Prevention and control strategy prediction method and device, equipment and storage medium

Abstract

The application discloses a kind of prevention and control strategy prediction method, device, equipment and storage medium, prevention and control strategy prediction method is applied to public transport, prevention and control strategy prediction method includes: obtaining public transport failure prediction data to be analyzed, wherein the data to be analyzed includes: environmental data and station configuration data, according to environmental data and station configuration data, prediction model is constructed, and failure prediction model is obtained, to the preset evacuation strategy optimization model is dual conversion to failure prediction model, and target evacuation model is obtained, according to the preset cut plane rule, target evacuation model is solved and handled, and target evacuation strategy is obtained.The application can predict the failure occurred in public transport, and the corresponding prevention and control strategy is obtained according to the failure.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, and storage medium for predicting prevention and control strategies. Background Technology

[0002] Railway malfunctions are inherently unpredictable, and once they occur, they can cause enormous damage. If personnel are not evacuated in a timely manner, they may even result in casualties. Therefore, how to develop appropriate prevention and control strategies after predicting malfunctions is an urgent problem to be solved. Summary of the Invention

[0003] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a prevention and control strategy prediction method, which can predict faults occurring in public transportation and derive corresponding prevention and control strategies based on the faults.

[0004] The present invention also proposes a prevention and control strategy prediction device.

[0005] This invention also proposes a prevention and control strategy prediction device.

[0006] The present invention also proposes a computer-readable storage medium.

[0007] In a first aspect, one embodiment of the present invention provides a prevention and control strategy prediction method applied to public transportation, the prevention and control strategy prediction method comprising:

[0008] Acquire the data to be analyzed for public transportation fault prediction; wherein the data to be analyzed includes: environmental data and station configuration data;

[0009] A fault prediction model is constructed based on the environmental data and the station configuration data.

[0010] The fault prediction model and the preset evacuation strategy optimization model are dualized to obtain the target evacuation model.

[0011] The target evacuation model is solved according to the preset cutting plane rules to obtain the target evacuation strategy.

[0012] The prevention and control strategy prediction method of this invention has at least the following beneficial effects: It acquires data to be analyzed for public transportation fault prediction, obtains environmental data and station configuration data to be analyzed, constructs a model for fault prediction based on the environmental data and station configuration data, obtains a fault prediction model, uses the prediction results of the fault prediction model as a set, inputs the set into a preset evacuation strategy optimization model for dual transformation, obtains a target evacuation model, and solves the target evacuation model according to preset cutting plane rules to obtain a target evacuation strategy. By constructing a target evacuation model and solving it according to preset cutting plane rules to obtain a target evacuation strategy, it is possible to predict faults occurring in public transportation and derive corresponding prevention and control strategies based on the faults.

[0013] According to other embodiments of the prevention and control strategy prediction method of the present invention, the environmental data includes rainfall data and rainfall location data, the station configuration data includes flood control configuration data and station location data, the fault prediction model includes a platform flooding prediction model, and the step of constructing a prediction model based on the environmental data and the station configuration data to obtain a fault prediction model includes:

[0014] A flood prediction model is constructed based on the rainfall data and the rainfall location data to obtain a flood disaster prediction model.

[0015] Based on the flood control configuration data, the station location data, and the flood disaster prediction model, an inflow model is constructed to obtain the platform inflow prediction model.

[0016] The prevention and control strategy prediction method according to other embodiments of the present invention further includes:

[0017] The evacuation strategy optimization model is constructed, specifically including:

[0018] Obtain sample data of evacuation strategies for model training;

[0019] The evacuation strategy optimization model is obtained by constructing a model based on the preset robust optimization rules and the sample data of the evacuation strategy.

[0020] According to other embodiments of the present invention, the method for predicting prevention and control strategies includes a target evacuation model comprising an upper-level evacuation model and a lower-level evacuation model. The step of solving the target evacuation model according to a preset cutting plane rule to obtain the target evacuation strategy includes:

[0021] The lower-level evacuation model is solved based on preset index data to obtain an initial evacuation strategy;

[0022] The upper-level evacuation model is iteratively solved according to the evacuation strategy to obtain the target evacuation strategy.

[0023] According to other embodiments of the prevention and control strategy prediction method of the present invention, the step of solving the lower-level evacuation model based on preset index data to obtain an initial evacuation strategy includes:

[0024] The lower-level evacuation model is solved based on the index data and the preset lower limit data to obtain the initial evacuation strategy; wherein the lower limit data is obtained by acquiring the pre-initialized lower limit parameters.

