Digital twinborn emergency rescue deduction optimization method for complex disaster scene

Abstract

The invention discloses a digital twinborn emergency rescue deduction optimization method for a complex disaster scene, and the method comprises the steps: constructing a composite random model for emergency resources, rescue task demands and the high dynamics and randomness of environmental elements in the disaster scene through employing a random process theory and a discretization method, and carrying out the deduction optimization of the digital twinborn emergency rescue. A deduction algorithm of a continuous state space and discrete time step length is constructed based on a digital twin deduction technology, bidirectional mapping of a physical space and a virtual space is realized, an emergency rescue optimization module based on a matching game mechanism is embedded in a digital twin deduction framework, a task demand gap and resource cost are taken as objective functions, and the task demand gap and the resource cost are taken as objective functions. And a global optimal solution is gradually approached by utilizing multi-round iteration and local strategy exchange, so that efficient optimization of resource allocation in a disaster scene is realized. According to the method, the future disaster trend can be predicted in real time, and dual optimization of task demand satisfaction and resource cost control can be realized.

Description

Technical Field

[0001] The present invention relates to the fields of digital twins and emergency management, and specifically to an emergency resource scheduling optimization method based on dynamic random modeling of multi-domain elements, digital twin deduction and matching game optimization in complex disaster scenarios. Background Art

[0002] In complex disaster emergency management, the evolution of disaster scenarios is highly dynamic and stochastic, with significant uncertainty in the dynamic changes of various elements (such as emergency rescue resources, rescue mission requirements, and objective environmental parameters). During the rescue mission, the energy and material resources of emergency agents, including emergency agents, are subject to nonlinear random attenuation due to mission efficiency and external environmental disturbances. Rescue mission requirements, such as communications, medical care, firefighting, and supplies, often exhibit exponential growth and sudden jumps due to the spread of the disaster and secondary events (such as aftershocks and heavy rain). Furthermore, objective environmental parameters (such as wind speed and temperature) exhibit both mean-reversion characteristics and random fluctuations. These elements do not evolve in isolation but rather influence each other through complex coupling mechanisms. Existing scheduling optimization methods, which are mostly based on static or quasi-static models, are unable to cope with dynamic scheduling problems involving multiple time steps and the coevolution of multiple variables. Traditional centralized planning approaches often suffer from computational delays and response rigidity when faced with minute-by-minute simulations and sudden disturbances, failing to meet the real-time intelligent decision-making needs of disaster sites. In addition, existing methods lack the ability to model the random coupling relationship between the three elements of resources, tasks and environment, and it is difficult to effectively reflect their evolutionary coordination in the temporal and spatial dimensions. Summary of the Invention

[0003] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a digital twin emergency rescue simulation optimization method for complex disaster scenarios.

[0004] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is: A digital twin emergency rescue simulation optimization method for complex disaster scenarios includes the following steps: Step 1: Obtain emergency rescue capabilities, mission requirements, and environmental factors for complex disaster scenarios. Based on these capabilities, mission requirements, and environmental factors, we establish emergency rescue factor models, rescue mission factor models, and objective environmental factor models for complex disaster scenarios. We also introduce the concept of time evolution and construct a multi-domain factor dynamic model based on a stochastic model.

[0005] Step 2: On the basis of the multi-domain element dynamic model based on the random model, a relationship dynamic model between multi-domain elements is constructed based on the time-varying relationship intensity matrix.

[0006] Step 3: The dynamic relationship model between multi-domain elements is discretized into an iterative formula using the Euler–Maruyama method, and the extended Kalman filter is used for state correction to achieve physical-virtual closed-loop mapping.

[0007] Step 4: During the state correction process of the extended Kalman filter, a gap-cost joint objective function is constructed for the task demand gap and resource supply at each time step. In the multi-round proposal-acceptance game process, based on the matching game idea, the global optimal resource allocation strategy is generated through bilateral preference order iteration.

