Method, system and equipment for dynamically scheduling ship helicopters based on tabu search

Abstract

The invention belongs to the technical field of task planning, and particularly relates to a taboo search-based ship helicopter dynamic scheduling method, system and equipment. The method comprises the following steps: calling a constructed ship helicopter operation task planning model; designing a dynamic scheduling strategy by adopting event demand simulation and generating a task coding initial value; carrying out optimization solution on the helicopter task coding sequence by adopting a dynamic scheduling algorithm based on tabu search; and outputting task dynamic planning scheme data. According to the method, a reference basis is provided for dynamic adjustment of a warship helicopter combat scheme in a high-intensity confrontation environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of mission planning, and in particular relates to a method, system and device for dynamic scheduling of ship helicopters based on taboo search. Background Art

[0002] In landing operations, ship-based helicopters that use amphibious assault ships and other forward bases as take-off and landing platforms can rely on flexible deployment, concealed and fast tactical characteristics to surpass the obstacles, minefields and solid fortresses set up by the anti-landing party on the waterside beach, deploy land combat forces in the shallow depth of the landing site and implement close-range ground fire support, coordinate with the plane landing forces to form a local advantage, and achieve the suddenness and rapidity of the landing operation. But at the same time, in actual amphibious landing operations, the high-intensity confrontation environment will inevitably lead to uncertainty in the execution of ship-based helicopter operation tasks, which requires the command system to dynamically dispatch the task plan in a timely manner.

[0003] At present, the research on dynamic scheduling at home and abroad mainly focuses on dynamic vehicle routing problems and dynamic traveling salesman problems. There is still a lack of research on aircraft operation task planning and aircraft flight task rescheduling in the context of amphibious operations. Similar to task planning, the solution methods for dynamic scheduling problems also include two categories: exact solution and heuristic algorithm. Among them, exact solution includes branch and bound method, column generation algorithm, adaptive dynamic programming algorithm, etc. However, these algorithms cannot avoid the exponential explosion problem and are only suitable for solving small-scale problems. Heuristic algorithms include taboo search algorithm, neighborhood search algorithm, ant colony algorithm, genetic algorithm and particle swarm algorithm. Existing research on the dynamic planning of ship helicopter operation tasks in the context of amphibious landing has the technical problem of being unable to accurately plan tasks in response to dynamic environmental interference. Summary of the invention

[0004] In view of the problems existing in the above-mentioned traditional methods, the present invention proposes a ship helicopter dynamic scheduling method based on taboo search, a ship helicopter dynamic scheduling system based on taboo search and a computer device, which can accurately plan tasks according to dynamic environmental interference.

[0005] One of the purposes of the present invention is to provide a method for dynamic scheduling of ship helicopters based on taboo search, which is characterized in that it includes the following steps:

[0006] Step 1: Call the constructed ship helicopter operation task planning model;

[0007] Step 2: Use event demand simulation to design a dynamic scheduling strategy and generate the initial value of the task code;

[0008] Step 3: Use a dynamic scheduling algorithm based on taboo search to optimize and solve the helicopter mission coding sequence;

[0009] Step 4: Output the dynamic planning data of the ship helicopter mission.

[0010] Preferably, the constraints of the ship helicopter operation task planning model are divided into two categories: available combat resource constraints and combat phase timing constraints. The available combat resource constraints are further divided into vertical landing team quantity constraints, ship helicopter quantity constraints, and LHD support resource constraints. The task phase timing constraints are divided into task phase timing constraints based on execution order and task phase timing constraints based on LHD deck scheduling.

[0011] The optimization objective functions of the ship helicopter operation mission planning model include: minimizing landing time, minimizing troop loss and minimizing helicopter risk.

[0012] Preferably, the event demand simulation proposes a dynamic scheduling strategy for ship helicopter combat missions, including a scheduling strategy based on a current wave window and a scheduling strategy based on an adjacent wave window.

[0013] Preferably, the process of generating the initial value of the task code includes:

[0014] Determine the selectable dynamic scheduling strategy based on the timing window of sudden demand;

[0015] Based on the generated ship helicopter operation task plan code, the optimization initial value of the corresponding scheduling algorithm is generated according to different dynamic scheduling strategies.

[0016] Preferably, the step of optimizing and solving the helicopter mission coding sequence by using a dynamic scheduling algorithm based on taboo search comprises:

[0017] The first step is to input the initial code and calculate the corresponding initial solution, which are set as the optimal code and the optimal solution respectively;

[0018] The second step is to select a gene position after the corresponding code of the damaged helicopter group, without changing its mission type segment, and randomly assign a value to its target sequence segment;

[0019] The third step is to determine the Pareto dominance relationship between the current solution and the optimal solution;

[0020] The fourth step is to determine whether the predetermined number of iterations has been reached; if so, exit the loop and return to the dominant solution library; if not, go to any gene position after the corresponding code of the damaged helicopter formation, do not change its mission type fragment, and randomly assign a value to its target sequence number fragment.

