Helicopter rescue scheduling method and system based on large neighborhood search algorithm

Abstract

The invention discloses a helicopter rescue scheduling method and system based on a large neighborhood search algorithm, and relates to the technical field of emergency rescue, and the method comprises the steps: obtaining the state parameter information of a helicopter, an affected area and a hospital; based on the state parameter information, constructing a helicopter rescue scheduling mixed integer programming model with the aim of minimizing the total rescue completion time; based on the state parameter information, generating an initial rescue scheduling scheme through a greedy algorithm; and based on the initial rescue scheduling scheme, carrying out iterative solution on the helicopter rescue scheduling mixed integer programming model, and ending iteration until an iteration stop condition is met to obtain a rescue scheduling scheme. By dynamically optimizing the scheduling scheme, the real-time change of the rescue scene is effectively responded, the total rescue completion time is remarkably shortened, and the helicopter rescue efficiency in the complex disaster scene is improved.

Description

Technical Field

[0001] The present application relates to the field of emergency rescue technology, and in particular to a helicopter rescue dispatching method and system based on a large neighborhood search algorithm. Background Art

[0002] Due to the suddenness and destructive nature of natural disasters, affected areas often face urgent needs for evacuation and rescue of the injured. However, ground-based rescue efforts struggle to respond quickly due to road damage or traffic congestion. Helicopter rescue, as a fast and efficient emergency response method, plays a vital role in disaster relief. Helicopters can quickly reach affected areas and transport the injured to hospitals promptly, buying valuable time for rescue efforts.

[0003] Currently, helicopter rescue dispatch typically relies on manual decision-making or simple dispatch algorithms. These methods have numerous shortcomings when faced with complex rescue scenarios. For example, manual dispatch is easily influenced by subjective factors, making it difficult to quickly and accurately formulate optimal rescue plans. Furthermore, dispatch algorithms used in related technologies cannot balance complex constraints with dynamic requirements, resulting in long rescue completion times and low rescue efficiency.

[0004] Therefore, there is an urgent need for a helicopter rescue dispatching method that can effectively shorten the total time to complete helicopter rescue, improve rescue efficiency, and better meet the emergency rescue needs of disaster-stricken areas. Summary of the Invention

[0005] The purpose of this application is to provide a helicopter rescue dispatching method and system based on a large neighborhood search algorithm, which can respond to dynamic changes in rescue scenarios in real time (such as the expansion of the disaster-stricken area and the surge in demand for the wounded), effectively shorten the total time to complete the rescue, improve the utilization rate of helicopter resources, and significantly enhance the efficiency and accuracy of emergency rescue in complex disaster scenarios.

[0006] To achieve the above objectives, this application provides the following solutions: In a first aspect, the present application provides a helicopter rescue dispatching method based on a large neighborhood search algorithm, comprising: Obtaining status parameter information of rescue entities; the rescue entities include helicopters, disaster-stricken areas, and hospitals; Based on the state parameter information, a helicopter rescue scheduling mixed integer programming model is constructed with the goal of minimizing the total time required to complete the helicopter rescue; Based on the state parameter information, generating an initial rescue scheduling plan through a greedy algorithm; Based on the initial rescue scheduling plan, the helicopter rescue scheduling mixed integer programming model is iteratively solved by a large neighborhood search algorithm until the iteration stops when the iterative stopping condition is met, and a rescue scheduling plan is obtained; the rescue scheduling plan includes the rescue mission sequence of each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity and helicopter flight time in each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time at each hospital; the rescue mission sequence includes the sequence of disaster-stricken areas and hospitals that the helicopter needs to visit in sequence.

[0007] In a second aspect, the present application provides a helicopter rescue dispatch system based on a large neighborhood search algorithm, comprising: A rescue data acquisition module is used to obtain status parameter information of rescue entities; the rescue entities include helicopters, disaster-stricken areas, and hospitals; A model building module is used to build a helicopter rescue scheduling mixed integer programming model based on the state parameter information with the goal of minimizing the total time to complete the helicopter rescue; An initial rescue scheduling plan generating module, configured to generate an initial rescue scheduling plan based on the state parameter information through a greedy algorithm; The rescue scheduling plan generation module is used to iteratively solve the helicopter rescue scheduling mixed integer programming model based on the initial rescue scheduling plan through a large neighborhood search algorithm until the iteration is terminated when the iteration stopping condition is met, thereby obtaining a rescue scheduling plan; the rescue scheduling plan includes the rescue mission sequence of each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity and helicopter flight time in each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time at each hospital; the rescue mission sequence includes the sequence of disaster-stricken areas and the sequence of hospitals that the helicopter needs to visit in sequence.

[0008] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the helicopter rescue dispatch method based on the large neighborhood search algorithm described above.

[0009] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned helicopter rescue dispatch methods based on a large neighborhood search algorithm.

[0010] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned helicopter rescue dispatch methods based on a large neighborhood search algorithm.

[0011] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a helicopter rescue dispatching method and system based on a large neighborhood search algorithm. By dynamically acquiring the status parameter information of helicopters, disaster-stricken areas, and hospitals, it accurately captures real-time changes in rescue scenarios (such as the expansion of disaster-stricken areas and the surge in demand for wounded people). It solves the problem of scheduling methods in related technologies that lag behind due to reliance on static data, and achieves rapid response to complex rescue scenarios. By using a mixed integer programming algorithm based on state parameter information to construct a scheduling model with the goal of minimizing the total time to complete the rescue, it incorporates multi-dimensional constraints such as helicopter flight time, rescue capacity, hospital unloading time, and refueling time into the optimization framework, solving the problem of no The method takes into account the defect of inefficiency caused by multiple factors and achieves the global optimality of the scheduling plan; based on the state parameter information, a greedy algorithm is used to generate an initial rescue scheduling plan; and based on the initial rescue scheduling plan, a large neighborhood search algorithm is used to iteratively solve the helicopter rescue scheduling mixed integer programming model, and a refined scheduling plan including the task sequence of each helicopter, the rescue time window of the disaster-stricken area and the time consumption of hospital operations is obtained. The problem that manual scheduling is highly subjective and difficult to quickly generate feasible plans is solved, and the efficient allocation and path optimization of rescue tasks are achieved, which significantly shortens the total rescue completion time and improves the utilization rate of helicopter resources and the emergency rescue efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, 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.

