Multi-domain element digital twin modeling method for complex disaster scene

Abstract

The invention discloses a multi-domain element digital twinborn modeling method for a complex disaster scene, and the method achieves the modeling of emergency rescue elements, rescue task elements and objective environment elements based on an improved ontology theory, achieves the unified representation of the elements, and introduces a four-dimensional hypergraph theory. And constructing a multi-level dynamic coupling relationship among tasks, resources and environments. By constructing a dynamic attribute reduction operator, the modeling efficiency and semantic consistency are improved; and a relation strength calculation model is introduced to realize quantitative expression of nonlinear interaction among multiple elements. According to the invention, disaster deduction, resource scheduling and auxiliary decision making can be effectively supported, and the intelligent level of emergency response in a multi-disaster scene is improved.

Description

Technical Field

[0001] The present invention relates to the fields of digital twins and complex network modeling, and specifically relates to a unified modeling and relationship deduction method for multi-source heterogeneous data in complex disaster scenarios based on ontology theory and four-dimensional hypergraph relationships. Background Art

[0002] In complex disaster scenarios, the interweaving and mutual influence of disaster types pose huge challenges to emergency management. Natural disasters often are not limited to a single event. Each disaster not only affects a specific area alone, but may also trigger a chain reaction with other disaster types, making the post-disaster recovery process more complex. Against this background, disaster emergency response not only needs to deal with various single disasters, but also must consider the interaction effects between disasters, the dynamic adjustment of resource scheduling, and the optimization of multi-party collaborative decision-making. This requires the emergency management system to be able to perceive the occurrence of different disasters in real time and, based on the interaction between disasters, make timely and effective resource allocation and decision responses. Therefore, building a multi-dimensional and multi-domain comprehensive model of disaster elements that can dynamically and accurately reflect the complex relationships and spatio-temporal characteristics between different disaster elements is the key to improving the efficiency of disaster emergency response. Summary of the Invention

[0003] Object of the Invention: The present invention aims to solve problems in the prior art such as poor modeling and expression ability, fuzzy relationship structure, and difficulty in data unification in disaster scenarios, and proposes an improved ontology modeling and four-dimensional hypergraph fusion modeling method to support semantic unified modeling and deduction analysis of rescue task elements, emergency rescue elements, and objective environment elements, provide the coupling modeling ability between multi-domain elements of disasters, and support intelligent scheduling and emergency decision-making.

[0004] Technical Solution: To achieve the above object, the technical solution adopted by the present invention is:

[0005] A multi-domain element digital twin modeling method for complex disaster scenarios, comprising the following steps:

[0006] Step 1, construct three element models of emergency rescue elements, rescue task elements, and objective environment elements for the relevant factors of complex disaster scenarios. The three element models include the mobility ability model, endurance ability model, and rescue ability model of emergency rescue elements, four task requirement models of rescue tasks, and the environmental factor model of the objective environment.

[0007] Step 2, by improving the ontology theory, construct a semantic-driven ontology model for the three elements to realize the representation of multi-domain heterogeneous elements in a unified semantic space.

[0008] Step 3, establish a multi-dimensional matching relationship between rescue tasks and emergency agents in a static scenario, and construct a rescue task element - emergency rescue element relationship hypergraph model.

[0009] Step 4. On the basis of Step 3, introduce objective environmental factors and construct a four-dimensional hypergraph model of the relationship among rescue mission elements - emergency rescue elements - objective environmental elements.

[0010] Step 5. Based on the various defined relationships in the construction of the hypergraph model, construct a multi-dimensional relationship strength network of mission - emergency - environment to quantify the non-linear influence relationship among various elements.

[0011] Preferably, the method for constructing the multi-dimensional relationship strength network of mission - emergency - environment in Step 5 is as follows:

[0012] Construct an optimization model according to the mission - emergency relationship strength, comprehensively consider the relationship between mission requirements and rescue capabilities, and the optimization objective is defined as follows:

[0013]

[0014] Among them, C jc represents the unit resource consumption cost for the emergency agent to execute the c-th type of mission, q ji represents the total amount of resources allocated by the emergency agent a j to the mission s i and o jic represents whether the emergency agent a j can provide resources for the c-th type of requirement of the mission s i . Construct an influence model according to the dynamic influence of environmental factors on emergency rescue capabilities:

[0015]

[0016] Among them, r k,j represents the intensity relationship influence factor of environmental factors on emergency rescue capabilities, respectively represent the variances of the characterization parameters of objective environmental factors and emergency rescue capabilities, λ represents the attenuation rate coefficient, and Δδ represents the influence time difference of environmental factors. Construct an influence model according to the influence of environmental factors on mission requirements:

[0017]

[0018] Among them, r k,i represents the intensity relationship influence factor of environmental factors on mission requirements, represents the characterization value of environmental parameters at time step t, represents the change amount of the characterization value of a certain mission requirement at time step t, and T represents the total number of discrete time steps of the time window. Thus, construct a relationship strength model among various elements.

[0019] Preferred: The modeling method of the four-dimensional hypergraph model of the rescue mission element - emergency rescue element - objective environment element relationship in step 4 is as follows:

[0020] Based on the hypergraph model of the rescue mission element - emergency rescue element relationship, a new environmental area node set L = {l1, l2,..., l K} is added, where each element contains various environmental elements within the area, and it is defined as the environmental impact factor vector F k =(f k1 , f k2 ,..., f kO ), where O represents the total number of environmental factors, and the four-dimensional hyperedge is defined as h ijk ={a j , s i , l k}, indicating that the emergency agent a j is affected by the environment of the area it passes through when completing the task s k in the environmental area l i . And the following constraints are given for the four-dimensional hypergraph:

[0021] Step 41, two satisfy

[0022] The path of the emergency agent during the execution of the task covers all the environmental areas it passes through, forming a many-to-many mapping relationship, that is:

[0023]

[0024] where L(i, j, k) represents the environment l j passed by the emergency agent a i during the execution of the task s k , and Path(a j , s i , l k ) represents the set of environments of the complete path experienced by the emergency agent a j from the starting point to the location of the task s i . For the complete relationship network formed by the four-dimensional hypergraph, any hyperedge h ijk needs to satisfy the following dynamic constraints:

[0025]

[0026] where Φ c (F k ) represents the loss coefficient of the environmental factors in the environment l k to the c-th type of emergency rescue ability.

[0027] Step 42, two definitions

[0028] Define the direction of the hyperedge as the spatio-temporal pointing from the emergency rescue agent to the task, that is, determined by the path order from the emergency agent to the task execution. The hyperedge strength is defined as follows:

[0029]

[0030] Among them, W l1 represents the total influence value of the emergency agent affected by the environment l k during the path experienced in the task execution process, W l2 represents the total distance of the emergency agent running for the task execution, W l3 represents the proportion of the path in each environmental area passed by the emergency agent in the whole path, EI k represents the influence value of the emergency agent affected by the environment l k during the task execution process, d k represents the running distance of the emergency agent in the environmental area l k and K represents the sequence of all environmental areas passed by the emergency agent during the task execution process.