[0021] The second object of the present invention is to provide a ship helicopter dynamic scheduling system based on taboo search, which is special in that it includes:

[0022] The model calling module is used to call the constructed ship helicopter operation task planning model;

[0023] Strategy initialization module, used to design dynamic scheduling strategy by event demand simulation and generate task coding initialization;

[0024] A search optimization module is used to optimize and solve the helicopter mission coding sequence using a dynamic scheduling algorithm based on taboo search;

[0025] The planning output module is used to output ship mission planning data.

[0026] The third object of the present invention is to provide a computer device, which is special in that it includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above-mentioned taboo search-based ship helicopter dynamic scheduling method when executing the computer program.

[0027] The present invention's method, system and device for dynamic scheduling of ship helicopters based on taboo search are designed to address the uncertainty and dynamic characteristics of the ship helicopter mission execution process under the background of amphibious landing, and a strategy-based dynamic scheduling optimization method is designed. First, the typical mission modes and characteristics of ship helicopters in landing campaigns are analyzed, and the specific constraints of the available combat resources of multiple platforms and the timing of the full-link mission stages of combat and support are considered to construct a mathematical model for dynamic scheduling. Then, a dynamic scheduling strategy is designed based on the idea of ​​disturbance event simulation, and the encoding initial value of the iterative algorithm is constructed on this basis. Based on the taboo search algorithm for individual iterative optimization, the constructed initial value is optimized and solved. It provides a reference for the dynamic adjustment of ship helicopter combat plans in high-intensity confrontation environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the conventional technology, the drawings required for use in the embodiments or the conventional technology descriptions are briefly introduced below. 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.

[0029] Figure 1 A schematic diagram is provided for describing the phases of the ship helicopter mission process;

[0030] Figure 2 This is a schematic diagram of a chromosome coding case;

[0031] Figure 3 This is a schematic diagram of a helicopter dispatch plan list example;

[0032] Figure 4 This is a schematic diagram of a chromosome coding case;

[0033] Figure 5 It is a schematic diagram of the optimization initial value corresponding to the scheduling strategy;

[0034] Figure 6 This is a schematic diagram of a landing battle scenario;

[0035] Figure 7 Schematic diagram of the Pareto frontier solution set obtained by algorithm iteration;

[0036] Figure 8 Schematic diagram of the algorithm iteration results under different encoding initial value conditions. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] Example 1

[0039] In view of the shortcomings of existing research on dynamic planning of ship-to-ship helicopter operation tasks in the context of amphibious landing, this embodiment proposes a strategy-based dynamic scheduling optimization method, including the following processing steps:

[0040] Step 1: Call the constructed ship helicopter operation task planning model;

[0041] Step 2: Use event demand simulation to design a dynamic scheduling strategy and generate the initial value of the task code;

[0042] Step 3: Use a dynamic scheduling algorithm based on taboo search to optimize and solve the helicopter mission coding sequence;

[0043] Step 4: Output the dynamic planning data of the ship helicopter mission.

[0044] Firstly, a mathematical model of dynamic programming is constructed to analyze the constraints and objectives of dynamic scheduling of ship helicopter missions. According to the proposed rescheduling strategy for disturbance events, the iterative initial value of the dynamic scheduling solution algorithm is constructed, and the model is solved using the taboo search algorithm based on individual evolution. Finally, a case study of the optimization results of dynamic scheduling is given in combination with a battalion-scale beach landing campaign scenario.

[0045] Event-driven dynamic scheduling strategy

[0046] Similar to the dynamic vehicle routing problem (DVRP), the dynamic planning of ship-helicopter operation tasks is mainly used to deal with sudden battlefield demands during landing operations. In the embodiment, the timing of sudden battlefield demands is collectively referred to as demand windows. Accordingly, according to the idea of ​​demand event simulation, the dynamic scheduling strategy of ship-helicopter operation tasks is determined according to the attributes of battlefield demands and the timing of demand windows. According to the different timings of sudden battlefield demands in landing operations, the demand windows can be summarized into two categories: current wave scheduling windows (referred to as current wave windows) and adjacent wave scheduling windows (referred to as adjacent wave windows).

[0047] 1. Scheduling strategy based on the current wave window

[0048] like Figure 1 The figure shows the subtask phase division of the helicopter formation of the ship in the wth wave on an amphibious assault ship after take-off and the subtask phase division of the helicopter formation of the ship in the w-1th wave on the deck for re-deployment preparation after landing. If the sudden battlefield demand caused by some reason occurs in the Class I stage between the time when the helicopter formation starts to take off and the time when it is put into battle, that is, for the helicopter formation of the ship in the wth wave, the time when the demand window appears is T d satisfy:

[0049]

[0050] Then the demand window is considered as the current wave window.

[0051] In the formula, Indicates when The corresponding start time of each sub-phase of the ship helicopter combat mission, They correspond to the K stages of a complete combat mission respectively; among them, It represents the number of helicopter groups dispatched from the wth wave of the Red Army on the lth LHD to perform the cth type combat mission against the pth target of the Blue Army. When c = 1, When c = 2, N represents a natural number; w∈{1,…,W}, represents the wave number of the Red side’s helicopter mission formation plan; l∈{1,…,L}, represents the Red side’s amphibious assault ship number; p∈{1,…,P}, represents the Blue side’s target number that the Red side plans to attack; c∈{1,2}, represents the Red side’s helicopter formation mission type, c=1 represents a vertical landing mission, and c=2 represents a close fire support mission.

