Offshore multi-target supply decision optimization method, system, equipment and product

Abstract

The invention discloses a maritime multi-target supply decision optimization method, system, device and product, and relates to the field of multi-target optimization, and the method comprises the steps: constructing a maritime multi-target supply optimization model; obtaining a seaborne supply target number, a seaborne supply ship number, a decision demand and a seaborne data set; constructing a basic decision pool according to the number of the seaborne supply targets and the number of the seaborne supply ships; determining an optimization strategy of the basic decision pool according to the decision demand; according to the offshore data set, the supply success probability of each offshore supply ship for each offshore supply target is determined; and based on the supply success probability and the offshore multi-target supply optimization model, optimizing the basic decision pool by adopting an optimization strategy to obtain an optimal decision pool, and supplying the offshore supply target according to the optimal decision pool. The marine multi-target supply efficiency and maneuverability can be improved.

Description

Technical Field

[0001] The present application relates to the field of multi-objective optimization, and in particular to a method, system, equipment and product for optimizing multi-objective replenishment decisions at sea. Background Art

[0002] With the increasing strategic importance of islands and the booming marine economy, island maritime replenishment has become a critical component in maintaining island security and stability and promoting sustained economic growth. However, traditional maritime replenishment models suffer from low efficiency and poor maneuverability. In peacetime, supply ships can deliver supplies and ammunition to multiple islands using a fixed deployment plan. However, in wartime, each island faces a unique battlefield environment and contributes significantly to the overall combat system. Furthermore, supply ships may be damaged, preventing them from being able to resupply, resulting in lower replenishment efficiency and reduced maneuverability. Furthermore, in wartime, replenishment may involve not only resupplying islands but also resupplying large naval vessels at sea. The diversity of supply targets, the complexity of the battlefield environment, and the uncertainty of the marine environment significantly increase the difficulty of replenishment and significantly reduce the probability of success. Traditional serial fixed deployment replenishment models cannot meet the demand for efficient, rapid, high-success, and low-damage replenishment in this context. Summary of the Invention

[0003] The purpose of this application is to provide a method, system, equipment and product for optimizing multi-objective replenishment decisions at sea, which can improve the efficiency and maneuverability of multi-objective replenishment at sea.

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

[0005] In a first aspect, the present application provides a multi-objective replenishment decision optimization method at sea, comprising:

[0006] Constructing a multi-objective replenishment optimization model at sea; the multi-objective replenishment optimization model at sea includes: a global optimization model and a local optimization model;

[0007] Obtain the number of replenishment at sea targets, number of replenishment at sea vessels, decision requirements, and maritime data sets;

[0008] Constructing a basic decision pool based on the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets;

[0009] Determine the optimization strategy of the basic decision pool according to the decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy;

[0010] Determine the probability of success of each sea supply ship in supplying each sea supply target based on the sea data set;

[0011] Based on the replenishment success probability and the maritime multi-objective replenishment optimization model, the optimization strategy is adopted to optimize the basic decision pool to obtain the optimal decision pool, and the maritime replenishment target is replenished according to the optimal decision pool.

[0012] In a second aspect, the present application provides a multi-objective replenishment decision optimization system at sea, comprising:

[0013] An offshore optimization model construction module is used to construct an offshore multi-objective replenishment optimization model; the offshore multi-objective replenishment optimization model includes: a global optimization model and a local optimization model;

[0014] An acquisition module is used to obtain the number of sea replenishment targets, the number of sea replenishment ships, decision requirements, and sea data sets;

[0015] A decision pool construction module is used to construct a basic decision pool according to the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets;

[0016] The offshore optimization decision module is used to determine the optimization strategy of the basic decision pool according to the decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy;

[0017] A replenishment success probability determination module is used to determine the replenishment success probability of each maritime replenishment ship to each maritime replenishment target based on the maritime data set;

[0018] The decision pool optimization module is used to optimize the basic decision pool based on the replenishment success probability and the maritime multi-objective replenishment optimization model using the optimization strategy to obtain the optimal decision pool, and replenish the maritime replenishment target according to the optimal decision pool.

[0019] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the above-mentioned methods for optimizing multi-objective replenishment decisions at sea.

[0020] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for optimizing multi-objective replenishment decisions at sea.

[0021] According to the specific embodiments provided in this application, this application has the following technical effects:

[0022] The present application provides a method, system, equipment and product for optimizing multi-objective replenishment decisions at sea. By establishing a global optimization model and a local optimization model, and selecting a global optimization strategy or a local optimization strategy according to decision-making requirements, decision makers can choose different maritime optimization decision strategies according to different decision-making requirements, thereby improving the mobility of replenishment. At the same time, selecting a global optimization strategy enables decision makers to quickly obtain the best global replenishment plan, thereby improving decision-making efficiency at the global level. Selecting a local optimization strategy enables decision makers to quickly obtain the best local replenishment plan, thereby improving the accuracy of decisions based on local decision-making requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A flowchart of a multi-objective replenishment decision-making optimization method at sea provided in one embodiment of the present application;

[0025] Figure 2 This is a schematic diagram of the probability circle for multiple supply ships supplying multiple targets;

[0026] Figure 3 This is a structural diagram of a computer device for this application. DETAILED DESCRIPTION

[0027] 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.

[0028] When multiple supply ships are deployed to supply multiple maritime targets in a combat scenario, numerous supply options are available. Decision-makers must comprehensively consider a vast amount of information, including the mission, time, space, environment, target characteristics, supply ship capabilities, and resource allocation. They must coordinate the use of all supply ships, quickly identify effective supply options, and find the optimal one, forming a variety of supply routes, such as one-to-one, one-to-many, and many-to-many. Consequently, maritime replenishment in this context is an optimal ship scheduling problem, a NP-hard problem. Due to the large amount of information data, its real-time nature, and the multitude of options, decision-makers struggle to quickly process and solve the problem manually. However, mathematical methods, aided by computer calculations, can quickly and accurately extract characteristic parameters, calculate key indicators, and assess the pros and cons of various options.

[0029] In order to solve the above decision-making problems, this application proposes a method for optimizing multi-objective replenishment decisions at sea. The method makes full use of information data related to marine replenishment, calculates the probability of successful replenishment of different marine supply ships for different marine supply targets, and uses the replenishment optimization model to find the optimal replenishment strategy in the replenishment decision pool, thereby reducing the burden of decision makers on processing data and judging pros and cons, and playing the role of assisting decision-making for many-to-many marine replenishment plans, so that decision makers can replenish marine supply targets according to the optimal decision pool, thereby improving replenishment efficiency and maneuverability.

[0030] 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.