[0031] Preferably: The hypergraph modeling method for the rescue task element - emergency rescue element relationship in step 3 is as follows:

[0032] First, define each element of the hypergraph. Define the rescue task node as S = {s1, s2,..., s N}, representing multiple rescue tasks, and each task is associated with a resource requirement vector D i = (D i1 , D i2 , D i3 , D i4 ), respectively representing the four requirements of communication, medical treatment, disaster relief, and materials. Define the emergency rescue node set as A = {a1, a2,..., a M}, representing multiple emergency agents, and the emergency resource vector carried by each emergency agent is defined as R j = (R j1 , R j2 , R j3 , R j4 ), representing the resource amounts of the four task requirements carried by the emergency agent a j . For the hyperedge e k = (A k , S k ) is defined as representing that the subset of agents jointly complete a subset of tasks and the following constraints are given for the hypergraph modeling:

[0033] Step 31, two that satisfy

[0034] For any task s i ∈S and agent a j ∈A, the set of tasks S that it participates in executing k The union of the required emergency rescue capabilities must be covered by the full set of the agent's resources, namely:

[0035]

[0036] Among them, Supp(D i ) indicates that the resource requirements for any task are converted into the corresponding emergency capacity set, Supp(R j ) represents agent a j The emergency capability resource set carried by . At the same time, the mutually exclusive conditions of rescue capability are defined:

[0037]

[0038] where c p and c q It represents the two rescue capabilities of the emergency agent required for a certain task. The above formula means that for an emergency, only one rescue capability can be contributed in a certain time period when performing a certain task. Resources that are not met by the task requirements need to be allocated to different emergency agents.

[0039] Step 32, two definitions

[0040] The direction of the hyperedge is determined by the arrival order of the agents, and the hyperedge strength constraint is defined:

[0041]

[0042] Where R jc Indicates the amount of resources required by various tasks carried by the agent, D ic Indicates the amount of resources required for a task, and c indicates several task requirements.

[0043] Preferably, the ontology model construction method based on improved ontology theory in step 2 includes the following steps:

[0044] First, we define multiple disaster scenarios in a single space-time, i.e., we construct a four-dimensional normalized coordinate system to ensure the consistency of multi-domain data in terms of space-time semantics. The entire coordinate system is defined as follows:

[0045] S total =(x,δ)

[0046] x=(x,y,z)

[0047]

[0048] Among them, S totalrepresents the overall coordinate space, δ∈[0,1] represents the normalized time parameter, and Δt is the total time window of disaster simulation. On this basis, the five-tuple form of the extended ontology model is defined as follows:

[0049] O=<C,P,A,R,D>

[0050] C represents the collection of classes in the ontology model, defined here as abstract categories of emergency rescue elements, rescue mission elements, and objective environment elements. P represents the attribute set, which includes the physical characteristics of each element. A represents the constraint rules, which are used to define the various constraints between attributes. R represents the relationship set, which quantifies the dynamic coupling relationships between multi-domain elements. D represents the dynamic operator space, which includes custom attribute reduction operators and spatial projection operators.

[0051] In view of the dynamic characteristics of the environment, an attribute reduction operator based on the attention mechanism is designed to construct the semantic alignment function Γ(X t ), the ontology mapping process is:

[0052] Γ(X t )=W t ·X t +b t

[0053]

[0054] Among them, X t Represents the original parameter matrix at time t, including the element and attribute feature matrix. t Represents the dynamic weight matrix based on the attention mechanism. Q, H, V represent the query matrix, key matrix and value matrix, which are mainly generated by historical data. k represents the feature dimension scaling factor, b t Represents the bias term, and its dimension is the same as the reduced dimension.

[0055] Preferably, the method for representing the multi-domain heterogeneous elements in step 2 in a unified semantic space is as follows:

[0056] The unified characterization standards for emergency rescue capability parameters are as follows:

[0057]

[0058] in It indicates the expression of emergency rescue capability after characterization. Represents the reference characteristic value of emergency rescue capability, μ emer ,σ emer They represent the mean and variance of each attribute feature in the emergency rescue element based on historical data, ζ emergency Represents the rescue conversion factor defined according to physical principles and ontology semantics.

[0059] For the task requirement parameters, the unified representation method is defined as:

[0060]

[0061] Where represents the expression of the rescue task requirement intensity after representation, D max , D min represents the theoretical value boundary of the characteristic parameters of various task requirements in the rescue task elements, ζ task represents the requirement conversion factor, which is determined by the ontology definition of the rescue task elements.

[0062] For the unified representation method of environmental parameters, it is defined as:

[0063]

[0064] Where represents the expression of the objective environmental parameters after representation, μ env , σ env represents the mean and standard deviation of various environmental factors calculated based on historical environmental data, ζ env represents the environmental conversion factor, which is determined by the ontology definition of the objective environmental elements.

[0065] Preferably: The construction methods of the three-element models in step 1 are as follows:

[0066] Step 11, the various influence ability models of the emergency rescue element model are as follows:

[0067] The emergency intelligent agent a in the emergency rescue elements j The maneuverability at the spatio-temporal position (x, t) is expressed as follows:

[0068]

[0069] Where, v j (t) represents the real-time speed of the emergency intelligent agent a j at time t, S j (x) refers to the task coverage area of the emergency intelligent agent a j at x = (x j , y j , z j ) position, and Δt(x) refers to the time window for task execution, refers to the total distance of obstacles encountered by the emergency intelligent agent a j in the spatio-temporal position (x, t), refers to the emergency intelligent agent a jThe total planned path distance in executing the task, α and β refer to weight parameters calibrated according to historical data.

[0070] Emergency agent a in emergency rescue elements j Endurance at the spacetime position (x, t) It is expressed as follows:

[0071]

[0072] in, Represents emergency agent a j The total energy reserve, Represents emergency agent a j Energy consumption under baseline conditions, g(x) represents the terrain slope at position x, w(x,t) represents the wind speed at the spatiotemporal position (x,t), T(x,t) represents the temperature at the spatiotemporal position, and k1, k2, and k3 represent the proportions of the three types of environmental factors, respectively.

[0073] Emergency agent a in emergency rescue elements j For the task point s at the space-time position (x, t) i Rescue capability required by mission c It is expressed as follows:

[0074]

[0075] Among them, R ijc Represents emergency agent a j Points for the task i The amount of resources required by the cth task, D ic (x,t) represents the task point s at the space-time position (x,t) i The demand intensity of the c-th task requirement, f ijc (x, t) represents the resource allocation policy function, λ ic Represents the task point s i is the priority weight of the cth task requirement, and ε represents the smoothing factor.

