Nuclear accident emergency three-layer coupling network model and construction method

Abstract

The invention discloses a nuclear accident emergency three-layer coupling network model and a construction method, and relates to the crossing field of nuclear accident emergency management and complex system risk simulation. The nuclear accident emergency three-layer coupling network model comprises an RMN layer which is used for simulating the dynamic diffusion process of radionuclides leaked by a nuclear accident in the atmosphere; the IN layer is used for constructing a spatial distribution and dependency relationship network of infrastructures; the EN layer is used for topologically defining a command or cooperation relationship of the emergency organization; according to the three-layer network coupling mechanism, when RDR of a middle grid is larger than or equal to RDRmax-1, IN-layer nodes in the middle grid at the same position are marked to be in a damaged state; and quantifying the bidirectional dependence density of the IN layer and the EN layer through the coupling strength epsilon. The method is used for predicting the direct risk of radioactive contamination in a nuclear accident and the secondary risk caused by emergency evacuation, revealing the risk evolution law, quantifying the risk diffusion scale, determining key influence factors, and providing a theoretical basis for formulating an emergency strategy.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of nuclear accident emergency management and complex system risk simulation, and in particular to a three-layer coupled network model for nuclear accident emergency response and its construction method. Background Technology

[0002] As an extreme public safety event, a nuclear accident is characterized by its suddenness, diffusion, and chain effects. The greatest challenge in its emergency response is not only the direct risks caused by radioactive materials (e.g., radioactive dust can spread with the wind, causing damage to human health, radiation damage to equipment, and environmental pollution), but also a series of secondary risks triggered by the direct risks (e.g., traffic paralysis: panic driving and fleeing causes traffic jams, preventing rescue vehicles from entering and people from dangerous areas from leaving; communication disruptions: radiation damages base stations, or too many people making phone calls causes network paralysis, preventing commands from being sent out and the situation on the ground from being transmitted back; command chaos: poor information flow and communication between emergency command departments lead to slow decision-making and delayed or erroneous resource allocation; social panic: the spread of rumors and irrational behavior by the public further exacerbates the chaos).

[0003] The above analysis shows that the prerequisite for accurate decision-making in nuclear accident emergency response is accurate modeling of the spread of radioactive contaminants, infrastructure functions, and the coordination of various departments. Based on accurate modeling, accurate prediction of risk evolution can be achieved, thereby optimizing / selecting the best control strategy.

[0004] However, most existing models have obvious shortcomings and blind spots, making it difficult to achieve relatively accurate predictions of risk evolution.

[0005] 1. "Seeing the trees but not the forest" - fragmented models that fail to consider chain reactions: For example, Gaussian plume models can only describe the physical concentration distribution of radioactive materials, but cannot assess the cascading effects of radioactive material diffusion on transportation or communication; for example, dependent network theory can only assess the topological characteristics of transportation or communication failures, but it does not consider that radiation first causes damage to a certain device, which in turn leads to transportation or communication failures (i.e., it does not couple the triggering mechanism of radioactive contamination); for example, organizational collaboration simulation models (such as agent-based modeling) can only simulate the decision-making process of emergency command departments, but it is unaware of the current situation of on-site traffic congestion, communication paralysis, and orders that cannot be transmitted (i.e., it is detached from physical risk input and facility failure feedback).

[0006] These models are like isolated "information islands." They cannot simulate how, in the real world, excessive radiation in one place (direct risk) leads to traffic congestion and communication paralysis (secondary risks), and how traffic congestion and communication paralysis delay rescue decisions, which in turn expose more people to radiation (risk amplification).

[0007] 2. "Marking the boat to find the sword" - Rigid models that ignore the "inertia" of risk transmission: Existing risk transmission models (such as the SIR infectious disease model) usually use integer-order calculus, which is like calculating speed by only considering the speed at the current instant, without considering "inertia". However, in a nuclear accident, risk transmission has "inertia": panic will not disappear immediately after the danger is over; it will continue to affect people's behavior (for example, even if the radiation level drops, people may still be afraid to go home); once traffic paralysis occurs, it takes time to clear the blockage; traffic will not immediately return to normal once the roads are open; organizational chaos also requires time to adjust and restore efficiency.

[0008] Existing risk propagation models cannot accurately describe the lasting impact and recovery lag of nuclear accidents, resulting in significant biases in predicting the long-term development and recovery process of risks (such as underestimating the duration of panic or the difficulty of restoring transportation).

[0009] 3. "Inaccurate Decision-Making" - Failure to Precisely Grasp Key Measures for Controlling the Risk Chain: Risks in a nuclear accident propagate in a leapfrog manner through a three-tiered network: "physical risk triggering - cascading failure of engineering facilities - organizational response delay / failure," accelerating each other (i.e., the risk propagation process involves complex chain reactions). Therefore, limited emergency resources (human, material, and time) should be prioritized for measures that best control the risk chain. For example, should priority be given to plugging the leak source (reducing the release of radioactive materials)? Or prioritizing the protection of key communication nodes (ensuring unimpeded command and control)? Or prioritizing the diversion of key intersections (opening up lifelines for rescue)? Or prioritizing the release of authoritative information to quell panic? This clearly requires a comprehensive, systematic, and quantitative parameter sensitivity analysis.

[0010] Existing models (such as those in points 1-2) suffer from fragmentation and neglect of inertia, leading to inaccurate predictions of subsequent risk evolution. This makes it difficult to assess which measures are most effective in ultimately controlling the entire risk chain (from its spread to transportation, communication, command, and panic). Consequently, emergency decision-making lacks quantitative basis, resource allocation may be inefficient or even erroneous, and the key measures to curb the vicious cycle of risk cannot be accurately identified. Summary of the Invention

[0011] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a three-layer coupled network model for nuclear accident emergency response and its construction method. It solves the problems of existing models failing to capture risk chain reactions, ignoring risk propagation inertia, and lacking accurate decision-making basis when applied to predicting the evolution of nuclear accident risks.

[0012] The technical solution of this invention is: a three-layer coupled network model for nuclear accident emergency response, comprising:

[0013] Nuclide Migration Network Layer (RMN layer): This layer simulates the dynamic diffusion process of radionuclides in the atmosphere following a nuclear accident. It is based on a dynamic Gaussian plume model and represents the nuclear accident-affected region Ω. XY Mesh the network into an RMN-layer spatial network at a resolution Δ. Its nodes are The grid cells in the model have node states that correspond to the radiation dose rate in the grid cells. The radiation dose rate is calculated and converted from the dynamic Gaussian plume model.