[0052] When sudden battlefield demands appear in the current wave window, since the current wave of ship-helicopter formations have not yet started the task implementation phase of vertical force projection or close fire support tasks, the Red Army can choose to dispatch the available helicopter formations in the current wave to perform support tasks for the nodes where the demands occur. Therefore, for sudden battlefield demands that appear in the current wave window, the dynamic scheduling strategies for the Red Army's ship-helicopter operation tasks can be roughly divided into two categories: mobilizing the current wave of helicopter formations to respond to demands and not responding to demands.

[0053] 2. Scheduling strategy based on adjacent wave windows

[0054] Similar to the judgment criteria of the current wave window, if the sudden battlefield demand caused by some reason occurs in the Class II stage between the moment when the helicopter formation dispatched in the current wave starts the mission implementation phase and the moment when the helicopter formation expected to be dispatched in the next wave starts to take off, that is, the moment when the battlefield demand window appears, T d satisfy

[0055]

[0056] Then the demand window is considered to be an adjacent wave window.

[0057] In the formula, Indicates when The corresponding start time of each sub-phase of the ship helicopter combat mission, They correspond to the K stages of a complete combat mission respectively; among them, It represents the number of helicopter groups dispatched from the wth wave of the Red Army on the lth LHD to perform the cth type combat mission against the pth target of the Blue Army. When c = 1, When c = 2, N represents a natural number; w∈{1,…,W}, represents the wave number of the Red side’s helicopter mission formation plan; l∈{1,…,L}, represents the Red side’s amphibious assault ship number; p∈{1,…,P}, represents the Blue side’s target number that the Red side plans to attack; c∈{1,2}, represents the Red side’s helicopter formation mission type, c=1 represents a vertical landing mission, and c=2 represents a close fire support mission.

[0058] When sudden battlefield demands appear in the near-wave window, since the current wave of ship-helicopter formations have already started their respective mission implementation phases, the Red Army can only choose to dispatch the next wave of ship-helicopter formations to perform support tasks for the node where the demand occurs. Therefore, for sudden battlefield demands that appear in the near-wave window, the Red Army's corresponding dynamic scheduling strategies can be roughly divided into two categories: calling the next wave of ship-helicopter combat resources to respond to the demand and not responding to the demand.

[0059] For example, if a ship-to-ship helicopter formation performing a vertical landing mission has several general-purpose helicopters shot down by enemy anti-aircraft firepower during the ultra-low-altitude flight phase in the beach area, this emergency will cause the mission execution capability of the above formation to be damaged, and then generate support requirements for the formation, that is, the need to dispatch other helicopter formations to launch vertical landing or close-range fire support missions to the attack nodes scheduled by the damaged formation. Since the above-mentioned sudden battlefield requirements appear in the current wave window, the Red Army can choose to dispatch other helicopter formations in the current wave to launch vertical landing or close-range fire support missions to the required nodes, or choose to ignore the support requirements for the nodes.

[0060] In addition, in the case where the helicopter of the vertical landing mission formation is shot down, the remaining helicopters of the formation choose to continue to perform the original mission or abandon the mission and withdraw from the battlefield, which will also affect the optimization initial value of the mission code as part of the red side's decision.

[0061] Step 1: Model construction

[0062] Constraint 1. Combat Resource Constraint

[0063] Constraint 1-1: Vertical landing team quantity constraint

[0064] In order to simultaneously deploy more vertical landing teams in one wave, the vertical landing teams should be dispersed and deployed on different LHDs. Since the total number of vertical landing teams that the Red Army can deploy in the landing campaign is limited, the number of vertical landing missions that each LHD can support must also meet the limit on the number of vertical landing teams loaded on the LHD. Specifically, The following constraints should be met

[0065]

[0066] In the formula, It represents the number of helicopter groups dispatched from the wth wave of the Red Army on the lth LHD to perform the cth type combat mission against the pth target of the Blue Army. When c = 1, When c = 2, N represents a natural number; w∈{1,…,W}, represents the wave number of the Red side helicopter mission formation plan; l∈{1,…,L}, represents the Red side amphibious assault ship number; p∈{1,…,P}, represents the Blue side target number that the Red side plans to attack; c∈{1,2}, represents the Red side helicopter formation mission type, c=1 represents the vertical landing mission, c=2 represents the close range fire support mission; represents the number of the g-th type vertical landing team sent by the Red side to carry out the vertical landing mission on the p-th target of the Blue side; u lgrepresents the number of the g-th type of vertical landing team carried on the Red side’s l-th amphibious assault ship; g∈{1,…,G} represents the type of the Red side’s vertical landing team.