[0031] In an exemplary embodiment, Figure 1 As shown, a method for optimizing multi-objective replenishment decision-making at sea 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 a server as an example, and includes the following steps S1 to S8. Among them:

[0032] Step S1: constructing a multi-objective replenishment optimization model at sea; the multi-objective replenishment optimization model at sea includes: a global optimization model and a local optimization model.

[0033] Furthermore, the global optimization model is to model and calculate all the sea supply ships and sea supply targets (the whole consisting of m sea supply ships and n sea supply targets) from the perspective of global optimization, and obtain a global value, denoted as F. The larger the F value, the higher the probability of global supply success, and the better the supply plan. The expression of the global optimization model is:

[0034]

[0035] The local optimization model is to optimize the specific j-th marine replenishment target to obtain the optimal replenishment plan for j, and the obtained local value is recorded as B. The background of using local optimization is generally that the replenishment target has a strong importance level (C j (larger), or priority replenishment is required in the combat plan or combat decision. The local optimization model for multi-objective optimal replenishment at sea is as follows, that is, to maximize the value of B. The larger the value, the higher the probability of successful local replenishment for j, and the more optimal the replenishment plan for j. The expression of the local optimization model is:

[0036]

[0037] Among them, F is the global value; is the probability of successful replenishment of the i-th sea supply ship to the j-th sea supply target; st is the constraint condition; x ij It refers to the judgment parameter of the i-th sea supply ship supplying the j-th sea supply target, which takes 0 or 1; μ ij is the supply safety probability of the i-th sea supply ship when it supplies the j-th sea supply target; ρ ij is the supply difficulty of the i-th sea supply ship to the j-th sea supply target; σ ij is the supply feasibility of the i-th sea supply ship to the j-th sea supply target; τ ij is the survival probability of the jth sea supply target when the i-th sea supply ship supplies the j-th sea supply target under the radio silence state of the supply ship; m is the number of sea supply ships; n is the number of sea supply targets; B is a local value.

[0038] Step S2: Obtain the number of maritime replenishment targets, the number of maritime replenishment ships, decision requirements, and maritime data sets. The maritime data sets include: temporal characteristic data, spatial characteristic data, replenishment ship and resource characteristic data, route data, weather environment data, replenishment target data, and replenishment mission data.

[0039] Step S3: constructing a basic decision pool according to the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets.

[0040] Specifically, multi-target replenishment at sea refers to the use of a total of m supply ships to supply n supply targets at sea. Each supply ship can supply (0 to n) supply targets at sea. The replenishment plan for each supply target at sea can include 0 to m supply ships, which can be expressed as:

[0041]

[0042] Where Sj is the subset of sea supply ships that supply the jth sea supply target, and S j It can be an empty set. According to the combination rules, there are 2 possible supply plans for each sea supply target. m There are 2 replenishment plans for n sea replenishment targets. m×n These 2 m×n The replenishment plans constitute the basic decision pool for multi-objective replenishment at sea, denoted as Θ:

[0043]

[0044] Where θ k is the kth supply plan, which is recorded as the following matrix:

[0045]

[0046] The row of the matrix is ​​the number of the sea supply ship i; the column of the matrix is ​​the number of the sea supply target j; ij It refers to the judgment parameter of whether the i-th sea supply ship supplies the j-th sea supply target, and it takes 0 or 1.

[0047]

[0048] Step S4: Determine the optimization strategy of the basic decision pool according to the decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy.

[0049] Furthermore, when the decision requirement is a global decision requirement, the optimization strategy of the basic decision pool is determined to be a global optimization strategy; when the decision requirement is a local decision requirement, the optimization strategy of the basic decision pool is determined to be a local optimization strategy.

[0050] Specifically, if the multi-objective replenishment decision is based on the comprehensive consideration of the replenishment needs of all objectives, and the goal of the decision is to obtain the best global solution, then the global optimization strategy is adopted to perform global optimization on the basic decision pool; if the multi-objective replenishment decision needs to distinguish the replenishment priorities of the objectives and obtain the best replenishment plan corresponding to each objective, then the local optimization strategy is adopted to perform local optimization on the basic decision pool.

[0051] Step S5: Determine the probability of success of each marine replenishment ship in replenishing each marine replenishment target based on the marine data set.

[0052] Furthermore, step S5 includes the following steps S51 to S55.

[0053] S51: Determine the supply feasibility of each maritime supply ship to each maritime supply target based on the time characteristic data, the spatial characteristic data, the supply ship and resource characteristic data, the route data and the weather environment data.

[0054] S52: Determine the difficulty of supplying each sea supply target by each sea supply ship according to the supply target data and the supply mission data, and the survival probability of each sea supply target when each sea supply ship supplies each sea supply target in a radio silence state.

[0055] S53: Determine, based on the route data and the weather environment data, the supply safety probability of each sea supply ship when each sea supply ship supplies each sea supply target.

[0056] S54: Calculate the probability of success of each sea supply ship in supplying each sea supply target based on the supply feasibility, the supply difficulty, the survival probability, and the supply safety probability.

[0057] Specifically, according to the probability of successful replenishment of m sea supply ships to n sea supply targets, a replenishment success probability matrix is ​​established, which is recorded as Φ:

[0058]

[0059] In the supply success probability matrix, each row of data represents the probability of a specific sea supply ship numbered i successfully supplying all sea supply targets, and each column of data in the supply success probability matrix represents the probability of a sea supply target numbered j being successfully supplied by all supply ships. The value of Φ is The calculation formula is:

[0060]

[0061] Where: μ ij is the supply safety probability of the i-th sea supply ship when it supplies the j-th sea supply target.

[0062] ρ ij is the supply difficulty of the i-th sea supply ship to the j-th sea supply target.

[0063] σ ij It represents the supply feasibility of the i-th sea supply ship to the j-th sea supply target.

[0064] τ ij is the survival probability of the jth sea supply target when the i-th sea supply ship supplies the j-th sea supply target under the radio silence state of the sea supply ship.

[0065] μ ij, ρ ij , σ ij , τ ij The value range is 0 to 1. The four-term multiplication is an implementation case, and can also be increased or decreased according to the actual implementation form.

[0066] (1) Supply safety probability μ ij

[0067] Supply safety probability μ ij The calculation method is:

[0068]

[0069] Where H si is the effective wave height of the current sea condition of the i-th offshore supply ship, V wi is the wind speed of the current sea conditions for the i-th offshore supply ship, and λ1 and λ2 are the offshore supply ship's sensitivity to wave height and wind speed. A larger λ value indicates a more significant impact of this factor on the probability of supply safety. Common values ​​for λ1 and λ2 are shown in Table 1.

[0070] Table 1 Typical reference values ​​of sensitivity

[0071]

[0072] (2) Supply difficulty ρ ij

[0073] ρ ij The calculation of mainly considers factors such as combat threats, tactical coordination and battlefield environment faced by sea supply ships when supplying sea supply targets. The specific formula is:

[0074]

[0075] Where, is the battlefield threat level encountered by the i-th sea supply ship during the supply process of the j-th sea supply target. The value selection method is shown in Table 2.