[0008] Preferably, the dynamic relationship model between the multi-domain elements in step 2 is constructed as follows: In order to show the relationship between various factors, the time-varying relationship intensity matrix is introduced , the formal definition is as follows:

[0009] in, Indicates the interaction intensity between mission requirements and emergency rescue capabilities. Represents the relationship between task requirements and environmental factors. Represents the interaction between rescue capabilities and mission requirements. Representing the mutual influence between emergency rescue capabilities and environmental factors, the final dynamic relationship model between multi-domain elements is:

[0010] in, represents the system state vector, A unified representation value representing emergency rescue capabilities, A unified representation value representing the intensity of task requirements, Represents the unified representation value of objective environmental parameters, drift term represents the deterministic evolution of the system, The diffusion term represents the random disturbance of the system and is defined as follows:

[0011] in, Represents an emergency agent is the space-time position The amount of resources provided by the task point, Indicates the resource base consumption rate caused by task execution per unit time. represents a standard normally distributed random number, Indicates the The natural growth coefficient of the time evolution of class task requirements, Indicates the resource satisfaction efficiency coefficient, Represents the dynamic weights of various environmental factors, represents the number of environmental factors, Indicates the intensity of environmental interference.

[0012] Preferably, step 3 includes the following steps:

[0013] Define the state space of the simulation system that integrates the three-dimensional dynamic characteristics of emergency rescue capabilities, rescue mission requirements, and objective environmental parameters as follows:

[0014] in, They represent the emergency rescue element set, task requirement set and objective environment set respectively. express The unified representation value of emergency rescue capability at any moment, express The unified representation value of the intensity of task demand at the moment, express The unified representation value of the objective environmental parameters at the moment, and the dynamic evolution equation are defined according to the continuous time evolution rules of the state variables:

[0015] in A dynamic model representing various elements, Represents various interference terms in the dynamic model of various elements, and on this basis, maps the virtual deduction results to physical sensor data to establish the observation function, which is defined as follows:

[0016] in represents the sensor mapping function, Indicates the error adjustment value.

[0017] The stochastic differential equation is discretized using Euler-Maruyama as follows:

[0018] in represents the discrete value of the random interference term. On this basis, we

[0019] Reconstruct the time-varying relationship strength matrix for the updated state variables , while receiving sensor observation data , the predicted state is corrected by the extended Kalman filter:

[0020] in represents the modified state space variables, represents the Kalman gain matrix, represents the temporal relationship intensity matrix, Indicates the next moment in time. Indicates at time The forecast state before the correction, express Sensor observation data at each moment, Represents the identity matrix, which is used to update the covariance matrix. represents the predicted covariance matrix, represents the corrected covariance matrix, represents the observation noise.

[0021] Preferably, step 4 includes:

[0022] To carry out the task Corresponding emergency agent No. Demand gap for emergency rescue capabilities Defined as:

[0023] in, Represents a binary assignment variable, judging the emergency agent Is it a task Provide resources, then the global resource gap Defined as:

[0024] in Indicates the priority weight of capabilities, reflecting the urgency of different task requirements. For emergency response, only the total execution cost of the task can be provided. To define:

[0025] in represents the resource consumption cost of the emergency agent, represents the path risk cost, Represents the agent allocation strategy. Finally, a multi-objective joint optimization function is constructed:

[0026] in, Indicates the construction of multi-objective joint optimization function, represents the total time step of the setting, Represents the cost sensitivity coefficient, determines the weight of the cost item, and finds the allocation matrix Make Maximum, that is:

[0027] in, represents the optimal allocation matrix.

[0028] Preferably, step 4 includes: defining a task preference order model, scoring task preferences To quantify, the specific form is:

[0029] in Indicates the current gap, Indicates the assumption that a new emergency agent is assigned For the emergency agent preference model, through the utility function Quantify it in the following form:

[0030] in, The utility function representing the preferences of the emergency agent, Indicates the priority weight of the task requirement, represents the resource consumption cost of the emergency agent, represents the path risk cost.

[0031] Preferably, the method for iteratively generating the global optimal resource allocation strategy according to the bilateral preference order in step 4 comprises the following steps:

[0032] Step 421, initialize the feasible solution, determine the allocation strategy of the initial emergency agent and task requirements according to the time-varying intensity matrix, and generate the initial allocation matrix , ensure that each task is covered by at least one emergency agent, and calculate the initial objective function value .

