A dynamic scheduling method for helicopter rescue based on rolling horizon strategy

Abstract

The present application discloses a dynamic scheduling method for helicopter rescue based on a rolling time domain strategy, which relates to the field of intelligent control technology. The method collects real-time disaster information of the disaster-stricken area; divides the helicopter rescue period according to a preset step size based on the rolling time domain strategy to determine multiple rolling windows; uses the real-time disaster information as the task data of the current rolling window, and updates the prediction model based on the task data of the current rolling window to determine the rescue plan of the current rolling window; executes the current rescue plan and feeds back the current execution result to determine the real-time disaster information after the execution of the current rescue plan; uses the real-time disaster information after the execution of the current rescue plan as the task data of the next rolling window; uses the task data of the next rolling window as the task data of the current rolling window, until all rolling windows are executed to complete the dynamic scheduling of helicopter rescue.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a dynamic scheduling method for helicopter rescue based on a rolling time domain strategy. Background Art

[0002] Existing approaches to helicopter disaster relief dispatching mostly address various aspects of static helicopter rescue dispatching. However, they face unresolved challenges in algorithmic innovation and system-level modeling, as well as adapting to multiple dynamic conditions and integrating complex constraints. Given the dynamic nature of disaster relief dispatching, related technologies often enhance emergency response effectiveness in dynamic disaster scenarios, but they have inherent limitations in dynamic adaptability. This means they struggle to adapt to sudden changes. Therefore, a dynamic helicopter rescue dispatching method is urgently needed to handle real-time, constantly changing dynamic conditions. Summary of the Invention

[0003] The purpose of this application is to provide a dynamic scheduling method, equipment, medium and product for helicopter rescue based on a rolling time domain strategy, which can take into account the event dynamics of disaster rescue scheduling, output rescue plans that are more suitable for the current disaster situation in real time and accurately, and improve the rescue capabilities to adapt to various dynamic conditions and complex constraints in the disaster-stricken area.

[0004] To achieve the above objectives, this application provides the following solutions.

[0005] In the first aspect, the present application provides a dynamic scheduling method for helicopter rescue based on a rolling time domain strategy, including: collecting real-time disaster information of the disaster-stricken area; dividing the helicopter rescue period according to a preset step size based on the rolling time domain strategy to determine multiple rolling windows; using the real-time disaster information as the task data of the current rolling window, and updating the prediction model based on the task data of the current rolling window to determine the rescue plan for the current rolling window; the prediction model is determined based on preset constraints; the rescue plan includes the helicopter rescue path and the number of transported wounded; executing the current rescue plan and feeding back the current execution result to determine the real-time disaster information after the execution of the current rescue plan; using the real-time disaster information after the execution of the current rescue plan as the task data of the next rolling window; using the task data of the next rolling window as the task data of the current rolling window, and returning to the step "updating the prediction model based on the task data of the current rolling window to determine the rescue plan for the current rolling window" until all rolling windows are executed and the dynamic scheduling of helicopter rescue is completed.

[0006] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described dynamic scheduling method for helicopter rescue based on the rolling time domain strategy.

[0007] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described dynamic scheduling method for helicopter rescue based on the rolling time domain strategy.

[0008] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned dynamic scheduling method for helicopter rescue based on the rolling time domain strategy.

[0009] According to the specific embodiments provided in this application, this application discloses the following technical effects.

[0010] The present application first collects real-time disaster information of the disaster-stricken area, and then divides the helicopter rescue period according to a preset step size based on a rolling time domain strategy, determines multiple rolling windows, and uses the real-time disaster information as the task data of the current rolling window to update the prediction model and determine the rescue plan for the current rolling window. The present application can output a rescue plan that is more suitable for the current disaster situation through a prediction model determined based on a rolling time domain strategy and preset constraints. Furthermore, the real-time disaster information after the current rescue plan is used as the task data of the next rolling window to further update the prediction model, and output the rescue plan for the next rolling window until all rolling windows are executed, thereby optimizing the rescue plan and completing the entire dynamic scheduling of helicopter rescue. The present application can take into account the event dynamics of disaster rescue scheduling, and output a rescue plan that is more suitable for the current disaster situation in real time and accurately, thereby improving the rescue capability to adapt to various dynamic conditions and complex constraints in the disaster-stricken area and avoiding the limitations of the rescue plans in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a flow chart of a dynamic scheduling method for helicopter rescue based on a rolling time domain strategy provided in an embodiment of the present application.

[0013] Figure 2 This is an example diagram of a rescue batch in an embodiment of the present application.

[0014] Figure 3 This is a flowchart of dynamic scheduling of helicopter rescue based on rolling time domain in an embodiment of the present application.

[0015] Figure 4This is an example diagram of the rescue scheduling plan before and after the dynamic event in Case 1 in the embodiment of this application.

[0016] Figure 5 This is an example diagram of the rescue scheduling plan before and after the dynamic event in Case 2 in the embodiment of this application.

[0017] Figure 6 This is an example diagram of the rescue scheduling plan before and after the dynamic event in Case 3 in the embodiment of this application.