[0025] According to another embodiment of the prevention and control strategy prediction method of the present invention, the step of iteratively solving the upper-level evacuation model based on the initial evacuation strategy to obtain the target evacuation strategy includes:

[0026] The upper-level evacuation model is solved based on the initial evacuation strategy to obtain candidate evacuation strategies;

[0027] The candidate evacuation strategies are subjected to strategy metric calculation to obtain strategy metric data;

[0028] The upper-level evacuation model is solved based on the policy metric data until the policy metric data converges, thus obtaining the target evacuation policy.

[0029] According to other embodiments of the prevention and control strategy prediction method of the present invention, the step of performing strategy measurement calculation on the candidate evacuation strategies to obtain strategy measurement data includes:

[0030] The difference between the solution lower limit data and the candidate evacuation strategy is calculated to obtain the strategy metric data.

[0031] Secondly, one embodiment of the present invention provides a prevention and control strategy prediction device for use in public transportation, the prevention and control strategy prediction device comprising:

[0032] The data acquisition module is used to acquire the data to be analyzed for public transportation fault prediction; wherein the data to be analyzed includes: environmental data and station configuration data;

[0033] The prediction model building module is used to build a prediction model based on the environmental data and the station configuration data to obtain a fault prediction model.

[0034] The dual transformation module is used to perform a dual transformation between the fault prediction model and the preset evacuation strategy optimization model to obtain the target evacuation model.

[0035] The strategy solving module is used to solve the target evacuation model according to the preset cutting plane rules to obtain the target evacuation strategy.

[0036] The prevention and control strategy prediction device of this invention has at least the following beneficial effects: The data acquisition module acquires the data to be analyzed for public transportation fault prediction, obtaining the environmental data and station configuration data to be analyzed. The prediction model construction module constructs a model for predicting faults based on the environmental data and station configuration data, obtaining a fault prediction model. The prediction results of the fault prediction model are used as a set. The dual transformation module inputs the set into a preset evacuation strategy optimization model for dual transformation, obtaining a target evacuation model. The strategy solving module solves the target evacuation model according to preset cutting plane rules, obtaining a target evacuation strategy. By constructing a target evacuation model and solving it according to preset cutting plane rules to obtain a target evacuation strategy, faults occurring in public transportation can be predicted, and corresponding prevention and control strategies can be derived based on the faults.

[0037] Thirdly, one embodiment of the present invention provides a prevention and control strategy prediction device, comprising:

[0038] At least one processor, and,

[0039] A memory communicatively connected to the at least one processor; wherein,

[0040] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the prevention and control strategy prediction method as described in the first aspect.

[0041] Fourthly, one embodiment of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer-executable instructions for causing a computer to perform the prevention and control strategy prediction method as described in the first aspect.

[0042] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description and the accompanying drawings. Attached Figure Description

[0043] Figure 1 This is a schematic flowchart of a specific embodiment of the prevention and control strategy prediction method in this invention.

[0044] Figure 2 yes Figure 1 A schematic flowchart of a specific embodiment of step S102;

[0045] Figure 3 This is a schematic flowchart of another specific embodiment of the prevention and control strategy prediction method in this invention;

[0046] Figure 4 yes Figure 1 A schematic flowchart of a specific embodiment of step S104;

[0047] Figure 5 yes Figure 4 A schematic flowchart of a specific embodiment of step S401;

[0048] Figure 6 yes Figure 4 A schematic flowchart of a specific embodiment of step S402;

[0049] Figure 7 yes Figure 6 A schematic diagram of a specific embodiment of step S602;

[0050] Figure 8 This is a block diagram of a specific embodiment of the prevention and control strategy prediction device in this invention.

[0051] Explanation of reference numerals in the attached figures:

[0052] Data acquisition module 801, prediction model construction module 802, dual transformation module 803, strategy solving module 804. Detailed Implementation

[0053] The following will describe the concept and technical effects of the present invention clearly and completely with reference to embodiments, so as to fully understand the purpose, features and effects of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention.

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0055] It should be noted that although the system diagram shows functional modules and the flowchart shows the logical order, in some cases, the steps shown or described may be executed in a different order than the module division in the system or the order in the flowchart.

[0056] In the description of this invention, unless otherwise explicitly defined, terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0057] In the description of the embodiments of the present invention, the term "several" means one or more, and the term "multiple" means two or more. The terms "greater than," "less than," and "exceeding" should be understood as excluding the stated number, while the terms "above," "below," and "within" should be understood as including the stated number. The terms "first" and "second" should be understood as distinguishing technical features, and not as indicating or implying relative importance, the number of indicated technical features, or the order of the indicated technical features.