[0076] Therefore, the comprehensive model of emergency rescue elements is defined, and the emergency agent a j For task points i The emergency rescue capability required by the cth mission The definition is as follows:

[0077]

[0078] Step 12: The rescue mission element model is defined according to the rescue mission demand intensity function, that is, the mission point s at the time and space position (x, t) i The intensity function of the cth task requirement

[0079]

[0080] in Represents the task point s i The initial demand intensity of the c-th task demand at the task point, g ic (t) represents the time evolution function, describing the change of task requirements over time, h ic (x, t) represents the spatial distribution function, which describes the dynamic distribution characteristics of task requirements in space. The definition of the time evolution function is as follows:

[0081]

[0082] where γ ic represents the time evolution coefficient, t0 represents the initial time point of the task requirement. For the spatial distribution function h ic (x,t) is defined as follows:

[0083]

[0084] Among them, x i (t) represents the task point s i The c-th task demand is at the center of time t, σ ic (t) represents the spatial distribution range parameter of the cth task requirement at time t.

[0085] Step 13: The objective environmental factor model is mainly quantified by the environmental state index, so the environmental factor parameter F at the spatiotemporal position (x, t) is env Defined as:

[0086] F env =ω T ·T norm (x,t)+ω H ·H norm (x,t)+ω W W(x,t)+ω G ·G(x)

[0087] Among them, T norm (x, t) represents the temperature influence coefficient, H norm (x, t) represents the humidity influence coefficient, W(x, t) represents the comprehensive evaluation parameter of weather factors, G(x) represents the terrain influence coefficient, ω T ,ω H ,ω W ,ω G Represents the weight parameters of various influencing factors. Among them, the temperature influence coefficient T for the mission target area norm(x,t) is defined as follows:

[0088]

[0089] Where T(x,t) represents the temperature of the region at the time and space position (x,t), T max ,T min They represent the temperature range that the emergency rescue equipment can withstand, T opt Indicates the optimal working temperature. For humidity influence coefficient H norm (x,t) is defined as follows:

[0090]

[0091] Among them, H(x,t) represents the humidity of the area at the time and space position (x,t), H max ,H min Indicates the humidity range that the equipment can tolerate. The comprehensive evaluation parameter W(x,t) for weather factors is defined as follows:

[0092]

[0093] Among them, I k (x, t) represents the quantitative value of the kth type of weather phenomenon, μ k represents the dynamic weight that is adaptively adjusted according to the task type, and K represents the total number of weather phenomena that have been counted.

[0094] Another object of the present invention is to provide a multi-domain element digital twin modeling system for complex disaster scenarios, which is used to implement the multi-domain element digital twin modeling method for complex disaster scenarios, including an input unit, three element model units, an ontology model unit, a rescue mission element-emergency rescue element relationship hypergraph modeling unit, a rescue mission element-emergency rescue element-objective environment element relationship four-dimensional hypergraph model unit, and a mission-emergency-environment multidimensional relationship strength network unit, wherein:

[0095] The input unit is used to input factors related to complex disaster scenarios.

[0096] The three-element model unit is used to construct three element models of emergency rescue elements, rescue mission elements and objective environment elements for factors related to complex disaster scenarios. The three element models include the mobility model, endurance model and rescue capability model of emergency rescue elements, four task requirement models of rescue missions and the environmental factor model of the objective environment.

[0097] The ontology model unit is used to construct an ontology model based on semantics-drivenness of three elements by improving ontology theory, so as to realize the representation of multi-domain heterogeneous elements in a unified semantic space.

[0098] The rescue mission element - emergency rescue element relationship hypergraph modeling unit is used to establish the multi - dimensional matching relationship between rescue missions and emergency agents in a static scenario, and construct a rescue mission element - emergency rescue element relationship hypergraph model.

[0099] The four - dimensional hypergraph model unit of rescue mission element - emergency rescue element - objective environment element is used to introduce objective environment factors on the basis of the rescue mission element - emergency rescue element relationship hypergraph model, and construct a four - dimensional hypergraph model of rescue mission element - emergency rescue element - objective environment element.

[0100] The task - emergency - environment multi - dimensional relationship strength network unit is used to construct a task - emergency - environment multi - dimensional relationship strength network based on various defined relationships in the construction of the hypergraph model, and quantify the non - linear influence relationship between various elements.

[0101] Preferably: In the task - emergency - environment multi - dimensional relationship strength network unit, an optimization model is constructed according to the task - emergency relationship strength, and the relationship between task requirements and rescue capabilities is comprehensively considered. The optimization objective is defined as follows:

[0102]

[0103] Among them, C jc represents the unit resource consumption cost for the emergency agent to execute the c - th type of task, q ji represents the total amount of resources allocated by the emergency agent a j to the task s i , and o jic represents whether the emergency agent a j can provide resources for the c - th type of requirement of the task s i . An influence model is constructed according to the dynamic influence of environmental factors on emergency rescue capabilities:

[0104]

[0105] Among them, r k,j represents the intensity relationship influence factor of environmental factors on emergency rescue capabilities, respectively represent the variances of the characterization parameters of objective environmental factors and emergency rescue capabilities, λ represents the attenuation rate coefficient, and Δδ represents the influence time difference of environmental factors. An influence model is constructed according to the influence of environmental factors on task requirements:

[0106]

[0107] Among them, r k,i represents the intensity relationship influence factor of environmental factors on task requirements, represents the environmental parameter characterization value at time step t, It represents the change in the characterization value of a certain task requirement at time step t, and T represents the total number of discrete time steps in the time window.

[0108] Preferably, in the four-dimensional hypergraph model unit of the relationship between rescue task elements - emergency rescue elements - objective environment elements, on the basis of the hypergraph model of the relationship between rescue task elements - emergency rescue elements, a new environmental area node set L = {l1, l2,..., l K} is added, where each element contains various environmental elements within the area, and it is defined as the environmental impact factor vector F k =(f k1 , f k2 ,..., f kO ), where O represents the total number of environmental factors, and the four-dimensional hyperedge is defined as h ijk ={a j , s i , l k}, indicating that the emergency agent a j passes through the environmental area l k and is affected by the environment of the passed area when completing the task s i . And the following constraints are given for the four-dimensional hypergraph:

[0109] Two satisfy

[0110] During the execution of the task, the path of the emergency agent covers all the environmental areas it passes through, forming a many-to-many mapping relationship, that is:

[0111]

[0112] Where L(i, j, k) represents the environment l j passed by the emergency agent a i during the execution of the task s k , and Path(a j , s i , l k ) represents the set of environments of the complete path experienced by the emergency agent a j from the starting point to the location of the task s i . For the complete relationship network formed by the four-dimensional hypergraph, any hyperedge h ijk needs to satisfy the following dynamic constraints:

[0113]

[0114] Where Φ c (F k ) represents the loss coefficient of the environmental factors in the environment l k to the c-th type of emergency rescue ability.