[0033] Step 422, the emergency agent proposal phase calculates the effectiveness of each emergency agent in performing tasks based on the current system state and allocation matrix. , select the task with the highest performance , and assign tasks to emergency agents and submit assignment proposals.

[0034] Step 423, task acceptance phase, the task is to collect a list of all proposed emergency agents and calculate the difference reduction corresponding to each proposal. , and update the allocation , other proposals are set to 0.

[0035] Step 423, objective function evaluation phase, recalculate the new objective function value using the updated allocation matrix , repeat the above steps, and stop the iteration if the iteration condition is met.

[0036] Preferably, the multi-domain element dynamic model based on the random model in step 1 includes a dynamic model of the emergency rescue element, and the dynamic model of the emergency rescue element is as follows:

[0037] in, Emergency Agent at all times The evolution function of emergency rescue capability, Indicates the resource base consumption rate caused by task execution per unit time. represents the random disturbance amplitude of environmental interference on resource consumption, Represents the emergency agent at time The unified representation value of Indicates the calibration coefficient of the energy consumption rate corresponding to unit maneuverability, Represents the emergency agent's position in space and time mobility, Represents an emergency agent Distance and space-time position when executing the task The obstacle distance in Represents an emergency agent At the moment The total obstacle distance when performing the task, represents the smoothing term, represents the obstacle disturbance coefficient, Representing space-time position The wind speed at represents the wind speed disturbance weight, represents a random interference term.

[0038] Preferably, the multi-domain element dynamic model based on the random model in step 1 includes a dynamic evolution model of the rescue mission element, and the dynamic evolution model of the rescue mission element is as follows:

[0039] in Indicates location in space and time Mission Point The first Class task requirements, Indicates the The natural growth coefficient of the time evolution of class task requirements, Indicates the resource satisfaction efficiency coefficient, Represents an emergency agent In space-time position For the task The amount of resources allocated to the task class.

[0040] Preferably, the multi-domain factor dynamic model based on the random model in step 1 includes a dynamic evolution model of objective environmental factors, and the dynamic evolution model of objective environmental factors is as follows:

[0041] in Representing space-time position The unified characterization value of the environmental parameters at Represents the dynamic weights of various environmental factors, Indicates the intensity of environmental interference, represents the random disturbance of the environment, It represents the sensitivity coefficient, which reflects the contribution of the change of reaction parameters to the global environmental state.

[0042] Another object of the present invention is to provide a digital twin emergency rescue simulation optimization system for complex disaster scenarios, which is used to implement a digital twin emergency rescue simulation optimization method for complex disaster scenarios. The system includes an input unit, a multi-domain element dynamic model unit, a relationship dynamic model unit, a physical-virtual closed-loop mapping unit, a global optimal resource allocation unit, and an output unit, wherein:

[0043] The input unit is used to input emergency rescue capabilities, mission requirements, and environmental factors of complex disaster scenarios.

[0044] The multi-domain element dynamic model unit is used to establish an emergency rescue element model, a rescue mission element model and an objective environment element model for complex disaster scenarios based on emergency rescue capabilities, task requirements and environmental factors, while introducing the concept of time evolution to construct a multi-domain element dynamic model based on a random model.

[0045] The relationship dynamics model unit is used to construct a relationship dynamics model between multi-domain elements based on a multi-domain element dynamics model based on a random model and a time-varying relationship intensity matrix.

[0046] The physical-virtual closed-loop mapping unit is used to discretize the relationship dynamics model between multi-domain elements into an iterative formula through the Euler-Maruyama method, and perform state correction in combination with the extended Kalman filter to achieve physical-virtual closed-loop mapping.

[0047] The global optimal resource allocation unit is used to extend the Kalman filter to perform state correction. For the task demand gap and resource supply at each time step, a gap-cost joint objective function is constructed. In the process of multiple rounds of proposal-acceptance game, the global optimal resource allocation strategy is generated through bilateral preference order iteration based on the matching game idea.