[0076] Table 2 Battlefield threat levels Value Table

[0077]

[0078]

[0079] The type of supply provided by the i-th offshore supply ship to the j-th offshore supply target. The value selection method is shown in Table 3.

[0080] Table 3 Supply types Value Table

[0081]

[0082] is the battlefield support capability of the i-th sea supply ship in the process of supplying the j-th sea supply target. The value selection method is shown in Table 4.

[0083] Table 4 Battlefield support capabilities Value Table

[0084]

[0085] (3) Survival probability τ in radio silence ij

[0086] τ ij The calculation method is:

[0087]

[0088] in, is the probability that the i-th sea supply ship is discovered when it is supplying the j-th sea supply target in radio silence, is the probability of the enemy launching an attack after discovering the i-th supply ship pair at sea, is the probability that the enemy attacks and destroys the i-th sea supply ship.

[0089] The calculation method is:

[0090]

[0091] Among them, λ d is the reconnaissance efficiency constant, usually taken as λ d =0.1~0.3; r i is the stealth coefficient of the supply ship, which represents the stealth capability of the supply ship numbered i, usually r i = 0.2~1, when a supply ship with stealth design and electronic jamming is conducting night supply, r i = 1, if ordinary ships are supplied during the day without cover, r i =0.2; t ij It is the time difference from the current time to the estimated completion of replenishment, in hours, usually (navigation time + docking time).

[0092] ISR is the enemy's surveillance intensity, which represents the enemy's reconnaissance capability per unit time. The enemy's surveillance means are denoted as G, and the number of devices corresponding to the g-th surveillance means is n. g , the ISR value of the device is ISR g , then the ISR is calculated as follows:

[0093]

[0094] ISR g The specific values ​​of are obtained by consulting combat regulations and tactical manuals, battlefield intelligence briefings, equipment performance databases, etc. Table 5 lists the ISR values ​​of typical reconnaissance methods.

[0095] Table 5 Typical ISR reference table

[0096]

[0097] For example, if there is one SAR satellite (ISR = 0.6) and two drones (ISR = 0.4) in the area, the total ISR = 0.6 + 0.4 × 2 = 1.4.

[0098] is the probability of the enemy launching an attack after discovering the i-th supply ship at sea, that is, the enemy's decision weight, which is related to the value of the supply target, the enemy's threat level to the battlefield, and the enemy's attack cost. The calculation method is:

[0099]

[0100] Where C j C is the importance level of the jth marine supply target, determined according to the specific combat plan or emergency plan. Generally, it is 0.8-1.0 for marine centers, 0.6-0.8 for radar stations, 0.6-0.8 for front-line airports, 0.4-0.6 for logistics transfer stations, and 0.2-0.5 for ordinary outposts. a The cost-benefit ratio of the enemy attacking the sea supply target numbered j can be calculated by referring to the cost-benefit analysis model in a certain country's Joint Fire Support Manual (JP3-09).

[0101] is the probability of the enemy attack hitting and destroying the supply ship numbered i, which is related to the close defense capability of the supply ship and the close defense capability of the supply target. The calculation method is:

[0102]

[0103] Where A i is the enemy's probability of hitting the i-th supply ship at sea, D ij The interception probability of one's own side (supply target and supply ship), E i is the environmental coefficient. i The calculation formula is:

[0104]

[0105] Among them, Ra is the inherent hit rate of the weapon, which can be obtained by looking up the table. Common values ​​are shown in Table 6.

[0106] Table 6 Common weapons inherent hit rate R a

[0107] Weapon Type Hit rate (no interference) Data Source Anti-ship missile (subsonic) 0.7～0.8 Jane's Missiles and Guided Weapons Anti-ship missile (supersonic) 0.6～0.7 DIA test report of a certain country torpedo 0.5～0.6 Naval Weapons Effectiveness Evaluation Manual

[0108] Δ is the maneuvering effect of the ship: Δ = 0.1 × evasion speed (knots) + 0.05 × maneuvering frequency (times / minute); λ a It is the correction coefficient, usually ranging from 1.5 to 2.0.

[0109] Where D ij is the interception probability of defense. According to the layered defense model, the interception probability of the p-th layer of defense is , then D ij for:

[0110]

[0111] The defensive interception probability values ​​can be found in documents such as Jane's Air Defense Systems Yearbook and a certain country's naval weapon system manual. Commonly used defensive interception probability values ​​are shown in Table 7.

[0112] Table 7 Defense interception probability

[0113] Defense layer Typical defense probability Modification conditions Electronic jamming (soft kill) 0.2～0.4 For every 1kW increase in interference power, the defense probability increases by 0.05 Close-in weapon system (CIWS) 0.7～0.9 If the enemy uses a supersonic missile, the defense probability is ×0.6 Air defense missiles (medium-range) 0.6～0.8 The enemy uses stealth weapons, defense probability × 0.5

[0114] E i is the environmental coefficient, and the value selection method is shown in Table 8.

[0115] Table 8 Environmental coefficient values

[0116] Environmental factors Coefficient range Physical mechanism clear sky 1.0 No attenuation rainstorm 0.3～0.5 Weakened radar / infrared guidance Waves > Level 6 0.6～0.8 Affects shipborne radar tracking Electromagnetic interference area 0.4～0.7 Reduced weapon communication link stability

[0117] (4) Supply feasibility σ ij

[0118] The supply weight factor is established based on the spatial dimension constraints, time dimension constraints, resource dimension constraints, prohibited area dimension constraints, and weather dimension constraints. The supply feasibility weight factor is:

[0119]

[0120] Among them, & represents the AND operation, and the operand value is 0 or 1. When the operand value is 0, it means that the task cannot be completed, and when the operand value is 1, the task can be completed; is the spatial dimension constraint, is the time dimension constraint, is the resource dimension constraint, For the dimension constraint of the prohibited area, It is the weather dimension constraint; the values ​​of the above five parameters are all 1 or 0.

[0121] Spatial dimension constraints The determination method is: based on the spatial feature data and the supply ship and resource feature data, establish the spatial dimension constraints of each supply ship. The spatial dimension constraints of each supply ship include whether the supply distance of the supply ship can meet the requirements. The spatial dimension constraints of the supply ship are:

[0122]

[0123] Where d is the spatial dimension identifier. The characteristic data includes the nautical mile data of the supply ship and the supply target displayed by the navigation software, and the supply ship data includes the nautical mile data of the remaining diesel fuel displayed on the supply ship instrument panel. The estimated travel distance of the supply ship fuel and the distance between the supply ship and the target are obtained; based on the size of these two distances, it is determined whether the remaining fuel of the supply ship can reach the target and determine value.