[0115] Two definitions

[0116] Define the direction of the hyperedge as the spatio-temporal pointing from the emergency rescue agent to the task, that is, determined by the path order from the emergency agent to the task execution. The following definitions are made for the hyperedge strength:

[0117]

[0118] Among them, W l1 represents the total influence value of the emergency agent affected by the environment l in the path experienced during the task execution, W k represents the total distance of the emergency agent running for task execution, W l2 represents the proportion of the path in each environmental area passed by the emergency agent in the whole path, EI l3 represents the influence value of the emergency agent affected by the environment l during the task execution, EI k represents the influence value of the emergency agent affected by the environment l during the task execution, d k represents the running distance of the emergency agent in the environmental area l k represents the running distance of the emergency agent in the environmental area l k within, and K represents the sequence of all environmental areas passed by the emergency agent during the task execution.

[0119] Compared with the prior art, the present invention has the following beneficial effects:

[0120] Based on the improved ontology theory, the present invention realizes the modeling and unified representation of emergency rescue elements, rescue task elements and objective environmental elements, introduces the four-dimensional hypergraph theory, and constructs a multi-level dynamic coupling relationship among tasks - resources - environment. By constructing a dynamic attribute reduction operator, the modeling efficiency and semantic consistency are improved; and a relationship strength calculation model is introduced to realize the quantitative representation of non-linear interaction among multiple elements. The present invention can effectively support disaster deduction, resource scheduling and auxiliary decision-making, and improve the intelligent level of emergency response in multi-disaster scenarios. Description of the Drawings

[0121] Figure 1 is the flowchart of the present invention;

[0122] Figure 2 is the output result of the relationship strength among multi-domain elements. Specific Embodiments

[0123] The following further clarifies the present invention in conjunction with the drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. After reading the present invention, various equivalent modifications made by those skilled in the art to the present invention fall within the scope defined by the appended claims of this application.

[0124] This embodiment provides a multi-domain element digital twin modeling method for complex disaster scenarios, as Figure 1-2 shown, and the specific operation steps are as follows:

[0125] Step 1: Construct mathematical models of emergency rescue elements, rescue mission elements, and objective environmental factors for complex disaster scenarios, including the mobility model, endurance model, and rescue capability model of emergency rescue elements, the four task requirement models of rescue missions, and the environmental factor model of the objective environment;

[0126] First, the three element models in complex disaster scenarios are constructed as follows:

[0127] (1) Emergency rescue element model Various impact capability models are as follows:

[0128] Emergency agent a in emergency rescue elements j Maneuverability at a spacetime location (x,t) It is expressed as follows:

[0129]

[0130] where v j (t) represents the emergency agent a j Real-time speed at time t, S j (x) refers to the emergency agent a j In x=(x j ,y j ,z j ) is the task coverage area at the location, Δt(x) refers to the time window of task execution, Refers to the emergency agent a j The total distance to obstacles encountered at the spacetime position (x,t), Refers to the emergency agent a j The total planned path distance in executing the task, α and β refer to weight parameters calibrated according to historical data.

[0131] Emergency agent a in emergency rescue elements j Endurance at the spacetime position (x, t) It is expressed as follows:

[0132]

[0133] in, Represents the emergency agent a j The total energy reserve, Represents the emergency agent a j Energy consumption under baseline conditions, g(x) represents the terrain slope at location x, w(x,t) represents the wind speed at the spatiotemporal location (x,t), T(x,t) represents the temperature at the spatiotemporal location, k1, k2, and k3 represent the proportions of the three environmental factors, respectively;

[0134] Emergency intelligent agent a in emergency rescue elements j For the task point s at the spatio-temporal position (x, t) i The rescue capacity for the c-th type of task requirement Is expressed as follows:

[0135]

[0136] Among them, R ijc Represents the emergency intelligent agent a j For the task point s i The amount of resources provided for the c-th type of task requirement, D ic (x, t) represents the task point s at the spatio-temporal position (x, t) i The demand intensity for the c-th type of task requirement, f ijc (x, t) represents the resource allocation strategy function, λ ic Represents the task point s i The priority weight for the c-th type of task requirement, and ε represents the smoothing factor. Thus, the comprehensive model of emergency rescue elements is defined, and the emergency intelligent agent a j For the task point s i The emergency rescue capacity for the c-th type of task requirement Is defined as follows:

[0137]

[0138] (2) The rescue task element model is defined according to the rescue task demand intensity function, that is, for the task point s at the spatio-temporal position (x, t) i The intensity function for the c-th type of task requirement

[0139]

[0140] Among them Represents the initial demand intensity of the c-th type of task requirement at the task point for the task point s i , g ic (t) represents the time evolution function, describing the variation law of the task requirement with time, h ic (x, t) represents the spatial distribution function, describing the dynamic distribution characteristics of the task requirement in space; The definition of the time evolution function is as follows:

[0141]

[0142] Among them, γ ic Represents the time evolution coefficient, and t0 represents the initial time point of the task requirement; For the spatial distribution function h ic (x, t) is defined as follows:

[0143]

[0144] Among them, x i (t) represents the task point s i The c-th task demand is at the center of time t, σ ic (t) represents the spatial distribution range parameter of the c-th task requirement at time t;

[0145] (3) The objective environmental factor model is mainly quantified by the environmental state index, so the environmental factor parameter F at the spatiotemporal position (x, t) is env Defined as:

[0146] F env =ω T ·T norm (x,t)+ω H ·H norm (x,t)+ω W W(x,t)+ω G ·G(x)

[0147] Where T norm (x, t) represents the temperature influence coefficient, H norm (x, t) represents the humidity influence coefficient, W(x, t) represents the comprehensive evaluation parameter of weather factors, G(x) represents the terrain influence coefficient, ω T ,ω H ,ω W ,ω G Represents the weight parameters of various influencing factors; among them, the temperature influence coefficient T for the mission target area norm (x,t) is defined as follows:

[0148]

[0149] Where T(x,t) represents the temperature of the region at the time and space position (x,t), T max ,T min They represent the temperature range that the emergency rescue equipment can withstand, T opt Indicates the optimal working temperature; for humidity influence coefficient H norm (x,t) is defined as follows:

[0150]

[0151] Among them, H(x,t) represents the humidity of the area at the time and space position (x,t), H max ,H min Indicates the humidity range that the equipment can tolerate. The comprehensive evaluation parameter W(x,t) for weather factors is defined as follows:

[0152]

[0153] Among them, I k (x, t) represents the quantitative value of the kth type of weather phenomenon, μ k represents the dynamic weight that is adaptively adjusted according to the task type, and K represents the total number of weather phenomena that have been counted.

