CN-121352567-B - Urban flood disaster dynamic emergency decision method, device, equipment and storage medium

CN121352567BCN 121352567 BCN121352567 BCN 121352567BCN-121352567-B

Abstract

The application discloses a dynamic emergency decision method, a device, equipment and a storage medium for urban flood disasters, and relates to the technical field of emergency management and response scheduling. The method comprises the steps of constructing a random differential game model for flood disaster emergency decision, solving a game saddle point strategy through a forward-backward scanning algorithm to obtain an optimal emergency prevention and control strategy, carrying out multi-dimensional simulation verification by taking a typical flood disaster event as a sample, and finally adjusting the strategy according to a simulation result to form a target emergency prevention and control strategy and using the target emergency prevention and control strategy for actual emergency decision. The application can give consideration to space-time dynamic characteristics and random uncertainty in emergency decisions of urban flood disasters, and realize dynamic adaptation and continuous efficient decisions of disaster systems and emergency response systems.

Inventors

CHEN XIAOHONG
HU HANG
WANG YANGJIE
LI SIZHE
LUO WENJIE

Assignees

中南大学

Dates

Publication Date: 20260505
Application Date: 20251219

Claims (7)

1. A dynamic emergency decision method for urban flood disasters, which is characterized by comprising the following steps: Establishing a safety state transition differential equation of a city region, dividing the city region into four safety states, and defining transition relations among the safety states, wherein the four safety states comprise a normal state, a disaster-stricken state, a recovery state and a paralysis state; Constructing a random differential game model of flood disaster emergency decision based on the safety state transfer differential equation, the four safety states and the transfer relation; Solving the random differential game model through a game saddle point strategy based on a forward-backward scanning algorithm to obtain an optimal emergency prevention and control strategy; taking a typical flood disaster event as a simulation sample, substituting the simulation sample into the optimal emergency prevention and control strategy to perform multidimensional simulation verification, and obtaining a simulation result; adjusting the optimal emergency prevention and control strategy according to the simulation result to obtain a target emergency prevention and control strategy, and carrying out emergency decision according to the target emergency prevention and control strategy; the step of constructing a random differential game model for flood disaster emergency decision based on the safety state transfer differential equation, the four safety states and the transfer relation comprises the following steps: Constructing a participant set, a time strategy set, a space behavior set, a strategy set and a state variable set; constructing a state transfer function set according to the calculation logic of the safety state transfer differential equation, the transfer relation and the change rate requirement of each variable in the state variable set; constructing a profit function set according to the transition quantity of each state in the state variable set and the resource consumption of each strategy in the strategy set; Constructing a random differential game model according to the participant set, the time strategy set, the space behavior set, the strategy set, the state variable set, the state transfer function set and the benefit function set; The step of establishing a security state transition differential equation of the urban area comprises the following steps: Acquiring historical flood disaster loss data and a weather hydrologic forecast model experimental result, and determining division basis of flood disaster intensity and emergency prevention and control intensity by combining urban emergency resource reserve scale and rescue team configuration information; dividing the intensity of the flood disaster into three types of high, medium and low according to the dividing basis, and dividing the emergency prevention and control intensity of an emergency decision body into three types of high, medium and low; Setting first selection probabilities of three types of high, medium and low intensity of the flood disaster at different moments based on occurrence frequencies of different intensity disasters in the historical disaster conditions; Setting second selection probabilities of three types of medium-high, medium-low and medium-low emergency prevention and control intensities at different moments based on urban emergency resource scheduling capability and disaster response level standards; Calculating a first actual expected comprehensive efficiency of the flood disaster at different moments and a second actual expected comprehensive efficiency of the emergency decision body based on random interference items, the disaster intensity, the emergency prevention and control intensity, the first selection probability and the second selection probability, and calculating a efficiency difference value of the first actual expected comprehensive efficiency and the second actual expected comprehensive efficiency; Establishing a safety state transfer differential equation of the urban area based on an SIR model in infectious disease dynamics, the disaster intensity, the emergency prevention and control intensity, the random interference item and the efficiency difference value; The step of constructing the participant set, the time strategy set, the space behavior set, the strategy set and the state variable set comprises the following steps: constructing a participant set according to a disaster system and the emergency decision-making main body; Constructing a time strategy set according to the disaster occurrence starting time, the emergency decision ending time, the difference value between the emergency response starting time and the disaster occurrence time and the emergency resource continuous release requirement; constructing a space behavior set according to the flood disaster intensity and the emergency prevention and control intensity; constructing a strategy set according to the disaster intensity occurrence probability data in the historical disaster condition, the control resource scheduling rule of the emergency decision body and each intensity behavior in the space behavior set; And constructing a state variable set according to the four security states and the area quantity of the four security states at each time.
2. The method of claim 1 wherein the random differential gaming model is defined as an octant expressed as follows: Wherein, the In order to participate in the collection of persons, As a function of the time variable, As a set of time policies, As a set of spatial behaviors, As a set of policies, As a set of state variables, As a set of state transfer functions, Is a collection of revenue functions.
3. The method of claim 1, wherein solving the random differential game model by a game saddle point strategy based on a forward-backward scanning algorithm to obtain an optimal emergency prevention and control strategy comprises: determining initial conditions of the random differential game model, wherein the initial conditions comprise the area number of the four security states at the initial moment, initial disaster intensity selection probability and initial prevention and control intensity selection probability; forward solving the safety state transfer differential equation to obtain a disaster evolution state track; Constructing a Hamiltonian of the emergency decision main body, and carrying out backward solution on the Hamiltonian to obtain common-state variables of a disaster system and the emergency decision main body; Iteratively optimizing a policy combination of the disaster system and the emergency decision body based on the disaster evolution state track and the common-mode variable; Calculating a first difference value between the disaster evolution state track obtained by the current iteration and the disaster evolution state track obtained by the previous iteration, a second difference value between the common-state variable obtained by the current iteration and the common-state variable obtained by the previous iteration, and a third difference value between the strategy combination obtained by the current iteration and the strategy combination obtained by the previous iteration, If any one of the first difference value, the second difference value and the third difference value is larger than or equal to a preset convergence threshold value, returning to the step of performing forward solution on the safety state transition differential equation to obtain a disaster evolution state track; and if the first difference value, the second difference value and the third difference value are smaller than the preset convergence threshold value, extracting a strategy set of the emergency decision main body from the strategy combination to obtain an optimal emergency prevention and control strategy.