[0067] At the same time, the total number of all vertical landing teams planned to be deployed in all vertical landing missions in a landing campaign should also meet the following constraints:

[0068]

[0069] In the formula, It represents the number of helicopter groups dispatched from the wth wave of the Red Army on the lth LHD to perform the cth type combat mission against the pth target of the Blue Army. When c = 1, When c = 2, N represents a natural number; w∈{1,…,W}, represents the wave number of the Red side helicopter mission formation plan; l∈{1,…,L}, represents the Red side amphibious assault ship number; p∈{1,…,P}, represents the Blue side target number that the Red side plans to attack; c∈{1,2}, represents the Red side helicopter formation mission type, c=1 represents the vertical landing mission, c=2 represents the close range fire support mission; represents the number of the g-th type vertical landing team sent by the Red side to carry out the vertical landing mission on the p-th target of the Blue side; u g represents the total number of g-th vertical landing teams that the Red Army can dispatch in a landing campaign; g∈{1,…,G} represents the category of the Red Army’s vertical landing teams.

[0070] Constraint 1-2: Constraint on the number of helicopters on ships

[0071] The total number of helicopters of a certain type required for the helicopter mission formations of adjacent waves on a certain LHD is limited by the total number of helicopters of this type carried by the LHD:

[0072]

[0073] In the formula, Indicates the total number of general-purpose helicopters loaded on the Red Side's first amphibious assault ship, Indicates the total number of armed helicopters loaded on the Red Army's first amphibious assault ship; the number of armed helicopters required for each individual mission and the number of general-purpose helicopters The ratio between them is determined by the mission type c. When c = 2, the helicopter group performs close fire support missions, and the mission group is composed only of armed helicopters. Therefore, the number of general helicopters When c = 1, the helicopter formation performs a vertical landing mission. The mission formation is composed of general-purpose helicopters and armed helicopters that provide accompanying cover. Usually, the number of armed helicopters is Should be no less than the number of general-purpose helicopters 1 / 2 of , so this article assumes:

[0074]

[0075] In the above formula Indicates that x is rounded up. In addition, the number of general helicopters required for each vertical landing mission is Depends on the number of various vertical landing teams required for the air assault on the Blue's pth node in the Red Army's pre-war staff plan

[0076]

[0077] In the formula, It indicates the number of general-purpose helicopters required by the Red Army to deploy the G-type vertical landing team.

[0078] Constraints 1-3: LHD security resource constraints

[0079] Limited by the number of take-off stands on the flight deck of each LHD, the number of helicopters in the helicopter mission formation of the same wave should meet the following formula:

[0080]

[0081] In the formula, It represents the total number of helicopter take-off slots that can be provided by all amphibious assault ships of the Red Army in the wth sortie wave.

[0082] In addition, helicopter groups assigned to the same task should be dispatched in the same wave or adjacent waves as much as possible, so in the formula The constraints should be satisfied:

[0083]

[0084] In the above formula, n w It represents the minimum number of waves required for all helicopter formations assigned to the c-th type of operation mission for target p to be fully deployed, that is, the sum of the number of two types of helicopters required by the mission formation divided by the total take-off positions of all LHDs of the Red Force, rounded up to the integer of the quotient.

[0085] Constraint 2: Task phase timing constraints

[0086] Constraint 2-1: Task phase timing constraints based on execution order

[0087] A complete ship helicopter operation mission consists of 9 parts: deck transportation phase, maintenance service support phase, take-off and departure phase, formation flight phase from LHD to coastline (cruising), formation flight phase from coastline to target node (ultra-low altitude penetration), mission execution phase, formation flight phase from target node to coastline (ultra-low altitude penetration), formation flight from coastline to LHD (cruising) and approach and landing phase. It is easy to know that in each independent ship helicopter operation mission, the start time and end time of each phase satisfy the following constraints:

[0088]

[0089] In the formula, and The following relationship is satisfied:

[0090]

[0091] In the formula, Indicates when The corresponding start time of each sub-phase of the ship helicopter combat mission, They correspond to the K stages of a complete combat mission respectively; Indicates when The corresponding end time of each sub-stage of the ship helicopter combat mission. The corresponding stage and same; It represents the time consumed by the Red Army's LHD in the wth wave to perform the cth type of operation task on the pth node target of the landing site. Among them, the time consumed by the shipboard transportation stage, the maintenance service support stage, the take-off and departure stage, and the approach and landing stage is an assumed fixed value, and the task execution phase takes time Depending on the mission type, the helicopter formation flight time It is obtained by dividing the planned optimal route length by the helicopter formation flight speed.

[0092] It is worth noting that although each ship helicopter operation task in this paper includes 9 sub-stages, not all helicopter formations assigned to the operation task will go through these 9 stages completely. The exception occurs in the first stage of shipboard transportation. If all the task formations of the wth and w+2th waves of a certain LHD need the same total number of helicopters of various types, then the helicopters of the wth wave can directly perform maintenance services at the take-off position after landing, and then directly take off to form the w+2th wave of task formations to perform the operation task. Therefore, the formula (11) Satisfy the following formula.

[0093]

[0094] Where T1 represents the assumed time required for shipboard transportation.

[0095] Constraint 2-2: Task phase timing constraints based on LHD deck scheduling

[0096] Due to the limited space on the LHD flight deck, the maintenance and service support of ship helicopters on the LHD usually occupies the take-off position, resulting in the ship helicopter formation preparing to land must wait until the LHD flight deck is cleared (that is, after the next wave of helicopters takes off) before landing. Therefore, for the ship helicopter formation dispatched in two adjacent waves on the same LHD, their take-off and landing times must meet the following constraints:

[0097]

[0098] In the formula, It indicates the landing time of the helicopter formation of the Red Army's LHD, which is dispatched in the w′th wave to perform the cth type of operation mission to the pth node target at the landing site. It indicates the take-off time of the ship-helicopter formation that performs the same mission and is dispatched in the w″th wave on the same LHD.