[0154] Step 2: By improving the ontology theory, a semantically driven ontology model of the three elements is constructed to achieve the representation of multi-domain heterogeneous elements in a unified semantic space;

[0155] First, we define multiple disaster scenarios in a single space-time, i.e., we construct a four-dimensional normalized coordinate system to ensure the consistency of multi-domain data in terms of space-time semantics. The entire coordinate system is defined as follows:

[0156] S total =(x,δ)

[0157] x=(x,y,z)

[0158]

[0159] Among them, S total represents the overall coordinate space, δ∈[0,1] represents the normalized time parameter, and Δt is the total time window of disaster simulation. On this basis, the five-tuple form of the extended ontology model is defined as follows:

[0160] O=<C,P,A,R,D>

[0161] Among them, C represents the various class sets in the ontology model, which are defined as abstract categories of emergency rescue elements, rescue mission elements, and objective environment elements; P represents the attribute set, which includes the physical characteristics of various elements; A represents the constraint rules, which are used to define various constraints between attributes; R represents the relationship set, which quantifies the dynamic coupling relationship between multi-domain elements; D represents the dynamic operator space, including custom attribute reduction operators and spatial projection operators. In view of the dynamic characteristics of the environment, the attribute reduction operator based on the attention mechanism is designed as follows to construct the semantic alignment function Γ(X t ), the ontology mapping process is:

[0162] Γ(X t )=W t ·X t +b t

[0163]

[0164] Among them, X t Represents the original parameter matrix at time t, including the element and attribute feature matrix; W trepresents the dynamic weight matrix based on the attention mechanism; Q, H, V represent the query matrix, key matrix and value matrix, which are mainly generated by historical data; d k represents the feature dimension scaling factor, b t Represents the bias term, and its dimension is the same as the simplified dimension. On this basis, the three basic elements are uniformly represented, and the unified representation standard of the emergency rescue capability parameters is as follows:

[0165]

[0166] in It indicates the expression of emergency rescue capability after characterization. Represents the reference characteristic value of emergency rescue capability, μ emer ,σ emer They represent the mean and variance of each attribute feature in the emergency rescue element based on historical data, ζ emergency Represents the rescue conversion factor defined according to physical principles and ontology semantics; for the task requirement parameters, the unified representation method is defined as:

[0167]

[0168] in Denotes the rescue mission requirement intensity after characterization, D max ,D min The theoretical value boundary of the characteristic parameters representing various task requirements in the rescue mission elements, ζ task It represents the demand conversion factor, which is determined by the ontology definition of the rescue mission elements; the unified representation method of environmental parameters is defined as:

[0169]

[0170] in Represents the objective environmental parameter expression after characterization, μ env ,σ env represents the mean and standard deviation of various environmental factors calculated based on historical environmental data, ζ env It represents the environmental conversion factor, which is determined by the ontological definition of the objective environmental elements.

[0171] Step 3: Establish a multi-dimensional matching relationship between rescue tasks and agent resources in a static scenario, and construct a hypergraph model of the relationship between rescue task elements and emergency rescue elements;

[0172] First, define each element of the hypergraph and define the rescue mission node as S={s1,s2,...,s N}, represents multiple rescue missions, each of which is associated with a resource demand vector D i =(D i1 ,Di2 , D i3 , D i4 ), respectively representing four types of requirements: communication, medical treatment, disaster relief, and materials. Define the emergency rescue node set as A = {a1, a2,..., a M}, representing multiple emergency agents. Define the emergency resource vector carried by each emergency agent as R j = (R j1 , R j2 , R j3 , R j4 ), indicating the resource quantities of the four task requirements carried by emergency agent a j . For the hyperedge e k = (A k , S k ), it is defined as representing a subset of agents jointly completing a subset of tasks and giving the following constraints for hypergraph modeling:

[0173] (1) For any task s

[0174] ∈ S and agent a i ∈ A, the union of the emergency rescue capabilities required by the set of tasks S j they participate in must be covered by the full set of resources of the agent, that is: k

[0175]

[0176] where Supp(D i ) represents converting the resource requirements for any task into the corresponding set of emergency capabilities, and Supp(R j ) represents the set of emergency capability resources carried by agent a j ; at the same time, define the mutual exclusion condition of rescue capabilities:

[0177]

[0178] where c p and c q represent two rescue capabilities of the emergency agents required for a certain task. The above formula means that for an emergency agent when performing a certain task, it can only contribute one rescue capability during a certain time period, and the resources not passed by the task requirements need to be allocated to different emergency agents.

[0179] (2) Two definitions

[0180] Define the direction of the hyperedge to be determined by the arrival order of the agents, and define the hyperedge strength constraint:

[0181] ​

[0182] Among them, c represents several task requirements.

[0183] Step 4: On the basis of Step 3, introduce objective environmental factors and construct a four-dimensional hypergraph model of the relationship among rescue task elements - emergency rescue elements - objective environmental elements;

[0184] Based on hypergraph modeling, a new environmental area node set L = {l1, l2,..., l K} is added. Each element contains various environmental elements within this area, and it is defined as the environmental impact factor vector F k = (f k1 , f k2 ,..., f kO ), where O represents the total number of environmental factors. And the four-dimensional hyperedge is defined as h ijk = {a j , s i , l k}, indicating that the emergency agent a j completes the task s k after passing through the environmental area l i and is affected by the environment of the passed area; and the following constraints are given to the four-dimensional hypergraph:

[0185] (1) Two

[0186] The paths of the emergency agents cover all the environmental areas they pass through during the task execution, forming a many-to-many mapping relationship, that is:

[0187]

[0188] Among them, L(i, j, k) represents the environment l j passed by the emergency agent a i during the execution of the task s k , and Path(a j , s i , l k ) represents the set of environments of the complete path experienced by the emergency agent a j from the starting point to the location of the task s i . For the complete relationship network formed by the four-dimensional hypergraph, any hyperedge h ijk needs to satisfy the following dynamic constraints:

[0189]

[0190] Among them, Φ c (F k ) represents the loss coefficient of the environmental factors in the environment l k to the c-th type of emergency rescue ability.

[0191] (2) Two definitions

[0192] Define the direction of the hyper - edge as the spatio - temporal pointing from the emergency rescue agent to the task, that is, determined by the path order from the emergency agent to the task execution. The following definitions are made for the hyper - edge strength:

[0193]

[0194] Where W l1 represents the total influence value of the emergency agent affected by the environment l k during the path experienced in the task execution process, W l2 represents the total distance of the emergency agent running for the task execution, W l3 represents the proportion of the path in each environmental area passed by the emergency agent in the whole path, EI k represents the influence value of the emergency agent affected by the environment l k during the task execution process, d k represents the running distance of the emergency agent in the environmental area l k and K represents the sequence of all environmental areas passed by the emergency agent during the task execution process.

[0195] Step 5: Based on the various defined relationships in the hyper - graph model construction, construct a multi - dimensional relationship strength network of task - emergency - environment, and quantify the non - linear influence relationship between various elements.