[0013] Figure 1 This is an application environment diagram of a helicopter rescue dispatching method based on a large neighborhood search algorithm in one embodiment of the present application.

[0014] Figure 2 A flowchart of a helicopter rescue dispatching method based on a large neighborhood search algorithm is provided in one embodiment of the present application.

[0015] Figure 3 A schematic diagram of a helicopter rescue post-disaster response provided in one embodiment of the present application.

[0016] Figure 4 A schematic diagram of a rescue batch provided in an embodiment of the present application.

[0017] Figure 5 This is a schematic diagram of the rescue situation at the initial moment provided by an embodiment of the present application; wherein, Figure 5 (a) Distribution map of disaster-affected areas and hospital locations; Figure 5 (b) is a diagram showing the number of people needed for rescue in each disaster-stricken area.

[0018] Figure 6 A schematic diagram of a helicopter rescue dispatching solution based on a large neighborhood search algorithm provided in one embodiment of the present application.

[0019] Figure 7 A schematic diagram of the objective function convergence curve provided in one embodiment of the present application.

[0020] Figure 8 A schematic diagram of a damage ratio change curve provided in an embodiment of the present application.

[0021] Figure 9 A schematic diagram of a curve showing a change in the probability of executing a global repair strategy provided in one embodiment of the present application.

[0022] Figure 10 A schematic diagram of the functional modules of a helicopter rescue dispatch system based on a large neighborhood search algorithm provided in one embodiment of the present application.

[0023] Figure 11 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0024] 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.

[0025] 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.

[0026] The helicopter rescue dispatching method based on the large neighborhood search algorithm provided in the embodiment of the present application can be applied to Figure 1In the application environment shown, the terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the state parameter information of the rescue entity to the server 104. After the server 104 receives the state parameter information of the rescue entity, the server 104 constructs a helicopter rescue scheduling mixed integer programming model based on the state parameter information with the goal of minimizing the total time to complete the helicopter rescue. Based on the state parameter information, an initial rescue scheduling plan is generated using a greedy algorithm. Based on the initial rescue scheduling plan, the helicopter rescue scheduling mixed integer programming model is iteratively solved using a large neighborhood search algorithm until the iteration stops when the iteration condition is met, thereby obtaining a rescue scheduling plan. The rescue scheduling plan includes a rescue mission sequence for each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity, and helicopter flight time for each disaster-stricken area; the time required to unload the wounded and the helicopter refueling time for each hospital; and the rescue mission sequence includes a sequence of disaster-stricken areas and a sequence of hospitals that the helicopter must visit in sequence. The server 104 can feed back the obtained rescue scheduling plan to the terminal 102. In addition, in some embodiments, the helicopter rescue scheduling method based on the large neighborhood search algorithm can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly generate a rescue scheduling plan based on the state parameter information of the rescue entity, or the server 104 can obtain the state parameter information of the rescue entity from a data storage system and generate a rescue scheduling plan based on the state parameter information of the rescue entity.

[0027] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0028] In an exemplary embodiment, Figure 2 As shown, a helicopter rescue dispatch method based on a large neighborhood search algorithm is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the process, including the following steps 201 to 204. Step 201: Obtain status parameter information for rescue entities, including helicopters, disaster-stricken areas, and hospitals. The helicopter's status parameter information includes its location coordinates, maximum flight time, helicopter's capacity for casualties, and rescue status. The disaster-stricken area's status parameter information includes its location coordinates, number of casualties, and emergency level. The hospital's status parameter information includes its location coordinates and its capacity for casualties.

[0029] Step 202 : Based on the state parameter information, a helicopter rescue scheduling mixed integer programming model is constructed with the goal of minimizing the total time required to complete the helicopter rescue.

[0030] Step 203: Based on the state parameter information, an initial rescue scheduling plan is generated through a greedy algorithm.

[0031] Step 204, based on the initial rescue scheduling plan, the helicopter rescue scheduling mixed integer programming model is iteratively solved by a large neighborhood search algorithm until the iteration stops when the iteration stop condition is met, and a rescue scheduling plan is obtained; the rescue scheduling plan includes a rescue mission sequence for each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity, and helicopter flight time for each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time for each hospital; the rescue mission sequence includes a sequence of disaster-stricken areas and a sequence of hospitals that the helicopter needs to visit in sequence.

[0032] By implementing the above steps 201 to 204, the present application can dynamically capture the real-time changes of the rescue scene and generate a globally optimized scheduling plan, significantly improving the efficiency of helicopter rescue and resource utilization.

[0033] In another exemplary embodiment of the present application, during the helicopter post-disaster rescue process, the coordinated operation of the disaster relief command center, helicopters, hospitals, disaster site rescue teams and engineering teams constitutes a dynamic three-dimensional post-disaster response network, such as Figure 3 shown.

[0034] Command Post: Based on disaster information, intelligent algorithms are used to plan rescue plans, optimize helicopter routes, and dynamically adjust transportation plans based on environmental changes.

[0035] Helicopters: Transport the wounded from various disaster-stricken areas to hospitals or temporary hospitals for treatment. Dynamically respond to adjustments to the command post's planning plan and quickly complete personnel distribution.

[0036] Hospital: Receive seriously injured patients transferred by helicopter through fast track.

[0037] Disaster-affected rescue teams: Build a ground search network, employ snake-like robots to penetrate the rubble, and use sonar and gas sensors to locate survivors. Centralize the management of the wounded, concentrating them in one area (the casualty concentration zone) to provide simplified injury management while facilitating helicopter rescue. Furthermore, for more remote disaster-stricken areas, separate casualty concentration zones will be established to facilitate direct helicopter rescue.

[0038] By combining the event-driven receding horizon strategy, the model is constructed as a mixed integer programming model for helicopter rescue scheduling.

[0039] Before building a mixed integer programming model for helicopter rescue dispatch, some model assumptions need to be made, including: (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.

[0040] (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.

[0041] (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.

