A whole-process collaborative resilience improvement method for an urban medical system

Abstract

The application relates to a whole-process collaborative resilience improvement method for an urban medical system, and relates to the field of urban resilience construction decision-making. The application is to solve the problem that a single-stage optimization strategy cannot comprehensively reflect the complexity of the whole process of an earthquake, thereby limiting the resilience improvement effect. The application correlates the basic information of the target urban traffic and medical system to construct a city medical system model; Monte Carlo simulation method is used to simulate post-earthquake uncertainty scenarios under different reinforcement strategies in the city medical system model; the Kmeans++ algorithm is used to cluster the post-earthquake uncertainty scenarios, and the cluster centers are extracted as representative damage scenarios; the reinforcement and recovery strategies under different representative damage scenarios are taken as individuals in the population, a genetic algorithm is used to find the optimal individual, and the reinforcement and recovery strategy corresponding to the optimal individual is used to realize the resilience improvement of the target urban medical system.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of urban resilience construction decision-making. BACKGROUND

[0002] Interdependent infrastructure systems play an indispensable role in modern cities, especially the linkage of transportation and medical systems, which is crucial in the face of extreme events such as earthquakes. With the acceleration of global urbanization, the complexity of infrastructure is increasing, and the dependence between these systems makes them vulnerable to disasters such as earthquakes. In previous earthquakes, phenomena such as hospital function interruption, limited medical services, and traffic network paralysis have frequently occurred, directly hindering the effective deployment of post-disaster rescue and recovery work. Especially for urban medical systems, the function cannot meet the social demand in emergency rescue and continuous medical services after disasters. These phenomena indicate that the seismic resilience of current urban medical systems is urgent and comprehensive improvement strategies must be taken to ensure the rapid recovery and continuous operation of the medical system after an earthquake.

[0003] Existing researches mainly focus on improving the seismic resilience of transportation systems, and the core methods include reinforcing and transforming key infrastructure (such as bridges, roads, etc.) to enhance the seismic capacity and rapid recovery ability of the system after an earthquake. However, these studies are mostly limited to the post-disaster recovery stage, and usually optimize and reinforce key components of the system under limited budget to improve the seismic performance of infrastructure. Some scholars use system-level importance analysis methods to consider the reinforcement priority of components in the transportation system from technical, social, and economic aspects. In addition, scholars have developed stochastic optimization models to maximize the connectivity of the transportation network after an earthquake, thereby effectively selecting reinforcement strategies under limited budget; other studies analyze the budget and benefit relationship of reinforcement through multi-objective genetic algorithm optimization methods. Although these methods improve the seismic capacity of the system, there are still problems of huge computational load and long time consumption in solving high-dimensional complex optimization problems. To address these challenges, some scholars have proposed limited sample analysis methods to determine the reinforcement priority of key components of the system by the correlation between component damage and system post-earthquake function in sample data. This method, which is between traditional importance ranking and stochastic optimization, improves the computational efficiency to some extent. In addition, some researches have proposed intelligent technology-based transportation system optimization methods, such as using multi-modal transportation networks to share post-disaster pressure and developing intelligent transportation systems to realize real-time monitoring and emergency management of transportation networks, thereby enhancing system resilience. By combining intelligent information technology with transportation management, it is helpful to further improve the seismic resilience of the transportation system to ensure the efficient deployment of post-earthquake rescue work.

[0004] In terms of medical systems, the ability of hospitals to quickly recover and provide emergency medical services after a disaster directly affects the success of post-disaster rescue. Therefore, it is particularly important to enhance the disaster resistance of hospitals. Scholars have significantly enhanced the disaster resistance of hospital buildings by using techniques such as seismic isolation and shock absorption, ensuring that hospitals can continue to operate during disasters. In addition, the development of seismic isolation systems based on optimization design methods is also advancing, including the use of meta-heuristic algorithms such as genetic algorithms and whale optimization algorithms to quickly find optimal solutions. The development of intelligent solution algorithms brings new possibilities for resilience optimization design in this field.