[0014] Infrastructure Network Layer (IN Layer): This layer is used to construct the spatial distribution and dependency network of infrastructure. It is a two-layer network that couples transportation and communication networks, built using a dependent network architecture. First, nodes are defined, including traffic intersections and communication base stations. Node states include susceptible (S), infected (I), immune (R), and damaged (D). Intra-network edges include roads and optical cables. Cross-network edges represent the functional dependencies between the transportation and communication networks. This is achieved through the adjacency matrix A. IN Define the physical connections between nodes in the IN layer; obtain network G through topology modeling. IN =(N IN E IN ), N IN E is the set of all nodes in the IN layer. IN The set of all edges in the IN layer; then the nuclear accident-affected region Ω. XY Mesh the network into an IN-layer spatial network based on resolution Δ. Finally, the geographic coordinates of all nodes in the IN layer are mapped to... In the corresponding grid cell;

[0015] Emergency Organization Network Layer (EN Layer): Its purpose is to topologically define the command and coordination relationships within an emergency organization. It includes four types of nodes: nuclear power plants, government departments, rescue agencies, and social organizations. Each type of node is further divided into multiple nodes based on their functions. This is achieved through an adjacency matrix A. EN Define the command or cooperative connection relationships between nodes in the EN layer; perform topology modeling on all nodes in the EN layer to obtain network G. EN =(N EN E EN ), N EN E is the set of all nodes in the EN layer. EN The set of all edges in the EN layer; the topology network G is calculated based on real-time communication quality, cooperative dependency strength, and connection response strength. EN The dynamic weights of each connected edge in the middle;

[0016] The coupling mechanism of the three-layer network in the model: I. When The current radiation dose rate (RDR) within the grid is greater than or equal to the general system threshold for radiation dose rate (RDR). max-1 hour, II. The IN layer nodes in the same location grid are marked as damaged state (D); II. The bidirectional dependency density between the IN layer and the EN layer is quantified by the coupling strength ε.

[0017] A further technical solution of the present invention is: in the RMN layer, the dynamic Gaussian plume model is obtained by introducing a diffusion time correction coefficient into the standard Gaussian plume model, discretizing the plume diffusion into multiple time periods, and assuming that the meteorological conditions are constant in each time period; the dynamic Gaussian plume model is shown in Formula 1.

[0018] Formula 1:

[0019] Where C(x,y,z,t) is the concentration of the radionuclide at spatial location (x,y,z) at time t; Q is the source strength; λ is the radionuclide decay constant; t n σ represents the duration of the nth time period; u represents the average wind speed downwind of the leak source during the current time period; σ represents the average wind speed downwind of the leak source during the current time period. x σ y σ z These are the diffusion coefficients in the downwind, crosswind, and vertical directions, respectively; h is the effective plume release height for the current period; x n ,y n ,z n Let be the coordinates of the center point of the plume diffusion area in the nth time period. In the dynamic spatial coordinate system in which it is located, the x-axis is parallel to the prevailing wind direction of the current time period, the y-axis is the horizontal direction perpendicular to the downwind direction of the current time period, and the z-axis is the vertical direction perpendicular to the ground. exp is an exponential function; γ1 and γ2 are the regression coefficients of the crosswind and vertical diffusion parameters, respectively; a1 and a2 are the regression exponents of the crosswind and vertical diffusion parameters, respectively; χ is the downwind straight-line distance from the leakage source to the receiving point; erf is the error function.

[0020] A further technical solution of the present invention is: in the RMN layer, the radiation dose rate value in the grid cell is calculated by formula 2;

[0021] Formula 2: RDR = C·DF

[0022] Where RDR is the radiation dose rate, C is the concentration of the radionuclide, and DF is the dose conversion factor determined by the characteristics of the radionuclide itself.

[0023] A further technical solution of the present invention is: in the IN layer:

[0024] I. Adjacency Matrix A IN Defined by Formula 3;

[0025] Formula 3:

[0026] Where N is the total number of nodes in the IN layer, and a is an element of the adjacency matrix;

[0027] II. The attributes of physical connections are determined by the weight matrix W = (w ij ) N×N Independent storage, where N is the total number of nodes in the IN layer, and w is the element of the weight matrix; when the physical connection between node i and node j is a road, w ij w is the road length; when the physical connection between node i and node j is an optical fiber, w ij For communication bandwidth;

[0028] III. The association rule between the adjacency matrix and the weight matrix is: if a ij =1, then w ij >0; if a ij =0, then w ij =0.

[0029] A further technical solution of the present invention is: In the IN layer, there is a spatial constraint rule for cross-network connections: when the distance d between two cross-network nodes is greater than d... max At that time, cross-network connections are not established.

[0030] A further technical solution of the present invention is: in the IN layer, Each grid cell in the equation has a state value defined by Equation 4;

[0031] Formula 4:

[0032] Where k is the node number in the IN layer, N is the total number of nodes in the IN layer, (x k ,y k (i,j) represents the geographic coordinates of node k; Δ represents the grid resolution; and (i,j) represents the grid cell index.

[0033] The logic of Formula 4 is as follows: Conditional branch 1 (with node mapping): If there exists a node k, whose coordinates (x... k ,y k If a given element is mapped to a grid cell (i,j), then the state value S of that grid cell is... ij The value is assigned to the node number k; Condition branch 2 (no node mapping): If the grid cell (i,j) has no node mapping, its state value is 0.

[0034] A further technical solution of the present invention is as follows: In the EN layer, the nodes under the nuclear power plant include the emergency command center, operation control group, safety protection group, network security group, equipment support group, public information group, emergency repair group, emergency secretariat group, fire protection group, repair support group, technical support group, logistics support group, emergency mobile equipment special group, construction and commissioning group, and China Nuclear Power Information Liaison Group; the nodes under the government departments include the National Nuclear Accident Emergency Office, the Provincial Nuclear Accident Emergency Office, the National Nuclear Safety Administration, the National Energy Administration, the East China Nuclear and Radiation Safety Supervision Station, the Provincial Meteorological Service Center, the East China Sea Forecasting Center of the State Oceanic Administration, the Provincial Earthquake Bureau, and the China Public Opinion Supervision Department; the nodes under the rescue agencies include the municipal armed police, the public security fire brigade, the municipal 120 emergency station, and the Nuclear Industry General Hospital; the nodes under the social organizations include China National Nuclear Corporation, China Nuclear Power, the mutual aid nuclear power plant, the command platform developer, the mechanical equipment supplier, the electrical equipment supplier, and the instrumentation and control equipment supplier.

[0035] A further technical solution of the present invention is: in the EN layer, the adjacency matrix A EN Defined by Formula 5;

[0036] Formula 5:

[0037] Where n is the total number of nodes in the EN layer, a is an element of the adjacency matrix, and the adjacency matrix A EN It has symmetry and no self-loop properties.

[0038] A further technical solution of the present invention is as follows: In the EN layer, the topology connection rules include the following points:

[0039] I. All members within a node are fully connected;

[0040] II. Within a node, there is a group leader with decision-making power. Nodes of the same type or with cooperative relationships maintain connections through the group leader.