[0048] The output unit is used to output the global optimal resource allocation strategy.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] This invention uses random process theory and discretization methods to construct a composite random model targeting the high dynamics and randomness of emergency resources, rescue mission requirements and environmental factors in disaster scenarios. The model is used to simulate the dynamic evolution and interaction of various factors to characterize the nonlinear evolution laws of emergency resource consumption, mission demand growth and environmental disturbances. Based on the digital twin deduction technology, a deduction algorithm of continuous state space and discrete time steps is constructed to achieve two-way mapping of physical space and virtual space. At the same time, an emergency rescue optimization module based on the matching game mechanism is embedded in the digital twin deduction framework. Based on the matching game idea, a bilateral dynamic matching mechanism between tasks and emergency resources is constructed. With the task demand gap and resource cost as the objective function, multiple rounds of iteration and local strategy exchange are used to gradually approach the global optimal solution, thereby achieving efficient optimization of resource allocation in disaster scenarios.

[0051] Based on the actual disaster evolution process, the present invention can dynamically model multi-domain elements, predict future disaster trends in real time, and conduct multiple rounds of resource optimization and matching through a game mechanism to achieve dual optimization of mission requirement satisfaction and resource cost control. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a flow chart of the present invention;

[0053] Figure 2 It is a schematic diagram of the dynamic evolution of multi-domain elements. DETAILED DESCRIPTION

[0054] The present invention is further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these examples are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention made by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0055] A digital twin emergency rescue simulation optimization method for complex disaster scenarios, such as Figure 1 As shown, the tool includes the following steps:

[0056] Step 1: For the emergency rescue element model, rescue mission element model, and objective environment element model of complex disaster scenarios, the concept of time evolution is introduced to construct a multi-domain element dynamic model based on a stochastic model;

[0057] The method for constructing a multi-domain element dynamic model based on a random model is as follows:

[0058] (1) Dynamic model of emergency rescue elements:

[0059] in, Emergency Agent at all times The evolution function of emergency rescue capability, Indicates the resource base consumption rate caused by task execution per unit time. represents the random disturbance amplitude of environmental interference on resource consumption, Represents the emergency agent at time The unified representation value of Indicates the calibration coefficient of the energy consumption rate corresponding to unit maneuverability, Represents the emergency agent's position in space and time mobility, Represents an emergency agent Distance and space-time position when executing the task The obstacle distance in Represents an emergency agent At the moment The total obstacle distance when performing the task, represents the smoothing term, represents the obstacle disturbance coefficient, Representing space-time position The wind speed at represents the wind speed disturbance weight, represents a random interference term.

[0060] (2) Dynamic evolution model of rescue mission elements:

[0061] in Indicates location in space and time Mission Point The first Class task requirements, Indicates the The natural growth coefficient of the time evolution of class task requirements, Indicates the resource satisfaction efficiency coefficient, Represents an emergency agent In space-time position For the task The amount of resources allocated to the task class.

[0062] (3) Dynamic evolution model of objective environmental factors:

[0063] in Representing space-time position The unified characterization value of the environmental parameters at Represents the dynamic weights of various environmental factors, Indicates the intensity of environmental interference, represents the random disturbance of the environment, It represents the sensitivity coefficient, which reflects the contribution of the change of reaction parameters to the global environmental state.

[0064] Step 2: Based on step 1, a dynamic relationship model between multi-domain elements is constructed based on the time-varying relationship intensity matrix;

[0065] First, in order to show the relationship between various factors, the time-varying relationship intensity matrix is introduced , the formal definition is as follows:

[0066] in It represents the interaction intensity between mission requirements and emergency rescue capabilities, reflecting the impact of rescue capabilities on mission requirements; It represents the relationship between task requirements and environmental factors, reflecting the impact of the environment on task execution; It represents the interaction between rescue capability and mission requirements, and reflects the consumption of resources after mission execution; It represents the mutual influence between emergency rescue capabilities and environmental factors, and reflects the impact of environmental changes on resource utilization efficiency. The final composite random model is:

[0067] in represents the system state vector, A unified representation value representing emergency rescue capabilities, A unified representation value representing the intensity of task requirements, Represents the unified representation value of objective environmental parameters, drift term represents the deterministic evolution of the system, The diffusion term represents the random disturbance of the system, and its main definition is as follows:

[0068] in Represents an emergency agent is the space-time position The amount of resources provided by the task point.