[0124] Time dimension constraints The determination method is as follows: Based on the spatial feature data and the offshore supply ship data in the dataset, the relationship between the supply demand time of the target within the supply range of the supply ship and the supply ship response time is determined, and the time dimension constraints of each supply ship are established. The time dimension constraints of each supply ship are:

[0125]

[0126] Among them, t i is the response time of the i-th sea supply ship to the j-th sea supply target, t j The time required to resupply the jth sea supply target. Time constraints must take into account sea conditions, which are categorized into nine levels, with varying replenishment response times. Target characteristic data includes, but is not limited to, navigation software displaying the sailing time between the supply ship and the target. Supply ship data includes, but is not limited to, the time it takes to reach the target based on the supply ship's current maintenance status. Supply ship maintenance status includes, but is not limited to, ship type data, a table showing speed, draft, and resistance in still water, a table showing wave drag transfer functions, a table showing wind drag coefficients, a table showing propeller thrust and torque coefficients, a table showing engine and steering gear performance parameters, a table showing diving drag coefficients, and the volume and dimensions of the replenishment items.

[0127] Resource dimension constraints The determination method is: Based on the supply mission data and supply ship and resource characteristic data in the dataset, the resource dimension constraints of each supply ship are established. The supply ship data includes but is not limited to ship type data, speed and draft resistance table in still water, wave resistance transfer function table, wind resistance coefficient table, propeller thrust torque coefficient table, engine and steering gear performance parameter table and diving resistance coefficient table, and the volume and size of the supply items. The resource dimension constraints of each supply ship are:

[0128]

[0129] Among them, Z i is the supply material carried by the i-th sea supply ship, Z j The current amount of supplies required by the jth sea supply target. Resource constraints also need to consider sea conditions. Sea conditions are categorized into nine levels, and different supply vessels are suitable for different sea conditions. The relationship between supply vessel data and sea condition data is as follows: for example, a ship capable of operating in sea condition level 8 may only be able to operate in sea condition level 7 if it is poorly maintained and too old. Forcing it to operate in sea condition level 8 risks the ship being destroyed and its crew members killed. The relationship between sea conditions and resource dimensions is as follows: for example, if the target requires 9 tons of diesel, and the supply vessel has enough diesel in its tank, but the current sea conditions prevent it from sailing, its resource dimension will be 0.

[0130] No-entry area dimension constraints The determination method is: Based on the route data in the dataset, establish the dimension constraints of the prohibited areas for each supply ship at sea. For example, some waters are very shallow, some supply ships can pass through, but some cannot. The dimension constraints of the prohibited areas between each supply ship and each supply target are:

[0131]

[0132] Obtain the waterline of the supply ship and the depth of the target waters; based on the waterline and depth, determine whether the supply ship can pass through the restricted area and determine The water depth near the target varies greatly and is also related to the tide. The waterline of the supply ship is related to the tonnage of the cargo carried. Therefore, the reliability of the supply needs to be compared with the supply ship's navigation manual by professional crew members. Experience and data can also be stored in the database for query.

[0133] Weather dimension constraints The determination method is: according to the weather environment data in the data set, establish the weather dimension constraints of each offshore supply ship. The weather dimension constraints of each offshore supply ship are:

[0134]

[0135] Different levels of offshore supply ships can sail in different levels of typhoons. Obtain the wind force level in the weather environment data; according to the ship stability manual, ship operation manual and SMS procedure documents, determine whether the offshore supply ship can sail in the wind force level (i.e. whether the weather is suitable) and determine value.

[0136] In order to improve the readability of the probability matrix Φ for successful replenishment, a probability circle is used to visualize Φ. The number of rows in the probability circle is the same as the number of rows in Φ, which is m. The row labels of the probability circle correspond to the number i of the sea supply ship. The number of columns in the probability circle is the same as the number of columns in Φ, which is n. The column labels of the probability circle correspond to the number j of the sea supply target. The diameter of the circle in each square (i-th row, j-th column) represents the replenishment relationship between the sea supply ship number i and the sea supply target number j, which is convenient for decision makers to understand intuitively. If there is a circular icon in the square (i-th row, j-th column), it means that the sea supply ship number i can supply the sea supply target number j. If there is no circular icon in the square, it means that the sea supply ship number i cannot supply the sea supply target number j. When the side length of the square is 1, the diameter of the circular icon in the square represents the supply probability, that is, The value of . Figure 2 For example, in the figure, 201, 202, 203, ..., 210 represent 10 supply ships, and 101, 102, 103, 104, 105 represent 5 supply targets. The diameter of the circle in each square represents the supply relationship between the supply ship and the supply target. When the diameter of the circle is 1, it means that the supply will be successful. For example, 3 in the figure means that the probability of 201 supplying 101 is successful. When the diameter of the circle is 0, it means that the supply will fail. When the diameter of the circle is 0.3, it means that the supply will fail. 4 means that the probability of successful replenishment between the sea supply ship and the sea supply target is 34%. In the figure, 4 represents the probability of successful replenishment of 201 to 102, 5 represents the probability of successful replenishment of 202 to 102, 6 represents the probability of successful replenishment of 204 to 104, 7 represents the probability of successful replenishment of 205 to 105, 8 represents the probability of successful replenishment of 206 to 105, and 9 represents the probability of successful replenishment of 208 to 105.

[0137] Step S6: Based on the replenishment success probability and the maritime multi-objective replenishment optimization model, the optimization strategy is used to optimize the basic decision pool to obtain an optimal decision pool, and the maritime replenishment target is replenished according to the optimal decision pool.

[0138] Furthermore, the replenishment target data includes the importance level of the marine replenishment target, and step S6 includes the following steps S61-S62.

[0139] Step S61: When the optimization strategy is a global optimization strategy, the basic decision pool is simplified based on the probability of successful replenishment, and the global value of the replenishment plan in the simplified basic decision pool is calculated based on the global optimization model. The replenishment plans in the simplified basic decision pool are sorted in descending order according to the global value. If there are replenishment plans with equal global values, they are sorted twice to obtain the optimal decision pool.

[0140] Specifically, according to the sub-items with zero values ​​in Φ, the basic decision pool Θ is simplified to obtain the decision pool Θ'; the simplification method is as follows: if there are sub-items in Φ That is, the supply ship No. l cannot supply the target No. h, then for all the targets in the decision pool that contain x lh = 1 sub-item is deleted to reduce the number of supply plans that cannot be implemented in the decision pool and improve the efficiency of subsequent optimization; all supply plans θ in the decision pool Θ' are deleted. k , and the supply probability matrix Φ are successively brought into the global optimization model to calculate and obtain the global value F corresponding to each supply scheme k ; According to the global value F k In descending order, for all replenishment plans θ in the decision pool Θ' k Re-sort and get the optimized decision pool Θ”; if there are different solutions θ in the optimized decision pool Θ” k But F k In the same situation, this type of supply plan is sorted again according to the local optimization strategy, and the sorting of other supply plans remains unchanged. Finally, a "global + local" hybrid optimization decision pool dominated by global optimization is obtained. The supply plan that is closer to the front in the decision pool is better, and it is given priority when making multi-objective supply decisions at sea.