[0041] III. Some nodes have hierarchical attributes, and the connection between the upper-level node and the lower-level node is maintained through the group leader;

[0042] IV. The members within a node include liaisons responsible for sending and receiving information. If the other three types of nodes, excluding nuclear power plants, have a cooperative relationship with a node in a nuclear power plant, they maintain a connection through the liaison.

[0043] A further technical solution of the present invention is: in the EN layer, nodes with connection relationships are assigned dynamic weights w(t), and the dynamic weights w(t) are defined by formula 6;

[0044] Formula 6:

[0045] In the dynamic weight w(t) formula: t is the time; a(t) is the communication quality index, used to quantify the real-time reliability of the communication link; β(t) is the cooperation dependency strength index, used to dynamically adjust the cooperation priority so that resource allocation focuses on key cooperation links; γ(t) is the response strength adjustment factor, realizing a triple emergency response based on radiation threat, time pressure and resource efficiency.

[0046] In the sub-formula for the communication quality index a(t): t is time; D suc (t) represents the number of data packets successfully transmitted at time t, and D req (t) represents the number of data packets requested to be transmitted at time t; D suc (t) and D req The ratio of (t) is used to reflect the degree of congestion in the communication network. When communication fails, it is determined that communication is paralyzed; SNR(t) is the signal-to-noise ratio at time t; Φ rad (t) is the radiation attenuation factor, representing the physical compensation for radiation damage to the base station. When radiation causes damage to the base station, Make Φ rad (t)=0, forcing a(t)=0; ξ is the base station radiation resistance coefficient;

[0047] In the formula for the cooperative dependency strength exponent β(t): t is time; β is the exponent. base The basic value for the strength of cooperative dependency; (1+α·RDR) norm ) represents the radiation risk amplification term; α is the risk sensitivity coefficient (taken as 0.2-0.5). The current radiation dose rate (RDR) and the universal system threshold RDR for radiation dose rate are used. max-1 The ratio; (1-e -k·PDR(t) ) represents the communication attenuation factor; k is the attenuation constant (taken as 2.0), and PDR(t) is the data packet arrival rate at time t;

[0048] In the sub-formula of the response intensity adjustment factor γ(t): k1, k2, and k3 are the radiation threat weight, time pressure weight, and resource efficiency weight, respectively, satisfying k1+k2+k3=1; This is a radiation threat term, reflecting the amplification effect of radiation levels on response intensity. The current radiation dose rate (RDR) and the device-specific threshold RDR for radiation dose rate. max-2 The ratio; k2·e -t / τ This is a time pressure term, used to reflect the urgency of the response due to the duration of the incident; e -t / τ τ is the time decay factor, which characterizes the urgency pressure that decays exponentially over time, and τ is the response half-life constant, which characterizes the decay rate of tissue response efficiency. This is a resource efficiency item, used to reflect the synergistic effect between resource efficiency and decision-making efficiency, η.r Resource availability rate represents the proportion of available resources to total demand, δ d This is the decision delay coefficient, representing the degree of delay in decision-making due to lack of information.

[0049] A further technical solution of the present invention is: in the coupling mechanism, the coupling strength ε is defined by formula 7;

[0050] Formula 7:

[0051] Where, N IE This is the number of nodes in the IN layer that depend on the EN layer; it is counted when an IN node requires an EN layer instruction to function. N I N represents the total number of nodes in the IN layer; EI This refers to the number of nodes in the EN layer that depend on the IN layer, counted when an EN node requires IN facility support to function; N E This represents the total number of nodes in the EN layer.

[0052] A further technical solution of the present invention is: in the coupling mechanism, the radiation dose rate universal system threshold RDR max-1 Set to 500 μSv / h.

[0053] The technical solution of this invention is: a method for constructing a three-layer coupled network model for nuclear accident emergency response, used to construct a three-layer coupled network model for nuclear accident emergency response, the steps of which are as follows:

[0054] S1, Construct the RMN layer:

[0055] A. The area affected by the nuclear accident Ω XY Mesh the network at resolution Δ to generate an RMN-layer spatial network.

[0056] B. Calculation using a dynamic Gaussian plume model The concentration of radionuclides in each grid cell;

[0057] C. Converting radionuclide concentration into radiation dose rate using the dose conversion factor DF;

[0058] S2, construct the IN layer:

[0059] A two-layer network coupling transportation and communication networks is constructed based on dependent networks. The construction process includes topology modeling and spatial modeling.

[0060] A. Topology Modeling: Define nodes including traffic intersections and communication base stations; node states include susceptible (S), infected (I), immune (R), and damaged (D); intra-network edges include roads and fiber optic cables; cross-network edges represent the functional dependencies between the traffic network and the communication network; establish the adjacency matrix A. INDefine the physical connections between nodes in the IN layer; obtain network G through topology modeling. IN =(N IN E IN ), N IN E is the set of all nodes in the IN layer. IN This is the set of all edges connected to the IN layer;

[0061] B. Spatial Modeling: First, define the nuclear accident impact area Ω XY Δ-meshation at resolution generates IN-layer spatial networks. Then map the geographic coordinates of all nodes in the IN layer to... In the corresponding grid cell;

[0062] S3, construct the EN layer:

[0063] A. Define four types of nodes: nuclear power plants, government departments, rescue agencies, and social organizations; each type of node is further divided into multiple nodes based on their functions; establish an adjacency matrix A. EN This represents the command or collaboration relationships between nodes in this layer.

[0064] B. Calculate the topology network G based on real-time communication quality, cooperative dependency strength, and connection response strength. EN The dynamic weights of each connected edge.

[0065] Compared with the prior art, the present invention has the following advantages:

[0066] 1. It constructs a nuclear accident emergency network model based on complex network theory. The model designs a three-layer network architecture according to the characteristics of nuclear emergency response, deeply analyzes the connection logic between network elements in each layer, and reveals the structural characteristics of the nuclear accident emergency network through spatial correlation and functional interaction, providing a complete network foundation for subsequent risk evolution modeling and optimal control strategy design.

[0067] 2. The three-layer coupled network model for nuclear accident emergency response adopts a three-layer coupled architecture of "radioactive contamination (RMN layer) → facility failure (IN layer) → organization response (EN layer)". The vertical transmission of risk between the RMN layer and the IN layer is realized through the rule of "excessive radiation dose rate in RMN layer → damage state of corresponding node in IN layer". The global dependency density between the IN layer and the EN layer is quantified by the coupling strength, thereby solving the problem of model fragmentation in the existing technology and realizing closed-loop simulation of risk chain.

[0068] 3. The three-layer coupled network model for nuclear accident emergency response incorporates an "inertia mechanism" to accurately describe the persistent impact of risk propagation and the lag in recovery, thus addressing the problem of existing technical models neglecting the inertia of risk propagation.