[0042] (4) Hospitals and temporary hospitals can achieve maximum rescue capacity.

[0043] (5) The helicopter's maximum load flight time, speed, and refueling rate are known.

[0044] (6) The speed at which a helicopter loads the wounded has nothing to do with the type of helicopter.

[0045] (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.

[0046] (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.

[0047] In another exemplary embodiment of the present application, after the rescue team in the disaster area gathers the wounded in a gathering area, they need to wait for a helicopter to respond to the needs of the disaster site, load the wounded onto the helicopter, and then transport them to the hospital for treatment. In order to reduce the waiting time for the wounded to be rescued from the gathering area and realize the dynamic adjustment of the weight as the rescue progresses, the objective function is set to minimize the weighted rescue waiting time of the wounded. The objective function of the helicopter rescue scheduling mixed integer programming model is: .

[0048] .

[0049] Where C represents the objective function value, represents the maximum rescue completion time of all rescue batches of the h-th helicopter; H represents the helicopter set; represents the rescue batch set of the h-th helicopter; r represents the hospital node passed by the h-th helicopter in the b-th rescue batch; represents the set of disaster-stricken area nodes and hospital nodes that the h-th helicopter has visited in the b-th rescue batch; R represents the set of hospital nodes; It represents the rescue completion time when the h-th helicopter of the b-th rescue batch returns to the hospital node r and unloads the wounded.

[0050] In another exemplary embodiment of the present application, the constraints of the helicopter rescue scheduling mixed integer programming model include: rescue batch constraints for the wounded in the disaster area, helicopter loading capacity constraints for the wounded, helicopter endurance constraints, helicopter operation timing constraints and hospital rescue capacity constraints for the wounded.

[0051] As an optional implementation, during the helicopter rescue process, all injured people in the disaster area must be rescued. Considering that the transportation capacity demand in the disaster area exceeds the carrying capacity limit of a single helicopter, multiple batches of helicopter rescue are required to complete the rescue of all injured people. The batch constraint for the rescue of injured people in the disaster area is shown in the following formula: .

[0052] in, represents the total number of injured people in the injured concentration area of the i-th disaster area node; h represents the h-th helicopter; H represents the helicopter set; b represents the b-th rescue batch; represents the rescue batch set of the h-th helicopter; D represents the set of nodes in the disaster area; It represents the number of injured people transported from the i-th disaster area node in the b-th rescue batch of the h-th helicopter.

[0053] A rescue batch diagram is as follows Figure 4 As shown in the figure, the number of wounded carried by a helicopter in a single batch cannot exceed the maximum capacity of this type of helicopter. That is, the helicopter's wounded loading capacity constraint is: .

[0054] .

[0055] .

[0056] .

[0057] .

[0058] in, An identifier variable indicating that the h-th helicopter is of the k-th type; represents the total number of wounded transported by the h-th helicopter in the b-th rescue batch; represents the maximum transport capacity of the kth type of helicopter; k represents the type of helicopter; K represents the set of helicopter types; represents the set of disaster-stricken area nodes and hospital nodes that the h-th helicopter has visited in the b-th rescue batch (e.g. Figure 4 shown); represents the arrival time of the h-th helicopter in the b-th rescue batch arriving at the i-th disaster area node; It represents the total number of injured persons remaining in the i-th disaster area node when the h-th helicopter arrives at the i-th disaster area node; represents the total time for loading the wounded in the i-th disaster area; R represents the set of hospital nodes; It means that the h-th helicopter is heading to the i-th disaster-stricken area node in the b-th rescue batch.

[0059] When a helicopter is performing a rescue transport mission, its flight time in the air cannot exceed the maximum endurance limit. The helicopter's flight time constraints are as follows: .

[0060] .

[0061] .

[0062] in, represents the air endurance time of the h-th helicopter in the b-th rescue batch; Indicates the remaining flight time of the h-th helicopter when it starts the b-th rescue batch; represents the maximum flight time of the kth type of helicopter; represents the time it takes for the h-th helicopter to travel from the j-th disaster-stricken area node to the i-th disaster-stricken area node; Indicates the speed at which the helicopter loads and unloads the wounded.

[0063] In particular, It represents the ratio of the fuel consumption rate of the helicopter during the process of loading the wounded in the i-th disaster area node to the fuel consumption rate during the cruising flight, which can be used to distinguish the way the helicopter loads the wounded in the disaster area. >1; When the helicopter landed in the disaster area where the injured were concentrated to load the injured, the engine was shut down. =0.

[0064] Helicopters are transporting injured people from multiple disaster areas back to hospitals. The timing constraints for arriving at different nodes within the same transport batch, i.e., the helicopter operation timing constraints, are as follows: .

[0065] .

[0066] .

[0067] Where, The variable representing the h-th helicopter in the b-th rescue batch, from the j-th disaster area node to the i-th disaster area node, i≠j; if the h-th helicopter in the b-th rescue batch, from the j-th disaster area node to the i-th disaster area node, =1; otherwise, =0; represents the time when the helicopter ends its rescue in the jth disaster area; and They represent the time when the helicopter starts and ends rescue in the i-th disaster area.

[0068] 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.

[0069] To characterize the constraints that need to be satisfied during the refueling process, we first define the refueling rate , which represents the increase in the helicopter's flight time for a certain k-type helicopter due to the fuel added during the r-th hospital unit time.

[0070] When a helicopter is refueling at a hospital, the refueling time must meet the following conditions: .

[0071] .

[0072] .

[0073] .

[0074] in, represents the refueling time of the h-th helicopter in the b-th rescue batch; represents the refueling speed of the rth hospital node of the kth type of helicopter, which represents the increase in the helicopter's flight time due to the fuel replenished per unit time; Indicates that the h-th helicopter is heading to the r-th hospital node in the b-th rescue batch; It represents the remaining flight time of the h-th helicopter after completing the rescue of the b-th rescue batch; represents the rth hospital node; represents the air endurance time of the h-th helicopter in the b+1-th rescue batch; represents the total time spent by the helicopter at the rth hospital node; represents the time it takes for a helicopter to unload the wounded at the rth hospital node; Indicates the remaining flight time of the h-th helicopter when it starts the b-th rescue batch; represents the air endurance time of the h-th helicopter in the b-th rescue batch.