[0018] Figure 7 This is an example diagram of the rescue scheduling plan before and after the dynamic event in Case 4 in the embodiment of this application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0020] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0021] For the helicopter disaster relief scheduling problem, a related paper proposed a hybrid particle swarm optimization greedy search algorithm for disaster relief helicopter task allocation, integrating helicopter performance parameters and mission characteristics to improve computational efficiency. Their approach, validated with real-world disaster data, demonstrated the effectiveness of optimizing resource management by balancing solution quality and computational speed. The related paper modeled helicopter rescue scheduling as a vehicle routing problem with time windows and employed an improved genetic algorithm to minimize flight distance using an adaptive local-global search strategy. The related paper focused on optimizing helicopter routes in disaster relief, leveraging a vehicle routing approach to improve resource allocation efficiency to maximize the number of rescued victims while controlling operating costs. By formalizing the trade-off between mission coverage and cost, their framework provided a data-driven approach for prioritizing high-impact rescue missions. The related paper studied flood disaster scheduling by developing a multi-objective model that integrates the bilateral fit between helicopter capabilities and mission requirements, as well as mission time satisfaction evaluated via soft time windows. The related paper proposed a hierarchical optimization framework for hybrid helicopter fleet scheduling, decomposing the problem into two stages: task allocation and route planning. This two-stage strategy reduces average scheduling time in static scenarios and highlights the advantages of modular heuristic integration. Ozdamar et al. introduced a planning system that combines mathematical models with route management procedures to coordinate the delivery of supplies and the evacuation of casualties, generating fuel- and capacity-feasible routes from static disaster data as a basic framework for large-scale operations. Related technologies constructed a conceptual model and multi-agent simulation system for flood rescue scheduling, using standardized task completion time as an evaluation metric to quantify the correlation between equipment capabilities and scheduling rules. Their analysis revealed bottlenecks in the efficiency of large helicopters due to loading / unloading limitations, emphasizing the need for adaptive mixed fleet strategies. Overall, these studies cover various aspects of static helicopter rescue scheduling, from algorithmic innovation to system-level modeling, while emphasizing unresolved challenges in adapting to multiple dynamic conditions and integrating complex constraints, thereby promoting progress in this key area.

[0022] Considering the dynamic nature of disaster relief dispatch, a robust optimization model for post-disaster rescue logistics under conditions of demand and transportation time uncertainty has been proposed. By utilizing interval-based uncertainty sets, a traceable deterministic correspondence is derived and validated using the Sichuan earthquake case study to enhance the robustness of resource mobilization and helicopter deployment decisions. A new mathematical model for disaster relief operations based on stochastic programming has been proposed, combining pre-disaster and post-disaster decision-making stages to optimize the transportation network and the allocation of relief supplies. This model considers multiple scenarios and employs a metaheuristic algorithm based on particle swarm optimization to improve the efficiency of rescue operations under uncertainty. A multi-agent reinforcement learning model has been proposed, treating rescue sites as collaborative agents and optimizing task allocation efficiency using a proximal policy optimization algorithm. Compared to proximity-based strategies, this approach achieves higher allocation rates and time satisfaction, effectively addressing the critical mismatch between fluctuating multi-site rescue demand and limited aircraft resources. However, while this framework enhances emergency response effectiveness in dynamic disaster scenarios, it suffers from inherent limitations in terms of dynamic adaptability. Specifically, its difficulty in adapting to sudden changes, such as the expansion of disaster-affected areas or the rapid establishment of temporary hospitals, highlights the need for more robust mechanisms to handle dynamic conditions that are changing in real time.

[0023] The Receding Horizon Strategy (RHS), originally derived from control theory, achieves real-time adaptability to evolving constraints by decomposing a time-varying environment into continuous finite-horizon subproblems, making it a revolutionary approach to solving dynamic scheduling problems. Notably, Yuan et al. proposed an RHS-based dynamic scheduling method for carrier-based aircraft support operations under uncertainty and demonstrated its effectiveness in handling dynamic disturbances in complex aircraft deck environments. Daofang et al. developed a dynamic rolling horizon decision-making strategy for container terminal yard crane scheduling and, through simulation modeling, improved the efficiency and adaptability of container terminal operations. Long et al. introduced an RHS-based emergency scheduling model for spatial surveillance networks. Their hierarchical distributed dynamic emergency scheduling algorithm significantly improved the efficiency and robustness of spatial surveillance network task scheduling under emergencies. Homsi et al. studied the dynamic matching and rematching problem in a ride-sharing system using RHS, proposed an expected value model, and evaluated different matching profit and mismatch penalty functions. This application proposes an RHS-based dynamic scheduling method for helicopter rescue operations.

[0024] like Figure 1 As shown, the present application provides a helicopter rescue dynamic scheduling method based on a rolling time domain strategy, including steps 11 to 16.

[0025] Step 11: Collect real-time disaster information of the disaster-stricken area.

[0026] Among them, real-time disaster information is determined based on real-time dynamic disturbance events.

[0027] In some embodiments, step 11 further includes steps 21 to 23.