[0069] Specifically, this is achieved through the following technical features:

[0070] 3.1 A four-state node model (S / I / R / D) was designed. The infected state (I) simulates a "transitional state" in risk propagation (e.g., traffic congestion at an intersection but not paralyzed), requiring continuous exposure time to reach a threshold before transitioning to the damaged state (D), reflecting "persistent impact." The immune state (R) represents equipment entering a risk-resistant state after temporary repair (e.g., activating backup circuits), but requires repair time (not instantaneous switching) and may revert to the infected state (I) due to a resurgence in radiation. The transition from damaged state (D) to immune state (R) requires manual repair time (e.g., 6 hours for base station repair), forcibly simulating "recovery lag." The transition from infected state (I) to damaged state (D) requires a cumulative radiation dose rate exceeding a threshold (e.g., 10 minutes) to avoid instantaneous failure. Example: At an intersection, panicked escapes from traffic jams (infected state (I)). Even if radiation decreases, vehicle evacuation still takes time (remaining in the infected state (I)) until traffic flow is restored (susceptible state (S)). Traditional SIR models cannot describe this process.

[0071] 3.2 The radiation attenuation factor Φ is embedded in the dynamic weighting formula (Formula 6). rad (t), reflecting the non-instantaneous communication recovery after base station damage (requiring repair time), and reflecting the inertia of facility recovery; the historical attenuation value of the data packet arrival rate PDR(t) affects the current communication quality index a(t), simulating the communication recovery delay; a radiation risk amplification term (1+α·RDR) is introduced. norm This reflects that even after the radiation level decreased, the cooperation weight remained high due to "lingering panic"; the time pressure term k2·e -t / τ This indicates that the longer the accident lasts, the higher the urgency of the response, reflecting the accumulation of psychological stress. Example: Even if radiation levels drop to a safe level (RDR < 500 μSv / h), the radiation risk amplification term (1 + α·RDR) in the cooperation dependence intensity index β(t) remains significant. norm The public information group's weight remains high, requiring continued release of reassuring information until β(t) decays naturally.

[0072] 3.3 The dynamic Gaussian plume model adopts a discretized time period design. Although the meteorological conditions are constant within each time period, the plume diffusion state is inherited between time periods, reflecting the continuous pollution effect. The error function erf describes the transient cumulative effect of the plume leading edge (such as the radiation concentration not immediately returning to zero after a sudden change in wind direction).

[0073] 4. The three-layer coupled network model for nuclear accident emergency response achieves a closed-loop quantitative decision-making chain from "risk key point location → measure priority ranking → global effect prediction" through full-chain topology simulation and the construction of dynamic weights in the EN layer. This solves the pain point in existing technologies where emergency resources cannot be deployed to the optimal measures due to the neglect of network topology leverage effect.

[0074] Specifically, this is achieved through the following technical features:

[0075] 4.1 Based on dynamic weight-driven resource priority ranking, the effectiveness of measures is quantified; in the dynamic weight formula (Formula 6):

[0076] Radiation risk amplification: Dynamically increase the decision priority of high-risk collaboration links to achieve resource reallocation driven by radiation threat, such as automatically increasing the weight of key groups such as medical rescue when radiation levels rise;

[0077] Communication attenuation factor: quantifies the inhibitory effect of communication mid-segment on collaboration efficiency, and constructs a communication quality-decision efficiency feedback loop. For example, when the communication quality drops to a certain level, the weight of the logistics group is automatically reduced.

[0078] 4.2 Identify network-level bottleneck types (such as "communication-dependent" or "command-dependent") by coupling strength (Formula 7), quantify the bidirectional dependency density between the IN and EN layers, diagnose the optimal measure type, and avoid resource mismatch.

[0079] The present invention will be further described below with reference to the figures and embodiments. Attached Figure Description

[0080] Figure 1 A schematic diagram of a three-layer coupled network for nuclear accident emergency response. Detailed Implementation

[0081] Example 1:

[0082] The three-layer coupled network model for nuclear accident emergency response includes a nuclide migration network layer, an infrastructure network layer, and an emergency organization network layer.

[0083] The nuclide migration network layer, or RMN layer for short, is used to simulate the dynamic diffusion process of radionuclides in the atmosphere after a nuclear accident leak; it is constructed based on a dynamic Gaussian plume model; and it represents the nuclear accident-affected region Ω. XY Mesh the network into an RMN-layer spatial network at a resolution Δ. Its nodes are The grid cells in the model have node states that correspond to the radiation dose rate in the grid cells. The radiation dose rate is calculated and converted from the dynamic Gaussian plume model.

[0084] In the RMN layer, the dynamic Gaussian plume model is obtained by introducing a diffusion time correction coefficient into the standard Gaussian plume model, discretizing the plume diffusion into multiple time periods, and assuming that the meteorological conditions are constant in each time period, so as to adapt to the dynamic response requirements of nuclear accident emergency decision-making; the dynamic Gaussian plume model is shown in Equation 1.

[0085] Formula 1:

[0086] Where C(x,y,z,t) is the concentration of radionuclides at spatial location (x,y,z) at time t (t represents the actual physical time elapsed continuously from the starting point of the nuclear accident leak); Q is the source strength (i.e., the activity release rate of radionuclides, representing the total activity of radionuclides released from the leak source into the environment per unit time during the nuclear accident); λ is the radionuclide decay constant; t n The duration of the nth time period; u represents the downwind direction of the leakage source in the current time period (x in the dynamic spatial coordinate system). n Average wind speed (in the axial direction); σ x σ y σ z These are the diffusion coefficients in the downwind, crosswind, and vertical directions, respectively; h is the effective release height of the plume at the current time (including the plume lifting effect); x n ,y n ,z n Let be the coordinates of the center point of the plume diffusion area in the nth time period. In the dynamic spatial coordinate system in which it is located, the x-axis is parallel to the prevailing wind direction of the current time period, the y-axis is the horizontal direction perpendicular to the downwind direction of the current time period, and the z-axis is the vertical direction perpendicular to the ground. exp is an exponential function; γ1 and γ2 are the regression coefficients of the crosswind and vertical diffusion parameters, respectively; a1 and a2 are the regression exponents of the crosswind and vertical diffusion parameters, respectively (both are determined by the atmospheric stability level); χ is the downwind straight-line distance from the leakage source to the receiving point; erf is the error function (used to describe the transient cumulative effect of the plume diffusion front).

[0087] In the RMN layer, the radiation dose rate in the grid cells is calculated using Equation 2:

[0088] Formula 2: RDR = C·DF

[0089] Where RDR is the radiation dose rate in the current time period, C is the concentration of radionuclides (calculated by the dynamic Gaussian plume model), and DF is the dose conversion factor determined by the characteristics of the radionuclides themselves.