[0099] The optimization objective function includes:

[0100] Minimize the time spent on landing: In an amphibious landing campaign, the defending side usually organizes mobile forces to counterattack when the attacking side is not yet firmly established. Whether the landing site can be quickly seized and effectively consolidated before the defending side launches a counterattack is an important factor in determining the success or failure of a landing campaign. Therefore, the model selects the total time spent by all combat units of the Red side to complete all the established tasks in the stage of opening up the landing site as the objective function.

[0101] z1=T fin -T0 (14)

[0102] Where T fin It indicates the end of the landing phase, that is, the moment when the first wave of landing troops of the Red Army completes the tasks required to open up and consolidate the beach landing site; T0 corresponds to the moment when the landing campaign is initiated, that is, the moment when the first wave of ships and helicopters of the Red Army begin to take off.

[0103] Minimize the loss of troops: The loss of troops of the Red Army includes the loss of troops for horizontal landing and the loss of troops for vertical landing, which can be specifically expressed as:

[0104]

[0105] In the formula, It represents the number of troops lost per unit time by all the Red troops fighting at the pth Blue node at time t. Indicates the start time of the battle on the pth blue node, T p Indicates the duration of the engagement in minutes.

[0106] This embodiment uses the Lanchester equation to calculate the loss of troops of the red and blue sides. When both sides use direct-aiming weapons such as rifles, small-caliber artillery, and anti-tank missiles, the Lanchester square law can be used to calculate the loss of troops of the warring sides:

[0107]

[0108] In the formula, r and b represent the number of combat units of the red and blue sides destroyed by each combat unit of the other side per unit time.

[0109] When one of the two warring parties uses indirect weapons such as howitzers, the casualties of the other party can be calculated using the Lanchester second linear rate.

[0110]

[0111] Where ξ represents the number of red units per unit area destroyed by each indirect combat unit of the blue side per unit time.

[0112] However, in actual combat, it is often difficult for the warring parties to maintain an ideal ISR level. Therefore, considering the limited situational awareness capabilities of the Red Army and the inability of the troops to be fully deployed, the above formula is modified.

[0113]

[0114] Where α′ represents the strength vector of the maximum force that the Red Army can deploy under the constraints of tactics, terrain, etc.; β0 represents the initial strength vector of the Blue Army; ε∈[0,1] represents the proportion of the Blue Army that can be accurately detected by the Red Army. When ε=1, the right side of the equation is similar to the Lanchester square law, and when ε=0, the right side of the equation is similar to the Lanchester second linear law; Φ and Ψ are both non-negative matrices whose column sums do not exceed 1, which are called the firepower allocation matrices of the warring parties.

[0115] In order to avoid the interference of tactical factors of the warring parties on Ψ and Φ, this paper assumes that both sides adopt the optimal strategy in the case of multi-arms combat. Under this strategy, each column of Ψ and Φ has only one element that is 1.

[0116] Minimizing Helicopter Risks:

[0117] To avoid making the model too complicated, this paper does not consider the impact of the Red Army helicopter being shot down on the two operational task force formations and the subsequent available helicopters on the amphibious assault ship, but only uses the cumulative threat of the Blue Army's air defense firepower during the helicopter's voyage to quantify the loss of the helicopter.

[0118]

[0119] In the formula, k Scale represents the proportionality coefficient, and They represent the formation assembly time after the helicopter formation takes off from the LHD deck and the time when it begins to approach and land after returning from the combat airspace. E(x,y,t) represents the sum of the ground air defense fire threat values ​​of the blue side to each helicopter of the red side at the coordinate (x,y) at time t. This paper simplifies the calculation of E(x,y,t) appropriately, considering only the air defense fire coefficient at each node of the blue side, the distance between the red helicopter and the node, and the shielding effect of the terrain on the blue side's air defense fire. E(x,y,t) can be calculated according to the formula.

[0120]

[0121] Where η SAM A Boolean variable indicating whether the node p and the coordinate (x, y) are blocked by the terrain. It is 1 if it is not blocked and 0 if it is blocked. represents the anti-aircraft firepower coefficient at the blue side's node p; D(x,y,p) represents the plane distance between the node p and the coordinate (x,y). The position (x,y) of each helicopter of the red side at time t is obtained by the path planning solution based on the Dijkstra algorithm in the decoding operation.

[0122] Step 2: Dynamic scheduling algorithm based on taboo search

[0123] This embodiment adopts the idea of ​​demand event simulation to design a dynamic scheduling strategy and generate initial values ​​of task codes. On this basis, a dynamic scheduling algorithm based on taboo search is designed to optimize the helicopter task code sequence.

[0124] Encoding: Figure 2 As shown, all satisfied The task is decomposed into A list of individual tasks, encoded as discrete two-row chromosomes:

[0125]

[0126] Where σ i represents the chromosome encoding gene, the first line c and the second line p represent the task type and target number respectively; Represents the chromosome length, that is, the total number of helicopter missions.