[0028] Step 21: Determine whether a dynamic event occurs after collecting real-time disaster information of the disaster-stricken area, and determine a first judgment result; the dynamic event includes at least: the number of newly added hospitals, the conditions for treating the injured in the newly added hospitals, and the number of newly added injured people.

[0029] Step 22: If the first judgment result is yes, update the real-time disaster information of the disaster-stricken area.

[0030] Step 23: If the first judgment result is no, the real-time disaster information of the disaster-stricken area is not updated.

[0031] Step 12: Based on the rolling time domain strategy, the helicopter rescue period is divided according to a preset step size to determine multiple rolling windows.

[0032] In actual applications, the preset step size is determined by real-time dynamic disturbance events, and the real-time disaster information is determined based on dynamic disturbance events; in actual applications, based on the rolling time domain strategy, the helicopter rescue period is divided in combination with dynamic disturbance events, and the rolling window is determined in real time according to dynamic disturbance events.

[0033] Step 13: Using the real-time disaster information as the task data for the current rolling window, and updating the prediction model based on the task data for the current rolling window, determine a rescue plan for the current rolling window; the prediction model is determined based on preset constraints; the rescue plan includes the helicopter rescue route and the number of injured people to be transported;

[0034] In some embodiments, step 13 specifically includes: inputting the task data of the current rolling window into the prediction model, minimizing the objective function of the prediction model as the goal, revising the preset constraints, and determining an updated prediction model; wherein, the real-time disaster information includes the current status of the helicopter, the current status information of the disaster-stricken area, and the operating status information of the hospital; the current status of the helicopter includes at least the helicopter position, speed, remaining fuel, and number of passengers; the current status information of the disaster-stricken area includes at least the number, distribution, and injury level of the wounded; the operating status information of the hospital includes at least the remaining capacity, treatment capacity, and resource reserves of the hospital; and determining the rescue plan for the current rolling window based on the metaheuristic algorithm and the updated prediction model.

[0035] In some embodiments, the objective function of the minimized prediction model is:

[0036] .

[0037] in, represents the weighted waiting time for the rescue of the wounded; A number indicating the affected area; Indicates the helicopter number; Indicates helicopter The transport batch set; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; Indicates helicopter No. transport batches in the disaster-stricken area The number of wounded people loaded requires rescue waiting time, Indicates helicopter No. transport batches in the disaster-stricken area The weight corresponding to the number of wounded loaded.

[0038] In some embodiments, before step 13, it also includes: defining the decision variables of the prediction model; the prediction model is a mixed integer programming model; determining preset constraints based on the decision variables; the preset constraints include disaster area transportation capacity constraints, helicopter capacity constraints, helicopter endurance constraints, helicopter operation timing constraints and hospital rescue capacity constraints; based on the preset constraints, constructing a prediction model.

[0039] In some embodiments, the disaster area transportation capacity constraint is: ;in, Indicates the affected area The total number of injured people in the injured concentration area, Indicates the helicopter number; Indicates helicopter The transport batch set; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; A number indicating the affected area.

[0040] The helicopter capacity constraint is: ;in, Indicates the helicopter type, Represents a set of helicopter types, represents the maximum transport capacity of a K-category helicopter, Indicates helicopter In the bThe total number of casualties transported in each transport batch; z h,k is the matching variable of helicopter type k. If the helicopter h for k type, z h,k is 1 if the value is set, otherwise it is 0.

[0041] The helicopter's endurance constraints are: ;in, Indicates helicopter In the b The air endurance time of each transport batch, Indicates helicopter The remaining flight time when the bth transport batch begins, express The maximum flight time of this type of helicopter.

[0042] Helicopter operation timing constraints: ;in, Indicates helicopter From the affected area Travel to the affected area time consuming, ; ; and Indicates that the helicopter is in the disaster area The time when rescue begins and ends; Indicates helicopter In the b transport batches from the affected areas Travel to the affected area ; Indicates helicopter In the b transport batches to the disaster-stricken areas ; Indicates helicopter In the b transport batches arrived in the disaster-stricken areas moment; Indicates that the helicopter is in the disaster area Loading the wounded takes time; Indicates helicopter From the affected area Travel to the affected area time consuming; Indicates helicopter In the b The set of disaster-stricken areas and hospital nodes to which each transport batch is sent; Indicates that the helicopter is in the disaster area The end of the rescue moment.

[0043] The hospital rescue capacity constraints are: ;in, Indicates hospital Maximum rescue capability; for the collection of hospitals; Indicates helicopter In the b Whether the transport batch goes to the hospital If yes, the indicator variable is 1, otherwise it is 0.

[0044] In some embodiments, in the case of step 22, that is, when the first judgment result is yes, the preset constraint condition also includes a dynamic constraint condition; the dynamic constraint condition is ;in, Indicates helicopter h In the b The duration of flight in the air during each transport batch; Indicates helicopter Start The remaining time of the flight when the transport batch is completed; Indicates the maximum flight time of a K-category helicopter; Indicates the helicopter type, Represents a set of helicopter types; Indicates the helicopter number; Indicates helicopter The transport batch set; represents the set of all helicopters; Indicates helicopter The collection of all shipping batches.