[0005] Although existing research has made some progress in enhancing the disaster resistance of transportation and medical systems, there is a common problem of stage fragmentation. Most methods focus on prevention and reinforcement before earthquakes or on recovery and management after earthquakes, lacking systematic and coordinated consideration of the entire earthquake process. However, urban transportation and medical systems are inherently interdependent and highly coupled, and this complex dependency relationship closely affects the performance of the system at various stages before and after an earthquake. Therefore, single-stage optimization strategies cannot fully reflect the complexity of the entire earthquake process, resulting in limited resilience enhancement effects. SUMMARY

[0006] To solve the problem that single-stage optimization strategies cannot fully reflect the complexity of the entire earthquake process, resulting in limited resilience enhancement effects, a whole-process coordinated resilience enhancement method for urban medical systems is provided.

[0007] A whole-process coordinated resilience enhancement method for urban medical systems, comprising:

[0008] Correlate the basic information of the target urban transportation and medical system, and construct an urban medical system model that considers the correlation between transportation and medical systems, as well as the correlation between hospital structures, non-structures, and medical equipment;

[0009] Use the Monte Carlo simulation method to simulate post-earthquake uncertainty scenarios under different reinforcement strategies in the urban medical system model, including the post-earthquake damage state of bridges, the available number of medical departments in hospitals, and the recovery time of bridges and hospitals;

[0010] Use the Kmeans++ algorithm to cluster the post-earthquake uncertainty scenarios and extract the cluster centers as representative damage scenarios;

[0011] Take the reinforcement and recovery strategies under different representative damage scenarios as individuals in the population, use the genetic algorithm to find the optimal individual, and use the reinforcement and recovery strategies corresponding to the optimal individual to enhance the resilience of the target urban medical system.

[0012] Further, the above-mentioned target city traffic and medical system is associated with the basic information of the city medical system model considering traffic and medical treatment, and the structure, non-structure and medical equipment in the hospital, including:

[0013] According to the GIS data and remote sensing image information in the target city range, the road information is obtained, and the traffic facilities and their connection relationship in the road information are abstracted as a topological structure to establish a post-earthquake traffic network topological model;

[0014] According to the population distribution information, the demand for medical services under the post-disaster situation is evaluated, and a medical demand model is constructed;

[0015] According to the structure type, construction year, location of each department and internal medical equipment configuration of the single hospital in the target city range, the vulnerability data of non-structure and medical equipment related disasters are integrated to establish a post-earthquake function quantization model of single hospital;

[0016] The post-earthquake traffic network topological model, the medical demand model and the post-earthquake function quantization model of single hospital jointly constitute the city medical system model.

[0017] Further, the objective function expression of the above-mentioned genetic algorithm is:

[0018]

[0019] Wherein, z represents the reinforcement state, including whether the bridge b is reinforced before the earthquake, whether the hospital h is reinforced in the structure level before the earthquake and whether the hospital h is reinforced in the non-structure structure level before the earthquake;

[0020] X represents the repairing state, including whether the bridge b is being repaired in the period t under the scene ξ and whether the hospital h is being repaired in the period t under the scene ξ;

[0021] A represents the completed repair state, including whether the bridge b is completed in the period t under the scene ξ and whether the hospital h is completed in the period t under the scene ξ;

[0022] Θ is a set of post-earthquake uncertainty scenarios under different reinforcement strategies of Monte Carlo simulation, and Ω is a set of representative damage scenarios;

[0023] E represents the expected optimization target under multiple uncertain scenarios, and F ξ is the weighted optimization target under the scene ξ.

[0024] Further, the expression of the above-mentioned weighted optimization target F ξ under the scene ξ is:

[0025]

[0026] wherein f1 ξ is the optimization objective based on recovery time under scenario ξ, f2 ξ is the optimization objective based on function improvement under scenario ξ, ω1 and ω2 are the weight coefficients of f1 ξ and f2 ξ respectively,

[0027]

[0028] WRT pre is the weighted recovery time of the pre-strengthening post-earthquake urban medical system, WRT ξ is the weighted recovery time of the post-strengthening post-earthquake urban medical system under scenario ξ, Q 0,pre is the function surplus of the pre-strengthening post-earthquake urban medical system at t0, Q0 ξ is the function surplus of the post-strengthening post-earthquake urban medical system at t0 under scenario ξ.

[0029] Further, the expected optimization objective E under the above scenario ξ is obtained by a proxy model calculation of the medical accessibility,

[0030] The input of the proxy model is the post-earthquake damage state of the bridges in the target area and the available number of medical departments in the hospital under the post-earthquake damage scenario, and the output is the post-earthquake medical accessibility.