[0075] Since the hospital's medical staff and space for treating injuries are prioritized, the hospital's rescue capacity constraints must be met during helicopter rescue. The hospital's rescue capacity constraints for injured patients are as follows: .

[0076] Where, It represents the maximum rescue capacity of the r-th hospital node, that is, the maximum casualty capacity.

[0077] The Large Neighborhood Search (LNS) algorithm is a heuristic algorithm for solving combinatorial optimization problems. Its core idea is to dynamically explore different regions of the solution space during the search process by alternating between "destruction" and "repair" operations, gradually approaching the optimal solution.

[0078] In another exemplary embodiment of the present application, considering that the number of injured in a disaster-stricken area may exceed the helicopter's carrying capacity, to ensure that all injured in the disaster-stricken areas can be rescued, the Class class method is used for decoding, and an initial rescue dispatch plan is constructed in conjunction with a greedy algorithm. This process mainly revolves around the status information of the helicopter, the disaster-stricken area, and the hospital. By continuously iteratively updating key information such as the rescue path list, helicopter capacity, and the number of injured in the disaster-stricken area and hospital, a feasible rescue dispatch plan is gradually constructed. Step 203 specifically includes: 1. Initial solution construction. Input: The scenario state within the current rolling window, including the helicopter state (such as location, endurance, current passenger capacity, etc.), the disaster area state (such as the number of injured and location), and the hospital state (such as the number of injured and location available). Based on these scenario states, an initial solution S is constructed, where the helicopter rescue path list is initially empty. This means that at the beginning, no specific rescue routes have been planned for any helicopter.

[0079] 2. Iterate the rescue process.

[0080] Cycle conditions: As long as there are injured people who have not been rescued, the cycle will continue.

[0081] Traverse the affected area node i: For each affected area node i, perform the following operations: Using a greedy strategy, we select the helicopter h closest to the disaster area node i for rescue. The “closest distance” here can be based on geographical distance or other appropriate distance metrics.

[0082] If the helicopter h has sufficient endurance (i.e., it can reach the disaster-stricken area node i and return to the hospital or other designated location after completing the rescue) and the helicopter has remaining carrying capacity, then the helicopter h will go to the disaster-stricken area node i for rescue, and the status of the helicopter h and the disaster-stricken area node i in S will be updated.

[0083] If the helicopter h has insufficient endurance or sufficient carrying capacity, it will first go to its hospital node r to refuel and drop off the injured on board, and update the status of the helicopter h and its hospital node r in S; after refueling, the helicopter h will go to the disaster-stricken area node i for rescue, and update the status of the helicopter h and the disaster-stricken area node i in S.

[0084] Each update of the status of the helicopter h and the node i in the disaster area in the initial solution S includes the following: (1) Rescue Path List Update: The helicopter’s departure time for disaster-stricken area node i, the flight time to disaster-stricken area node i, and the loading time at disaster-stricken area node i (or unloading time if i is a hospital node) are added to the helicopter’s rescue path list. This enables the rescue path list to record every action of the helicopter in detail.

[0085] (2) Helicopter capacity update: Based on the loading status of the helicopter at node i in the disaster-stricken area (or the unloading status at the hospital node), the remaining carrying capacity of the helicopter is updated.

[0086] (3) Update of the number of injured patients in the disaster-affected area and hospital: For node i in the disaster-affected area, update the number of injured patients remaining, that is, reduce the number of injured patients carried by helicopter h. If it is a hospital node, update the number of injured patients that the hospital can subsequently accept, and increase the number of injured patients unloaded from helicopter h.

[0087] After all the wounded have been rescued, the updated initial solution S is output. This solution is the initial rescue scheduling plan constructed by the greedy algorithm, which includes key content such as the rescue path of each helicopter and the status information of each node, providing a basis for subsequent optimization and adjustment.

[0088] Through the above process, the situation where the number of injured people in the disaster area exceeds the helicopter carrying capacity can be effectively handled, and a preliminary feasible rescue dispatch plan can be generated.

[0089] In another exemplary embodiment of the present application, step 204 specifically includes: S1. Take the objective function value corresponding to the initial rescue scheduling plan as the current optimal solution.

[0090] S2. Randomly select a preset number of nodes to be removed from each helicopter rescue mission sequence in the initial rescue scheduling plan to obtain multiple sets of removed nodes; the nodes include disaster-stricken area nodes and hospital nodes.

[0091] S3. Remove the nodes in the set of removed nodes from the corresponding helicopter rescue mission sequences respectively to obtain multiple helicopter rescue partial scheduling sequences.

[0092] In LNS, the "destruction" operation removes some nodes (i.e., rescue tasks) from the current solution, generating a "partial solution." This step aims to disrupt the structure of the current solution, preventing the algorithm from falling into a local optimum. It also creates opportunities for subsequent repair operations, which can explore more optimal scheduling solutions by reinserting the removed nodes.

[0093] In the multi-helicopter rescue scheduling problem, the random destruction strategy is implemented as follows: (1) Determine the destruction ratio: Set a destruction ratio , represents the ratio of the number of nodes removed from each helicopter rescue mission sequence in the current solution to the total number of nodes. At the same time, the total number of nodes that need to be removed from each helicopter rescue mission sequence is calculated. .

[0094] Where, represents the preset number of nodes to be removed in the h-th helicopter rescue mission sequence; Represents the helicopter rescue mission sequence of helicopter h in the current solution S, which is represented by All nodes in Combined according to the order of access; represents the number of helicopter rescue mission sequences calculated for the h-th helicopter; It means that the decimal is rounded up, which can ensure that at least one node is destroyed in each helicopter rescue mission sequence.

[0095] (2) Randomly select the node to be removed. For each helicopter path , randomly selected from them with equal probability Nodes (including disaster-stricken area nodes and hospital nodes) are removed to generate a "partial solution" (i.e., the partial scheduling sequence of helicopter rescue), and the removed nodes are recorded to obtain a set of removed nodes.

[0096] S4. Based on the set of removed nodes, a probabilistic selection strategy method is used to select repair strategies for the helicopter rescue scheduling sequence to obtain a current repair scheduling plan; the repair strategies include local repair strategies and global repair strategies.