[0045] Step 14: Execute the current rescue plan and feedback the current execution result to determine the real-time disaster information after executing the current rescue plan.

[0046] Step 15: Use the real-time disaster information after executing the current rescue plan as the task data for the next rolling window.

[0047] Step 16: Use the task data of the next rolling window as the task data of the current rolling window, and return to step "update the prediction model based on the task data of the current rolling window and determine the rescue plan for the current rolling window" until all rolling windows are executed and the dynamic scheduling of helicopter rescue is completed.

[0048] In practical applications, this application combines an event-driven rolling horizon strategy to model the helicopter rescue dispatch dynamic model (i.e., the prediction model) as a mixed integer programming model. This model mainly includes four parts: model assumptions, decision variable definitions, constraints, and objective functions.

[0049] First, model assumptions.

[0050] Before introducing the prediction model, some assumptions are made as follows.

[0051] (1) For each disaster-stricken area, information about the wounded who have been rescued by the rescue team and concentrated at the helicopter rescue point can be transmitted back to the command post in a timely manner; however, the command post is not aware of the situation of the wounded who have not been discovered by the rescue team.

[0052] (2) The rescue teams in the disaster-stricken areas can transfer the rescued wounded to the wounded concentration area within a certain period of time.

[0053] (3) There should be sufficient helicopter landing and take-off points around the hospital and temporary hospital, and the helicopters should be able to be deployed again to provide support operations while dropping off and transporting the wounded.

[0054] (4) Hospitals and temporary hospitals can achieve maximum rescue capacity.

[0055] (5) The helicopter's maximum load flight time, speed, and refueling rate are known.

[0056] (6) The speed at which a helicopter loads the wounded has nothing to do with the type of helicopter.

[0057] (7) In view of the fact that helicopters usually carry the wounded at their maximum load during rescue operations, the impact of the load on their flight time is not considered during the rescue process.

[0058] (8) Information such as the construction of new casualty concentration areas and temporary hospitals, and the transfer of new casualties to concentration areas by rescue teams in the disaster-stricken areas can be transmitted back to the command post in a timely manner.

[0059] Second, decision variables.

[0060] The decision variables used in the prediction model are defined as follows.

[0061] :Indicates helicopter In the bth transport batch, from the disaster-stricken area Travel to the affected area If the helicopter In the bth transport batch, from the disaster-stricken area Travel to the affected area , Not equal to , =1. Otherwise, =0.

[0062] :Indicates helicopter In the bth transport batch to the disaster-stricken area .

[0063] : Indicates helicopter When the bth transport batch arrives at the disaster-stricken area moment.

[0064] : Indicates helicopter The bth transport batch is in the disaster-stricken area Number of wounded loaded.

[0065] : Indicates helicopter Refueling time for the bth transport batch.

[0066] Third, model constraints.

[0067] In order to optimize the dynamic scheduling problem of helicopter rescue of wounded patients, the preset constraints that need to be considered mainly include: helicopter capacity constraint, helicopter flight time constraint, hospital capacity constraint, casualty change constraint in the disaster area and dynamic event constraint.

[0068] 3.1 Transport capacity constraints in disaster areas.

[0069] During the helicopter rescue process, all injured people in the disaster area must be rescued. Considering that the transportation demand in the disaster area is greater than the carrying capacity limit of a single helicopter, multiple batches of helicopters are needed to complete the rescue of all injured people. The corresponding constraint can be expressed as formula (1).

[0070] (1)

[0071] in, Indicates the affected area The total number of injured people in the injured concentration area, Indicates the helicopter number; Indicates helicopter The transport batch set; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; A number indicating the affected area.

[0072] 3.2 Helicopter capacity constraints.

[0073] The number of wounded transported by a helicopter in a single batch cannot exceed the maximum capacity of that type of helicopter.

[0074] (2)

[0075] (3)

[0076] (4)

[0077] (5)

[0078] (6)

[0079] in, Indicates the helicopter type, Represents a set of helicopter types, express Maximum transport capacity of class helicopters, Indicates helicopter In the b The total number of casualties transported in each transport batch; Indicates helicopter In the b The set of disaster-stricken areas and hospitals visited in each transport batch (including the disaster-stricken areas reached and the hospitals reached, such as Figure 2 As shown, Indicates helicopter In the b transport batches arrived in the disaster-stricken areas moment, Indicates helicopter Arrival at the disaster area The total number of wounded remaining in the area at the time; Indicates that the helicopter is in the disaster area Loading the wounded takes time; R Represents the set of disaster-affected areas of all hospitals.

[0080] 3.3 Helicopter endurance constraints.

[0081] When a helicopter is performing a rescue and transport mission, its flight time in the air cannot exceed the maximum endurance limit.