[0058] First, let's analyze some of the terms used in this application:

[0059] Robust optimization: a field of optimization theory that deals with uncertain optimization problems by seeking robust measures. Uncertainty mainly includes the uncertainty of model parameters or the problem itself, while robust measures represent the best outcome among all uncertainties, i.e., the min-max problem.

[0060] Mixed-integer linear programming: Mixed-integer linear programming is a linear programming problem that simultaneously involves continuous and integer variables. This problem is typically NP-complete, meaning it cannot be solved in polynomial time. However, it has been studied extensively, and readily available solvers (such as CPLEX) can be used to obtain relatively accurate solutions (with errors less than a given value) within a specified time.

[0061] Benders decomposition algorithm: The Benders decomposition algorithm is used to solve mixed integer programming problems. The main idea of ​​this algorithm is to fix the complex variables and solve the optimization problem (subproblem) that only includes simple variables. Then, new constraints are generated in the original optimization problem (main problem) and solved. The new solution is then substituted into the subproblem for a new round of solving. Through continuous iteration, it eventually converges to the optimal solution.

[0062] Cutting plane method: The cutting plane method is a method for solving integer programming problems. Its principle is to continuously increase the number of cutting planes, making the relaxed linear optimization problem increasingly closer to the solution of the original integer programming problem. The cutting plane method is an important component of the Benders decomposition algorithm.

[0063] With the progress of modern urbanization and the booming development of the rail transit industry, while creating significant value for society, certain safety hazards also exist. If safety plans are prepared in advance before a disaster (or malfunction), and if rapid and orderly evacuation of people is implemented during a disaster, and if proper post-disaster recovery efforts are carried out and orderly operations restored promptly, casualties and economic losses caused by disasters can be reduced. Railway malfunctions are inherently unpredictable, and once they occur, they can cause enormous damage. Delayed evacuation can also lead to casualties. Therefore, developing appropriate prevention and control strategies after a malfunction is predicted is a pressing issue that needs to be addressed.

[0064] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a prevention and control strategy prediction method, which can predict faults occurring in public transportation and derive corresponding prevention and control strategies based on the faults.

[0065] Please refer to Figure 1 , Figure 1 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, the prevention and control strategy prediction method is applied to public transportation, and the method specifically includes, but is not limited to, steps S101 to S104.

[0066] Step S101: Obtain the data to be analyzed for public transportation fault prediction; the data to be analyzed includes: environmental data and station configuration data;

[0067] Step S102: Construct a prediction model based on environmental data and station configuration data to obtain a fault prediction model;

[0068] Step S103: Perform dual transformation between the fault prediction model and the preset evacuation strategy optimization model to obtain the target evacuation model.

[0069] Step S104: Solve the target evacuation model according to the preset cutting plane rules to obtain the target evacuation strategy.

[0070] In steps S101 to S104 of this embodiment, data to be analyzed for public transportation fault prediction is obtained, including environmental data and station configuration data. A fault prediction model is constructed based on the environmental and station configuration data. The prediction results from the fault prediction model are used as a set, and this set is input into a preset evacuation strategy optimization model for dual transformation, resulting in a target evacuation model. The target evacuation model is solved according to preset cutting plane rules to obtain a target evacuation strategy. By constructing a target evacuation model and solving it according to preset cutting plane rules to obtain a target evacuation strategy, faults occurring in public transportation can be predicted, and corresponding prevention and control strategies can be derived based on the faults.

[0071] In some embodiments, before step S101, a two-stage optimization model is constructed. The optimization model is the overall model framework before the fault occurs and serves as the evacuation strategy optimization model within the model framework. Let y represent the decision variable for pre-emptive train shutdown during normal operation. The two-stage optimization model is shown in the following formula (1):

[0072]

[0073] (1)

[0075] Formula (1) is the minimax objective function set based on robust optimization theory. In the objective function, function f is the objective function for the second stage, where f is related to the optimized commuter flow result of the emergency route after the fault. The optimal value of y is limited to its feasible set Y under the preset constraints. Here, y represents the decision of whether the platform should be shut down, and is the decision variable for the first stage; the feasible set... ρ represents the set of all feasible values ​​that y can take; ρ represents whether a certain station is flooded. This represents the uncertainty set regarding whether any of the platforms will be flooded.

