the edge E(t), and construct the adjacency matrix A based on whether there is an edge ij , which is expressed as follows: Among them, R(B i ) represents the influence radius of the infrastructure node, B i Indicates the type of node i. The power infrastructure node will affect a circular area with a radius of 500 meters. The communication base station coverage area has an influence radius of 1 kilometer. The adjacency matrix A ij It includes both the inter-layer connections of single-order networks and the associated edges of high-order networks. The edge priority is power E>communication C>transport T; The disaster propagation path of the disaster propagation model is fixed, the propagation of the disaster is immediate, and the associated node set is updated as the spatiotemporal flood field is updated. The associated node list of node i at time t is as follows: in, is the set of associated nodes, is the actual geographical distance between node i and communication node j, R(B) is the influence distance of the infrastructure node. When the node attribute is a power node, it will be associated with the communication nodes and transportation nodes within the distance range. When the node attribute is a communication node, it will be associated with the transportation nodes within the distance range.

8. The method according to claim 7, characterized in that According to the disaster propagation model, for each infrastructure node's associated node set, when the power node fails at time t, the communication and transportation nodes at time t will also be marked as failed; when the communication node fails at time t, the transportation node at time t will also be marked as failed; when transportation node i fails at time t, all edges associated with i will be marked as failed at time t. Based on the associated node set, the dynamic failure state and adjacency matrix of the node are constructed to obtain the cascading failure model, which is expressed as: V F (t)={v i ∈V|φ i (t)=0} (15) Among them, the node state φ i When (t) is 1, it indicates that the submerged depth of the node is less than 0.5, and the node is not invalid. At this time, the associated nodes with higher priority around the node v j If the state of is also 1, the node remains in the non-failed state, otherwise it is updated to failed; N i (t) is the associated node set of node i, and the network topology state is represented by A ij (t) indicates that f(B j ,B i ) is expressed as the priority of the node: the priority relationship of power E is greater than communication C and greater than transportation T. The node with a higher priority will affect the failure status of other nodes; V F (t) represents all failed nodes; The cascading failure model comprehensively considers the direct destructive effects of the spatiotemporal flood field on nodes under flood disasters and the cross-system cascading failure process caused by it. On the basis of the three-order infrastructure network topology model, the extreme rainstorm and flood model is superimposed, and the disaster propagation mechanism is considered to construct a cascading failure model of power E>communication C>transportation T. The model analyzes the dynamic failure set of direct failure nodes and cascading failure nodes, systematically characterizes the failure characteristics of multiple types of infrastructure nodes, network coupling relationships and failure dynamics under the action of floods, quantitatively analyzes the response process of key infrastructure to flood disasters in the continuous time domain, and realizes the dynamic propagation prediction of disasters in the power-communication-transportation coupling network.

9. The method according to claim 1, characterized in that Said S4 specifically includes: S41, infrastructure node failure prediction; Based on the results of numerical simulations, the flooded water depth of all infrastructure nodes at each time point was extracted, and a node failure threshold h was set. When the flooded water depth was greater than h, the node failed. The number of all failed nodes and the number of failed nodes of different types of infrastructure were counted, and the dynamic changes in the failure of urban infrastructure nodes over time under urban flood disasters were analyzed. S42, independent infrastructure network failure prediction; Based on the extreme rainstorm flood model and the independent critical infrastructure network, failure nodes at time nodes are extracted. When a node fails, if the node is a transportation node, then the associated links of the node will fail. If the node is a communication node or a power node, then the associated links of the node and the area within the affected radius will also fail. ArcGIS or Python is used to visualize the time evolution state and failure status of the two-dimensional independent network. S43, prediction of cascading failure of critical urban infrastructure; Based on the third-order infrastructure network topology model, considering the disaster propagation model and cascading failure model, predict other nodes affected by the failed node and count the number of infrastructure nodes under cascading failure; A three-dimensional network structure was constructed using Python, and the extreme rainfall inundation depth was superimposed. Based on the cascading failure model, a node association adjacency matrix was constructed to predict all failed nodes, failed links, and failed areas, and to visualize the dynamic evolution of cascading failures of critical infrastructure.

10. An infrastructure cascading failure prediction system based on flood disaster scenarios, characterized by: The system comprises: A scenario construction module is used to construct extreme rainstorm and flood disaster scenarios based on territorial characteristics analysis and historical extreme rainstorm case studies using numerical simulation methods. The extreme rainstorm and flood disaster scenarios include the inundation depth of urban infrastructure affected by extreme rainstorms; Independent network building modules for building independent network topologies for transportation infrastructure, power infrastructure, and communication infrastructure; A third-order network construction module is used to analyze the cascading relationship and disaster propagation mechanism of power, communication, and transportation infrastructure under the impact of extreme rainstorms. Based on the independent network topology structures of the three infrastructures, a third-order network model of cascading failure of urban critical infrastructure is constructed. The third-order network model of cascading failure of urban critical infrastructure includes a third-order infrastructure network topology model, an extreme rainstorm and flooding model, a disaster propagation model, and a cascading failure model. The prediction module is used to set infrastructure failure conditions based on the flooding depth of the urban infrastructure affected by extreme rainstorms, predict infrastructure node failures, and then predict independent infrastructure network failures, as well as predict cascading failures of urban key infrastructure based on the third-order network model of cascading failures of urban key infrastructure.

Images