[0090] The Infrastructure Network Layer (IN layer) is used to construct the spatial distribution and dependency network of infrastructure. It is a two-layer network coupling transportation and communication networks, built using a dependent network architecture. First, nodes are defined, including traffic intersections and communication base stations. Node states include vulnerable (S) (meaning the facility is functional but lacks protective capabilities), infected (I) (meaning it is partially disabled due to indirect risks, such as a congested but not paralyzed intersection), immune (R) (meaning it is in a resilient state after temporary repairs, such as a substation with redundant circuitry), and damaged (D) (meaning it is unusable due to direct radiation damage, such as monitoring equipment damaged by ionization). Intra-network edges include roads and optical cables; cross-network edges represent the functional dependencies between the transportation and communication networks. These dependencies are defined using the adjacency matrix A.IN Define the physical connections between nodes in the IN layer; obtain network G through topology modeling. IN =(N IN E IN ), N IN E is the set of all nodes in the IN layer. IN The set of all edges in the IN layer; then the nuclear accident-affected region Ω. XY Mesh the network into an IN-layer spatial network based on resolution Δ. ( and (With the same resolution and complete overlap), finally, the geographic coordinates of all nodes in the IN layer are mapped to... The center point of the corresponding grid cell.

[0091] In the IN layer:

[0092] I. Adjacency Matrix A IN Defined by Formula 3;

[0093] Formula 3:

[0094] Where N is the total number of nodes in the IN layer, and a is an element of the adjacency matrix;

[0095] II. The attributes of physical connections are determined by the weight matrix W = (w ij ) N×N Independent storage, where N is the total number of nodes in the IN layer, and w is the element of the weight matrix; when the physical connection between node i and node j is a road, w ij w is the road length; when the physical connection between node i and node j is an optical fiber, w ij For communication bandwidth;

[0096] III. The association rule between the adjacency matrix and the weight matrix is: if a ij =1, then w ij >0; if a ij =0, then w ij =0.

[0097] In the IN layer, there is a spatial constraint rule for cross-network connections (crossing transportation and communication networks): when the distance d between two cross-network nodes (crossing transportation and communication networks) is greater than d... maxAt this time, cross-network connections / links are not established to ensure that functional dependencies conform to engineering reality. The reasons for setting the constraint rules are mainly based on the following two points: ①. Reasonableness of risk propagation: To avoid introducing unrealistic long-distance dependencies into the model and ensure that the risk propagation path conforms to real-world logic (e.g., maintenance vehicles cannot cross 50km to support base stations) and engineering constraints; ②. Engineering constraints: In real-world scenarios, the functional dependencies between traffic nodes (intersections) and communication nodes (base stations) are limited by physical distance (e.g., the maintenance of communication base stations requires maintenance support from the traffic network; if the distance is too far, timely response to faults is impossible; for example, traffic signal control relies on the communication network to transmit commands; if the distance between the base station and the intersection exceeds the effective signal coverage range, real-time control cannot be achieved); d max The principle for determining the value of d is: for two cross-network nodes that rely on physical maintenance, d max Set to a distance that the vehicle can reach in one hour, i.e., 30-50km; for the two cross-network nodes that real-time control depends on, d max Set the maximum distance for wireless signal coverage, i.e., 300-1000m.

[0098] In the IN layer, Each grid cell in the equation has a state value defined by Equation 4:

[0099] Formula 4:

[0100] Where k is the node number in the IN layer, N is the total number of nodes in the IN layer, (x k ,y k (i,j) represents the geographic coordinates of node k; Δ represents the grid resolution; and (i,j) represents the grid cell index.

[0101] The logic of Formula 4 is as follows: Conditional branch 1 (with node mapping): If there exists a node k, whose coordinates (x... k ,y k If a given element is mapped to a grid cell (i,j), then the state value S of that grid cell is... ij The value is assigned to the node number k; Condition branch 2 (no node mapping): If the grid cell (i,j) has no node mapping, its state value is 0.

[0102] The Emergency Organization Network Layer (EN Layer) aims to topologically define the command and coordination relationships of emergency organizations. It includes four types of nodes: nuclear power plants, government departments, rescue agencies, and social organizations. Each type of node is further divided into multiple nodes based on their functions. This is achieved through an adjacency matrix A. EN Define the command or cooperative connection relationships between nodes in the EN layer; perform topology modeling on all nodes in the EN layer to obtain network G. EN =(N EN EEN ), N EN E is the set of all nodes in the EN layer. EN The set of all edges in the EN layer. The topology network G is calculated based on real-time communication quality, cooperative dependency strength, and connection response strength. EN The dynamic weights of each connected edge.

[0103] In the EN layer, the nodes under the nuclear power plant include the Emergency Command Center, Operation Control Group, Safety Protection Group, Network Support Group, Equipment Support Group, Public Information Group, Emergency Repair Group, Emergency Secretariat Group, Fire Protection Group, Repair Support Group, Technical Support Group, Logistics Support Group, Emergency Mobile Equipment Special Group, Construction and Commissioning Group, and China Nuclear Power Information Liaison Group; the nodes under the government departments include the National Nuclear Accident Emergency Office, Provincial Nuclear Accident Emergency Office, National Nuclear Safety Administration, National Energy Administration, East China Nuclear and Radiation Safety Supervision Station, Provincial Meteorological Service Center, East China Sea Forecasting Center of the State Oceanic Administration, Provincial Earthquake Bureau, and China Public Opinion Supervision Department; the nodes under the rescue agencies include the Municipal Armed Police, Public Security Fire Brigade, Municipal 120 Emergency Station, and Nuclear Industry General Hospital; the nodes under the social organizations include China National Nuclear Corporation, China Nuclear Power, Mutual Aid Nuclear Power Plant, Command Platform Developer, Mechanical Equipment Supplier, Electrical Equipment Supplier, and Instrumentation and Control Equipment Supplier.

[0104] In the EN layer, the adjacency matrix A EN Defined by Formula 5;

[0105] Formula 5:

[0106] Where n is the total number of nodes in the EN layer, a is an element of the adjacency matrix, and the adjacency matrix A EN It has symmetry and no self-loop properties.

[0107] In the EN layer, the topology connection rules include the following:

[0108] I. All members within a node are fully connected; for example, the five members of a safety protection group (group leader, radiation monitor, protective equipment manager, on-site coordinator, and data analyst) form a fully connected subnet to ensure real-time information sharing within the group;

[0109] II. Within a node, there is a group leader with decision-making authority. Nodes of the same type or with cooperative relationships maintain connections through the group leader. For example, the equipment support group leader and the emergency repair support group leader may need to coordinate the allocation of heavy machinery and establish a two-way collaborative link.