[0097] In the "repair" phase, two repair strategies are designed: Local Repair Strategy (LRS) and Global Repair Strategy (GRS).

[0098] In another exemplary embodiment of the present application, the local repair strategy in S5 specifically includes: The positions of the nodes in the removed node set are randomly exchanged multiple times to obtain multiple new removed node sets.

[0099] Traverse all new removal node sets respectively. If the current node in the current new removal node set is a disaster-stricken area node, insert the current node into the first preset position in the corresponding helicopter rescue part scheduling sequence; the first preset position is the position that increases the air operation time of the corresponding helicopter rescue part scheduling sequence the least.

[0100] If the current node in the newly removed node set is a hospital node, a hospital node is randomly selected in the corresponding helicopter rescue partial scheduling sequence to be exchanged with the current node.

[0101] Until all new removed node sets are traversed, a repair scheduling plan is obtained.

[0102] Specifically, the local repair strategy focuses on reinserting the nodes removed from the helicopter operation sequence after being “damaged” to optimize the scheduling solution. The specific process is as follows: Enter the "partial solution" after destruction and remove the node set. For each helicopter h in the "partial solution" corresponding to the operation sequence , first disturb the set of removed nodes D deleted from the sequence h This step is to try different insertion orders when reinserting nodes later to find a better job sequence.

[0103] Next, traverse the D after the disordered order h If the current node is a node in the disaster area, Select the position that increases the minimum time of helicopter operation sequence corresponding to the current node, insert the current node into this position and update After insertion, the time change needs to be calculated considering factors such as the distance between nodes and flight speed. If the current node is a hospital node, A hospital node is randomly selected to replace the current node in order to adjust the operation sequence, which may affect operations such as refueling and unloading of wounded patients.

[0104] After completing the node insertion and replacement operations, the helicopter operation sequence The feasibility of the operation is modified to ensure that various constraints such as the helicopter's carrying capacity (the number of wounded carried at each stage does not exceed the maximum capacity), endurance (the remaining endurance meets the needs of subsequent flights), and time sequence (all operations are in a reasonable order) are met.

[0105] Finally, the output is a repaired and adjusted operation sequence for each helicopter that satisfies the constraints and minimizes the duration of the helicopter's aerial operations. After the above steps are processed, the repaired and adjusted operation sequence for each helicopter is obtained. These sequences minimize the helicopter's aerial operation time while satisfying the constraints, thereby improving the efficiency of rescue dispatch.

[0106] The global repair strategy in S5 includes: The positions of the nodes in the removed node set are randomly exchanged multiple times and integrated to obtain an integrated removed node set.

[0107] Traverse the integrated removal node set. If the current node is a disaster area node, insert the current node into the second preset position; the second preset position is the corresponding position in the helicopter rescue part scheduling sequence that increases the aerial operation time the least.

[0108] If the current node is a hospital node, randomly select any hospital node in the helicopter rescue part scheduling sequence to exchange with the current node.

[0109] Until the traversal of the integrated removed node set is completed, the repair scheduling plan is obtained.

[0110] Specifically, the global repair strategy optimizes rescue scheduling and reduces the duration of helicopter aerial operations by redistributing the nodes deleted from the helicopter operation sequence during the "destruction" phase. The specific process is as follows: First, prepare the input, that is, the "partial solution" after destruction and the node sequence corresponding to each helicopter. Then, shuffle the node sequence removed from each helicopter sequence and integrate it into the sequence D s middle.

[0111] When D s Loop operation when it is not empty: if D s The current node is the node in the disaster area, so the job sequence of h The selected helicopter that increases the air operation time the least is inserted into the current node position in the dispatch sequence. When inserting, the distance between nodes and the flight speed should be considered to calculate the change in duration, and the update should be made. If the current node is a hospital node, any hospital node in the random helicopter rescue scheduling sequence is selected and exchanged with the current node, which may affect the refueling, unloading of the wounded and other links.

[0112] After processing the current node, remove it from D s Delete. Finally, for each helicopter operation sequence Perform feasibility corrections to meet the constraints of helicopter carrying capacity (the number of wounded carried at each stage does not exceed the maximum capacity), endurance (the remaining endurance meets the needs of subsequent flights), time sequence (each operation is in a reasonable order), etc., and output the helicopter operation sequence that meets the constraints and minimizes the duration of aerial operations after repair and adjustment.

[0113] S5. If the objective function value corresponding to the current repair scheduling plan is greater than or equal to the objective function value corresponding to the current optimal solution, the initial rescue scheduling plan is updated through the annealing mechanism.

[0114] S6. If the objective function value corresponding to the current repair scheduling plan is less than the objective function value corresponding to the current optimal solution, the initial rescue scheduling plan is updated using the current repair scheduling plan.

[0115] S7. If the iteration stopping conditions are met, the currently updated initial rescue scheduling plan will be used as the rescue scheduling plan; the iteration stopping conditions include reaching a preset maximum number of iterations, or the change in the objective function value corresponding to the current optimal solution within a preset number of consecutive times is less than a preset change threshold, or the algorithm running time exceeds a preset running time threshold.

[0116] S8. If the iteration stopping condition is not met, return to step "taking the objective function value corresponding to the initial task planning scheme as the current optimal solution".

[0117] The LNS mainly includes two processes: the "destruction" operation and the "repair" strategy. This application introduces a probability selection mechanism to achieve the adaptive selection of the repair strategy and the global repair strategy. At the same time, the simulated annealing strategy is introduced, which can select non-optimal solutions with a certain probability to improve the exploration ability of the algorithm.

[0118] The probability selection mechanism realizes the adaptive selection of the local repair strategy (LRS) and the global repair strategy (GRS). During the algorithm iteration process, by setting the selection probability P1, it is randomly determined which repair strategy to adopt in each iteration. When the random number is less than P1, the global repair strategy is selected; otherwise, the local repair strategy is selected. This adaptive selection method can flexibly apply different repair strategies according to different problem instances and iteration situations, which helps to more comprehensively search the solution space and find better solutions.