[0069] Step 3: Discretize the composite stochastic model into an iterative formula using the Euler–Maruyama method, and perform state correction in combination with the extended Kalman filter to achieve physical-virtual closed-loop mapping;

[0070] First, the state space of the simulation system is defined, which integrates the three-dimensional dynamic characteristics of emergency rescue capabilities, rescue mission requirements, and objective environmental parameters. as follows:

[0071] in They represent the set of emergency rescue elements, the set of mission requirements, and the set of objective environments respectively; the dynamic evolution equation is defined according to the continuous time evolution rule of the state variables:

[0072] in Represent the dynamic models of various elements, represent the various interference terms in the dynamic models of various elements, and on this basis, map the virtual deduction results to physical sensor data to establish the observation function, which is defined as follows:

[0073] in represents the sensor mapping function, Represents the error adjustment value; at the same time, to ensure the dynamic randomness of the model and achieve efficient simulation, the stochastic differential equation is discretized into the following form using Euler-Maruyama:

[0074] in Represents the discrete value of the random interference term; on this basis, we reconstruct the time-varying relationship intensity matrix for the updated state variables , while receiving sensor observation data , the predicted state is corrected by the extended Kalman:

[0075] in represents the modified state space variables, Represents the Kalman gain matrix, which determines the weight distribution of model decision error and observation noise. represents the error covariance matrix, represents the observation noise.

[0076] Step 4: Construct a gap-cost joint objective function for the task demand gap and resource supply at each time step. In the process of multiple rounds of proposal-acceptance game, based on the matching game idea, generate the global optimal resource allocation strategy through bilateral preference order iteration.

[0077] First, perform the task Corresponding emergency agent No. Demand gap for emergency rescue capabilities Defined as:

[0078] in, Represents a binary assignment variable, judging the emergency agent Is it a task Provide resources, then the global resource gap Defined as:

[0079] in Indicates the priority weight of capabilities, reflecting the urgency of different task requirements; for emergency response, only the total execution cost of the task can be provided. To define:

[0080] in Indicates the resource consumption cost of the emergency agent, which is negatively correlated with the endurance. represents the path risk cost, Represents the agent allocation strategy; finally, a multi-objective joint optimization function is constructed:

[0081] in Represents the cost sensitivity coefficient, determines the weight of the cost item, and finds the allocation matrix Make Maximum, that is:

[0082] Further define the task preference order model, through task preference scoring To quantify, the specific form is:

[0083] in Indicates the current gap, Indicates the assumption that a new emergency agent is assigned New gap after; for the emergency agent preference model, through the utility function Quantify it in the following form:

[0084] The global optimal resource allocation strategy is generated iteratively based on the bilateral preference order. The specific algorithm implementation process is as follows:

[0085] (1) Initialize the feasible solution, determine the allocation strategy of the initial emergency agent and task requirements according to the time-varying intensity matrix, and generate the initial allocation matrix , ensure that each task is covered by at least one emergency agent, and calculate the initial objective function value ;

[0086] (2) The emergency agent proposal phase calculates the effectiveness of each emergency agent in performing tasks based on the current system state and allocation matrix. , select the task with the highest performance , and assign tasks to emergency agents and submit assignment proposals;

[0087] (3) Task acceptance stage: The task is to collect the list of emergency agents of all proposals and calculate the difference reduction corresponding to each proposal. , and update the allocation , other proposals are set to 0;

[0088] (4) Objective function evaluation phase: recalculate the new objective function value using the updated allocation matrix , repeat the above steps, and stop the iteration if the iteration condition is met;

[0089] Step 5: Based on the above steps, build a digital twin simulation model for complex disaster scenarios.