[0082] (7)

[0083] (8)

[0084] (9)

[0085] in, Indicates helicopter h In the b The duration of flight in the air during each transport batch; Indicates helicopter Start b The remaining time of the flight when the transport batch is completed; express The maximum flight time of the helicopter Indicates helicopter From the affected area Travel to the affected area j time consuming; Indicates the speed at which helicopters load and unload casualties; It represents the ratio of the fuel consumption rate of the helicopter during loading the wounded to the fuel consumption rate during cruising flight; It can be used to distinguish the way helicopters load the wounded in disaster areas. When helicopters hover in the air to load the wounded, >1; When the helicopter landed in the disaster area where the injured were concentrated to load the injured, due to the engine being shut down, =0.

[0086] 3.4 Helicopter operation timing constraints.

[0087] Helicopters are transporting injured people from multiple disaster-stricken areas back to hospitals. The timing constraints for arriving at different disaster-stricken areas in the same transport batch include:

[0088] (10)

[0089] (11)

[0090] (12)

[0091] in, Indicates helicopter From the affected area Travel to the affected area time consuming, ; ;s i and e i Indicates that the helicopter is in the disaster area The time when rescue begins and ends; Indicates helicopter In the b transport batches from the affected areas Travel to the affected area ; Indicates helicopter In the b transport batches to the disaster-stricken areas ; Indicates helicopter In the b transport batches arrived in the disaster-stricken areas moment; and Indicates that the helicopter is in the disaster area The time when rescue begins and ends; Indicates that the helicopter is in the disaster area Loading the wounded takes time; Indicates helicopter From the affected area Travel to the affected area time consuming.

[0092] After the helicopter transports the injured back to the hospital, the injured need to be unloaded and transferred to the hospital by hospital medical staff for treatment. At the same time, the helicopter's accompanying maintenance personnel or the maintenance personnel at the hospital will check the condition of the helicopter and refuel as appropriate.

[0093] To characterize the constraints that need to be satisfied during the refueling process, we first define the refueling speed: : For a certain type of helicopter, the amount of time the helicopter can fly due to the fuel added per unit time.

[0094] When a helicopter refuels at a hospital, the refueling time must meet the following constraints.

[0095] (13)

[0096] (14)

[0097] (15)

[0098] (16)

[0099] in, Indicates helicopter h The amount of fuel replenished cannot exceed the maximum endurance. Indicates helicopter h The amount of fuel replenished needs to meet the endurance of the next batch of rescue. represents the refueling time of helicopter h in the bth transport batch; for Class helicopter refueling rate at hospital r; for Matching variables for helicopters, if helicopter h for type, z h,k is 1, otherwise 0; Indicates helicopter h Complete the b +1 transport batch's remaining endurance after rescue; For helicopters h Start bThe endurance of each transport batch before rescue; For helicopters h Complete the b The endurance required for a rescue transport batch; Indicates helicopter h Complete the b The remaining endurance of a transport batch after rescue; Said it took time for helicopters to unload the wounded at the hospital; represents the time the helicopter spends at the hospital; , and Indicates that the helicopter has arrived at the hospital r The moment of leaving the hospital r moment.

[0100] 3.5 Constraints on hospital rescue capabilities.

[0101] Since the hospital’s medical staff and space for treating injuries are prioritized, the hospital’s rescue capacity constraints need to be met during the helicopter rescue process.

[0102] (17)

[0103] in, Indicates hospital Maximum rescue capability; Indicates helicopter In the b transport batches to the hospital ; Collection for hospital.

[0104] 3.6 Dynamic condition constraints.

[0105] In the process of constructing the dynamic dispatch model of helicopter rescue, the dynamic factors considered include: rescue teams in the disaster area have discovered and transported a certain number of injured people to the concentrated area of ​​injured people, the project has successfully built a temporary hospital and met the conditions for treating the injured, and new disaster relief areas have been added. T When dynamic events occur at any time, the following constraints need to be added to the model.

[0106] (18)

[0107] (19)

[0108] (20)

[0109] Among them, T + and T - Respectively TThe corresponding moments before and after the dynamic event occurs. Indicates the disaster-affected area i in T The number of wounded is increasing all the time. express T The number of hospitals is increasing all the time. express T The ever-increasing collection of disaster-affected areas; Indicates that after a dynamic event occurs, T + The affected areas at the time ； Indicates that T before a dynamic event occurs - The collection of disaster-affected areas at the moment; Indicates that after a dynamic event occurs, T + Hospital collection at the moment; T represents the time before the dynamic event occurs - Hospital collection at the moment; Indicates that after a dynamic event occurs, T + Disaster-affected areas at all times the number of remaining casualties; Indicates that T before a dynamic event occurs - Disaster-affected areas at all times the number of remaining casualties; Indicates the remainder, It is the first letter of the English word left, which means remainder.

[0110] Fourth, objective function.

[0111] After the rescue teams in the affected area gather the injured in the injured collection area, they need to wait for helicopters to respond to the disaster site, load the injured onto the helicopters, and then transport them to the hospital for treatment. To reduce the waiting time for the injured from the injured collection area to receive rescue at the hospital, and to achieve dynamic adjustment of the weights as the rescue progresses, the objective function is set to minimize the weighted rescue waiting time for the injured.