[0127] According to the constraints in formula (5), if the total number of a certain type of helicopters on the LHD is not enough to ensure the continuous call of two adjacent waves of operation tasks, the limited ship helicopter resources must be allocated to one wave, and the other wave can only arrange tasks that require other types of helicopters or no tasks. Add vacant genes to indicate the idle state of some takeoff positions of a certain LHD in a certain wave. To represent this idle state, it is assumed that the number of idle take-off positions is the greatest common factor of the number of helicopters required for all operational missions.

[0128] The neighborhoods where the genes in the chromosome code are located are divided into two categories: the genes corresponding to the vertical landing missions that require a variable number of helicopters (hereinafter collectively referred to as Class 1 missions) are located in the Class 1 neighborhood, and the genes corresponding to the close fire support missions that require a fixed number of helicopters (hereinafter collectively referred to as Class 2 missions) are located in the Class 2 neighborhood. It is easy to see that the change in chromosome length is caused by the change in Class 1 missions in the code.

[0129] Decoding: The input parameter of the decoding operation is a discrete task list The decoding process is as follows: First, the discrete code representing the task sequence is Converted into a sortie list representing the helicopter sortie plan for each ship pass List of the launch times of the vertical landing missions and close fire support missions of the Red Army against the Blue Army's targets Combined with the predetermined combat plan of the plane landing force, the improved Lanchester model is used to calculate the change of the forces of the two warring parties at each target node of the landing site over time, and then the landing site node state vector Δ is obtained. The landing site situation transition model is used to deduce the direction of the landing campaign, and finally determine whether the Red Army can finally achieve the campaign objectives and the set of operational task effectiveness indicators.

[0130] exist and The first step in the solution phase is to calculate The total number of Airborne Landing Units (ALUs) deployed for each vertical landing mission u i , and the number of helicopters required for each vertical landing mission and close fire support mission h i (Lines 1-2); Step 2, according to u i and h i Select Job Task List The type 1 task that satisfies the constraints of formulas (3)-(9) is given by formula (22): Assign values ​​to the elements in the list and add the executed tasks to the list in the order of the original list In the third step, all the tasks of type 1 that meet the constraints are inserted Then, search according to the same principle as step 2. The two types of tasks that meet the constraints in and

[0131] Assuming the Red side has 3 LHDs and sufficient ships and helicopters, and dispatches 3 waves in total, then Figure 3 Chromosomal coding in Can be transcribed as Figure 3 Ship helicopter sortie schedule shown is a w line, A matrix of columns, It represents the task performed by the ship helicopter taking off from the jth take-off position of the lth LHD in the wth wave. Its value is shown in formula (22).

[0132]

[0133] exist The first step in the solution phase is to use the helicopter dispatch plan The goal of each task in i , and combined with the known or assumed distribution of enemy air defense firepower at the landing site, the Dijkstra algorithm is used to plan the helicopter to each mission target p i The set of routes with the least risk The second step is to find the starting time of the task execution phase of each task (for the vertical landing task, it is the time when the vertical landing team starts to land, and for the close fire support task, it is the time when the armed helicopter starts to attack the ground)

[0134] exist In the first step of the solution phase, it is assumed that at the time T0 when the landing campaign is launched, The landing site node state vector Δ(p) = 0; the second step is to solve the risk of each helicopter formation mission of the Red Army during the flight process; the third step is to calculate the time when each ship helicopter mission is initiated Combined with the predetermined combat plan of the plane landing force, the landing site nodes where fighting is taking place at each simulation moment are found, and the force loss and combat time of the two sides at each node are calculated using formulas (14)-(20), and t stepThe strength vector of the Red and Blue forces at each target node is numerically integrated for the simulation step length until Δ meets the predetermined target of the Red landing campaign in the landing site opening stage or the Red side determines that it cannot achieve the campaign goal; finally, the total time consumed in the landing site opening stage z1, the cumulative force loss z2 of the Red side's vertical landing team and plane landing force, and the risk z3 borne by all ships and helicopters performing operational tasks are calculated to form the target vector

[0135] Step 3: Generate optimization initial value based on scheduling strategy

[0136] From the previous analysis, we can know that the scheduling strategy that the Red Army can choose can be determined according to the timing window of sudden battlefield demand. In this section, we will adopt the idea of ​​demand event simulation, based on the generated ship helicopter operation task plan coding, and generate the corresponding optimization initial value of the scheduling algorithm according to different dynamic scheduling strategies.

[0137] like Figure 4 As shown in the figure, assuming that in the ship helicopter operation task plan corresponding to the chromosome encoding, the helicopter group corresponding to gene σ1 has two general helicopters and one armed helicopter shot down by the blue side's pre-hidden air defense missiles during the ultra-low altitude flight to the blue side's target node, then according to the previous conclusion, Figure 4 The emergency described in will lead to the need for reinforcement of the blue side's 6# node, and this need occurs in the current wave window. It can be seen that the red side faces two decisions: whether to continue the mission execution of the damaged wave and whether to dispatch the available ship and helicopter formations of the current wave or the next wave to support the damaged formation. The strategy combination results of the red side in the above two dimensions are as follows: Figure 5 shown.