[0076] Construct a relational model of the decision variables regarding whether to shut down, as shown in formula (2) below:

[0077]

[0078] In formula (2), it can be understood that if platform j stops operating at time t-1, then platform j will also stop operating at time t. For example, at time t0, the disaster has just begun, but the decision on whether to stop operating has already been made, that is, the value of y has been determined. At time t>t0, ρ i, This indicates whether platform i is flooded at time t, so y is not affected by ρ. i, Influence.

[0079] In step S101 of some embodiments, referring to Table 1, the data to be analyzed also includes the data in Table 1 below, and the data includes: sets, parameters and variables. The sets, parameters and variables in Table 1 are explained as follows.

[0080] Table 1

[0081]

[0082]

[0083]

[0084] Please refer to Figure 2 , Figure 2 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, environmental data includes rainfall data and rainfall location data, station configuration data includes flood control configuration data and station location data, fault prediction model includes platform flooding prediction model, and step S102 specifically includes, but is not limited to, steps S201 to S202.

[0085] Step S201: Construct a flood prediction model based on rainfall data and rainfall location data to obtain a flood disaster prediction model;

[0086] Step S202: Based on the flood control configuration data, station location data, and flood disaster prediction model, an inflow model is constructed to obtain the platform inflow prediction model.

[0087] In steps S201 to S202 of this embodiment, a flood prediction model is constructed based on rainfall data and rainfall location data to obtain a flood disaster prediction model. The data to be analyzed is input into the flood disaster prediction model for prediction, and the predicted data is obtained. An inundation model is constructed based on flood control configuration data, station location data, and the data predicted by the flood disaster prediction model to obtain a platform inundation prediction model. The flood disaster prediction model constructed using rainfall data and rainfall location data can be used to predict flood disasters. The platform inundation prediction model constructed based on flood control configuration data, station location data, and the flood disaster prediction model can predict the degree of platform inundation failure.

[0088] In step S201 of some embodiments, in order to better characterize flood disasters under uncertain scenarios, the following modeling is required: Since rainfall is generally regional, a parameter z with a value of 0 or 1 is introduced. r,t This represents whether rainfall is ongoing in region r at time t. If rainfall is ongoing in region r at time t, then z... r,t If z is set to 1, and there is no rainfall in region r at time t, then z r,t Take 0. Through z r,t The magnitude of rainfall can be obtained, for example, the maximum number of areas experiencing continuous rainfall simultaneously within the same time period (such as a certain quarter). The flood disaster prediction model is modeled as follows: (3)

[0089]

[0090] In formula (3), γ represents the area with the maximum number of rainfall events, and Γ t Let represent the set of all possible scenarios with faults at time t. This refers to the total time period. Since the flooding of the platform is a continuous process, therefore... This represents all times from the initial water inflow time t0 to the final time T, with a specific time t representing a particular moment. From another perspective, the logic of the flood disaster prediction model can be used to model the maximum duration of rainfall in each region, as shown in the following formula (4):

[0091]

[0092] Therefore, by introducing the parameter z r, It can depict floods of different magnitudes in different scenarios during different seasons.

[0093] In step S202 of some embodiments, a parameter ρ is further introduced to describe the state of the platform, ρ i, That is, whether platform i is flooded at time t. ρ i,t The modeling is as follows: (5)

[0094]

[0095] In formula (5), the parameter δ i, This describes whether platform i is within region r, with parameter δ. i, The value of δ is 0 or 1. If platform i is within region r, then δ i, If platform i is not within region r, then δ is set to 1. i, Set the parameter h to 0. r, This describes the rainfall intensity at time t in region r. Δ i The minimum flood resistance of platform i is related to its construction standards, structural features, and materials used. Therefore, ρ i, The model states that if the rainfall received by station i exceeds the minimum flood control capacity, then station i loses its ability to operate normally at time t, i.e., ρ takes the value of 1; otherwise, ρ takes the value of 1 and continues to operate normally.

[0096] (6)

[0098] Formula (6) constructs an uncertain set of ρ. Here, M represents a very large positive number; for example, for ease of calculation, M can be set to 1000 or 10000. In Formula (6) All {ρ∈{0,1} that simultaneously satisfy the two inequality constraints on the right-hand side. N* The result of this is that these two inequality constraints are equivalent transformations of formula (5) (restricting ρ to a 0-1 variable).

[0099] Please refer to Figure 3 , Figure 3A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, the prevention and control strategy prediction method further includes constructing an evacuation strategy optimization model, which includes, but is not limited to, steps S301 to S302.

[0100] Step S301: Obtain evacuation strategy sample data for model training;

[0101] Step S302: Based on the preset robust optimization rules and evacuation strategy sample data, a model is constructed to obtain the evacuation strategy optimization model.