[0119] The introduction of the simulated annealing strategy enables the algorithm to select non-optimal solutions with a certain probability, thereby improving the exploration ability of the algorithm. When the objective function value of the repair scheduling scheme is greater than or equal to the objective function value corresponding to the current optimal solution, instead of directly discarding the scheme, the initial rescue scheduling scheme is updated through the simulated annealing mechanism. This strategy allows the algorithm to accept worse solutions under certain conditions, avoiding the algorithm from prematurely falling into the local optimal solution and increasing the possibility of exploring the global optimal solution in the solution space.

[0120] The specific steps of the overall architecture of the algorithm are as follows: (1) Initial solution generation: Record the current CPU time t1, and use the greedy algorithm to generate the initial solution S, which contains the operation sequences of each helicopter. Set the initial solution as the optimal solution S best = S, and calculate its objective function value C best = C(S best )). At the same time, set the initial solution as the current solution S0 = S, and calculate the objective function value C0 = C best .

[0121] (2) Iteration process: a. Destruction operation: For each iteration number ite , take the current solution S0 as the input, perform the destruction operation according to the given destruction ratio, and generate the "partial solution" and the removed node sequence corresponding to each helicopter.

[0122] b. Repair strategy selection and execution: Global repair strategy: If the random number rand() < P1, use the global repair strategy (GRS) to repair the "partial solution" to obtain S GS , and calculate its corresponding objective function C(S GS ). If C(S GS ) < C0, update the current solution S0 = SGS , and update the current objective function value \(C_0 = C(S GS ); If \(C(S GS ) < C best , then update the optimal solution \(C best = C(S GS ), S best = S GS . Otherwise, update \(S_0\) through the simulated annealing mechanism.

[0123] The described simulated annealing mechanism is expressed as: If \(C(S GS ) < C_0\), then first calculate the difference of the objective function ; Secondly, calculate the inferior value acceptance probability ; Thirdly, randomly generate a random number \(rand()\) in the interval \((0, 1)\). If \(rand() < , then update the current solution \(S_0 = S LS , and update the current objective function value \(C_0 = C(S GS ).

[0124] Local repair strategy: If \(rand() \geq P_1\), use the local repair strategy (LRS) to repair the "partial solution" to obtain \(S LS , and calculate its corresponding objective function \(C(S LS ). If \(C(S LS ) < C_0\), then update the current solution \(S_0 = S LS , and update the current objective function value \(C_0 = C(S GS ); If \(C(S LS ) < C best , then update the optimal solution \(C best = C(S LS ), S best = S LS . Otherwise, update \(S_0\) through the simulated annealing mechanism.

[0125] The described simulated annealing mechanism is expressed as: If \(C(S LS ) < C_0\), then first calculate the difference of the objective function ; Secondly, calculate the inferior value acceptance probability ; Thirdly, randomly generate a random number \(rand()\) in the interval \((0, 1)\). If \(rand() < , then update the current solution \(S_0 = S LS , and update the current objective function value \(C_0 = C(S LS ).

[0126] Parameter update: record the current CPU time t2, calculate the algorithm runtime delt = t2 − t1, and update the runtime ratio rate = delt / TC. Update the damage ratio, selection probability P1, and annealing temperature T0 based on rate (where the damage ratio update formula is =0.6*(1-1 / (1+exp(6*(1-2*rate)))), the selection probability update formula is P1=0.7-0.5*sin(rate* ), the annealing temperature update formula is T0=T0∗alpha).

[0127] Termination condition judgment: If delt>TC (that is, the algorithm running time exceeds the set CPU running time TC), then jump out of the loop.

[0128] Output result: The final output is the optimal solution S best .

[0129] In another embodiment of the present application, a simulation case is constructed with 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 wounded) has been completed. Each hospital is equipped with two rescue helicopters to form a basic transfer network. The helicopter position coordinates are (0, 0), (50, 50) and (0, 50) respectively. In the simulation of actual rescue scenarios, by constructing multi-scale simulation cases, it is intended to fully simulate the scheduling problems under different rescue complexities, and provide a rich data foundation for subsequent rescue scheduling strategy research. The geographical coordinates of each disaster-stricken area are randomly generated within the range of x, y∈[0, 100] km on the two-dimensional plane. This setting fully simulates the scattered and irregular distribution characteristics of the disaster-stricken areas after the earthquake, and is closer to the real disaster scene. The number of wounded in each area is randomly generated in the interval of [20, 80], which further increases the diversity and complexity of the simulation cases. 20 simulation cases of the scale of the disaster-stricken area are generated. The rescue situation from the perspective of the disaster relief command post is as follows: Figure 5 As shown. Among them, Figure 5 (a) Distribution map of disaster-affected areas and hospital locations; Figure 5 (b) is a diagram showing the number of people needed for rescue in each disaster-stricken area.

[0130] In actual rescue operations, given the unique circumstances of some affected areas, helicopters were unable to land in some locations. Therefore, a touch-and-hover method was used to load the injured. In this case, the helicopter's fuel consumption increased to 1.15 relative to cruise control. The helicopter used had the following parameters: a speed of 210 km / h, a capacity of 50 passengers, and a maximum endurance of 240 minutes (initial endurance was set at 200 minutes). During refueling, the helicopter was refueled according to its maximum endurance to ensure sufficient endurance for subsequent rescue missions. The algorithm terminated after 120 seconds of CPU runtime. This setting ensured sufficient runtime to find the optimal solution while preventing excessive computational time from impacting the timeliness of rescue decisions. Furthermore, the helicopter was loaded and unloaded at a rate of 2 passengers per minute, a parameter crucial for calculating the helicopter's dwell time at each node and the overall rescue scheduling schedule.

[0131] In order to verify the feasibility of the LNS algorithm, Python coding was used to implement LNS to solve the helicopter rescue dispatch problem in the rescue case, and the helicopter rescue dispatch solution was obtained as follows: Figure 6 shown. Figure 6 In the figure, the unnumbered light grey area indicates the time it takes for the helicopter to transfer between nodes; the dark grey rectangle indicates the process of the helicopter unloading the wounded and refueling at the hospital node; the black rectangle indicates the process of the helicopter loading the wounded in the disaster area. Figure 6 It can be seen that the helicopter rescue dispatching scheme obtained by the LNS method fully meets all kinds of constraints, which fully demonstrates the feasibility and effectiveness of the solution method.