[0196] Construct an optimization model according to the task - emergency relationship strength, comprehensively consider the relationship between task requirements and rescue capabilities, and the optimization objective is defined as follows:

[0197]

[0198] Among them, C jc represents the unit resource consumption cost of the emergency agent for executing the c - type task, q ji represents the total amount of resources allocated by the emergency agent a j to the task s i and o jic represents whether the emergency agent a j can provide resources for the c - type requirement of the task s i ; construct an influence model according to the dynamic influence of environmental factors on emergency rescue capabilities:

[0199]

[0200] Among them, r k,j represents the influence factor of the environmental factor on the strength relationship of emergency rescue capabilities, They represent the variance of the characterization parameters of objective environmental factors and emergency rescue capabilities, λ represents the attenuation rate coefficient, and Δδ represents the time difference of the impact of environmental factors. An impact model is constructed based on the impact of environmental factors on task requirements:

[0201]

[0202] where r k,i Indicates the influence factor of the intensity relationship between environmental factors and task requirements, The environmental parameter representation value at time step t, It represents the change in the representation value of a task requirement at time step t, and T represents the total number of discrete time steps in the time window; thus, a relationship strength model between various elements is constructed.

[0203] In another embodiment, a multi-domain element digital twin modeling system for complex disaster scenarios is provided, which is used to implement the multi-domain element digital twin modeling method for complex disaster scenarios, including an input unit, three element model units, an ontology model unit, a rescue mission element-emergency rescue element relationship hypergraph modeling unit, a rescue mission element-emergency rescue element-objective environment element relationship four-dimensional hypergraph model unit, and a mission-emergency-environment multidimensional relationship strength network unit, wherein:

[0204] The input unit is used to input factors related to complex disaster scenarios.

[0205] The three-element model unit is used to construct three element models of emergency rescue elements, rescue mission elements and objective environment elements for factors related to complex disaster scenarios. The three element models include the mobility model, endurance model and rescue capability model of emergency rescue elements, four task requirement models of rescue missions and the environmental factor model of the objective environment.

[0206] The ontology model unit is used to construct an ontology model based on semantics-drivenness of three elements by improving ontology theory, so as to realize the representation of multi-domain heterogeneous elements in a unified semantic space.

[0207] The rescue mission element-emergency rescue element relationship hypergraph modeling unit is used to establish a multi-dimensional matching relationship between rescue missions and emergency agents in a static scene, and to construct a rescue mission element-emergency rescue element relationship hypergraph model.

[0208] The four-dimensional hypergraph model unit of the relationship between rescue mission elements, emergency rescue elements and objective environmental elements is used to introduce objective environmental factors on the basis of the hypergraph model of the relationship between rescue mission elements and emergency rescue elements, and to construct a four-dimensional hypergraph model of the relationship between rescue mission elements, emergency rescue elements and objective environmental elements.

[0209] The task-emergency-environment multidimensional relationship strength network unit is used to construct a task-emergency-environment multidimensional relationship strength network based on various definition relationships in the hypergraph model construction, and quantify the nonlinear influence relationship between various factors.

[0210] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-domain element digital twin modeling method for complex disaster scenarios, characterized in that, It includes the following steps: Step 1: Construct three element models for emergency rescue elements, rescue task elements, and objective environment elements for the relevant factors of complex disaster scenarios. The three element models include the mobility ability model, endurance ability model, and rescue ability model of emergency rescue elements, four task requirement models of rescue tasks, and the environmental factor model of the objective environment; Step 2: By improving the ontology theory, construct a semantic-driven ontology model for the three elements to achieve the representation of multi-domain heterogeneous elements in a unified semantic space; Step 3: Establish a multi-dimensional matching relationship between rescue tasks and emergency agents in a static scenario, and construct a hypergraph model of the relationship between rescue task elements - emergency rescue elements; Step 4: On the basis of Step 3, introduce objective environmental factors to construct a four-dimensional hypergraph model of the relationship between rescue task elements - emergency rescue elements - objective environment elements; Step 5: Based on various defined relationships in the construction of the hypergraph model, construct a multi-dimensional relationship strength network of task - emergency - environment to quantify the non-linear influence relationship between various elements.

2. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 1, characterized in that: The method for constructing the multi-dimensional relationship strength network of task - emergency - environment in Step 5 is as follows: Construct an optimization model according to the task - emergency relationship strength, comprehensively consider the relationship between task requirements and rescue capabilities, and the optimization objective is defined as follows: Among them, C jc represents the unit resource consumption cost for the emergency agent to execute the c -th type of task, q ji represents the total amount of resources allocated by the emergency agent a j to the task s i , and o jic represents whether the emergency agent a j can provide resources for the i -th type of requirement of the task s c . A impact model is constructed based on the dynamic impact of environmental factors on the emergency rescue ability: where r k,j represents the impact factor of the environmental factor on the intensity relationship of the emergency rescue ability, respectively represent the variances of the objective environmental factor and the characterization parameters of the emergency rescue ability, λ represents the attenuation rate coefficient, and Δδ represents the influence time difference of the environmental factor; an influence model is constructed according to the influence of the environmental factor on the task requirements: where r k,i represents the influence factor of the environmental factor on the intensity relationship of task requirements, and represents the characterization value of the environmental parameter at time step t, represents the change amount of the characterization value of a certain task requirement at time step t, and T represents the total number of discrete time steps of the time window; thus, a relationship intensity model among various elements is constructed.

3. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 2, wherein: The modeling method of the four-dimensional hypergraph model of the relationship between rescue task elements - emergency rescue elements - objective environment elements in Step 4 is as follows: Based on the hypergraph model of the relationship between rescue mission elements and emergency rescue elements, a new environmental area node set L = {l1, l2,..., l K} is added, where each element contains various environmental elements within the area, and it is defined as the environmental impact factor vector F k = (f k1 , f k2 ,..., f kO ), where O represents the total number of environmental factors, and the four-dimensional hyperedge is defined as h ijk = {a j , s i , l k}, indicating that the emergency agent a j is affected by the environment of the area passed through when completing the task s i after passing through the environmental area l k ; and the following constraints are given for the four-dimensional hypergraph: Step 41, two satisfy During the process of emergency agents executing tasks, the paths cover all environmental areas they pass through, forming a many-to-many mapping relationship, that is: where L(i, j, k) represents the emergency agent a j while performing task s i through the environment l k , Path(a j , s i , l k ) represents the set of environments of the complete path experienced by the emergency agent a j from the starting point to the location where task s i is located; for the complete relational network composed of four-dimensional hypergraphs, any hyperedge h ijk needs to satisfy the following dynamic constraints: where Φ c (F k ) represents the loss coefficient of environmental factors in environment l k on the emergency rescue capacity of the c th category; Step 42, two definitions Define the direction of the hyperedge as the spatio-temporal direction from the emergency rescue agent to the task, that is, determined by the path order of the emergency agent to execute the task; the following definitions are made for the hyperedge strength: Among which W l1 represents the total influence value of the environment l k on the path experienced by the emergency agent during the mission execution, W l2 represents the total distance traveled by the emergency agent during the mission execution, W l3 represents the proportion of the path in each environmental area passed by the emergency agent during the mission execution to the whole path, EI k represents the influence value of the environment l k on the emergency agent during the mission execution, d k represents the running distance of the emergency agent in the environmental area l k and K represents the sequence of all environmental areas passed by the emergency agent during the mission execution.

4. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 3, characterized in that: The hypergraph modeling method of the relationship between rescue task elements - emergency rescue elements in Step 3 is as follows: First, define each element of the hypergraph. Define the rescue task nodes as S = {s1, s2,..., s N}, representing multiple rescue tasks. Each task is associated with a resource requirement vector D i = (D i1 , D i2 , D i3 , D i4 ), which respectively represent the four requirements of communication, medical treatment, disaster relief, and materials. Define the emergency rescue node set as A = {a1, a2,..., a M}, representing multiple emergency agents. The emergency resource vector carried by each emergency agent is defined as R j = (R j1 , R j2 , R j3 , R j4 ), indicating the amount of resources for the four task requirements carried by the emergency agent a j . For the hyperedge e k = (A k , S k ) is defined as representing that a subset of agents jointly complete a subset of tasks and give the following constraints for hypergraph modeling: Step 31, two satisfy For any task s i ∈ S and agent a j ∈ A, the union of the required emergency rescue capabilities of the task set S k in which it participates must be covered by the entire resource set of the agent, that is: where Supp(D i ) represents the conversion of the resource requirements for any task into the corresponding set of emergency capabilities, and Supp(R j ) represents the set of emergency capability resources carried by the agent a j ; at the same time, the mutual exclusion condition of rescue capabilities is defined as: where c p and c q represent two rescue capabilities of the required emergency agents for a certain task. The above formula means that for an emergency agent when performing a certain task, it can only contribute one rescue capability during a certain time period, and resources that do not meet the task requirements need to be allocated to different emergency agents; Step 32, two definitions Define the hyperedge direction to be determined by the arrival order of the agents, and define the hyperedge strength constraint: Among them, R jc represents the resource quantity of various task requirements carried by the agent, D ic represents the resource quantity of the task requirement, c represents several task requirements.

5. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 4, characterized in that: The method for constructing the ontology model based on the improved ontology theory in Step 2 includes the following steps: First, define multiple disaster scenarios in one space-time, that is, construct a four-dimensional normalized coordinate system to ensure the consistency of multi-domain data in space-time semantics; the entire coordinate system is defined as follows: S total = (x, δ) x=(x,y,z) Among them, S total represents the overall coordinate space, δ∈[0,1] represents the normalized time parameter, and Δt is the total time window for disaster deduction. On this basis, the five-tuple form of the extended ontology model is defined as follows: O = <C,P,A,R,D> Among them, C represents various class sets in the ontology model, which is defined here as the abstract categories of emergency rescue elements, rescue task elements, and objective environment elements; P represents the attribute set, which includes the physical characteristics of various elements; A represents the constraint rules, which are used to define various constraints between attributes; R represents the relationship set, which quantifies the dynamic coupling relationship between multi-domain elements; D represents the dynamic operator space, including user-defined attribute reduction operators and space projection operators; For the dynamic characteristics of the environment, an attribute reduction operator based on the attention mechanism is designed as follows. Construct a semantic alignment function Γ(X t ), and the ontology mapping process is as follows: Γ(X t ) = W t ·X t + b t Among them, X t represents the original parameter matrix at time t, including the feature and attribute feature matrix; W t represents the dynamic weight matrix based on the attention mechanism; Q, H, and V represent the query matrix, key matrix, and value matrix, which are mainly generated from historical data; d k represents the feature dimension scaling factor, b t represents the bias term, and its dimension is the same as the dimension after reduction.

6. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 5, characterized in that: The method for representing multi-domain heterogeneous elements in a unified semantic space in Step 2 is as follows: The unified representation standard for emergency rescue ability parameters is as follows: Among them represents the expression of emergency rescue ability after characterization represents the reference characteristic value of emergency rescue ability, μ emer , σ emer respectively represent the mean and variance generated based on historical data for each attribute characteristic in the emergency rescue elements, ζ emergency represents the rescue conversion factor defined according to physical principles and ontology semantics; For task requirement parameters, the unified representation method is defined as: Among them represents the expression of the intensity of the rescue mission requirements after characterization, D max , D min represents the theoretical value boundary of the characteristic parameters of various mission requirements in the rescue mission elements, ζ task represents the demand conversion factor, which is determined by the ontology definition of the rescue mission elements; For the unified representation method of environmental parameters is defined as: Among them represents the objective environmental parameter expression after characterization, μ env , σ env represents the mean and standard deviation of various environmental factors calculated based on historical environmental data, ζ env represents the environmental conversion factor, which is determined by the ontology definition of objective environmental elements.

7. The multi-domain element digital twin modeling method for complex disaster scenarios according to claim 6, characterized in that: The construction method of the three element models in Step 1 is as follows: Step 11, the various impact ability models of the emergency rescue element model are as follows: Emergency intelligent agent a in emergency rescue elements j Maneuverability at spatio-temporal position (x, t) Is expressed as follows: Among them, v j (t) represents the real-time speed of the emergency agent a j at time t, S j (x) refers to the emergency agent a j at x = (x j , y j , z j ) the task coverage area at the position, Δt(x) refers to the time window for task execution, refers to the total distance of the obstacles encountered by the emergency agent a j in the spatio-temporal position (x, t), [[ID=,19]] refers to the total planned path distance of the emergency agent a j in the task execution, α and β refer to the weight parameters calibrated according to historical data; Emergency intelligent agent a in emergency rescue elements j Endurance at spatio-temporal position (x, t) It is expressed as follows: Among them, represents the total energy reserve of the emergency agent a j , represents the energy consumption of the emergency agent a j under the reference conditions, g(x) represents the terrain slope at position x, w(x,t) represents the wind speed at the spatio-temporal position (x,t), T(x,t) represents the temperature at the spatio-temporal position, and k1, k2, and k3 respectively represent the weights of the three types of environmental factors; Emergency intelligent agent a in emergency rescue elements j For the task point s at the spatio-temporal position (x, t) i The rescue capacity for the c-th type of task requirement Is expressed as follows: Among them, R ijc represents the amount of resources provided by the emergency intelligent agent a j for the c-th task requirement of the task point s i , D ic (x, t) represents the demand intensity of the c-th task requirement of the task point s i at the spatio-temporal position (x, t), f ijc (x, t) represents the resource allocation strategy function, λ ic represents the priority weight of the c-th task requirement of the task point s i , and ε represents the smoothing factor; Therefore, a comprehensive model of emergency rescue elements is defined, and the emergency intelligent agent a j For the task point s i The emergency rescue capability for the c-th type of task requirement is defined as follows: Step 12, the rescue mission element model is defined according to the rescue mission demand intensity function, that is, the intensity function of the c-th type of mission demand at the mission point s at the spatio-temporal position (x, t) i of the c-th type of mission demand where represents the sth task point i The initial demand intensity of the cth task requirement at the task point, g ic (t) represents the time evolution function, describing the variation law of task requirements with time, h ic (x, t) represents the spatial distribution function, describing the dynamic distribution characteristics of task requirements in space; the definition of the time evolution function is as follows: where γ ic represents the time evolution coefficient, and t0 represents the initial time point of the task requirement; for the spatial distribution function h ic (x, t) is defined as follows: Among them, x i (t) represents the central position of the c-th task requirement of the task point s i at time t, and σ ic (t) represents the spatial distribution range parameter of the c-th task requirement at time t; Step 13. The objective environmental factor model is mainly quantified by the environmental state index. Therefore, the environmental factor parameter F at the spatio-temporal position (x, t) env is defined as: F env = ω T · T norm (x, t) + ω H · H norm (x, t) + ω W · W(x, t) + ω G · G(x) Among them, T norm (x, t) represents the temperature influence coefficient, H norm (x, t) represents the humidity influence coefficient, W(x, t) represents the comprehensive evaluation parameter of weather factors, G(x) represents the terrain influence coefficient, ω T 、ω H 、ω W 、ω G represent the weight parameters of various influencing factors; among them For the temperature influence coefficient T of the task target area norm (x, t) is defined as follows: where T(x,t) represents the temperature of the region at the spatio-temporal position (x,t), T max , T min respectively represent the temperature range that the emergency rescue equipment can withstand, T opt represents the optimal working temperature; for the humidity influence coefficient H norm (x,t) is defined as follows: where, H(x,t) represents the humidity of the area at the space-time position (x,t), H max ,H min represents the humidity range that the device can tolerate; the comprehensive evaluation parameter W(x,t) for weather factors is defined as follows: Among them, I k (x, t) represents the quantization value of the k-th type of weather phenomenon, and μ k represents the dynamic weight adaptively adjusted according to the task type, and K represents the total number of weather phenomena obtained through statistics.