[0138] In addition, considering the situation where sudden battlefield demands appear near the wave window, the strategies that the Red side can choose are the same as Figure 5 (5) is similar to (6), that is, since there is no helicopter formation available in the current wave, the Red side can only choose to dispatch the next wave of helicopter formations to support the demand node.

[0139] Step 4: Design a dynamic scheduling algorithm based on taboo search

[0140] The taboo search algorithm avoids circuitous search by introducing an additional storage structure and corresponding taboo criteria, and has higher search efficiency. Compared with intelligent optimization algorithms such as genetic algorithms based on population evolution, the taboo search algorithm based on individual evolution requires less computing resources and can obtain optimization results more quickly.

[0141] In order to further optimize the initial encoding built based on the above scheduling strategy, this section designs a dynamic scheduling algorithm based on taboo search and introduces decentralized search and centralized search strategies.

[0142] Step 4-1: Input the initial code, calculate the corresponding initial solution, and set them as the optimal code and optimal solution respectively.

[0143] Step 4-2, select a gene position after the corresponding code of the damaged helicopter group, do not change its mission type segment, and randomly assign a value to its target sequence segment. The above operation is called the current neighborhood operation, including the selected gene sequence and the target sequence value assigned to the gene target sequence segment. The updated chromosome code is called the current code, and the corresponding target value is called the current solution.

[0144] Step 4-3: Determine the Pareto dominance relationship between the current solution and the optimal solution.

[0145] If the current solution dominates the optimal solution, the optimal code is replaced by the current code, the optimal solution is replaced by the current solution, and a centralized search operation is triggered. According to the centralized search operation, the taboo list and the dominant solution library are cleared, the current operation is inserted into the taboo list, and the updated optimal code and optimal solution are inserted into the dominant solution library.

[0146] If the current solution is dominated by the optimal solution, the scatter search counter is first accumulated, and then the taboo list is further determined to be full. If the taboo list is not full, the current operation is inserted into the taboo list; if the taboo list is full, no operation is performed.

[0147] If the current solution and the optimal solution cannot dominate each other, first accumulate the scattered search counter, insert the current code and the current solution into the dominated solution library, and then further determine whether the taboo list is full. If the taboo list is not full, insert the current operation into the taboo list; if the taboo list is full, further determine whether the current operation has appeared in the taboo list. If not, insert the current operation from the bottom of the taboo list and delete the first element in the list; if yes, do not perform any operation.

[0148] Determine whether the scatter search counter reaches the scatter search threshold. If so, reset the counter to zero and change the task type fragment of the selected gene in the next neighborhood operation; if not, do not perform the operation.

[0149] Step 4-4: Determine whether the predetermined number of iterations has been reached. If so, exit the loop and return to the dominant solution library; if not, go to step 2.

[0150] Case simulation

[0151] Landing Operation Scenario

[0152] like Figure 6As shown in Figure 1, a typical battalion-level amphibious joint landing operation scenario is given. The deployment of each target node at the landing site is shown in Figure 1. Figure 6 As shown in (1), the beachhead terrain at the landing site is as follows Figure 6 (2). Assume that the air defense firepower deployed by the Blue Force on the beachhead consists of two short-range air defense missile systems and individual air defense missiles at each node. Then, affected by the terrain, assuming that the Red Force helicopter flies at an ultra-low altitude of 10m, the air defense firepower threat of the Blue Force at the landing site can be calculated by formula (20) as follows: Figure 6 (3) as shown.

[0153] Assuming that the Red Army's Amphibious Ready Group has three LHDs, and ideally loaded, each LHD can be equipped with 12 general-purpose helicopters, 12 armed helicopters, and 6 helicopter takeoff positions (i.e., any type of helicopter can independently meet the dual-wave cyclic deployment requirements) for vertical landing missions and close-range fire support missions; assuming that the Red Army's landing forces are divided into the left-wing plane assault group (No. ①), the right-wing plane assault group (No. ②), the deep assault group (No. ③), and the air assault group (No. ④), of which the air assault group is further divided into five vertical landing units of three types, namely, infantry unit 1 (deployed in LHD1), infantry unit 2, fire unit 1 (deployed in LHD2), infantry unit 3, and reconnaissance unit 1 (deployed in LHD3). Among the above vertical landing units, six general-purpose helicopters are required to carry each infantry unit and fire unit, and four helicopters are required for the reconnaissance unit.

[0154] Case simulation results analysis

[0155] choose Figure 5 The code in the dynamic scheduling algorithm is used as the initial value for optimization. Assume that the depth of the taboo list is 10, the scattered search threshold is 10, and each initial value is iterated 100 times. The simulation environment is Core TM A portable computer with a 2.6GHz processor and 32GB memory. The algorithm took 60.58 seconds to iterate 100 times. All the solutions and the Pareto frontier solution set were obtained as follows: Figure 7 shown.

[0156] For further analysis Figure 5 The impact of different initial encoding values ​​on the algorithm iteration results will be Figure 7 The individuals in the Pareto dominated solution set in are differentiated according to their initial values ​​of iteration, and the results are as follows Figure 8 shown.