[0110] III. Some nodes have hierarchical attributes, and the connection between upper-level and lower-level nodes is maintained through the group leader; for example, the emergency command center (upper-level) → operation control group leader (lower-level) constitutes a command chain for issuing instructions to the radiation isolation zone;

[0111] IV. The members within a node include liaisons responsible for sending and receiving information. If the other three types of nodes, excluding nuclear power plants, have a cooperative relationship with a node in a nuclear power plant, they maintain a connection through the liaison. For example, the liaison of the National Nuclear Safety Administration → assistant of the safety protection group transmits radiation monitoring data to the national regulatory platform.

[0112] In the EN layer, nodes with connections are assigned a dynamic weight w(t), which is defined by Equation 6:

[0113] Formula 6:

[0114] In the dynamic weight w(t) formula: t is time (representing the real physical time continuously elapsed from the starting point of the nuclear accident leak); a(t) is the communication quality index, used to quantify the real-time reliability of the communication link (when a(t) > 0.7: represents a green channel (command transmission delay < 1s); when 0.4 ≤ a(t) ≤ 0.7: represents a yellow warning (data retransmission required); when a(t) < 0.4: represents a red interruption (satellite backup link should be activated)); β(t) is the collaboration dependency strength index, used to dynamically adjust collaboration priorities, focusing resource allocation on key collaboration links (e.g., automatically increasing the weight of the medical rescue team when radiation increases, and automatically reducing the weight of the logistics team when communication is interrupted or weakened); β(t) changes in real time with the accident level and communication status, automatically increasing the weight when the accident escalates and automatically decreasing the weight when communication quality declines; it is coupled with physical risk, incorporating radiation levels into the assessment of collaboration urgency; it is coupled with functional weights, when radiation levels rise, β... base The higher the value of a node, the more significant the weight increase, β. base The weight of nodes with lower values ​​decreases more significantly; γ(t) is a response intensity adjustment factor to achieve a triple emergency response based on radiation threat, time pressure and resource efficiency.

[0115] In the communication quality index a(t) sub-formula: t is time (representing the actual physical time continuously elapsed from the starting point of the nuclear accident leak); D suc (t) represents the number of data packets successfully transmitted at time t, and D req (t) represents the number of data packets requested to be transmitted at time t; D suc (t) and D req The ratio of (t) is used to reflect the degree of congestion in the communication network. When the signal is lost, communication is considered to be paralyzed; SNR(t) is the signal-to-noise ratio at time t (used to reflect the signal quality of the communication network); Φ rad (t) is the radiation attenuation factor, representing the physical compensation for radiation damage to the base station. When radiation causes damage to the base station... Make Φ rad(t)=0, forced a(t)=0; ξ is the base station radiation resistance coefficient.

[0116] In the sub-formulation of the cooperation dependency strength index β(t): t is time (representing the real physical time continuously elapsed from the starting point of the nuclear accident leak); the index β base The base value for collaboration dependency strength reflects the inherent importance level of different collaboration links (lifeline collaboration is 0.95 (e.g., medical rescue team - 120 emergency station); critical control collaboration is 0.9 (e.g., emergency command center - security protection team); resource support collaboration is 0.8 (e.g., equipment support team - machinery and equipment supplier); information collaboration is 0.75 (e.g., public information team - public opinion monitoring department); routine support collaboration is 0.6 (e.g., secretariat team - logistics team)); (1+α·RDR) norm The radiation risk amplification term is used to dynamically increase the decision priority of high-risk collaborative links, thereby achieving resource reallocation driven by radiation threats; α is the risk sensitivity coefficient (ranged from 0.2 to 0.5). The current radiation dose rate (RDR) and the universal system threshold RDR for radiation dose rate are used. max-1 The ratio; (1-e -k·PDR(t) ) is the communication attenuation factor, which is used to quantify the inhibitory effect of the communication mid-segment on the cooperation efficiency and to construct the communication quality-decision efficiency feedback loop; k is the attenuation constant (taken as 2.0), and PDR(t) is the data packet arrival rate at time t.

[0117] In the sub-formula of the response intensity adjustment factor γ(t): k1, k2, and k3 are the radiation threat weight, time pressure weight, and resource efficiency weight, respectively, satisfying k1+k2+k3=1; This is a radiation threat term, used to reflect the amplification effect of radiation levels on response intensity. The current radiation dose rate (RDR) and the device-specific threshold RDR for radiation dose rate. max-2 The ratio (radiation dose rate device-specific threshold RDR) max-2 (The critical value for ionization damage of the equipment); k2·e -t / τ This is a time pressure factor, used to reflect the urgency of the response due to the duration of the incident. Its design aims to maximize efficiency when resources are sufficient and decision-making is efficient; and to forcibly reduce resource weight when information is lacking, avoiding ineffective resource allocation. -t / τ τ is the time decay factor, which characterizes the urgency pressure that decays exponentially over time, and τ is the response half-life constant, which characterizes the decay rate of tissue response efficiency. This is a resource efficiency item, used to reflect the synergistic effect between resource efficiency and decision-making efficiency, η. r δ represents resource availability, which is the proportion of available resources to total demand. d This is the decision delay coefficient, representing the degree of delay in decision-making due to lack of information.

[0118] The coupling mechanism of the three-layer network in the model: I. When The current radiation dose rate (RDR) of the grid is greater than or equal to the universal system threshold for radiation dose rate (RDR). max-1 hour, II. Mark the IN layer nodes in the same location grid as damaged state (D); ii. Quantify the bidirectional dependency density between IN layer and EN layer (EN layer node pointing to IN layer node is command function dependency, IN layer node pointing to EN layer node is state feedback dependency) by coupling strength ε.

[0119] The coupling strength ε is defined by Equation 7:

[0120] Formula 7:

[0121] Where, N IE This is the number of nodes in the IN layer that depend on the EN layer. It is counted when an IN node requires an EN layer command to function (e.g., a traffic lightless intersection requires an EN layer traffic control command); N I N represents the total number of nodes in the IN layer; EI This refers to the number of nodes in the EN layer that depend on the IN layer, counted when an EN node requires IN infrastructure support to function (e.g., an emergency command center relies on an IN layer base station to transmit video surveillance); N E This represents the total number of nodes in the EN layer.

[0122] The RDR max-1 The radiation dose rate is a general system threshold of 500 μSv / h, which is mainly set based on the radiation failure threshold of semiconductor devices or the exposure limit of emergency personnel in a single mission.

[0123] Figure 1 The diagram shows the structure of a three-layer coupled network for nuclear accident emergency response. From top to bottom, the three planes represent the radionuclide migration network layer (RMN layer), the infrastructure network layer (IN layer), and the emergency organization network layer (EN layer). The diagram illustrates the spatial network of the RMN layer. With IN layer spatial network The resolution is consistent and the layers are aligned vertically. The black dots in the RMN layer represent leakage sources, and the multiple circles surrounding the black dots represent the diffusion range of radionuclides at different times. These circles are projected into the IN layer. The multiple black dots in the IN layer represent nodes representing traffic intersections or communication base stations, which are marked according to their geographical coordinates. In the IN layer, the nodes within the projection circle are those located within the radionuclide coverage area of ​​the corresponding time period. The blue lines in the EN layer represent the command or collaborative connections between the nodes in the EN layer. The dashed lines connecting EN layer nodes and IN layer nodes represent the cross-layer dependencies between the EN and IN layers, including command function dependencies (EN nodes (e.g., emergency command centers) depend on IN nodes (communication base stations) to transmit instructions) and status feedback dependencies (IN→EN, for example, IN nodes (traffic intersection cameras) transmit real-time status back to EN nodes (operation control groups)).