4. A method according to any one of claims 1 to 3, wherein the step of substituting the optimal emergency prevention and control strategy into the simulated sample for multidimensional simulation verification to obtain a simulation result comprises: taking a typical flood disaster event as a simulation sample, and acquiring historical disaster conditions, emergency resources and geospatial data corresponding to the simulation sample; Setting simulation initial parameters based on the historical disaster condition data, wherein the simulation initial parameters comprise the number of the four safety state areas, the initial value of disaster intensity probability, the cost coefficient of prevention and control intensity and random interference intensity at the initial moment; substituting the optimal emergency prevention and control strategy and the simulation initial parameters into a simulation scene, and starting simulation according to a state transition rule of the random differential game model; When simulation is started, monitoring and recording the state variable set, the strategy implementation cost of the emergency decision body and the disaster control effect at each moment to obtain monitoring data; and analyzing the monitoring data from three dimensions of time, space and efficiency to obtain simulation results of the state evolution curve, the cost change curve and the effect evaluation table.
5. An urban flood disaster dynamic emergency decision device, characterized in that the device comprises: The state modeling module is used for establishing a safety state transfer differential equation of the urban area, dividing the urban area into four safety states, defining transfer relations among the four safety states, wherein the four safety states comprise a normal state, a disaster-affected state, a recovery state and a paralysis state, the step of establishing the safety state transfer differential equation of the urban area comprises the steps of acquiring historical flood loss data and a weather and hydrologic prediction model experimental result, combining urban emergency resource reserve scale and rescue team configuration information, determining division basis of flood disaster intensity and emergency prevention and control intensity, dividing the flood disaster intensity into three types of high, medium and low according to the division basis, dividing the emergency prevention and control intensity of an emergency decision main body into three types of high, medium and low, setting first selection probability of the three types of the flood disaster intensity at different moments based on the occurrence frequency of different types of the historical disaster, setting the first selection probability of the high, medium and low types of the flood disaster intensity at different moments based on an emergency resource scheduling capability and a weather hydrologic prediction model experiment result, combining the urban emergency resource reserve scale and rescue team configuration information, calculating a second selection probability of the emergency disaster intensity and the emergency control intensity at different moments based on the first random selection probability of the emergency disaster intensity, the emergency disaster intensity and the actual disaster intensity and the second selection probability of the emergency decision main body and the emergency disaster intensity, and the actual disaster control probability of the emergency decision main body and the actual disaster control probability are calculated based on the first random selection probability and the actual comprehensive performance and the expected performance and the actual comprehensive performance and the expected performance, establishing a safety state transition differential equation of the urban area; The game modeling module is used for constructing a random differential game model for flood disaster emergency decision based on the safety state transfer differential equation, the four safety states and the transfer relation; the step of constructing a random differential game model for flood disaster emergency decision based on the safety state transfer differential equation, the four safety states and the transfer relation comprises constructing a participant set, a time strategy set, a space behavior set, a strategy set and a state variable set; the method comprises the steps of constructing a state transfer function set according to calculation logic of a safety state transfer differential equation, transfer relations and change rate requirements of variables in the state variable set, constructing a benefit function set according to transfer quantity of the states in the state variable set and resource consumption of strategies in the strategy set, constructing a random differential game model according to the participant set, the time strategy set, the space behavior set, the strategy set, the state variable set, the state transfer function set and the benefit function set, constructing a participant set, a time strategy set, the space behavior set, the strategy set and the state variable set, constructing a participant set according to a disaster system and an emergency decision main body, constructing a time strategy set according to a disaster occurrence starting time, an emergency decision ending time, a difference value between emergency response starting time and disaster occurrence time and the requirement for continuously throwing emergency resources, constructing a space behavior set according to the flood disaster intensity and emergency prevention intensity, constructing disaster probability data according to the strength decision in history, the prevention and control scheduling rules of the emergency main body, constructing a strategy set according to the four safety states and the area quantity of the four safety states at each time, and constructing a state variable set; The strategy solving module is used for solving the random differential game model through a game saddle point strategy based on a forward-backward scanning algorithm to obtain an optimal emergency prevention and control strategy; The simulation verification module is used for substituting the typical flood disaster event as a simulation sample into the optimal emergency prevention and control strategy to perform multidimensional simulation verification to obtain a simulation result; and the decision optimization module is used for adjusting the optimal emergency prevention and control strategy according to the simulation result to obtain a target emergency prevention and control strategy and carrying out emergency decision according to the target emergency prevention and control strategy.
6. An urban flood disaster dynamic emergency decision device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the urban flood disaster dynamic emergency decision method according to any one of claims 1 to 4.
7. A storage medium, characterized in that the storage medium is a computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the urban flood disaster dynamic emergency decision method according to any one of claims 1 to 4.