[0141] Furthermore, if there are supply plans with equal global values, a secondary sorting is performed, specifically including: sorting the supply plans with the same global values ​​in descending order according to the importance level of the maritime supply targets, and calculating the local values ​​of the supply plans with the same global values ​​based on the local optimization model, and sorting the supply plans with the same global values ​​in descending order according to the local values ​​of the supply plans with the same global values ​​to complete the secondary sorting.

[0142] Specifically, the method of secondary sorting according to the local optimization strategy is as follows: sort the maritime replenishment targets in descending order of importance, and the ones that are closer to the front will be given priority in meeting the needs; according to the importance level sorting, the replenishment target with the largest importance level (denoted as q1) is selected as the first local optimal object, and the q1 column of the replenishment plan and the replenishment probability matrix Φ are calculated in turn using the local optimization model to obtain the local value B corresponding to each replenishment plan. q1 According to B q1The replenishment plans are reordered in descending order of the values ​​and recorded in the new decision pool Θ"'; q1 If the values ​​are still the same, sort them according to the importance of the supply targets and calculate the local supply success results B of these supply plans for the less important target (denoted as q2) according to the above steps. q2 , and re-sort and record the results in descending order to obtain a new decision pool Θ""; if there is still B q2 In the same situation, the order is recursively pushed back according to the target importance level, local values ​​are calculated, and re-ordered and recorded until all supply plans are sorted.

[0143] Step S62: When the optimization strategy is a local optimization strategy, the supply plans in the basic decision pool are sorted in descending order and simplified based on the importance level of the offshore supply target and the probability of successful supply. Based on the local optimization model, the local values ​​of the supply plans in the sorted and simplified basic decision pool are calculated, and the supply plans whose local values ​​are greater than or equal to the threshold of the probability of successful supply are screened out. The screened supply plans are sorted in descending order according to the local values ​​to obtain the optimal decision pool.

[0144] Specifically, according to the importance level ranking, a replenishment plan classification decision pool Θ is established based on the basic decision pool Θ * , classification decision pool Θ * Each individual decision pool The ordering is always the same:

[0145]

[0146] Classification decision pool Θ * It is a secondary classification of the basic decision pool Θ according to the importance of the replenishment target, with a total of n subsets; among them, Represents a single decision pool for target j, including (2 m ) classified supply plans.

[0147]

[0148] For the marine replenishment target j, the classification replenishment plan is denoted as θ jk :

[0149] θ jk =[x j1 … x jm ] T .

[0150] Furthermore, step S62 includes the following steps S621 to S65.

[0151] Step S621: Sort the replenishment plans in the basic decision pool in descending order according to the importance level of the marine replenishment targets to obtain a classification decision pool.

[0152] Step S622: Simplify the replenishment plans with a replenishment success probability of zero in the classification decision pool to obtain a sorted and simplified basic decision pool.

[0153] Specifically, according to the sub-items with zero values ​​in Φ, the basic decision pool Θ is simplified * , get the decision pool Θ ** ; The simplified method is, if there are sub-items in Φ That is, the supply ship No. l cannot supply the target No. h, then for all the targets in the decision pool that contain x lh =1 sub-item is deleted to reduce the number of unimplementable replenishment plans in the decision pool and improve subsequent optimization efficiency.

[0154] Step S623: Based on the probability of successful replenishment, a local optimization model is used to calculate the local value of each replenishment plan in the sorted and simplified basic decision pool.

[0155] Specifically, the classification decision pool Θ will be simplified ** All classified supply plans θ jk , and the probability of successful replenishment are successively brought into the local optimization model to calculate and obtain the classified replenishment plan θ jk Local value of supply target j

[0156] Step S624: Calculate the replenishment success probability threshold of each marine replenishment target based on the replenishment target data and the replenishment mission data.

[0157] Step S625: Filter out the replenishment plans whose local values ​​are greater than or equal to the replenishment success probability threshold of the corresponding sea replenishment target, and sort them in descending order according to the local values ​​to obtain the optimal decision pool.

[0158] Specifically, according to the requirements of the combat system or combat plan, the supply success probability threshold δ of the supply target j is extracted j ; Determine whether target j can be resupplied according to the following formula.

[0159]

[0160] In a single decision pool In the above example, delete the solution that cannot meet the target j supply, and Greater than the threshold δ j The classified supply plan is based on Rearrange the values ​​in descending order and record them in the new single decision pool In the process, a new classification decision pool Θ is formed*** .

[0161] In a specific embodiment, based on the above method, three sea supply ships are used to supply two sea supply targets. Each sea supply ship can supply (0 to 2) sea supply targets, and the supply plan for each sea supply target can include 0 to 3 sea supply ships, which are recorded as:

[0162]

[0163] Where S j is the subset of sea supply ships that supply sea supply target j, and S j Can be an empty set.

[0164] According to the combination rules, there are eight possible replenishment plans for the No. 1 maritime replenishment target, including:

[0165] 1) Ship 1 does not supply the target, Ship 2 does not supply the target, and Ship 3 does not supply the target;

[0166] 2) Ship 1 supplies the target, while Ship 2 and Ship 3 do not supply the target;

[0167] 3) Ship 1 does not supply the target, ship 2 supplies the target, and ship 3 does not supply the target;

[0168] 4) Ship 1 does not supply the target, ship 2 does not supply the target, and ship 3 supplies the target;

[0169] 5) Ship 1 supplies the target, ship 2 supplies the target, and ship 3 does not supply the target;

[0170] 6) Ship 1 supplies the target, Ship 2 does not supply the target, and Ship 3 supplies the target;

[0171] 7) Ship 1 does not supply the target, Ship 2 supplies the target, and Ship 3 supplies the target;

[0172] 8) Ship No. 1 supplies the target, Ship No. 2 supplies the target, and Ship No. 3 supplies the target.

[0173] There are 64 replenishment plans for the two targets, and these 64 replenishment plans constitute the basic decision pool for multi-target replenishment at sea, denoted as Θ:

[0174] Θ=(θ1,…,θ k ,…θ 64 ).

[0175] Where θ k is the kth supply plan, which is recorded as the following matrix:

[0176]

[0177] The row labels of the matrix are the numbers of the sea supply ships; the column labels of the matrix are the numbers of the sea supply targets; ij It refers to the judgment parameter of whether the i-th sea supply ship supplies the j-th sea supply target, and it takes 0 or 1.