Claims

1. A three-layer coupled network model for nuclear accident emergency response, characterized by: include: Nuclide migration network layer: abbreviated as RMN layer, used to simulate the dynamic diffusion process of radioactive nuclides in the atmosphere after a nuclear accident leak; It is constructed based on a dynamic Gaussian plume model; the nuclear accident-affected area Ω XY Mesh the network into an RMN-layer spatial network at a resolution Δ. Its nodes are The grid cells in the model have node states that correspond to the radiation dose rate in the grid cells. The radiation dose rate is calculated and converted from the dynamic Gaussian plume model. Infrastructure Network Layer (IN layer): This layer is used to build the spatial distribution and dependency network of infrastructure. It is a two-layer network consisting of a traffic network and a communication network, constructed using a dependent network architecture. First, nodes are defined as traffic intersections and communication base stations, with node states including susceptible (S), infected (I), immune (R), and damaged (D). Intra-network edges include roads and optical cables; cross-network edges represent the functional dependencies between the traffic network and the communication network; and adjacency matrix A is used to define these dependencies. IN Define the physical connections between nodes in the IN layer; obtain network G through topology modeling. IN =(N IN E IN ), N IN E is the set of all nodes in the IN layer. IN The set of all edges in the IN layer; then the nuclear accident-affected region Ω. XY Mesh the network into an IN-layer spatial network based on resolution Δ. Finally, the geographic coordinates of all nodes in the IN layer are mapped to... In the corresponding grid cell; Emergency Organization Network Layer (EN Layer): Its purpose is to define the command or collaboration relationships of emergency organizations in a topological manner. It includes four types of nodes: nuclear power plants, government departments, rescue agencies, and social organizations; each type of node is further divided into multiple nodes based on their functions; and is connected via an adjacency matrix A. EN Define the command or cooperative connection relationships between nodes in the EN layer; perform topology modeling on all nodes in the EN layer to obtain network G. EN =(N EN E EN ), N EN E is the set of all nodes in the EN layer. EN The set of all edges in the EN layer; the topology network G is calculated based on real-time communication quality, cooperative dependency strength, and connection response strength. EN The dynamic weights of each connected edge in the middle; The coupling mechanism of the three-layer network in the model: I. When The current radiation dose rate (RDR) in the grid is greater than or equal to the general system threshold for radiation dose rate (RDR). max-1 hour, IN layer nodes within the same location grid are marked as damaged (D); II. Quantify the bidirectional dependency density between the IN and EN layers using coupling strength ε.

2. The three-layer coupled network model for nuclear accident emergency response as described in claim 1, characterized in that: In the RMN layer, the dynamic Gaussian plume model is obtained by introducing a diffusion time correction coefficient into the standard Gaussian plume model, discretizing the plume diffusion into multiple time periods, and assuming that the meteorological conditions are constant in each time period. The dynamic Gaussian plume model is shown in Equation 1; Formula 1: Where C(x,y,z,t) is the concentration of the radionuclide at spatial location (x,y,z) at time t; Q is the source strength; λ is the radionuclide decay constant; t n σ represents the duration of the nth time period; u represents the average wind speed downwind of the leak source during the current time period; σ represents the average wind speed downwind of the leak source during the current time period. x σ y σ z These are the diffusion coefficients in the downwind, crosswind, and vertical directions, respectively; h is the effective plume release height for the current period; x n ,y n ,z n Let be the coordinates of the center point of the plume diffusion area in the nth time period. In the dynamic spatial coordinate system in which it is located, the x-axis is parallel to the prevailing wind direction of the current time period, the y-axis is the horizontal direction perpendicular to the downwind direction of the current time period, and the z-axis is the vertical direction perpendicular to the ground. exp is an exponential function; γ1 and γ2 are the regression coefficients of the crosswind and vertical diffusion parameters, respectively; a1 and a2 are the regression exponents of the crosswind and vertical diffusion parameters, respectively; χ is the downwind straight-line distance from the leakage source to the receiving point; erf is the error function.

3. The three-layer coupled network model for nuclear accident emergency response as described in claim 2, characterized in that: In the RMN layer, the radiation dose rate in the grid cell is calculated by Equation 2; Formula 2: RDR = C·DF Where RDR is the radiation dose rate, C is the concentration of the radionuclide, and DF is the dose conversion factor determined by the characteristics of the radionuclide itself.

4. The three-layer coupled network model for nuclear accident emergency response as described in claim 3, characterized in that: In the IN layer: I. Adjacency Matrix A IN Defined by Formula 3; Formula 3: Where N is the total number of nodes in the IN layer, and a is an element of the adjacency matrix; II. The attributes of physical connections are determined by the weight matrix W = (w ij ) N×N Independent storage, where N is the total number of nodes in the IN layer, and w is the element of the weight matrix; when the physical connection between node i and node j is a road, w ij w is the road length; when the physical connection between node i and node j is an optical fiber, w ij For communication bandwidth; III. The association rule between the adjacency matrix and the weight matrix is: if a ij =1, then w ij >0; if a ij =0, then w ij =0.

5. The three-layer coupled network model for nuclear accident emergency response as described in claim 4, characterized in that: In the IN layer, there is a spatial constraint rule for cross-network connections: when the distance d between two cross-network nodes is greater than d... max At that time, no cross-network connection is established; In the IN layer, Each grid cell in the equation has a state value defined by Equation 4; Formula 4: Where k is the node number in the IN layer, N is the total number of nodes in the IN layer, (x k ,y k (i,j) represents the geographic coordinates of node k; Δ represents the grid resolution; and (i,j) represents the grid cell index. The logic of Formula 4 is as follows: Conditional branch 1: If there exists a node k, whose coordinates (x... k ,y k If a given element is mapped to a grid cell (i,j), then the state value S of that grid cell is... ij The value is assigned to the node number k; Condition branch 2: If no node is mapped to the grid cell (i,j), then its state value is 0.