[0102] In steps S301 to S302 of this embodiment, sample data required for model training is obtained, evacuation strategy sample data is acquired, an initial model is constructed according to preset robust optimization rules, and the evacuation strategy sample data is input into the initial model for training to obtain an optimized evacuation strategy model. By constructing the initial model using robust optimization rules and training it with the acquired evacuation strategy sample data, an optimized evacuation strategy model is obtained, thus constructing an optimized evacuation strategy model capable of deriving evacuation strategies from the data to be tested.

[0103] It should be noted that the optimal evacuation plan f(y,ρ) can be obtained by solving the following linear programming evacuation strategy optimization model, where ω represents a certain OD pair; Ω represents the set of all OD pairs; eω) represents the destination of a certain OD pair; and sω) represents the starting point of a certain OD pair. Let represent the set of platforms where passengers can transfer by walking. The evacuation strategy optimization model is shown in the following formulas (7a) to (7j):

[0104]

[0105] st

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] q,l,a,b,d,o≥0

[0115] (7j)

[0116] The objective function in formula (7a) is to maximize the outbound passenger flow after the accident. Minimizing negative passenger flow is used here to facilitate finding the dual problem later. Formula (7b) represents the maximum number of passengers exiting each platform at each moment after the accident. Formula (7c) states that the number of passengers exiting cannot exceed the platform capacity. Formula (7d) represents the capacity limit for train passenger capacity. Formula (7e) states that passengers cannot queue at flooded platforms. Formulas (7f) to (7j) represent the passenger movement logic within the system.

[0117] In step S103 of some embodiments, the overall model in formula (1) is actually a three-level optimization model of maximum-min-maximum (max-min-max model).

[0118] First, the evacuation strategy optimization model is transformed into a target evacuation model using the duality theory of linear models. The target evacuation model is shown in equations (8a) to (8k) below:

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130] The target evacuation model is derived from the evacuation strategy optimization model through dual transformation; the two are completely equivalent. In formulas (8a) to (8b), g represents the dual variable, and the superscripts of g (7b) to (7i) correspond to (7b) to (7i) of the evacuation strategy optimization model, respectively. (8b), (8c), (8d), (8e), and (8f) are respectively about The constraints; (8g) and (8h) are about The constraints; (8i) and (8j) are the constraints at t=T regarding and The constraints. The objective evacuation model is a linear programming problem, which can be solved using standard solvers (such as CPLEX).

[0131] Please refer to Figure 4 , Figure 4 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, the target evacuation model includes an upper-level evacuation model and a lower-level evacuation model. The target evacuation model is solved according to a preset cutting plane rule to obtain the target evacuation strategy, including but not limited to steps S401 to S402.

[0132] Step S401: Solve the upper-level evacuation model based on the preset index data to obtain the initial evacuation strategy;

[0133] Step S402: Iteratively solve the lower-level evacuation model according to the initial evacuation strategy to obtain the target evacuation strategy.

[0134] In steps S401 to S402 of this embodiment, preset indicator data is obtained. This indicator data and data from when a fault occurs are input into the upper-level evacuation model of the target evacuation model. The upper-level evacuation model is solved to obtain an initial evacuation strategy. This initial evacuation strategy is then input into the lower-level evacuation model of the target evacuation model, and the lower-level evacuation model is iteratively solved to obtain the target evacuation strategy. By inputting indicator data into the upper-level evacuation model to obtain the initial evacuation strategy, and then inputting the initial evacuation strategy into the lower-level evacuation model to obtain the target evacuation strategy, corresponding prevention and control strategies can be derived based on faults occurring in public transportation.

[0135] It should be noted that the upper-level evacuation model is a max-min (robust optimization framework), while the lower-level evacuation model is a max. Such models are difficult to handle; therefore, duality theory is used to transform the lower-level max problem into a min problem, and further into a single-level min problem, thus simplifying the processing. The solution f(y,ρ) of the lower-level evacuation model can be obtained through the target evacuation model. Substituting the solution of the lower-level evacuation model into the upper-level evacuation model, the solution of the overall model can be obtained. The lower-level evacuation model corresponds to the evacuation strategy optimization model.

[0136] Please refer to Figure 5 , Figure 5 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, the upper-level evacuation model is solved based on preset index data to obtain an initial evacuation strategy, including but not limited to step S501.

[0137] Step S501: Solve the lower-level evacuation model based on the index data and the preset solution lower limit data to obtain the initial evacuation strategy; wherein, the solution lower limit data is obtained by acquiring the pre-initialized lower limit parameters.