[0178]

[0179] For example, there is a replenishment plan θ1 in the basic decision pool Θ:

[0180]

[0181] This means that the three sea supply ships do not supply any targets, so x ij All are recorded as 0;

[0182] Another supply plan θ 10 :

[0183]

[0184] This means that target 1 is supplied by ship No. 1, and target 2 is supplied by ships No. 1 and No. 2; from the perspective of the supply ship, ship No. 1 needs to supply targets No. 1 and No. 2, ship No. 2 supplies target No. 2, and ship No. 3 does not participate in the supply.

[0185] According to the success probability of three sea supply ships supplying two sea supply targets, a supply probability matrix is ​​established, which is recorded as Φ:

[0186]

[0187] Supply safety probability u ij :This embodiment uses a large military supply ship, according to the weather data set data: H s1 =1m, H s2 =2m,H s3 =1m; V w1 =4m / s, V w2 =8m / s, V w3 =4.9m / s.

[0188] The wave height and wind speed sensitivities of the three offshore supply ships are the same, namely: λ1 = 0.02, λ2 = 0.01.

[0189] By calculation, we can get u ij For: u 11 =u 12 =0.8353,u 21 =u 22 =0.4868,u 31 =u32 =0.7711.

[0190] Supply Difficulty ρ ij In this embodiment, based on the replenishment mission data, replenishment target data, and replenishment vessel data, no battlefield threats are detected in the three maritime replenishment vessels and their surrounding navigation areas. No battlefield threats are currently detected near target 2, but there are enemy air threats (low-density drone patrols) or possible unmanned submarines near target 1. Therefore, any vessel supplying there may encounter a battlefield threat, which is recorded as:

[0191] Target 1 needs to be resupplied with food and fuel, and Target 2 needs to be resupplied with ammunition and medical supplies, so, The values ​​are:

[0192] Since target 1 is more important, all ships supplying target 1 can be escorted. Ships 1 and 2 can be escorted by destroyers, while ship 3 can only be escorted by coast guard ships (with lower capabilities than frigates). There are no escort ships when supplying target 2. From the perspective of the supply ship, ships 1 and 2 have air defense systems, and all three ships have stealth designs, so the calculation is for:

[0193] Calculate the supply difficulty ρ ij is: 11 =0.86;ρ 12 =0.77;ρ 21 =0.86;ρ 22 =0.77;ρ 31 =0.84;ρ 32 =0.77.

[0194] Survival probability τ in radio silence ij In this embodiment, according to the replenishment mission data, replenishment target data, replenishment ship data, route data, etc., the specific replenishment time is uncertain. The three supply ships at sea are all stealth-capable, and ship No. 1 can actively implement electronic jamming. Therefore, r1 = 0.8, r2 = r3 = 0.4; the reconnaissance efficiency constant λ d Take 0.1; the estimated replenishment completion time is: t 11 =6;t 12 =10; t 21 =14;t 22 =10; t 31 =14;t 32 =4.

[0195] The enemy surveillance intensity is the same throughout the entire area, with one SAR satellite (ISR = 0.6) and two drones (ISR = 0.4), so the total ISR = 0.6 + 0.4 × 2 = 1.4.

[0196] From this, it is concluded for:

[0197] is the probability of the enemy launching an attack after discovering the supply ship i, that is, the enemy's decision weight, which is related to the value of the supply target, the enemy's threat level to the battlefield, and the enemy's attack cost. The calculation method is:

[0198]

[0199] Where C j C is the importance level of supply target j, which is determined according to the specific combat plan or emergency plan. Generally, it is 0.8-1.0 for maritime centers, 0.6-0.8 for radar stations, 0.6-0.8 for front-line airports, 0.4-0.6 for logistics transfer stations, and 0.2-0.5 for ordinary outposts. aj is the cost-benefit ratio of the enemy attacking target j, which can be calculated by referring to the cost-benefit analysis model in a certain country's Joint Fire Support Manual (JP3-09).

[0200] In this embodiment, according to the supply mission data and supply target data, target 1 is a maritime sub-center with an importance level of C1=0.8, and target 2 is a maritime relay station with an importance level of C2=0.4; battlefield threat Calculated; Based on the enemy's possible attack cost-benefit ratio calculated in the mission document, C a1 =0.7, C a2 =0.2. Calculation for:

[0201] is the probability of the enemy attack hitting and destroying the supply ship i, which is related to the close defense capability of the supply ship and the close defense capability of the supply target, and is calculated as follows:

[0202]

[0203] Where A i is the enemy's probability of hitting the supply ship i, D ij The interception probability of one's own side (supply target and supply ship), E i is the environmental coefficient. i The calculation formula is:

[0204]

[0205] Among them, R a is the inherent hit rate of the weapon, which can be obtained by looking up the table. Common values ​​are shown in Table 6. a is the correction factor, usually ranging from 1.5 to 2.0. Δ is the ship's maneuvering effect: Δ = 0.1 × evasion speed (knots) + 0.05 × maneuvering frequency (times / minute).

[0206] D ij is the interception probability of defense. According to the layered defense model, the interception probability of the p-th layer of defense is , then D ij for:

[0207]

[0208] The defensive interception probability value can be found in documents such as Jane's Air Defense Systems Yearbook and a certain country's naval weapon system manual.

[0209] E i is the environmental coefficient, and the value method is shown in Table 8. In this embodiment, according to the supply mission data, supply target data, weather data, relevant literature and the above formula, it is determined that: A1 = 0.21, A2 = 0.39, A3 = 0.41, D 11 =D 21 =D 31 =0.7, D 12 =D 22 =D 32 =0.5.

[0210] Ship No. 1 can carry out electromagnetic interference, E1=0.5; Ships No. 2 and 3 are both in clear skies, without electromagnetic interference, E2=E3=1.

[0211] Calculated for:

[0212] Calculate τ ij is: 11 =0.9881; τ 12 =0.9887; τ 21 =0.9326; τ 22 =0.9508; τ 31 =0.9292; τ 32 =0.9598.

[0213] Supply feasibility σ ij In this embodiment, according to the spatial feature data, supply ship data and supply mission data, it is determined that all supply ships always have sufficient fuel. Even if the same ship supplies two targets, its fuel can meet the requirements. Therefore, all are all 1; based on the time characteristic data, supply ship data, and supply target data, it is determined that the response time of all ships supplying all targets is less than the target's supply time, that is: According to the supply ship and resource situation and supply mission data, since the fuel tank capacity of ship No. 3 is small, it cannot meet the supply needs of target No. 1. The supply and demand between the remaining ships and the target can be met, that is: other According to the route data, supply ship data and supply mission data, since the depth of the target waters No. 2 is relatively shallow, the No. 1 ship has a deep draft and cannot berth; there are no prohibited areas in the rest of the supply routes. the remaining According to the weather environment data and supply ship data, all supply ships can adapt to the current weather conditions, i.e.