Description

Urban flood disaster dynamic emergency decision method, device, equipment and storage medium Technical Field The application relates to the technical field of emergency management and response scheduling, in particular to a dynamic emergency decision method, device and equipment for urban flood disasters and a storage medium. Background The traditional urban flood disaster emergency decision method is mostly based on static or discrete models, and takes the occurrence and development processes of the flood disasters as relatively static environments or focuses decisions on strategy selection at discrete time points. The model can capture strategy interaction among multiple subjects, but cannot reflect the continuous, state-dependent and real-time feedback dynamic game relationship between the disaster system and the emergency response system, so that the capability of supporting real-time and dynamic emergency decisions is limited. The differential game theory is applied in the field of emergency management, but the application in the field of urban flood disaster emergency decision is still in an exploration stage. The urban flood disaster is taken as a typical dynamic complex system, the evolution process of the urban flood disaster has the characteristics of high dynamic property, time variability, space heterogeneity and the like, and the emergency response behavior under the scene is continuous time strategy adjustment and interactive game carried out by a decision-making main body under the boundary conditions of incomplete information, resource constraint rigidity, environment uncertainty and the like. Therefore, how to consider the space-time dynamic characteristics and the random uncertainty in the emergency decision of urban flood disasters and realize the dynamic adaptation and continuous efficient decision of a disaster system and an emergency response system becomes a problem to be solved urgently. Disclosure of Invention The application aims to provide a dynamic emergency decision method, device, equipment and storage medium for urban flood disasters, and aims to solve the technical problems of considering time-space dynamic characteristics and random uncertainty in the emergency decision of urban flood disasters and realizing dynamic adaptation and continuous efficient decision of a disaster system and an emergency response system. In order to achieve the above purpose, the application provides a dynamic emergency decision method for urban flood disasters, which comprises the following steps: Establishing a safety state transition differential equation of a city region, dividing the city region into four safety states, and defining transition relations among the safety states, wherein the four safety states comprise a normal state, a disaster-stricken state, a recovery state and a paralysis state; Constructing a random differential game model of flood disaster emergency decision based on the safety state transfer differential equation, the four safety states and the transfer relation; Solving the random differential game model through a game saddle point strategy based on a forward-backward scanning algorithm to obtain an optimal emergency prevention and control strategy; taking a typical flood disaster event as a simulation sample, substituting the simulation sample into the optimal emergency prevention and control strategy to perform multidimensional simulation verification, and obtaining a simulation result; and adjusting the optimal emergency prevention and control strategy according to the simulation result to obtain a target emergency prevention and control strategy, and carrying out emergency decision according to the target emergency prevention and control strategy. In an embodiment, the step of constructing a random differential game model for flood disaster emergency decision based on the safety state transition differential equation, the four safety states and the transition relation includes: Constructing a participant set, a time strategy set, a space behavior set, a strategy set and a state variable set; constructing a state transfer function set according to the calculation logic of the safety state transfer differential equation, the transfer relation and the change rate requirement of each variable in the state variable set; constructing a profit function set according to the transition quantity of each state in the state variable set and the resource consumption of each strategy in the strategy set; and constructing a random differential game model according to the participant set, the time strategy set, the space behavior set, the strategy set, the state variable set, the state transfer function set and the benefit function set. In one embodiment, the step of constructing the participant set, the temporal policy set, the spatial behavior set, the policy set, and the state variable set includes: constructing a participant set according to the disaster system and the emergenc