[0138] In step S501 of some embodiments, pre-initialized lower bound parameters are obtained to get solution lower bound data. The index data and solution lower bound data are input into the lower-level evacuation model, and the lower-level evacuation model is solved to obtain the initial evacuation strategy. By solving the lower-level evacuation model using the index data and solution lower bound data, the result of the first iteration of the lower-level evacuation model can be obtained.

[0139] It should be noted that formula (1) needs to be appropriately transformed. First, (8a) can be written as φ(y,ρ)=β 0 g+β 1 gy+β 2 gρ, where β 2 ≤0,β 0 andβ 1 `free` represents the coefficient vector in the objective function at the time of the fault. The feasible dual solution set of expressions (8b) to (8l) is expressed, and let Let represent the feasible solution set of formula (2). Then formula (1) can be transformed into the following formula (9):

[0140]

[0141] Please refer to Figure 6 , Figure 6 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, the upper-level evacuation model is iteratively solved based on the initial evacuation strategy to obtain the target evacuation strategy, including but not limited to steps S601 to S603.

[0142] Step S601: Solve the upper-level evacuation model according to the initial evacuation strategy to obtain candidate evacuation strategies;

[0143] Step S602: Calculate the strategy metric for the candidate evacuation strategy to obtain strategy metric data;

[0144] Step S603: Solve the upper-level evacuation model based on the strategy measurement data until the strategy measurement data converges to obtain the target evacuation strategy.

[0145] In steps S601 to S603 of this embodiment, the initial evacuation strategy is input into the upper-level evacuation model, and the upper-level evacuation model is solved to obtain candidate evacuation strategies. Strategy metrics are calculated on the candidate evacuation strategies to obtain strategy metric data. It is then determined whether the strategy metric data has converged. If the strategy metric data indicates non-convergence, the strategy metric data is input into the upper-level evacuation model, and the upper-level evacuation model is solved again until the strategy metric data converges, thus obtaining the target evacuation strategy. By iteratively solving the upper-level evacuation model using the initial evacuation strategy, the target evacuation strategy can be obtained, enabling the development of corresponding prevention and control strategies based on faults occurring in public transportation.

[0146] It should be noted that, Let G denote all the poles in equation (9), since all equations are linear, and let G denote the set of these poles. Then equation (9) can be equivalently transformed into equation (10), and equations (10a) to (10c) are as follows:

[0147] max y α(y)

[0148] (10a)

[0149]

[0150]

[0151] In formulas (10a) to (10c), α(y) represents a constant function of y, where, Represents the dual variable. The main meaning of formulas (10a) to (10c) is to linearize the max-min in formula (9), ensuring that the target value remains unchanged during the transformation.

[0152] Please refer to Figure 7 , Figure 7 A flowchart illustrating the prevention and control strategy prediction method in an embodiment of the present invention is shown. In some embodiments, strategy measurement calculations are performed on candidate evacuation strategies to obtain strategy measurement data, including but not limited to step S701.

[0153] Step S701: The difference between the solution lower limit data and the candidate evacuation strategies is calculated to obtain the strategy measurement data.

[0154] In step S701 of some embodiments, the lower bound data and the candidate evacuation strategy are subtracted to obtain the difference between the lower bound data and the candidate evacuation strategy, thus obtaining the strategy metric data. The strategy metric data obtained by solving the lower bound data and the candidate evacuation strategy is used to determine whether the strategy metric data has converged and can be used to obtain the optimized target evacuation strategy.

[0155] In some embodiments, for example, let LB represent the lower bound, UB represent the upper bound, m represent the iteration count of the algorithm, and use a certain positive constant ε>0 as the optimal solution gap.

[0156] definition This includes all indicator data across all disaster scenarios, where each indicator... Corresponding to one The combination, and G is The feasible set, Γ is ∈ l The feasible set. Since flooding may cause water to enter different platforms, there are many different disaster scenarios. Therefore, all disaster scenarios can be defined as... Each metric Indicates the first A disaster scenario. Therefore Indicates the first The values ​​of dual variables and the optimal gap size under a disaster scenario. For example, defining a subset of readily available indicator data. Some feasible solutions y are obtained by solving formulas (8a) to (8k). Initialization Let m denote the effective lower bound of any one of formulas (8a) to (8k), and let the iteration count m←1.

[0157] Solve the following relaxation problem using the following formulas (11a) to (11c):

[0158] max y α(y)

[0159] (11a)

[0160]

[0161]

[0162] To obtain the solution for the first stage in the m-th iteration, let y ← y m ,

[0163] Add constraints on ρ, i.e., equation (6), to equations (8a) to (8k) of the subproblem, and solve equations (8a) to (8k). The resulting optimal dual solution and worst-case failure scenario are used... Indicates. Update in Indicates from To the indicator space The mapping. If Then update And y * ←ym , where y * It stores the best solution generated so far.