[0031] Further, the method for obtaining the post-earthquake damage state of the bridges in the target area and the available number of medical departments in the hospital under the above post-earthquake damage scenario is:

[0032] By the Monte Carlo simulation method, the post-earthquake damage scenario is simulated in the urban medical system model, and the function reduction of the bridges, the available number of medical departments in the hospital and the repair time of the bridges and the hospital are evaluated under each post-earthquake damage scenario.

[0033] Further, the constraints of the above objective function include: pre-strengthening strategy constraints, cost constraints and repair resource constraints.

[0034] Further, the pre-strengthening strategy constraints include:

[0035]

[0036] wherein, represents the pre-strengthening hospital h reinforcement strategy selection variable, h∈H, H is the set of single hospitals in the urban medical system, is a binary variable of whether the pre-strengthening hospital h structure is reinforced and has is a binary variable of whether the pre-strengthening hospital h non-structure is reinforced and has

[0037] Further, the above cost constraints include:

[0038]

[0039] wherein c b is the cost of reinforcing the bridge b, z b is a binary variable indicating whether the bridge b is reinforced before the earthquake and has z b ∈{0, 1}, Cost B is the cost of reinforcing the transportation system, is the cost of reinforcing the structure of the hospital h, is the cost of reinforcing the non-structure of the hospital h, Cost H is the cost of reinforcing the medical system, b∈B, B is the set of bridges in the city's medical system.

[0040] Further, the above resource constraints include:

[0041]

[0042] wherein x b,t represents whether the bridge b is being repaired in the time period t, x h,t represents whether the hospital h is being repaired in the time period t, n ET,b and n ET,h are the number of repair engineering teams for bridges and hospitals after the earthquake, respectively, t∈T, T is the set of long-term repair time periods after the earthquake.

[0043] The beneficial effects of the present application are:

[0044] 1. The present application considers the correlation between the transportation and medical systems in terms of function and recovery, quantifies the key influence of pre-earthquake reinforcement decisions on post-earthquake recovery, and establishes a random optimization model for whole-process resilience enhancement. The model realizes comprehensive decision-making for reinforcement and repair of correlated systems under resource constraints, maximizes the reduction of post-earthquake function loss, and shortens the recovery time.

[0045] 2. The present application combines a fast function calculation agent model and integrates a random damage scenario reduction method to design an efficient intelligent solution algorithm, which significantly reduces the decision-making calculation cost and improves the solution efficiency of the random optimization problem.

[0046] 3. The present application optimizes the reinforcement and recovery decisions of correlated systems, not only improves the practical application value of the model, but also provides convenience for the promotion of resilience enhancement strategies. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a flowchart of a whole-process collaborative resilience enhancement method for a city's medical system;

[0048] Figure 2The topology graph of the urban medical system;

[0049] Figure 3 The medical demand graph of the urban medical system;

[0050] Figure 4 The PGA distribution graph in the region when the earthquake scenario M = 7.8;

[0051] Figure 5 The bridge vulnerability curve graph;

[0052] Figure 6 The Bayesian network graph of the key medical department;

[0053] Figure 7 The training process graph of the surrogate model;

[0054] Figure 8 The loss curve graph of the surrogate model;

[0055] Figure 9 The difference and correlation test result diagram between the predicted value and the actual value based on the surrogate model. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0057] Based on the problems in the prior art, it is necessary to integrate complex disaster scenarios, coordinate defense and recovery strategies at different stages, maximize the resilience of the urban medical system, and propose efficient intelligent decision-making algorithms to improve the optimization solving efficiency. It provides a new idea for the construction of future resilient cities and provides strong technical support for decision-makers when formulating urban disaster resistance planning.

[0058] Reference Figures 1 to 9 To specifically describe the present embodiment, the whole-process collaborative resilience improvement method of the urban medical system according to the present embodiment comprises:

[0059] Step 1, collect the relevant basic information data of the target urban medical system, and perform associated modeling on the traffic and medical system to obtain an urban medical system model considering the traffic and medical system and the structure, non-structure and medical equipment in the hospital.

[0060] Specifically, it comprises:

[0061] Step 1.1, Establishing the traffic network topology model: Aggregate the available GIS (Geographic Information System) data and remote sensing image information within the target area to obtain road information, and perform cleaning, sorting, and processing to ensure availability. Abstract the traffic facilities and connection relationships into a topological structure, and assign attributes such as coordinates, length, capacity, and speed limits to nodes and edges to establish a post-earthquake traffic network topology model.