[0157] It can be seen that Figure 5 (4) and Figure 5 (5) The optimization effect of individuals with initial encoding value is better in the dimension of combat time consumption, while Figure 5 (5) and Figure 5 (6) The optimization effect of the individuals with the initial coding value in the dimension of troop loss is better. Figure 5 (4) Figure 5 (5) and Figure 5 (6) Select the three individuals with the best performance in the two target dimensions among the iterative results obtained as the initial encoding value, and observe their encoding characteristics.

[0158] Combination Figure 8 As shown in Table 1, in the simulation case of the first wave of vertical landing mission helicopter formation damaged given in this section, the effect of the Red Army choosing to support it in subsequent waves is better than supporting it in the current wave, which shows that the execution priority of this task is lower than the tasks in the first two waves in the original operation task plan. In addition, all three better solutions in Table 1 arranged the close fire support task for node 5# at the front position of the code, which may indicate that this task is more important and more urgent in time.

[0159] Table 1

[0160]

[0161] Conclusion and Analysis

[0162] This application studies the problem of dynamic scheduling of ship helicopter operating tasks in joint landing operations. In view of the situation where ship helicopter operating tasks encounter sudden battlefield demands during execution, a strategy-based dynamic scheduling method is proposed. Drawing on the idea of ​​demand event simulation, the optimization initial values ​​of scheduling strategies and operating task encoding are constructed. Based on this, a dynamic scheduling algorithm based on taboo search is designed. The algorithm has high computational efficiency and good convergence effect, which provides a theoretical reference for the dynamic scheduling of ship helicopter operating tasks in landing operations.

[0163] In one embodiment, a ship helicopter dynamic scheduling system based on taboo search is also provided, comprising:

[0164] The model calling module is used to call the constructed ship helicopter operation task planning model;

[0165] Strategy initialization module, used to design dynamic scheduling strategy by event demand simulation and generate task coding initialization;

[0166] A search optimization module is used to optimize and solve the helicopter mission coding sequence using a dynamic scheduling algorithm based on taboo search;

[0167] The planning output module is used to output ship mission planning data.

[0168] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for dynamic scheduling of ship helicopters based on taboo search are implemented.

[0169] The technical features of the above embodiments may 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.

[0170] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and all of them belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A ship-helicopter dynamic scheduling method based on taboo search, characterized in that: Includes steps: Step 1: Call the constructed ship helicopter operation task planning model; Step 2: Use event demand simulation to design a dynamic scheduling strategy and generate the initial value of the task code; Step 3: Use a dynamic scheduling algorithm based on taboo search to optimize and solve the helicopter mission coding sequence; Step 4: Output the dynamic planning data of the ship helicopter mission.

2. The method for dynamic scheduling of ship helicopters based on taboo search according to claim 1 is characterized in that: The constraints of the ship helicopter operation task planning model are divided into two categories: available combat resource constraints and combat phase timing constraints. Available combat resource constraints are further divided into vertical landing team quantity constraints, ship helicopter quantity constraints, and LHD support resource constraints. The task phase timing constraints are divided into task phase timing constraints based on execution order and task phase timing constraints based on LHD deck scheduling. The optimization objective functions of the ship helicopter operation mission planning model include: minimizing landing time, minimizing troop loss and minimizing helicopter risk.

3. The ship-helicopter dynamic scheduling method based on taboo search according to claim 1 is characterized in that: The event demand simulation proposes a dynamic scheduling strategy for ship helicopter combat missions, including a scheduling strategy based on the current wave window and a scheduling strategy based on the adjacent wave window.

4. The method for dynamic scheduling of ship helicopters based on taboo search according to claim 1, characterized in that: The process of generating the initial value of the task code includes: Determine the selectable dynamic scheduling strategy based on the timing window of sudden demand; Based on the generated ship helicopter operation task plan code, the optimization initial value of the corresponding scheduling algorithm is generated according to different dynamic scheduling strategies.

5. The ship-helicopter dynamic scheduling method based on taboo search according to claim 1 is characterized in that: The step of optimizing and solving the helicopter mission coding sequence by using a dynamic scheduling algorithm based on taboo search includes: The first step is to input the initial code and calculate the corresponding initial solution, which are set as the optimal code and the optimal solution respectively; The second step is to select a gene position after the corresponding code of the damaged helicopter group, without changing its mission type segment, and randomly assign a value to its target sequence segment; The third step is to determine the Pareto dominance relationship between the current solution and the optimal solution; The fourth step is to determine whether the predetermined number of iterations has been reached; if so, exit the loop and return to the dominant solution library; if not, go to any gene position after the corresponding code of the damaged helicopter formation, do not change its mission type fragment, and randomly assign a value to its target sequence number fragment.

6. The ship-helicopter dynamic scheduling system based on taboo search is characterized by: include: The model calling module is used to call the constructed ship helicopter operation task planning model; Strategy initialization module, used to design dynamic scheduling strategy by event demand simulation and generate task coding initialization; A search optimization module is used to optimize and solve the helicopter mission coding sequence using a dynamic scheduling algorithm based on taboo search; The planning output module is used to output ship mission planning data.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the ship helicopter dynamic scheduling method based on taboo search as described in any one of claims 1 to 5 are implemented.

Images