[0164] If UB-LB≤∈, then stop iterating and return the optimal solution with gap ∈. Otherwise, continue iterating, let m←m+1, and return to solving the relaxation problem.

[0165] Since the set of all flood scenarios Γ is a finite set, it takes at most a finite number of steps to generate the optimal cutting plane for formulas (11a) to (11c). In general, the cutting plane algorithm can generate the optimal solution for formula (1) in at most N*T iterations.

[0166] In addition, this application also discloses a prevention and control strategy prediction device, please refer to... Figure 8 , Figure 8 This invention discloses a block diagram of a prevention and control strategy prediction device according to an embodiment of the present invention. Applied to public transportation, it can implement the aforementioned prevention and control strategy prediction method. The prevention and control strategy prediction device includes: a data acquisition module 801, a prediction model construction module 802, a dual transformation module 803, and a strategy solving module 804. The data acquisition module 801, prediction model construction module 802, dual transformation module 803, and strategy solving module 804 are all communicatively connected.

[0167] The data acquisition module 801 acquires the data to be analyzed for public transportation fault prediction; this data includes environmental data and station configuration data. The prediction model construction module 802 constructs a prediction model based on the environmental data and station configuration data to obtain a fault prediction model. The dual transformation module 803 performs a dual transformation between the fault prediction model and a preset evacuation strategy optimization model to obtain a target evacuation model. The strategy solution module 804 solves the target evacuation model according to preset cutting plane rules to obtain the target evacuation strategy.

[0168] The data acquisition module 801 acquires the data to be analyzed for public transportation fault prediction, obtaining the environmental data and station configuration data to be analyzed. The prediction model construction module 802 constructs a model for predicting faults based on the environmental data and station configuration data, obtaining a fault prediction model. Using the prediction results from the fault prediction model as a set, the dual transformation module 803 inputs the set into a preset evacuation strategy optimization model for dual transformation, obtaining a target evacuation model. The strategy solving module 804 solves the target evacuation model according to preset cutting plane rules, obtaining a target evacuation strategy. By constructing the target evacuation model and solving it according to preset cutting plane rules to obtain the target evacuation strategy, faults occurring in public transportation can be predicted, and corresponding prevention and control strategies can be derived based on the faults.

[0169] The operation process of the prevention and control strategy prediction device in this embodiment is specifically described above. Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 The steps S101 to S104, S201 and S202, S301 and S302, S401 and S402, S501, S601 to S603 and S701 of the prevention and control strategy prediction method are not described in detail here.

[0170] Another embodiment of the present invention discloses a prevention and control strategy prediction device, comprising: at least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform, for example... Figure 1 Control method steps S101 to S104 Figure 2 Control method steps S201 and S202 Figure 3 Control method steps S301 and S302 Figure 4 Control method steps S401 and S402 Figure 5 Control method steps S501 Figure 6 The control method steps S601 to S603 and Figure 7 The control method step S701 in the control method is the prevention and control strategy prediction method.

[0171] Another embodiment of the present invention discloses a storage medium, the storage medium comprising: storing computer-executable instructions for causing a computer to perform... Figure 1 Control method steps S101 to S104 Figure 2 Control method steps S201 and S202 Figure 3 Control method steps S301 and S302 Figure 4 Control method steps S401 and S402 Figure 5 Control method steps S501 Figure 6 The control method steps S601 to S603 and Figure 7 The control method step S701 in the control method is the prevention and control strategy prediction method.

[0172] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0173] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0174] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. Furthermore, the embodiments of the present invention and the features thereof can be combined with each other unless otherwise specified.