[0112] (twenty one)

[0113] (twenty two)

[0114] (twenty three)

[0115] (twenty four)

[0116] (25)

[0117] (26)

[0118] in, represents the weighted waiting time for the rescue of the wounded; A number indicating the affected area; Indicates the helicopter number; Indicates helicopter The transport batch set; For helicopters No. transport batches in the disaster-stricken area the number of casualties loaded; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; Indicates helicopter No. transport batches in the disaster-stricken area The number of wounded people loaded requires rescue waiting time, Indicates helicopter No. transport batches in the disaster-stricken area The weight corresponding to the number of wounded loaded.

[0119] in, It consists of three parts: scale weight (Indicates that the number of people rescued in the disaster-stricken area i by this batch accounts for the maximum number of injured people in all disaster-stricken areas response weight (indicates the weight gain of the process of waiting for helicopter rescue); transportation weight (representing the weight gain of the process of transporting the rescued wounded to the hospital); Indicates the waiting time of the wounded in the wounded concentration area, It means that it takes time to transport the injured to the hospital; are weight gain coefficients, and They represent the ideal maximum response waiting time and maximum transportation time respectively (artificially set parameters).

[0120] The specific implementation process of the helicopter rescue dynamic scheduling method based on the rolling time domain strategy provided in this application is as follows.

[0121] In dynamic scenarios involving multiple helicopters rescuing casualties from multiple disaster zones, traditional static scheduling methods struggle to adapt to changing rescue needs, such as sudden increases in casualties, newly discovered areas of concentrated casualties, and changes in the opening hours of temporary hospitals. This application proposes a rolling time-domain strategy as the core framework for dynamic scheduling, enabling rapid response to dynamic events and real-time optimization and adjustment of scheduling plans.

[0122] The framework of the dynamic scheduling method is as follows.

[0123] In the dynamic scheduling method of helicopter rescue based on the rolling time domain optimization framework, when an emergency disaster event or real-time environmental disturbance occurs, the command organization needs to reconstruct the multi-helicopter collaborative rescue plan within a limited time. Unlike traditional static scheduling, an event-driven optimization strategy is adopted. At each dynamic event, that is, at each decision moment, a newly established mixed integer programming model is used based on the latest environmental information and helicopter status to optimize and generate helicopter rescue paths, task timing and resource allocation plans. Specifically, the dynamic scheduling of helicopter rescue based on the rolling time domain strategy mainly includes four key steps, such as Figure 3 shown.

[0124] The first is disaster-stricken area monitoring, which obtains real-time disaster information of the disaster-stricken area through multi-source data collection, including helicopter status information, disaster-stricken area status information and hospital status information.

[0125] The second step is updating the prediction model. Before the start of each prediction window (the first rescue dispatch is considered the first prediction window, i.e., the current rolling window), real-time disaster information in the affected area is updated based on monitoring data. This includes the location and condition of newly injured patients, the real-time status of helicopters, and the operational status of temporary hospitals. Based on this updated mission data, the original rescue dispatch model is parameterized and constraints modified to rebuild the optimization model for the current window; the original rescue dispatch model serves as the prediction model.

[0126] The third step is to formulate a rescue plan for each phase. This involves running a metaheuristic algorithm for decision-making within a rolling window to plan the rescue dispatch plan for the current phase.

[0127] Fourth, rescue plan execution. During the execution of the rescue plan, sensors and communication equipment continuously collect mission progress data, such as the actual flight position of the helicopter, the progress of patient transfers in the affected area, and remaining hospital capacity. This feedback information serves as an important basis for updating the next time-domain model, forming a closed-loop dynamic scheduling process of "information collection - model optimization - decision execution - feedback adjustment."

[0128] The dynamic response mechanism is as follows.

[0129] It's worth emphasizing that the dynamic rescheduling process must consider the operational feasibility of helicopter mission execution. Specifically, the real-time dispatch feasibility of each helicopter is determined by its current mission execution phase: helicopters in critical rescue operations (such as casualty loading, no remaining capacity, hospital unloading, and fuel refueling) have mission non-interruptibility, and the execution of new instructions must be delayed until the safe termination point of the current operation; helicopters on route with sufficient remaining capacity have immediate response capabilities.

[0130] Therefore, when implementing the dynamic scheduling method for executor rescue based on the rolling horizon strategy, the following principles need to be followed.

[0131] ① If a helicopter is currently loading wounded patients in a disaster-stricken area and a rescue team from that area brings back new wounded patients, the helicopter will continue to load new patients at its maximum capacity as long as there is still space available.

[0132] ② If the helicopter is on the route and is heading to the hospital to unload the wounded, the helicopter will continue to execute the original unloading plan.

[0133] ③ When the helicopter is on route and there is still spare capacity, it will immediately respond to the new scheduling plan.

[0134] ④ If the helicopter is unloading the wounded at the hospital, it will respond to the new dispatch plan immediately after completing the unloading of the wounded and refueling.

[0135] The dynamic scheduling simulation based on the event-driven rolling horizon strategy is as follows.