[0062] Step 1.2, Building a medical demand model: Collect population distribution information to determine regional differences in medical demand. Evaluate the demand for medical services in post-disaster situations to ensure the rational allocation of medical resources.

[0063] Step 1.3, Collecting information on individual hospitals: Collect key information on the structural type, construction year, location of each department, and internal medical equipment configuration of individual hospitals within the target area. Integrate non-structural and medical equipment-related disaster vulnerability data to establish a post-earthquake functional quantification model for individual hospitals.

[0064] The topological network of the urban medical system model and the urban medical demand are shown in Figure 2 and Figure 3 . Based on historical earthquakes and considering ground motion attenuation, determine the PGA (Peak Ground Acceleration) size of the target area, as shown in Figure 4 .

[0065] Step 2, Considering the uncertainty of the damage state, functional reduction, and recovery time of components within the post-earthquake urban medical system, generate potential post-earthquake damage scenarios in the urban medical system model established in Step 1 through Monte Carlo simulation. Use the bridge seismic vulnerability model as shown in Figure 5 to simulate the damage state of bridges after an earthquake, and use the Bayesian network as shown in Figure 6 to determine the availability probability of each key medical department within the hospital. Then, under each damage scenario, evaluate the functional reduction of bridges, the available number of hospital departments, and the repair time of bridges and hospitals to estimate the traffic system travel time and the treatment capacity of individual hospitals under each post-earthquake damage scenario. Specifically, it includes:

[0066] Step 2.1, First, based on the obtained road information, the road connection relationship is topological and relevant attributes are assigned, and traffic system modeling is performed. Subsequently, based on daily traffic data and post-earthquake travel demand characteristics, the travel OD matrix in the recovery period is predicted. Then, the damage of key components (such as roads, bridges, tunnels) after the earthquake and the influence of street building debris on traffic capacity are analyzed, and the traffic capacity reduction of the system is evaluated. Combined with traffic assignment method, considering the demand change and traffic capacity loss, the traffic flow distribution of each road section is calculated. Using the shortest path search algorithm, the travel time from different demand points to medical supply points is predicted, and finally the overall analysis of the traffic system function level is carried out.

[0067] Step 2.2, Establish a single hospital function analysis method. First, based on the collected basic data about the hospital, the single hospital modeling is performed. Second, the seismic response analysis of the hospital structure is carried out to evaluate the damage state of the building structure and estimate the repair time. At the same time, the vulnerability data of non-structural and medical equipment in the hospital are integrated to analyze their influence on the hospital function. Then, combined with the correlation of structure, non-structure and medical equipment, the availability of departments is evaluated, and the recovery time of non-structural components is estimated. Based on the number of medical resources and the demand of different types of wounded, the treatment capacity of the hospital is analyzed, and finally the comprehensive treatment capacity of the hospital after the earthquake is evaluated by quantifying the urgency of different types of wounded.

[0068] Step three, build a fast calculation of urban medical system function proxy model based on artificial neural network.

[0069] Combined with the traffic system travel time and the treatment capacity of each single hospital in the post-earthquake damage scenario generated in step two, an artificial neural network proxy model of urban medical system function is constructed. The input data of the artificial neural network proxy model is the post-earthquake damage state of the bridge in the target area and the available number of each medical department in the hospital in the post-earthquake damage scenario generated in step two, and the output data is the normalized post-earthquake medical accessibility. Through training of the above input and output data, a high-precision urban medical system function prediction model is obtained. The training process of the artificial neural network proxy model is shown in Figure 7 , and the difference between the model prediction value and the actual value and the correlation test result are shown in Figure 8 . The step three is specifically:

[0070] Step 3.1, training set data generation of neural network. Based on the uncertainty damage scenario generated in step two, the post-earthquake traffic system travel time and the treatment capacity of each single hospital are calculated, so as to determine the function level of post-earthquake urban medical system, and generate training data set.

[0071] Step 3.2, data normalization. The input data for the neural network model is the bridge post-earthquake damage state and the available number of hospital departments, and the output data is the normalized post-earthquake medical accessibility. Through training, a high-precision urban medical system function calculation agent model is obtained.

[0072] Step 3.3, the internal parameters of the neural network are set as shown in Table 1. In order to avoid underfitting and overfitting problems in the training process, 10,000 random samples are used to train the agent model. The data set is divided into training set, validation set and test set in the ratio of 0.7:0.15:0.15 to ensure the generalization ability of the model.