8. A modeling system for implementing the multi-domain element digital twin modeling method for complex disaster scenarios described in claim 1, characterized in that: It includes an input unit, three element model units, an ontology model unit, a rescue task element - emergency rescue element relationship hypergraph modeling unit, a rescue task element - emergency rescue element - objective environment element relationship four - dimensional hypergraph model unit, and a task - emergency - environment multi - dimensional relationship strength network unit, where: The input unit is used to input factors related to complex disaster scenarios; The three element model units are used to construct three element models of emergency rescue elements, rescue task elements, and objective environment elements for factors related to complex disaster scenarios. The three element models include a mobility ability model, a endurance ability model, and a rescue ability model of emergency rescue elements, four task requirement models of rescue tasks, and an environmental factor model of the objective environment; The ontology model unit is used to construct a semantic - driven ontology model of the three elements by improving ontology theory to realize the representation of multi - domain heterogeneous elements in a unified semantic space; The rescue task element - emergency rescue element relationship hypergraph modeling unit is used to establish a multi - dimensional matching relationship between rescue tasks and emergency agents in a static scenario and construct a rescue task element - emergency rescue element relationship hypergraph model; The rescue task element - emergency rescue element - objective environment element relationship four - dimensional hypergraph model unit is used to introduce objective environmental factors on the basis of the rescue task element - emergency rescue element relationship hypergraph model and construct a rescue task element - emergency rescue element - objective environment element relationship four - dimensional hypergraph model; The task - emergency - environment multi - dimensional relationship strength network unit is used to construct a task - emergency - environment multi - dimensional relationship strength network based on various defined relationships in the construction of the hypergraph model to quantify the non - linear impact relationships between elements.

9. The system according to claim 8, wherein: In the task - emergency - environment multi - dimensional relationship strength network unit, an optimization model is constructed according to the task - emergency relationship strength, and the relationship between task requirements and rescue capabilities is comprehensively considered. The optimization objective is defined as follows: Among them, C jc represents the unit resource consumption cost for the emergency agent to execute the c th type of task, q ji represents the total amount of resources allocated by the emergency agent a j to the task s i , and o jic represents whether the emergency agent a j can provide resources for the i th type of requirement of the task s c . An impact model is constructed based on the dynamic impact of environmental factors on the emergency rescue ability: where r k,j represents the influence factor of the environmental factor on the intensity relationship of the emergency rescue ability, respectively represent the variances of the objective environmental factor and the characterization parameters of the emergency rescue ability, λ represents the attenuation rate coefficient, and Δδ represents the influence time difference of the environmental factor; an influence model is constructed according to the influence of the environmental factor on the task requirements: where r k,i represents the impact factor of the environmental factor on the intensity relationship of task requirements, and represents the characterization value of the environmental parameter at time step t, represents the change amount of the characterization value of a certain task requirement at time step t, and T represents the total number of discrete time steps of the time window.

10. The system according to claim 9, wherein: In the four-dimensional hypergraph model unit of the relationship among rescue mission elements - emergency rescue elements - objective environment elements, based on the hypergraph model of the relationship between rescue mission elements and emergency rescue elements, a new environmental area node set \(L = \{l_1, l_2, \cdots, l\) K \}\) is added, where each element contains various environmental elements within the area, and it is defined as the environmental impact factor vector \(F k =(f k1 , f k2 , \cdots, f kO ), where \(O\) represents the total number of environmental factors, and the four-dimensional hyperedge is defined as \(h ijk =\{a j , s i , l k \}\), indicating that the emergency intelligent agent \(a j \) is affected by the environment of the area it passes through when completing the task \(s i \) after passing through the environmental area \(l k \); and the following constraints are given for the four-dimensional hypergraph: Two satisfactions During the process of emergency agents executing tasks, the paths cover all environmental regions they pass through, forming a many - to - many mapping relationship, that is: where L(i, j, k) represents the emergency agent a j when performing task s i through the environment l k , Path(a j , s i , l k ) represents the set of environments of the complete path experienced by the emergency agent a j from the starting point to the location where task s i is located; for the complete relational network composed of four-dimensional hypergraphs, any hyperedge h ijk needs to satisfy the following dynamic constraints: Among which Φ c (F k ) represents the loss coefficient of the environmental factor in the environment l k for the emergency rescue capacity of the c th category; Two definitions Define the direction of the hyper - edge as the spatio - temporal pointing from the emergency rescue agent to the task, that is, determined by the path order of the emergency agent to execute the task; the following definitions are made for the hyper - edge strength: Among which W l1 represents the total influence value of the environment l k on the path experienced by the emergency agent during the task execution, W l2 represents the total distance traveled by the emergency agent during the task execution, W l3 represents the proportion of the path in each environmental area passed by the emergency agent during the task execution to the whole path, EI k represents the influence value of the environment l k on the emergency agent during the task execution, d k represents the running distance of the emergency agent in the environmental area l k , and K represents the sequence of all environmental areas passed by the emergency agent during the task execution.

Images