[0136] In order to evaluate the dynamic scheduling method of helicopter rescue based on the rolling time domain strategy, a simulation case was constructed with the large-scale earthquake disaster rescue as the background. In the early stage of post-earthquake rescue, the deployment of three temporary rescue hospitals (with a capacity of 1,000 casualties) has been completed. Each hospital is equipped with two rescue helicopters to form a basic transfer network. The coordinates of the three temporary rescue hospitals are (0, 0), (0, 50) and (50, 50) respectively. In the actual rescue scenario, by constructing simulation cases, the scheduling problem under different dynamic disturbance events is simulated. The geographical coordinates of each region are plotted on a two-dimensional plane. x , y The number of injured patients in each region at the current moment is randomly generated within the interval [20, 80]. Dynamic factors considered in the case study include: the number of injured patients awaiting transfer in the affected area may increase suddenly as rescue teams advance in the area; new affected areas are added; and engineers build and operate a temporary hospital (with a capacity of 400). Key parameter settings for each case are shown in Table 1.

[0137] Table 1 Key parameter settings for each case

[0138]

[0139] As shown in Table 1, the initial disaster-affected area information in Case 2 to Case 4 is the same as that in Case 1. The main difference lies in the different types of dynamic events considered, which are used to analyze the impact of different dynamic times on the rescue scheduling process.

[0140] When a dynamic event occurs, the helicopter rescue dispatch plan is replanned based on the RHS. To analyze the impact of different dynamic events on the dynamic scheduling problem, the four constructed task cases were executed sequentially, with the results shown in Table 2. The difference (Gap) between the post-dynamic event scheduling objective function and the static scheduling objective function was used as an indicator to evaluate the impact of different dynamic event task sizes.

[0141] Table 2 Example of simulation results for dynamic scheduling problem

[0142]

[0143] In Table 2, from the perspective of rescue scale, when the dynamic events are the same, as the task scale increases, the difference in the objective function of the dynamic scheduling scheme increases exponentially. From the perspective of the type of dynamic events, the rescue scheduling schemes before and after the dynamic time in different task cases are obtained as follows: Figures 4 to 7 As shown in the figure, D1 to D20 represent disaster-affected areas 1 to 20; Hosp1 to Hosp4 represent hospitals 1 to 4; and Heli 1 to Heli 4 represent helicopters 1 to 4. Given the same initial rescue dispatch plan, adding rescue hospitals can improve rescue efficiency. This is because increasing the number of hospitals (Case 2) shortens the helicopter rescue path, thereby reducing the objective function value and shortening the rescue mission completion time. Meanwhile, increasing the number of disaster-affected areas (Case 3) and the number of injured in an affected area (Case 4) both delay the rescue mission completion time.

[0144] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above method when executing the computer program.

[0145] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the above method when executed by a processor.

[0146] In an exemplary embodiment, a computer program product is provided, including a computer program, which implements the above method when executed by a processor.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0148] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRdM), magnetic random access memory (MRdM), ferroelectric random access memory (FRdM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RdM) or external cache memory, etc. By way of illustration and not limitation, RdM may be in various forms, such as static random access memory (SRdM) or dynamic random access memory (DRdM).

[0149] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0150] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A helicopter rescue dynamic scheduling method based on a rolling time domain strategy, characterized in that: include: Collect real-time disaster information in the disaster-stricken areas; Based on the rolling time domain strategy, the helicopter rescue period is divided according to the preset step size to determine multiple rolling windows; The real-time disaster information is used as the task data of the current rolling window, and the prediction model is updated based on the task data of the current rolling window to determine the rescue plan of the current rolling window, which specifically includes: inputting the task data of the current rolling window into the prediction model, and revising the preset constraints with the goal of minimizing the objective function of the prediction model to determine the updated prediction model; determining the rescue plan of the current rolling window according to the metaheuristic algorithm and the updated prediction model; wherein the real-time disaster information includes the current state of the helicopter, the current state information of the disaster-stricken area and the operating state information of the hospital; the current state of the helicopter includes at least the position, speed, remaining fuel and number of passengers of the helicopter; the current state information of the disaster-stricken area includes at least the number, distribution and injury level of the wounded; the operating state information of the hospital includes at least the remaining capacity, treatment capacity and resource reserve of the hospital; the prediction model is determined based on preset constraints; the preset constraints include the disaster area transportation capacity constraint, helicopter capacity constraint, helicopter endurance constraint, helicopter operation timing constraint and hospital rescue capacity constraint; the rescue plan includes the helicopter rescue path and the number of transported wounded; The objective function of the minimization prediction model is: ； in, represents the weighted waiting time for the rescue of the wounded; A number indicating the affected area; Indicates the helicopter number; Indicates helicopter The shipping batch number; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; Indicates helicopter No. transport batches in the disaster-stricken area The number of wounded people loaded requires rescue waiting time, Indicates helicopter No. transport batches in the disaster-stricken area The weight corresponding to the number of wounded loaded; Execute the current rescue plan and feedback the current execution results, and determine the real-time disaster information after the execution of the current rescue plan; Use the real-time disaster information after executing the current rescue plan as the task data for the next rolling window; The task data of the next rolling window is used as the task data of the current rolling window, and the process returns to step "update the prediction model based on the task data of the current rolling window and determine the rescue plan for the current rolling window" until all rolling windows are executed, completing the dynamic scheduling of helicopter rescue.