[0073] Table 1 Internal parameter settings of neural network

[0074]

[0075] Step 3.4, the training process of neural network. First, build the neural network model, determine the number of layers, the number of neurons in each layer and the activation function, and randomly initialize the weight parameters. Then, perform forward propagation, pass the input data layer by layer to the output layer and generate the predicted output. Next, calculate the loss function value by comparing the predicted results with the true values to measure the error of the model. Then, through back propagation, the loss value is transmitted layer by layer, and the gradient of each layer parameter is calculated. According to the optimization algorithm (such as gradient descent method), the parameters of the model are updated step by step until the predetermined convergence condition is reached or other stopping criteria are met. Finally, use the validation set to evaluate the model performance, and adjust the hyperparameters if necessary to further optimize the prediction ability of the model.

[0076] Step 3.5, establish a high-precision function agent model based on neural network. The training process of the agent model is as shown in Figure 8 . Figure 9 The difference between the predicted value and the actual value based on the agent model and the correlation test results are shown. Through this step, the final neural network agent model can predict the performance of the urban medical system with high precision.

[0077] Step four, considering various uncertainties and complexities after the earthquake, the correlation between traffic and medical care in the urban medical system is analyzed, and a two-stage comprehensive and collaborative decision-making model for pre-disaster reinforcement and post-disaster repair is established. Specifically, it includes:

[0078] Step 4.1, stage I optimization of pre-earthquake reinforcement scheme. Under limited economic conditions, pre-earthquake reinforcement decisions are usually modeled as a 0 / 1 knapsack problem. The pre-earthquake reinforcement model aims to select whether to reinforce the components in the system under limited resources to minimize the loss of system function, while quantifying the impact of reinforcement decisions on potential post-earthquake damage scenarios.

[0079] Step 4.2, Based on the consideration of the impact of reinforcement options on post-earthquake damage scenarios, Phase II determines the optimal parallel repair sequence of the urban medical system under potential damage scenarios based on reinforcement schemes. The recovery sequence decision is usually modeled as a TSP problem (Traveling Salesman Problem), considering the limitations of repair resources, planning the repair sequence of components within the system to shorten the weighted recovery time after the earthquake to improve system resilience. The sets, parameters, and decision variables in the mathematical model of specific resilience improvement are shown in Tables 2 to 4:

[0080] Table 2 Definition of sets in the resilience improvement stochastic optimization model

[0081]

[0082] Table 3 Definition of parameters in the resilience improvement stochastic optimization model

[0083]

[0084]

[0085]

[0086] Table 4 Decision variables in the resilience improvement stochastic optimization model

[0087]

[0088]

[0089] The objective and constraints of the specific mathematical model are shown in Equations (1) to (35):

[0090] Equation (1) defines the optimization objective of the model, representing the expected optimization objective under the representative damage scenario:

[0091]

[0092] where z represents the reinforcement state, including z b 、 and x represents the state of repair, including and A represents the completed repair state, including and

[0093]

[0094] Equations (5) to (16) define the selection of the pre-earthquake reinforcement strategy, considering the limitation of reinforcement cost and reflecting the influence of different reinforcement strategies on the function surplus and repair time of bridges and hospitals. Equation (5) is to determine the reinforcement strategy of hospital h according to whether the hospital is reinforced at the structural and non-structural levels, and different reinforcement strategies will affect the post-earthquake treatment capacity and repair time of the hospital, and further affect the recovery decision. Equations (6) to (9) define the undetermined coefficients corresponding to the four strategies of no reinforcement of structure and non-structure, only structural reinforcement, only non-structural reinforcement, and reinforcement of structure and non-structure. After selecting a reinforcement strategy, only the undetermined coefficient before the treatment capacity and repair time corresponding to the strategy is not 0, and the rest are 0, so as to realize the corresponding relationship between different reinforcement strategies and treatment capacity and repair time.

[0095]

[0096] Equations (10) and (11) define the calculation of the treatment capacity and repair time of hospital h under the uncertain damage scenario ξ after the earthquake according to the different reinforcement strategies of the hospital.

[0097]

[0098] Equations (12) and (13) define the influence of whether the bridge b is reinforced on its traffic capacity and repair time under the uncertain damage scenario ξ after the earthquake.

[0099]

[0100] Equation (14) constrains the type of reinforcement decision variable:

[0101]

[0102] Equations (15) and (16) constrain the reinforcement cost invested in the transportation system and the medical system, respectively.