[0214] According to the formula and parameter results, the recharge feasibility σ is obtained ij is: 11 =1,σ 12 =0,σ 21 =1,σ 22 =1,σ 31 =0,σ 32 =1.

[0215] According to the calculated u ij , ρ ij , σ ij , τ ij The supply probability matrix Φ is calculated as follows:

[0216]

[0217] If global optimization is performed, when simplifying the basic decision pool, there may be a situation where the probability of successful replenishment is zero. For example, there is a plan θ9 in the decision pool:

[0218]

[0219] There is also a supply plan θ 10 :

[0220]

[0221] By comparison, θ9 and θ 10 The difference is that θ 10 x in 12 =1, and x in θ9 12 =0, considering It can be determined that θ 10 In actual implementation, it is the same as θ9, and in the subsequent calculation of F k, there will be a numerical overlap, so the interference scheme θ is deleted. 10 .

[0222] When calculating global values, for example, there is a replenishment plan θ1 in the decision pool:

[0223]

[0224] All its x ij If both are 0, then F1=0;

[0225] Supply plan θ9:

[0226]

[0227] Among them, x 11 =1,x 22 =1, and the rest are 0. Bring it into the global optimization model for calculation, and we have:

[0228]

[0229] Another supply plan θ 48 :

[0230]

[0231] Bringing this solution into the global optimization model for calculation, we have:

[0232]

[0233]

[0234] With F1, F9 and F 48 For example, the supply plan θ can be determined according to the value 48 It is better than the supply plan θ9, and the supply plan θ9 is better than the supply plan θ1. Therefore, it is re-ranked according to the principle of the more superior, the higher the priority, and recorded in the optimized decision pool Θ". The other supply plans are also calculated in the above way, and finally the optimal decision pool is obtained.

[0235] If local optimization is performed, for example, the above supply plan θ9:

[0236]

[0237] For target 1, its classification supply plan is recorded as θ 1,9 , recorded middle;

[0238] θ 1,9 =[1 0 0] T .

[0239] For target 2, its classification supply plan is recorded as θ 2,9 , recorded middle.

[0240] θ 2,9 =[0 1 0] T .

[0241] In this embodiment, the replenishment success threshold for target 1 is δ1=0.7, and the replenishment success threshold for target 2 is δ2=0.6.

[0242] For example, the above classified supply plan θ 1,9 :θ 1,9 =[1 0 0] T .

[0243] Classified supply plan θ 2,9 :θ 2,9 =[0 1 0] T .

[0244] Bring it into the local optimal optimization model and calculate it:

[0245]

[0246] Obviously, due to B 1,9 >δ1,θ 1,9 Can be recorded in a new single decision pool In; Due to B 2,9 <δ2, solution θ 2,9 delete.

[0247] The above supply plan θ 48 :

[0248]

[0249] Its classification supply plan for target 2 is: θ 2,48 =[0 1 1] T .

[0250] Bring it into the local optimal optimization model and calculate it:

[0251] It can be seen that the classified supply plan θ 2,48 It can meet the supply needs of target 2 and also realize the decision of multiple ships supplying multiple targets. Other supply plans are also calculated using the above method, and finally the optimal decision pool is obtained.

[0252] The beneficial effects of the multi-objective marine replenishment decision-making optimization method proposed in this application are mainly manifested in:

[0253] (1) The calculated probability of successful replenishment takes into account factors such as mission, time, space, weather, route, supply ship and resource characteristics, and supply target characteristics, effectively integrating features from different dimensions and improving the accuracy of the prediction of the probability of successful replenishment.

[0254] (2) The discrete quantity of resupply feasibility is used in the resupply success probability matrix, which in turn affects the value of the resupply success probability, provides a criterion for determining invalid resupply plans in the decision pool, and improves the optimization calculation speed of the decision pool.

[0255] (3) By establishing a global optimization model for multi-objective replenishment decision-making at sea and proposing a global optimization strategy for the basic decision pool, decision makers can quickly obtain the best global replenishment plan, thereby improving decision-making efficiency at the global level.

[0256] (4) By establishing a local optimization model and proposing a local optimization strategy for the basic decision pool, decision makers can quickly obtain the optimal supply plan for maritime supply targets of different importance levels, thereby improving the accuracy of supply plan decisions for a certain maritime supply target.

[0257] (5) The probability of successful replenishment is plotted in the matrix of successful replenishment in the form of a circle and shown graphically to improve the readability of the probability of successful replenishment and provide decision makers with a means to intuitively view the relationship between the supply and the probability of success.

[0258] Based on the same inventive concept, the present application also provides a system for implementing the aforementioned multi-objective replenishment decision-making optimization system. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more of the following embodiments of the multi-objective replenishment decision-making optimization system can be found in the above-mentioned limitations of the multi-objective replenishment decision-making optimization system method, and will not be further elaborated here.

[0259] In an exemplary embodiment, a multi-objective replenishment decision optimization system at sea is provided, comprising:

[0260] The offshore optimization model construction module is used to construct an offshore multi-objective replenishment optimization model; the offshore multi-objective replenishment optimization model includes: a global optimization model and a local optimization model.

[0261] The acquisition module is used to obtain the number of maritime supply targets, the number of maritime supply ships, decision requirements and maritime data sets.

[0262] The decision pool construction module is used to construct a basic decision pool according to the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets.

[0263] The offshore optimization decision module is used to determine the optimization strategy of the basic decision pool according to decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy.

[0264] The replenishment success probability determination module is used to determine the replenishment success probability of each maritime replenishment ship to each maritime replenishment target based on the maritime data set.

[0265] The decision pool optimization module is used to optimize the basic decision pool based on the replenishment success probability and the maritime multi-objective replenishment optimization model using the optimization strategy to obtain the optimal decision pool, and replenish the maritime replenishment target according to the optimal decision pool.

[0266] 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 3 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through 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 the optimal decision pool. 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 through a network connection. When the computer program is executed by the processor, a method for optimizing multi-objective replenishment decisions at sea is implemented.

[0267] Those skilled in the art will understand that Figure 3 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.

[0268] 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.

[0269] 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.

[0270] 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.

[0271] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and 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 may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0272] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0273] 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.

[0274] 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 multi-objective replenishment decision optimization method at sea, characterized in that: The multi-objective replenishment decision optimization method at sea includes: Constructing a multi-objective replenishment optimization model at sea; the multi-objective replenishment optimization model at sea includes: a global optimization model and a local optimization model; Obtain the number of replenishment at sea targets, number of replenishment at sea vessels, decision requirements, and maritime data sets; Constructing a basic decision pool based on the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets; Determine the optimization strategy of the basic decision pool according to the decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy; Determine the probability of success of each sea supply ship in supplying each sea supply target based on the sea data set; Based on the replenishment success probability and the maritime multi-objective replenishment optimization model, the optimization strategy is adopted to optimize the basic decision pool to obtain the optimal decision pool, and the maritime replenishment target is replenished according to the optimal decision pool.