Claims

1. A method for predicting prevention and control strategies, characterized in that, Applied to public transportation, the prevention and control strategy prediction method includes: Acquire the data to be analyzed for public transportation fault prediction; wherein the data to be analyzed includes: environmental data and station configuration data; A fault prediction model is constructed based on the environmental data and the station configuration data. The fault prediction model and the preset evacuation strategy optimization model are dualized to obtain the target evacuation model. The construction of the evacuation strategy optimization model includes the following steps: obtaining evacuation strategy sample data for model training; constructing the model according to the preset robust optimization rules and evacuation strategy sample data to obtain the evacuation strategy optimization model. The target evacuation model includes an upper-level evacuation model and a lower-level evacuation model. The upper-level evacuation model is max-min, and the lower-level evacuation model is max. The target evacuation model is solved according to a preset cutting plane rule to obtain a target evacuation strategy, including: solving the lower-level evacuation model according to preset index data to obtain an initial evacuation strategy; and iteratively solving the upper-level evacuation model according to the initial evacuation strategy to obtain the target evacuation strategy. The process of performing a dual transformation between the fault prediction model and the preset evacuation strategy optimization model to obtain the target evacuation model includes: The results predicted by the fault prediction model are used as a set. The set is input into the evacuation strategy optimization model and the duality theory of the linear model is used to transform the evacuation strategy optimization model into the target evacuation model. The duality theory is used to transform the lower-level max problem into a min problem, and then the overall model is transformed into a single-level max problem.

2. The prevention and control strategy prediction method according to claim 1, characterized in that, The environmental data includes rainfall data and rainfall location data; the station configuration data includes flood control configuration data and station location data; the fault prediction model includes a platform flooding prediction model; the construction of the prediction model based on the environmental data and station configuration data to obtain the fault prediction model includes: A flood prediction model is constructed based on the rainfall data and the rainfall location data to obtain a flood disaster prediction model. Based on the flood control configuration data, the station location data, and the flood disaster prediction model, an inflow model is constructed to obtain the platform inflow prediction model.

3. The prevention and control strategy prediction method according to claim 1, characterized in that, Also includes: The evacuation strategy optimization model is constructed, specifically including: Obtain sample data of evacuation strategies for model training; The evacuation strategy optimization model is obtained by constructing a model based on the preset robust optimization rules and the sample data of the evacuation strategy.

4. The prevention and control strategy prediction method according to any one of claims 1 to 3, characterized in that, The step of solving the lower-level evacuation model based on preset index data to obtain an initial evacuation strategy includes: The lower-level evacuation model is solved based on the index data and the preset lower limit data to obtain the initial evacuation strategy; wherein the lower limit data is obtained by acquiring the pre-initialized lower limit parameters.

5. The prevention and control strategy prediction method according to claim 4, characterized in that, The step of iteratively solving the upper-level evacuation model based on the initial evacuation strategy to obtain the target evacuation strategy includes: The upper-level evacuation model is solved based on the initial evacuation strategy to obtain candidate evacuation strategies; The candidate evacuation strategies are subjected to strategy metric calculation to obtain strategy metric data; The upper-level evacuation model is solved based on the policy metric data until the policy metric data converges, thus obtaining the target evacuation policy.

6. The prevention and control strategy prediction method according to claim 5, characterized in that, The step of performing policy metric calculations on the candidate evacuation strategies to obtain policy metric data includes: The difference between the solution lower limit data and the candidate evacuation strategy is calculated to obtain the strategy metric data.

7. A prevention and control strategy prediction device, characterized in that, The prevention and control strategy prediction device, applied to public transportation, includes: The data acquisition module is used to acquire the data to be analyzed for public transportation fault prediction; wherein the data to be analyzed includes: environmental data and station configuration data; The prediction model building module is used to build a prediction model based on the environmental data and the station configuration data to obtain a fault prediction model. The dual transformation module is used to perform a dual transformation between the fault prediction model and the preset evacuation strategy optimization model to obtain the target evacuation model. The construction of the evacuation strategy optimization model includes the following steps: obtaining evacuation strategy sample data for model training; constructing the model according to the preset robust optimization rules and evacuation strategy sample data to obtain the evacuation strategy optimization model. The target evacuation model includes an upper-level evacuation model and a lower-level evacuation model. The upper-level evacuation model is max-min, and the lower-level evacuation model is max. The strategy solving module is used to solve the target evacuation model according to preset cutting plane rules to obtain the target evacuation strategy, including: solving the lower-level evacuation model according to preset index data to obtain an initial evacuation strategy; and iteratively solving the upper-level evacuation model according to the initial evacuation strategy to obtain the target evacuation strategy. The process of performing a dual transformation between the fault prediction model and the preset evacuation strategy optimization model to obtain the target evacuation model includes: The results predicted by the fault prediction model are used as a set. The set is input into the evacuation strategy optimization model and the duality theory of the linear model is used to transform the evacuation strategy optimization model into the target evacuation model. The duality theory is used to transform the lower-level max problem into a min problem, and then the overall model is transformed into a single-level max problem.

8. A prevention and control strategy prediction device, characterized in that, include: At least one processor, and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the prevention and control strategy prediction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the prevention and control strategy prediction method as described in any one of claims 1 to 6.

Images