[0103]

[0104] Equations (17) to (32) define the mathematical model of post-earthquake recovery decision, and describe in detail the repair state of bridges and hospitals, resource constraints, and the process of functional recovery of bridges and single hospitals. Equations (17) and (18) establish the relationship between the start and end times of the repair of bridges b and hospitals h. Equations (19) to (22) define whether the bridge b or the hospital h is repaired in period t under the uncertain damage scenario ξ.

[0105]

[0106] Equations (23) and (24) constrain the number of bridges b and hospitals h that can be repaired at the same time, reflecting the limitation of repair resources.

[0107]

[0108] Equations (25) to (28) define whether the bridge b or hospital h is repaired within the time period t under the uncertain post-earthquake damage scenario ξ. Equations (29) and (30) constrain the type of decision variables in the recovery phase.

[0109]

[0110] Equations (31) and (32) represent the functional recovery of the bridge b or hospital h to the pre-earthquake level after the repair under the post-earthquake damage scenario ξ, which realizes the real-time update of the bridge and hospital functional state in the post-earthquake recovery process.

[0111]

[0112] Equations (33) to (35) define the calculation of the functional level and the recovery time weighting method of different functional demand points in the recovery process of the urban medical system under the uncertain post-earthquake damage scenario. Equation (33) calculates the post-earthquake functional level using the surrogate model constructed in step three. Equations (34) and (35) calculate the weighted recovery time considering the weights of different functional demand points.

[0113]

[0114] Step five, extract representative scenarios from a large number of potential damage scenarios as inputs for the random optimization model through the Kmeans++ based clustering algorithm. Establish a random optimization intelligent solving algorithm that integrates uncertainty damage scenario reduction technology. The specific step five is:

[0115] Step 5.1, generate post-earthquake uncertainty scenarios through MCS simulation.

[0116] Through multiple Monte Carlo simulation methods, simulate the post-earthquake structural damage state, medical function reduction, and recovery time under different reinforcement strategies in the urban medical system model constructed in step one as the post-earthquake uncertainty scenario. For a bridge, the post-earthquake uncertainty scenario includes: damage state, residual traffic capacity, and recovery time. After reinforcement, the damage state threshold changes, resulting in different damage states, function reductions, and recovery times before and after reinforcement under the same random conditions. For a hospital, the post-earthquake uncertainty scenario includes: structural damage state (post-earthquake damage state of bridges and hospitals), available department number (function reduction), and recovery time. After reinforcement, the corresponding parameters also change. Number each uncertainty scenario to clearly indicate the damage state, function reduction, or available department number, and recovery time, etc.

[0117] Step 5.2, reduction of post-earthquake uncertainty scenarios.

[0118] Post-earthquake uncertainty scenarios generated by MCS are normalized (bridge capacity, available hospital departments, recovery time scaled to 0-1). The post-earthquake uncertainty scenarios generated by Monte Carlo simulation are clustered by Kmeans++ algorithm, and the cluster centers are extracted as representative damage scenarios to achieve effective reduction of post-earthquake uncertainty scenarios. The specific process is as follows:

[0119] Step 5.2.1, input post-earthquake uncertainty scenarios.

[0120] Step 5.2.2, normalization.

[0121] The post-earthquake structure damage state, medical function reduction and recovery time in the post-earthquake uncertainty scenario are arranged into a matrix, with rows representing damage scenarios and columns representing data. Normalize the data in each column to ensure that the data range is between 0 and 1, and form a standard sample set for clustering.

[0122] Step 5.2.3, initialize cluster center. Randomly select an initial cluster center from the sample set, calculate the selection probability of the next cluster center based on the distance between each data point and the selected cluster center, and select k cluster centers in turn.

[0123] Step 5.2.4, clustering by K-means algorithm. Assign data points to the nearest cluster center and update the position of each cluster center to complete clustering. Repeat the above operation until the cluster center position no longer changes.

[0124] Step 5.2.5, output reduced representative damage scenarios. Reverse normalize the cluster centers obtained by Kmeans++ algorithm to finally form a set of representative damage scenarios, including the damage state of bridges, remaining capacity, recovery time, and the damage state of hospitals, available number of departments and recovery time.