6. The three-layer coupled network model for nuclear accident emergency response as described in claim 5, characterized in that: In the EN layer, the nodes under the nuclear power plant include the Emergency Command Center, Operation Control Group, Safety Protection Group, Network Support Group, Equipment Support Group, Public Information Group, Emergency Repair Group, Emergency Secretariat Group, Fire Protection Group, Repair Support Group, Technical Support Group, Logistics Support Group, Emergency Mobile Equipment Special Group, Construction and Commissioning Group, and China Nuclear Power Information Liaison Group; the nodes under the government departments include the National Nuclear Accident Emergency Office, Provincial Nuclear Accident Emergency Office, National Nuclear Safety Administration, National Energy Administration, East China Nuclear and Radiation Safety Supervision Station, Provincial Meteorological Service Center, East China Sea Forecasting Center of the State Oceanic Administration, Provincial Earthquake Bureau, and China Public Opinion Supervision Department; the nodes under the rescue organizations include the Municipal Armed Police, Public Security Fire Brigade, Municipal 120 Emergency Station, and Nuclear Industry General Hospital; the nodes under the social organizations include China National Nuclear Corporation, China Nuclear Power, Mutual Aid Nuclear Power Plant, Command Platform Developer, Mechanical Equipment Supplier, Electrical Equipment Supplier, and Instrumentation and Control Equipment Supplier. In the EN layer, the adjacency matrix A EN Defined by Formula 5; Formula 5: Where n is the total number of nodes in the EN layer, a is an element of the adjacency matrix, and the adjacency matrix A EN It has symmetry and no self-loop properties.

7. The three-layer coupled network model for nuclear accident emergency response as described in claim 6, characterized in that: In the EN layer, the topology connection rules include the following: I. All members within a node are fully connected; II. Within a node, there is a group leader with decision-making power. Nodes of the same type or with cooperative relationships maintain connections through the group leader. III. Some nodes have hierarchical attributes, and the connection between the upper-level node and the lower-level node is maintained through the group leader; IV. The members within a node include liaisons responsible for sending and receiving information. If the other three types of nodes, excluding nuclear power plants, have a cooperative relationship with a node in a nuclear power plant, they maintain a connection through the liaison.

8. The three-layer coupled network model for nuclear accident emergency response as described in claim 7, characterized in that: In the EN layer, nodes with connections are assigned a dynamic weight w(t), which is defined by Equation 6. Formula 6: In the dynamic weight w(t) formula: t is the time; a(t) is the communication quality index, used to quantify the real-time reliability of the communication link; β(t) is the cooperation dependency strength index, used to dynamically adjust the cooperation priority so that resource allocation focuses on key cooperation links; γ(t) is the response strength adjustment factor, realizing a triple emergency response based on radiation threat, time pressure and resource efficiency. In the sub-formula for the communication quality index a(t): t is time; D suc (t) represents the number of data packets successfully transmitted at time t, and D req (t) represents the number of data packets requested to be transmitted at time t; D suc (t) and D req The ratio of (t) is used to reflect the degree of congestion in the communication network. When communication fails, it is determined that communication is paralyzed; SNR(t) is the signal-to-noise ratio at time t; Φ rad (t) is the radiation attenuation factor, representing the physical compensation for radiation damage to the base station. When radiation causes damage to the base station, Make Φ rad (t)=0, forcing a(t)=0; ξ is the base station radiation resistance coefficient; In the formula for the cooperative dependency strength exponent β(t): t is time; β is the exponent. base The basic value for the strength of cooperative dependency; (1+α·RDR) norm ) represents the radiation risk amplification term; α is the risk sensitivity coefficient (taken as 0.2-0.5). The current radiation dose rate (RDR) and the universal system threshold RDR for radiation dose rate are used. max-1 The ratio; (1-e -k·PDR(t) ) represents the communication attenuation factor; k is the attenuation constant (taken as 2.0), and PDR(t) is the data packet arrival rate at time t; In the sub-formula of the response intensity adjustment factor γ(t): k1, k2, and k3 are the radiation threat weight, time pressure weight, and resource efficiency weight, respectively, satisfying k1+k2+k3=1; This is a radiation threat term, reflecting the amplification effect of radiation levels on response intensity. The current radiation dose rate (RDR) and the device-specific threshold RDR for radiation dose rate. max-2 The ratio; k2·e -t / τ This is a time pressure term, used to reflect the urgency of the response due to the duration of the incident; e -t / τ τ is the time decay factor, which characterizes the urgency pressure that decays exponentially over time, and τ is the response half-life constant, which characterizes the decay rate of tissue response efficiency. This is a resource efficiency item, used to reflect the synergistic effect between resource efficiency and decision-making efficiency, η. r Resource availability rate represents the proportion of available resources to total demand, δ d This is the decision delay coefficient, representing the degree of delay in decision-making due to lack of information.

9. The three-layer coupled network model for nuclear accident emergency response as described in claim 8, characterized in that: In the coupling mechanism, the coupling strength ε is defined by Equation 7; Formula 7: Where, N IE This is the number of nodes in the IN layer that depend on the EN layer; it is counted when an IN node requires an EN layer instruction to function. N I N represents the total number of nodes in the IN layer; EI This refers to the number of nodes in the EN layer that depend on the IN layer, counted when an EN node requires IN facility support to function; N E This represents the total number of nodes in the EN layer. In the coupling mechanism, the universal system threshold RDR for radiation dose rate max-1 Set to 500 μSv / h.

10. A method for constructing a three-layer coupled network model for nuclear accident emergency response, used to construct the three-layer coupled network model for nuclear accident emergency response as described in claim 9, characterized in that the steps are as follows: as follows: S1, Construct the RMN layer: A. The area affected by the nuclear accident Ω XY Mesh the network at resolution Δ to generate an RMN-layer spatial network. B. Calculation using a dynamic Gaussian plume model The concentration of radionuclides in each grid cell; C. Converting radionuclide concentration into radiation dose rate using the dose conversion factor DF; S2, construct the IN layer: A two-layer network coupling transportation and communication networks is constructed based on dependent networks. The construction process includes topology modeling and spatial modeling. A. Topology Modeling: Define nodes including traffic intersections and communication base stations; node states include susceptible (S), infected (I), immune (R), and damaged (D); intra-network edges include roads and fiber optic cables; cross-network edges represent the functional dependencies between the traffic network and the communication network; establish the adjacency matrix A. IN Define the physical connections between nodes in the IN layer; obtain network G through topology modeling. IN =(N IN E IN ), N IN E is the set of all nodes in the IN layer. IN This is the set of all edges connected to the IN layer; B. Spatial Modeling: First, define the nuclear accident impact area Ω XY Generate an IN-layer spatial network by Δ meshing at resolution. Then map the geographic coordinates of all nodes in the IN layer to... In the corresponding grid cell; S3, construct the EN layer: A. Define four types of nodes: nuclear power plants, government departments, rescue agencies, and social organizations; each type of node is further divided into multiple nodes based on their functions; establish an adjacency matrix A. EN It represents the command or collaboration relationships between nodes in this layer; B. Calculate the topology network G based on real-time communication quality, cooperative dependency strength, and connection response strength. EN The dynamic weights of each connected edge.

Images