2. The helicopter rescue dynamic scheduling method based on the rolling time domain strategy according to claim 1 is characterized in that: Before using the real-time disaster information as the task data for the current rolling window and updating the prediction model based on the task data for the current rolling window to determine the rescue plan for the current rolling window, the following steps are also included: Define the decision variables of the prediction model; the prediction model is a mixed integer programming model; Determine preset constraints based on decision variables; Build a prediction model based on preset constraints.

3. The helicopter rescue dynamic scheduling method based on the rolling time domain strategy according to claim 1 is characterized in that: The transport capacity constraint in the disaster area is: ;in, Indicates the affected area The total number of injured people in the injured concentration area, Indicates the helicopter number; Indicates helicopter The shipping batch number; represents the set of disaster-affected areas; represents the set of all helicopters; Indicates helicopter The collection of all transport batches; A number indicating the affected area; Indicates helicopter The bth transport batch is in the disaster area the number of casualties loaded; The helicopter capacity constraint is: ; ； in, Indicates the helicopter type, Represents a set of helicopter types, express Maximum transport capacity of class helicopters, Indicates helicopter In the b The total number of casualties transported in each transport batch; z h,k for Matching variables for helicopters, if helicopter h for k type, z h,k is 1, otherwise 0; Indicates helicopter The bth transport batch is in the disaster area the number of casualties loaded; The helicopter's endurance constraints are: ; ； ； in, Indicates helicopter In the b The air endurance time of each transport batch, Indicates helicopter The remaining flight time when the bth transport batch begins, express Maximum flight time for helicopters of this type; Indicates helicopter From the affected area Travel to the affected area j time consuming; Indicates the speed at which helicopters load and unload casualties; It represents the ratio of the fuel consumption rate of the helicopter during loading the wounded to the fuel consumption rate during cruising flight; It can be used to distinguish the way helicopters load the wounded in disaster-stricken areas; when helicopters hover in the air to load the wounded, >1; When the helicopter landed in the disaster area where the injured were concentrated to load the injured, due to the engine being shut down, =0; Helicopter operation timing constraints: ;in, Indicates helicopter From the affected area Travel to the affected area time consuming, ; ; and Indicates that the helicopter is in the disaster area The time when rescue begins and ends; Indicates helicopter In the b transport batches from the affected areas Travel to the affected area ; Indicates helicopter In the b transport batches arrived in the disaster-stricken areas moment; Indicates helicopter From the affected area Travel to the affected area time consuming; Indicates that the helicopter is in the disaster area Loading the wounded takes time; Indicates helicopter In the b The transport batches arrived at the disaster-stricken areas and hospitals; Indicates that the helicopter is in the disaster area The end of the rescue moment; The hospital rescue capacity constraints are: ;in, Indicates hospital Maximum rescue capability; Indicates helicopter In the b Whether the transport batch goes to the hospital indicator variables; Collection for hospital.

4. The helicopter rescue dynamic scheduling method based on the rolling time domain strategy according to claim 1 is characterized in that: After collecting real-time disaster information of the disaster-stricken area, it also includes: Determining whether a dynamic event occurs after collecting real-time disaster information about the disaster-stricken area, and determining a first judgment result; the dynamic event includes at least: the number of newly added hospitals, the conditions for treating the injured in the newly added hospitals, and the number of newly added injured; If the first judgment result is yes, updating the real-time disaster information of the disaster-stricken area; If the first judgment result is no, the real-time disaster information of the disaster-stricken area is not updated.

5. The helicopter rescue dynamic scheduling method based on the rolling time domain strategy according to claim 4 is characterized in that: When the first judgment result is yes, the preset constraint condition further includes a dynamic condition constraint; The dynamic condition constraints are: ； ； ； Among them, T + and T - Respectively T The corresponding moments before and after the dynamic event occurs. Indicates the disaster-affected area i in T The number of wounded is increasing all the time. express T The number of hospitals is increasing all the time. express T The ever-increasing collection of disaster-affected areas; Indicates that after a dynamic event occurs, T + The affected areas at the time ； Indicates that T before a dynamic event occurs - The collection of disaster-affected areas at the moment; Indicates that after a dynamic event occurs, T + Hospital collection at the moment; T represents the time before the dynamic event occurs - Hospital collection at the moment; Indicates that after a dynamic event occurs, T + Disaster-affected areas at all times the number of remaining casualties; Indicates that T before a dynamic event occurs - Disaster-affected areas at all times the number of remaining casualties; Indicates remainder; Represents the set of affected areas.

6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the helicopter rescue dynamic scheduling method based on the rolling time domain strategy according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the helicopter rescue dynamic scheduling method based on the rolling horizon strategy described in any one of claims 1 to 5 is implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the helicopter rescue dynamic scheduling method based on the rolling horizon strategy described in any one of claims 1 to 5 is implemented.

Images