[0125] Step 5.3, efficient intelligent solution algorithm for fusion of representative damage scenarios and agent model. Combine the functional fast calculation agent model constructed in step three and the representative damage scenarios obtained in step 5.2, substitute the representative damage scenarios into the objective function of the random optimization model, i.e. equation (1), and calculate the average value of the objective function under the representative damage scenarios as the target value of random optimization. Use adaptive genetic algorithm to solve the two-stage random optimization model problem, and dynamically adjust the algorithm parameters to avoid local optimal solution. Further obtain the feasible solution within a certain time range as the decision-making scheme for pre-earthquake reinforcement and post-earthquake recovery.

[0126] The specific process is as follows:

[0127] Step 5.3.1, initialization phase, randomly generate a population of strategies (reinforcement and recovery strategies), each individual representing a potential solution of a strategy, which is {x, A, z}, and each strategy can be connected to different representative damage scenarios.

[0128] Step 5.3.2, calculate fitness. Based on the representative damage scenarios and the strategy population, the average loss of function and weighted recovery time of the post-earthquake urban medical system are calculated by the proxy model, and then the maximum expectation is calculated as the fitness value using formula (1).

[0129] Step 5.3.3, genetic evolution cycle. Select the initial potential solution as the parent individual for crossover to generate new offspring individuals and perform mutation operation. Evaluate the fitness value of the offspring and update the population.

[0130] Step 5.3.4, dynamically adjust algorithm parameters. According to the fitness value of the population, dynamically adjust the parameters such as crossover rate and mutation rate of genetic algorithm to avoid local optimal solution.

[0131] Step 5.3.5, check termination condition. If the maximum number of iterations is reached or the set satisfactory solution is found, terminate the algorithm and output the optimal solution or its approximate solution.

[0132] The purpose of the embodiment is to analyze the impact of pre-earthquake reinforcement on post-earthquake potential damage scenarios, consider the complex coupling relationship between traffic and medical care, establish a two-stage random optimization model based on pre-earthquake component reinforcement selection and post-earthquake repair sequence optimization, and construct a pre-earthquake reinforcement and post-earthquake recovery collaborative resilience improvement method. At the same time, the embodiment also solves the problem of long time-consuming in traditional evaluation methods, improves the sampling efficiency through the post-earthquake uncertain damage scenario reduction method. By reducing the sampling bias, further combined with the high-precision proxy model based on neural network, the efficient intelligent solving algorithm for the whole process resilience improvement optimization problem is developed. This algorithm can quickly solve the reinforcement and recovery strategy while ensuring accuracy, and quantify the improved system resilience level, significantly improving the calculation efficiency. Taking M7.8 earthquake as an example, after optimization decision, the post-earthquake function remaining is improved by about 32%, the recovery time is shortened by 12%, the weighted recovery time is shortened by 30%, and the solving time is reduced to 0.3% of the traditional method, effectively solving the problem of high calculation cost in the random optimization solving process.

[0133] While the application has been described with reference to particular embodiments thereof, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present application. It will be apparent to those skilled in the art that numerous modifications can be made within the scope of the present application as defined by the appended claims. It is intended that all such modification fall within the spirit and scope of the present application. It will be understood that the features described in connection with one embodiment can be used in connection with another embodiment.

Claims

1. A method for improving the whole-process collaborative resilience of an urban medical system, characterized in that, The application relates to a method for improving the resilience of a target city medical system. The method comprises the following steps: correlating the basic information of the target city traffic and medical system, and constructing a city medical system model considering the correlation between traffic and medical treatment and the correlation between the structure and non-structure in the hospital and medical equipment; simulating the post-earthquake uncertainty scenarios under different reinforcement strategies in the city medical system model by using a Monte Carlo simulation method, wherein the post-earthquake uncertainty scenarios include the post-earthquake damage state of the bridge, the available number of medical departments in the hospital and the recovery time of the bridge and the hospital; clustering the post-earthquake uncertainty scenarios by using a Kmeans++ algorithm, and extracting the cluster centers as representative damage scenarios; taking the reinforcement and recovery strategies under different representative damage scenarios as individuals in a population, finding an optimal individual by using a genetic algorithm, and realizing the resilience improvement of the target city medical system by using the reinforcement and recovery strategy corresponding to the optimal individual; ， wherein, represents the state of reinforcement, including pre-earthquake bridges whether reinforced, pre-earthquake hospitals whether reinforced at the structural level and pre-earthquake hospitals whether reinforced at the non-structural level; represents that the status is being repaired, including the scenario lower bridge whether being repaired and the scenario within the time period lower hospital whether being repaired within the time period lower hospital whether being repaired within the time period represents a completed repair status, including scenario lower bridge whether repair is completed within a time period and scenario lower hospital whether repair is completed within a time period and scenario a set of post-earthquake uncertainty scenarios for different reinforcement strategies for Monte Carlo simulation, a set of representative damage scenarios; representing an optimization objective expected under a plurality of uncertain scenarios, for an optimization objective weighted in a scenario under the scenario; The scenario The optimization objective with the lower weight The expression is: ， wherein is an optimization objective based on recovery time in a scenario wherein is an optimization objective based on performance improvement in a scenario wherein and are weight coefficients for and respectively, ， ， To reinforce the weighted recovery time of the urban healthcare system after the main shock, To reinforce the weighted recovery time of the urban healthcare system after the main shock, To reinforce the weighted recovery time of the urban healthcare system after the main shock, To reinforce the functional surplus of the urban healthcare system after the main shock, To reinforce the functional surplus of the urban healthcare system after the main shock, To reinforce the functional surplus of the urban healthcare system after the main shock, To reinforce the functional surplus of the urban healthcare system after the main shock, To reinforce the functional surplus of the urban healthcare system after the main shock.