2. The multi-objective replenishment decision optimization method at sea according to claim 1, characterized in that: The expression of the global optimization model is: The expression of the local optimization model is: Among them, F is the global value; is the probability of successful replenishment of the i-th sea supply ship to the j-th sea supply target; st is the constraint condition; x ij It refers to the judgment parameter of the i-th sea supply ship supplying the j-th sea supply target, which takes 0 or 1; μ ij is the supply safety probability of the i-th sea supply ship when it supplies the j-th sea supply target; ρ ij is the supply difficulty of the i-th sea supply ship to the j-th sea supply target; σ ij is the supply feasibility of the i-th sea supply ship to the j-th sea supply target; τ ij is the survival probability of the jth sea supply target when the i-th sea supply ship supplies the j-th sea supply target under the radio silence state of the supply ship; m is the number of sea supply ships; n is the number of sea supply targets; B is a local value.

3. The multi-objective replenishment decision optimization method at sea according to claim 1, characterized in that: Determine the optimization strategy of the basic decision pool based on decision requirements, including: When the decision requirement is a global decision requirement, the optimization strategy of the basic decision pool is determined to be a global optimization strategy; When the decision requirement is a local decision requirement, the optimization strategy of the basic decision pool is determined to be a local optimization strategy.

4. The multi-objective replenishment decision optimization method at sea according to claim 1, characterized in that: The maritime data set includes: time characteristic data, spatial characteristic data, supply ship and resource characteristic data, route data, weather environment data, supply target data and supply mission data; Based on the maritime data set, determine the probability of success of each maritime supply ship in supplying each maritime supply target, including: Determining the feasibility of each marine supply ship supplying each marine supply target based on the time characteristic data, the spatial characteristic data, the supply ship and resource characteristic data, the route data, and the weather environment data; Determining, based on the replenishment target data and the replenishment mission data, the difficulty of replenishing each supply target by each supply ship at sea, and the survival probability of each supply target at sea when each supply ship at sea replenishes each supply target at sea in a radio silence state; determining, based on the route data and the weather environment data, a supply safety probability of each sea supply ship when each sea supply ship supplies each sea supply target; The supply success probability of each sea supply ship to each sea supply target is calculated based on the supply feasibility, the supply difficulty, the survival probability and the supply safety probability.

5. The multi-objective replenishment decision optimization method at sea according to claim 4 is characterized in that: The replenishment target data includes the importance level of the marine replenishment target; Based on the replenishment success probability and the multi-objective replenishment optimization model at sea, the optimization strategy is used to optimize the basic decision pool to obtain the optimal decision pool, including: When the optimization strategy is a global optimization strategy, the basic decision pool is simplified based on the probability of successful replenishment, and the global value of the replenishment plan in the simplified basic decision pool is calculated based on the global optimization model. The replenishment plans in the simplified basic decision pool are sorted in descending order according to the global value. If there are replenishment plans with the same global value, they are sorted again to obtain the optimal decision pool. When the optimization strategy is a local optimization strategy, the supply plans in the basic decision pool are sorted in descending order and simplified based on the importance level and the probability of successful replenishment, and based on the local optimization model, the local values ​​of the supply plans in the sorted and simplified basic decision pool are calculated, and the supply plans whose local values ​​are greater than or equal to the threshold of the probability of successful replenishment are screened out. The screened supply plans are sorted in descending order according to the local values ​​to obtain the optimal decision pool.

6. The multi-objective replenishment decision optimization method at sea according to claim 5, characterized in that: If there are supply plans with equal global values, a secondary sorting is performed, including: According to the importance level, the supply plans with the same global values ​​are sorted in descending order, and based on the local optimization model, the local values ​​of the supply plans with the same global values ​​are calculated. According to the local values ​​of the supply plans with the same global values, the supply plans with the same global values ​​are sorted in descending order to complete the secondary sorting.

7. The multi-objective replenishment decision optimization method at sea according to claim 5, characterized in that: When the optimization strategy is a local optimization strategy, the replenishment plans in the basic decision pool are sorted in descending order and simplified based on the importance level and the probability of successful replenishment. Based on the local optimization model, the local values ​​of the replenishment plans in the sorted and simplified basic decision pool are calculated, and the replenishment plans whose local values ​​are greater than or equal to the threshold of the probability of successful replenishment are screened out. The screened replenishment plans are sorted in descending order according to the local values ​​to obtain the optimal decision pool, which specifically includes: Sort the replenishment plans in the basic decision pool in descending order according to the importance level to obtain a classification decision pool; Simplify the replenishment plans with zero probability of successful replenishment in the classification decision pool to obtain a sorted and simplified basic decision pool; Based on the success probability of replenishment, a local optimization model is used to calculate the local value of each replenishment plan in the sorted and simplified basic decision pool; Calculating a replenishment success probability threshold for each maritime replenishment target based on the replenishment target data and the replenishment mission data; The replenishment plans whose local values ​​are greater than or equal to the replenishment success probability threshold of the corresponding sea replenishment target are screened out, and the plans are sorted in descending order according to the local values ​​to obtain the optimal decision pool.

8. A multi-objective replenishment decision optimization system at sea, characterized by: The multi-objective optimization system for replenishment of multiple supply ships at sea includes: An offshore optimization model construction module is used to construct an offshore multi-objective replenishment optimization model; the offshore multi-objective replenishment optimization model includes: a global optimization model and a local optimization model; An acquisition module is used to obtain the number of sea replenishment targets, the number of sea replenishment ships, decision requirements, and sea data sets; A decision pool construction module is used to construct a basic decision pool according to the number of maritime supply targets and the number of maritime supply ships; the basic decision pool includes multiple supply plans for maritime supply ships to supply maritime targets; The offshore optimization decision module is used to determine the optimization strategy of the basic decision pool according to the decision requirements; the decision requirements include: global decision requirements and local decision requirements; the optimization strategy includes: global optimization strategy and local optimization strategy; A replenishment success probability determination module is used to determine the replenishment success probability of each maritime replenishment ship to each maritime replenishment target based on the maritime data set; The decision pool optimization module is used to optimize the basic decision pool based on the replenishment success probability and the maritime multi-objective replenishment optimization model using the optimization strategy to obtain the optimal decision pool, and replenish the maritime replenishment target according to the optimal decision pool.

9. 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 offshore multi-objective replenishment decision-making optimization method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for optimizing multi-objective replenishment decisions at sea according to any one of claims 1 to 7 is implemented.

Images