[0090] The overall process of the digital twin deduction model is as follows:

[0091] Step 1: Data collection, collecting the resource status, task requirements and environmental parameters of the emergency agent and related sensors;

[0092] Step 2: Dynamic modeling, dynamic modeling of multi-domain elements based on random models;

[0093] Step 3: Twin deduction, which is used to perform closed-loop deduction of Euler–Maruyama discretization and extended Kalman filtering, and build a twin deduction mechanism;

[0094] Step 4: Decision optimization, calculating and outputting the global resource allocation strategy through the matching game mechanism;

[0095] Step 5: Control feedback, input the optimization results into the deduction module, iterate until the preset termination conditions are met, and output the final resource allocation strategy.

[0096] In another embodiment of the present invention, a digital twin emergency rescue simulation optimization system for complex disaster scenarios is provided, which is used to implement a digital twin emergency rescue simulation optimization method for complex disaster scenarios, including an input unit, a multi-domain element dynamic model unit, a relationship dynamic model unit, a physical-virtual closed-loop mapping unit, a global optimal resource allocation unit, and an output unit, wherein:

[0097] The input unit is used to input emergency rescue capabilities, mission requirements, and environmental factors of complex disaster scenarios.

[0098] The multi-domain element dynamic model unit is used to establish an emergency rescue element model, a rescue mission element model and an objective environment element model for complex disaster scenarios based on emergency rescue capabilities, task requirements and environmental factors, while introducing the concept of time evolution to construct a multi-domain element dynamic model based on a random model.

[0099] The relationship dynamics model unit is used to construct a relationship dynamics model between multi-domain elements based on a multi-domain element dynamics model based on a random model and a time-varying relationship intensity matrix.

[0100] The physical-virtual closed-loop mapping unit is used to discretize the relationship dynamics model between multi-domain elements into an iterative formula through the Euler-Maruyama method, and perform state correction in combination with the extended Kalman filter to achieve physical-virtual closed-loop mapping.

[0101] The global optimal resource allocation unit is used to extend the Kalman filter to perform state correction. For the task demand gap and resource supply at each time step, a gap-cost joint objective function is constructed. In the process of multiple rounds of proposal-acceptance game, the global optimal resource allocation strategy is generated through bilateral preference order iteration based on the matching game idea.

[0102] The output unit is used to output the global optimal resource allocation strategy.

[0103] The simulation relationship between the three elements of emergency rescue capability, mission requirement intensity, and environmental parameters in this embodiment is as follows: Figure 2 As shown, the present invention uses random process theory and discretization methods to construct a composite random model based on the high dynamics and randomness of emergency resources, rescue mission requirements and environmental factors in disaster scenarios. The model is used to simulate the dynamic evolution and interaction of various factors to characterize the nonlinear evolution laws of emergency resource consumption, mission demand growth and environmental disturbances. Based on the digital twin deduction technology, a deduction algorithm of continuous state space and discrete time steps is constructed to achieve two-way mapping of physical space and virtual space. At the same time, an emergency rescue optimization module based on the matching game mechanism is embedded in the digital twin deduction framework. Based on the matching game idea, a bilateral dynamic matching mechanism between tasks and emergency resources is constructed. With the task demand gap and resource cost as the objective function, multiple rounds of iteration and local strategy exchange are used to gradually approach the global optimal solution, thereby achieving efficient optimization of resource allocation in disaster scenarios.

[0104] 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 digital twin emergency rescue simulation optimization method for complex disaster scenarios, characterized by: The following steps are involved: Step 1: Obtain the emergency rescue capabilities, mission requirements, and environmental factors of complex disaster scenarios; establish emergency rescue factor models, rescue mission factor models, and objective environmental factor models for complex disaster scenarios based on these factors, while introducing the concept of time evolution and constructing a multi-domain factor dynamic model based on a stochastic model; Step 2: On the basis of the multi-domain element dynamic model based on the random model, a dynamic model of the relationship between multi-domain elements is constructed based on the time-varying relationship intensity matrix; Step 3: Discretize the dynamic relationship model between multi-domain elements into an iterative formula using the Euler–Maruyama method, and perform state correction in combination with the extended Kalman filter to achieve physical-virtual closed-loop mapping. Step 4: During the state correction process of the extended Kalman filter, a gap-cost joint objective function is constructed for the task demand gap and resource supply at each time step. In the multi-round proposal-acceptance game process, based on the matching game idea, the global optimal resource allocation strategy is generated through bilateral preference order iteration.

2. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 1 is characterized by: The dynamic relationship model between the multi-domain elements in step 2 is constructed as follows: In order to show the relationship between various factors, the time-varying relationship intensity matrix is introduced , the formal definition is as follows: in, Indicates the interaction intensity between mission requirements and emergency rescue capabilities; Represents the relationship between task requirements and environmental elements; represents the interaction between rescue capabilities and mission requirements; Representing the mutual influence between emergency rescue capabilities and environmental factors, the final dynamic relationship model between multi-domain elements is: in, represents the system state vector, A unified representation value representing emergency rescue capabilities, A unified representation value representing the intensity of task requirements, Represents the unified representation value of objective environmental parameters, drift term represents the deterministic evolution of the system, The diffusion term represents the random disturbance of the system and is defined as follows: in, Represents an emergency agent is the space-time position The amount of resources provided by the task point, Indicates the resource base consumption rate caused by task execution per unit time. represents a standard normally distributed random number, Indicates the The natural growth coefficient of the time evolution of class task requirements, Indicates the resource satisfaction efficiency coefficient, Represents the dynamic weights of various environmental factors, represents the number of environmental factors, Indicates the intensity of environmental interference.

3. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 2 is characterized by: Step 3 includes the following steps: Define the state space of the simulation system that integrates the three-dimensional dynamic characteristics of emergency rescue capabilities, rescue mission requirements, and objective environmental parameters as follows: in, They represent the emergency rescue element set, task requirement set and objective environment set respectively. express The unified representation value of emergency rescue capability at any moment, express The unified representation value of the intensity of task demand at the moment, express The unified representation value of the objective environmental parameters at the moment, and the dynamic evolution equation are defined according to the continuous time evolution rules of the state variables: in A dynamic model representing various elements, Represents various interference terms in the dynamic model of various elements, and on this basis, maps the virtual deduction results to physical sensor data to establish the observation function, which is defined as follows: in represents the sensor mapping function, Indicates the error adjustment value; The stochastic differential equation is discretized using Euler-Maruyama as follows: in Represents the discrete value of the random interference term; on this basis we Reconstruct the time-varying relationship strength matrix for the updated state variables , while receiving sensor observation data , the predicted state is corrected by the extended Kalman filter: in represents the modified state space variables, represents the Kalman gain matrix, represents the temporal relationship intensity matrix, Indicates the next moment in time. Indicates at time The forecast state before the correction, express Sensor observation data at each moment, Represents the identity matrix, which is used to update the covariance matrix. represents the predicted covariance matrix, represents the corrected covariance matrix, represents the observation noise.

4. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 3 is characterized by: Step 4 includes: To carry out the task Corresponding emergency agent No. Demand gap for emergency rescue capabilities Defined as: in, Represents a binary assignment variable, judging the emergency agent Is it a task Provide resources, then the global resource gap Defined as: in Indicates the priority weight of capabilities, reflecting the urgency of different task requirements; for emergency response, only the total execution cost of the task can be provided. To define: in represents the resource consumption cost of the emergency agent, represents the path risk cost, Represents the allocation strategy of the emergency agent; finally, a multi-objective joint optimization function is constructed: in, Indicates the construction of multi-objective joint optimization function, represents the total time step of the setting, Represents the cost sensitivity coefficient, determines the weight of the cost item, and finds the allocation matrix Make Maximum, that is: in, represents the optimal allocation matrix.

5. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 4 is characterized by: Step 4 includes: defining a task preference order model, scoring task preferences To quantify, the specific form is: in Indicates the current gap, Indicates the assumption that a new emergency agent is assigned New gap after; for the emergency agent preference model, through the utility function Quantify it in the following form: in, The utility function representing the preferences of the emergency agent, Indicates the priority weight of the task requirement, represents the resource consumption cost of the emergency agent, represents the path risk cost.

6. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 5 is characterized by: The method for iteratively generating the global optimal resource allocation strategy based on the bilateral preference order in step 4 includes the following steps: Step 421, initialize the feasible solution, determine the allocation strategy of the initial emergency agent and task requirements according to the time-varying intensity matrix, and generate the initial allocation matrix , ensure that each task is covered by at least one emergency agent, and calculate the initial objective function value ; Step 422, the emergency agent proposal phase calculates the effectiveness of each emergency agent in performing tasks based on the current system state and allocation matrix. , select the task with the highest performance , and assign tasks to emergency agents and submit assignment proposals; Step 423, task acceptance phase, the task is to collect a list of all proposed emergency agents and calculate the difference reduction corresponding to each proposal. , and update the allocation , other proposals are set to 0; Step 423, objective function evaluation phase, recalculate the new objective function value using the updated allocation matrix , repeat the above steps, and stop the iteration if the iteration condition is met.

7. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 6 is characterized by: The multi-domain element dynamic model based on the random model in step 1 includes the dynamic model of the emergency rescue element. The dynamic model of the emergency rescue element is as follows: in, Emergency Agent at all times The evolution function of emergency rescue capability, Indicates the resource base consumption rate caused by task execution per unit time. represents the random disturbance amplitude of environmental interference on resource consumption, Represents the emergency agent at time The unified representation value of Indicates the calibration coefficient of the energy consumption rate corresponding to unit maneuverability, Represents the emergency agent's position in space and time mobility, Represents an emergency agent Distance and space-time position when executing the task The obstacle distance in Represents an emergency agent At the moment The total obstacle distance when performing the task, represents the smoothing term, represents the obstacle disturbance coefficient, Representing space-time position The wind speed at represents the wind speed disturbance weight, represents a random interference term.

8. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 7 is characterized by: The multi-domain element dynamic model based on the stochastic model in step 1 includes the dynamic evolution model of the rescue mission elements. The dynamic evolution model of the rescue mission elements is as follows: in Indicates location in space and time Mission Point The first Class task requirements, Indicates the The natural growth coefficient of the time evolution of class task requirements, Indicates the resource satisfaction efficiency coefficient, Represents an emergency agent In space-time position For the mission point The amount of resources allocated to the task class.

9. The digital twin emergency rescue simulation optimization method for complex disaster scenarios according to claim 8 is characterized by: The multi-domain factor dynamic model based on the stochastic model in step 1 includes a dynamic evolution model of objective environmental factors. The dynamic evolution model of objective environmental factors is as follows: in Representing space-time position The unified characterization value of the environmental parameters at Represents the dynamic weights of various environmental factors, Indicates the intensity of environmental interference, represents the random disturbance of the environment, It represents the sensitivity coefficient, which reflects the contribution of the change of reaction parameters to the global environmental state.

10. An optimization system for implementing the digital twin emergency rescue simulation optimization method for complex disaster scenarios as described in claim 1, characterized in that: It includes an input unit, a multi-domain element dynamic model unit, a relationship dynamic model unit, a physical-virtual closed-loop mapping unit, a global optimal resource allocation unit, and an output unit, wherein: The input unit is used to input emergency rescue capabilities, task requirements, and environmental factors of complex disaster scenarios; The multi-domain element dynamic model unit is used to establish an emergency rescue element model, a rescue mission element model, and an objective environment element model for complex disaster scenarios based on emergency rescue capabilities, mission requirements, and environmental factors. At the same time, the concept of time evolution is introduced to construct a multi-domain element dynamic model based on a random model. The relationship dynamics model unit is used to construct a relationship dynamics model between multi-domain elements based on a multi-domain element dynamics model based on a random model and a time-varying relationship intensity matrix; The physical-virtual closed-loop mapping unit is used to discretize the relationship dynamic model between multi-domain elements into an iterative formula through the Euler-Maruyama method, and perform state correction in combination with the extended Kalman filter to achieve physical-virtual closed-loop mapping; The global optimal resource allocation unit is used to construct a gap-cost joint objective function for the task demand gap and resource supply at each time step during the state correction process of the extended Kalman filter, and to generate a global optimal resource allocation strategy through bilateral preference order iteration based on the matching game concept in multiple rounds of proposal-acceptance game. The output unit is used to output the global optimal resource allocation strategy.

Images