[0132] During the execution of the algorithm, the convergence curve of the objective function value is as follows: Figure 7 The algorithm has completed convergence after 290 iterations. This result shows that the algorithm has good convergence and can find a better solution within a relatively small number of iterations, which improves the computational efficiency of the algorithm.

[0133] The curve of the destruction ratio in the “destruction” stage changing with the number of iterations is as follows: Figure 8 As shown in the figure, a larger destruction probability is selected at the beginning of the algorithm. This allows for a wide-scale global search, expanding the search space and increasing the likelihood of finding the global optimal solution. As the algorithm continues, the destruction probability adaptively decreases with execution time (number of iterations). This adjustment facilitates algorithm convergence and avoids excessive destruction of the already found optimal solution structure.

[0134] The probability curve of using the global repair strategy for repair in the "repair" stage is as follows: Figure 9As shown in the figure, at the beginning of the algorithm, a global repair strategy is executed with a high probability. This strategy helps improve the algorithm's global exploration capabilities, enabling it to search for potentially better solutions in a wider solution space. As the algorithm progresses, a local search is performed within the helicopter's operating path with a high probability. This accelerates the algorithm's convergence and allows for fine-tuning of the found better solutions.

[0135] By adaptively adjusting the damage ratio and the probability of executing the global repair strategy, the LNS algorithm achieves a good balance between exploration and exploitation during the optimization process. During the exploration phase, the algorithm actively searches for new solutions to avoid being trapped in local optima. During the exploitation phase, the algorithm focuses on optimizing and improving the already found optimal solutions, thereby improving the overall quality of the rescue dispatch plan. In summary, the LNS algorithm demonstrates good performance and adaptability for the helicopter rescue dispatch problem, providing effective decision support for helicopter dispatch in large-scale earthquake disaster relief.

[0136] The present application also provides an application scenario, which applies the above-mentioned helicopter rescue dispatching method based on the large neighborhood search algorithm. Specifically: The helicopter rescue dispatching method based on the large neighborhood search algorithm provided in this embodiment can be applied in large-scale natural disaster emergency rescue scenarios. Large-scale natural disaster emergency rescue scenarios include a disaster assessment link, a rescue resource allocation link, and an on-site rescue execution link; the disaster information enters the rescue resource allocation link from the disaster assessment link, and after comprehensive analysis and calculation of information such as the number of casualties, the distribution of disaster-stricken areas, and the available rescue helicopter resources, a helicopter rescue dispatching plan is obtained, and then enters the on-site rescue execution link.

[0137] The helicopter rescue dispatch method based on the large neighborhood search algorithm provided in this embodiment belongs to the rescue resource allocation phase of large-scale natural disaster emergency rescue. Specifically, in this phase, this method uses detailed disaster information transmitted from the disaster assessment phase, such as the geographic location of each affected area, the number of injured, and the severity of their injuries, combined with available helicopter resources (including the number of helicopters, performance parameters (speed, endurance, passenger capacity), and current location), to generate an optimal helicopter rescue dispatch plan using relevant optimization algorithms such as the LNS algorithm. This plan specifies each helicopter's flight route, the affected area to be addressed, and the timing for refueling and unloading patients at the hospital, ensuring that rescue resources are efficiently and rationally allocated to each affected area and maximizing the chance of saving lives.

[0138] Based on the same inventive concept, embodiments of the present application also provide a large neighborhood search algorithm-based helicopter rescue dispatch system for implementing the aforementioned large neighborhood search algorithm-based helicopter rescue dispatch method. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the large neighborhood search algorithm-based helicopter rescue dispatch system provided below can be found in the above-mentioned limitations of the large neighborhood search algorithm-based helicopter rescue dispatch method, and will not be further elaborated here.

[0139] In an exemplary embodiment, Figure 10 As shown, a helicopter rescue dispatch system based on a large neighborhood search algorithm is provided, including: The rescue data acquisition module 301 is used to obtain state parameter information of rescue entities; the rescue entities include helicopters, disaster-stricken areas, and hospitals.

[0140] The model building module 302 is used to build a helicopter rescue scheduling mixed integer programming model based on the state parameter information with the goal of minimizing the total time to complete the helicopter rescue.

[0141] The initial rescue scheduling plan generating module 303 is configured to generate an initial rescue scheduling plan based on the state parameter information by using a greedy algorithm.

[0142] The rescue scheduling plan generation module 304 is used to iteratively solve the helicopter rescue scheduling mixed integer programming model based on the initial rescue scheduling plan through a large neighborhood search algorithm until the iteration is terminated when the iteration stopping condition is met, thereby obtaining a rescue scheduling plan; the rescue scheduling plan includes the rescue mission sequence of each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity and helicopter flight time in each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time at each hospital; the rescue mission sequence includes the sequence of disaster-stricken areas and the sequence of hospitals that the helicopter needs to visit in sequence.

[0143] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 11As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store rescue task sequence processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a helicopter rescue dispatch method based on a large neighborhood search algorithm is implemented.

[0144] Those skilled in the art will understand that Figure 11 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.

[0145] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0146] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[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 understand 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 embodiments of the above-mentioned methods. Among them, 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. The database involved in the embodiments provided in this application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on blockchain, etc., but is not limited to this. The processor involved in the embodiments provided in this application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but is not limited to this.

[0149] 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.

[0150] 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 dispatching method based on a large neighborhood search algorithm, characterized in that: The helicopter rescue dispatching method based on the large neighborhood search algorithm includes: Obtaining status parameter information of rescue entities; the rescue entities include helicopters, disaster-stricken areas, and hospitals; Based on the state parameter information, a helicopter rescue scheduling mixed integer programming model is constructed with the goal of minimizing the total time required to complete the helicopter rescue; Based on the state parameter information, generating an initial rescue scheduling plan through a greedy algorithm; Based on the initial rescue scheduling plan, the helicopter rescue scheduling mixed integer programming model is iteratively solved by a large neighborhood search algorithm until the iteration stops when the iterative stopping condition is met, and a rescue scheduling plan is obtained; the rescue scheduling plan includes the rescue mission sequence of each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity and helicopter flight time in each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time at each hospital; the rescue mission sequence includes the sequence of disaster-stricken areas and hospitals that the helicopter needs to visit in sequence.