2. The method of claim 1, wherein, the target function expression of the genetic algorithm is as follows: The method comprises the following steps: obtaining road information according to GIS data and remote sensing image information in the target city, abstracting the traffic facilities and their connection relationship in the road information into a topological structure, and establishing a post-earthquake traffic network topological model; evaluating the demand for medical services under the post-disaster situation according to the population distribution information, and constructing a medical demand model; integrating the vulnerability data of the non-structure and medical equipment related disasters according to the structure type, construction year, location of each department and internal medical equipment configuration of the monomer hospital in the target city, and establishing a post-earthquake function quantification model of the monomer hospital; 3. The method for enhancing the collaborative resilience of the entire urban healthcare system according to claim 1, characterized in that, The scenario The desired optimization goal Healthcare accessibility obtained by the proxy model calculation, The post-earthquake traffic network topological model, the medical demand model and the post-earthquake function quantification model of the monomer hospital jointly constitute the city medical system model.

4. The method of claim 3, wherein, The input of the surrogate model is the post-earthquake damage state of the bridge in the target area under the post-earthquake damage scenario and the available number of each medical department in the hospital, and the output is the post-earthquake medical accessibility. The method for obtaining the post-earthquake damage state of the bridge in the target area under the post-earthquake damage scenario and the available number of each medical department in the hospital is as follows:

5. The method for enhancing the collaborative resilience of the entire process of an urban medical system according to claim 1, characterized in that, simulating the post-earthquake damage scenarios in the city medical system model by using a Monte Carlo simulation method, and evaluating the function reduction of the bridge, the available number of the hospital department and the repair time of the bridge and the hospital under each post-earthquake damage scenario.

6. The method of claim 5, wherein the method is a method of improving the resilience of a whole process of a city medical system, and the method comprises the following steps: The constraints of the target function include a pre-earthquake reinforcement strategy constraint, a cost constraint and a repair resource constraint. ​ ， wherein, represents a pre-earthquake hospital reinforcement strategy selection variable, , is a collection of individual hospitals within a city's healthcare system, represents a pre-earthquake hospital a binary variable of whether the structure is reinforced and has , represents a pre-earthquake hospital a binary variable of whether the non-structure is reinforced and has .

7. The method of claim 6, wherein the method is a method of improving the resilience of a whole process of a city medical system, and the method comprises the following steps: The pre-earthquake reinforcement strategy constraint includes: ​ ， ， wherein, is the bridge cost of retrofit needed, is the pre-earthquake bridge a binary variable whether the bridge was retrofitted or not and has , is the cost of retrofit investment for the transportation system, is the hospital cost of structural retrofit needed, is the hospital cost of non-structural retrofit needed, is the cost of retrofit investment for the medical system, , is the set of bridges within the city medical system.

8. The method of claim 7, wherein the method is a method of improving the resilience of a whole process of a city medical system, and the method comprises the following steps: The cost constraint includes: The resource constraint includes: ​ ， ， wherein, represents a bridge is being repaired during a time period, is being repaired during a time period, represents a hospital is being repaired during a time period, is being repaired during a time period, and are the number of repair teams for bridges and hospitals, respectively, , is the set of long-term repair time periods after the earthquake.

Images