2. The helicopter rescue dispatching method based on the large neighborhood search algorithm according to claim 1 is characterized in that: The objective function of the helicopter rescue dispatch mixed integer programming model is: ； ； Where C represents the objective function value, represents the maximum rescue completion time of all rescue batches of the h-th helicopter; H represents the helicopter set; represents the rescue batch set of the h-th helicopter; r represents the hospital node passed by the h-th helicopter in the b-th rescue batch; represents the set of disaster-stricken area nodes and hospital nodes that the h-th helicopter has visited in the b-th rescue batch; R represents the set of hospital nodes; It represents the rescue completion time when the hth helicopter of the bth rescue batch returns to hospital r and unloads the wounded.

3. The helicopter rescue dispatching method based on the large neighborhood search algorithm according to claim 1 is characterized in that: The constraints of the mixed integer programming for helicopter rescue scheduling include: rescue batch constraints for the wounded in the disaster area, helicopter loading capacity constraints for the wounded, helicopter flight time constraints, helicopter operation time sequence constraints, and hospital rescue capacity constraints for the wounded.

4. The helicopter rescue dispatching method based on the large neighborhood search algorithm according to claim 1 is characterized in that: Based on the initial rescue scheduling plan, the helicopter rescue scheduling mixed integer programming model is iteratively solved by the large neighborhood search algorithm until the iteration stops when the iterative stopping condition is met. The rescue scheduling plan is obtained, which specifically includes: The objective function value corresponding to the initial rescue dispatch planning scheme is taken as the current optimal solution; Randomly selecting a preset number of nodes to be removed from each helicopter rescue mission sequence in the initial rescue scheduling plan to obtain multiple sets of removed nodes; the nodes include disaster-stricken area nodes and hospital nodes; Removing the nodes in the removed node set from the corresponding helicopter rescue mission sequence respectively to obtain multiple helicopter rescue partial scheduling sequences; Based on the set of removed nodes, a probabilistic selection strategy method is used to select a repair strategy to repair the helicopter rescue scheduling sequence and obtain the current repair scheduling plan; the repair strategy includes a local repair strategy and a global repair strategy; If the objective function value corresponding to the current repair scheduling plan is greater than or equal to the objective function value corresponding to the current optimal solution, the initial rescue scheduling plan is updated through the annealing mechanism; If the objective function value corresponding to the current repair scheduling plan is less than the objective function value corresponding to the current optimal solution, the initial rescue scheduling plan is updated using the current repair scheduling plan; If the iteration stopping conditions are met, the currently updated initial rescue scheduling plan is used as the rescue scheduling plan; the iteration stopping conditions include reaching a preset maximum number of iterations, or the change in the objective function value corresponding to the current optimal solution within a preset number of consecutive times is less than a preset change threshold, or the algorithm running time exceeds a preset running time threshold; If the iteration stopping condition is not met, return to step "taking the objective function value corresponding to the initial task planning scheme as the current optimal solution".

5. The helicopter rescue dispatching method based on the large neighborhood search algorithm according to claim 4 is characterized in that: Local repair strategies include: Perform multiple random exchange operations on the positions of nodes in the removed node set to obtain multiple new removed node sets; Traversing all new removal node sets respectively, if the current node in the current new removal node set is a disaster area node, inserting the current node into the first preset position in the corresponding helicopter rescue part scheduling sequence; the first preset position is the position that minimizes the increase in the air operation time of the corresponding helicopter rescue part scheduling sequence; If the current node in the newly removed node set is a hospital node, a hospital node is randomly selected in the corresponding helicopter rescue part scheduling sequence to be exchanged with the current node; Until all new removed node sets are traversed, a repair scheduling plan is obtained.

6. The helicopter rescue dispatching method based on the large neighborhood search algorithm according to claim 4 is characterized in that: Global repair strategy, including: Perform multiple random swap operations on the positions of the nodes in the removed node set and integrate them to obtain an integrated removed node set; Traverse the set of nodes for integration and removal. If the current node is a node in the disaster-stricken area, insert the current node into a second preset position; the second preset position is the corresponding position in the helicopter rescue scheduling sequence that minimizes the increase in the duration of the aerial operation. If the current node is a hospital node, randomly select any hospital node in the helicopter rescue scheduling sequence and exchange it with the current node; Until the traversal of the integrated removed node set is completed, the repair scheduling plan is obtained.

7. A helicopter rescue dispatch system based on a large neighborhood search algorithm, characterized in that: The helicopter rescue dispatch system based on the large neighborhood search algorithm includes: A rescue data acquisition module is used to obtain status parameter information of rescue entities; the rescue entities include helicopters, disaster-stricken areas, and hospitals; A model building module is used to build a helicopter rescue scheduling mixed integer programming model based on the state parameter information with the goal of minimizing the total time to complete the helicopter rescue; An initial rescue scheduling plan generating module, configured to generate an initial rescue scheduling plan based on the state parameter information through a greedy algorithm; The rescue scheduling plan generation module is used to iteratively solve the helicopter rescue scheduling mixed integer programming model based on the initial rescue scheduling plan through a large neighborhood search algorithm until the iteration is terminated when the iteration stopping condition is met, thereby obtaining a rescue scheduling plan; the rescue scheduling plan includes the rescue mission sequence of each helicopter; the helicopter rescue start time, helicopter rescue end time, helicopter rescue capacity and helicopter flight time in each disaster-stricken area; the time taken to unload the wounded and the helicopter refueling time at each hospital; the rescue mission sequence includes the sequence of disaster-stricken areas and the sequence of hospitals that the helicopter needs to visit in sequence.

8. 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 dispatch method based on the large neighborhood search algorithm according to any one of claims 1 to 6.

9. 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 dispatching method based on a large neighborhood search algorithm described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the helicopter rescue dispatching method based on a large neighborhood search algorithm described in any one of claims 1 to 6 is implemented.

Images