Flexible solution acquisition method and device based on multi-objective emergency task allocation

Abstract

The present invention provides a flexible solution acquisition method and device based on multi-objective emergency task allocation, the method includes: based on at least one possible scenario in the development process of an emergency at sea, determining multiple emergency tasks and types of emergency tasks related to the possible scenario; determining candidate emergency subjects that meet task parameter constraints according to multiple emergency goals that need to be achieved; applying a differential evolution algorithm to assign corresponding emergency subjects to each emergency task from candidate emergency subjects according to the task type and multiple emergency goals that need to be achieved, and obtaining the optimal task allocation scheme under the possible scenario; applying an intelligent optimization algorithm based on the optimal task allocation scheme under each possible scenario, obtaining a flexible task allocation scheme that meets the maximum flexible goals of multiple scenarios. Through the above methods, the present invention can achieve flexible, fast, and low-cost solution adjustment for any scenario, and provide scientific and flexible decision support for various marine emergencies.

Description

Technical Field

[0001] The present invention belongs to the technical field of marine rescue, and in particular relates to a method and device for obtaining a flexible solution based on multi-objective emergency task allocation. Background Art

[0002] More than 90% of global commodity trade freight is transported by sea. Complex and intensive maritime economic activities, while driving economic prosperity, are also more prone to various maritime emergencies. From the perspective of the types of ships affected by maritime emergencies, the probability of accidents in bulk cargo ships, sand and gravel ships, and fishing boats is higher, while the probability of accidents in passenger ships is relatively low. Once an accident occurs, it is a major maritime emergency, causing huge harm and impact. Major maritime emergencies such as the collision of the "Sanchi", the explosion of the Nigerian oil tanker, and the grounding of the "Wakashio" have occurred frequently, causing serious impacts on the marine ecological environment and people's lives and property. The causes of maritime emergencies are complex and the types of derivative accidents are diverse. They will involve a complex emergency response task system, covering multiple emergency tasks including lifesaving, firefighting, medical rescue, pollution monitoring, pollution recovery, environmental restoration, accident assessment, etc. Therefore, it is necessary to cooperate with various professional rescue teams formed by local governments, the military, and various departments, as well as the coordinated response of multiple emergency entities such as merchant ships and fishing boats. After the maritime search and rescue center receives the distress signal, it is crucial for decision makers to formulate a scientific and reasonable task allocation plan.

[0003] Maritime emergencies are characterized by multiple types of emergency tasks, changing scenarios, and difficult program adjustments. In addition, due to the influence of complex accident factors and the changing marine environment, maritime emergencies often involve multiple scenarios that evolve over time. For example, a collision of oil tankers may cause an oil spill, which may cause an explosion if the temperature rises. A fire will then occur, resulting in casualties. The evolution of maritime emergency scenarios requires adjustments to the task allocation plan, but program changes for major operations are time-consuming and expensive. Therefore, the decision-making strategy for maritime emergencies must follow the principle of flexibility. In order to ensure that the task allocation plan meets basic emergency requirements while allowing for faster and more cost-effective adjustments, it is necessary for decision makers to develop flexible task allocation plans.

[0004] Flexibility is the execution capability of the supply chain to cope with changing environments. However, there are differences in the definition and measurement of flexibility in different fields, which is related to the characteristics of specific fields. Flexibility theory has been widely used in finance, manufacturing and other fields. The existing task allocation for maritime emergencies is mostly focused on a single accident scenario, rather than on maritime emergencies where multiple accident scenarios occur simultaneously. So far, no one has introduced flexibility theory into the task allocation problem of maritime emergencies. In order to meet the emergency needs of maritime emergencies, it is urgent to carry out research on flexible task allocation schemes. Summary of the invention

[0005] The present invention provides a flexible solution acquisition method and device based on multi-objective emergency task allocation, which solves the problem that the task allocation of marine emergencies is mostly concentrated on a single accident scene.

[0006] Based on the above purpose, the present invention proposes a flexible solution acquisition method based on multi-objective emergency task allocation, including: based on at least one possible scenario in the development process of an emergency at sea, and determining multiple emergency tasks related to any of the possible scenarios and the task type to which each of the emergency tasks belongs; for any of the possible scenarios, determining each candidate emergency subject that meets the task parameter constraints according to the multiple emergency goals that need to be achieved in the possible scenario; for any of the possible scenarios, applying a differential evolution algorithm to assign corresponding emergency subjects to each of the emergency tasks from all of the candidate emergency subjects according to the task type to which each of the emergency tasks belongs and the multiple emergency goals that need to be achieved in the possible scenario, and obtaining the optimal task allocation scheme under the possible scenario; based on the optimal task allocation scheme under each of the possible scenarios, applying an intelligent optimization algorithm to obtain a flexible task allocation scheme that maximizes flexible goals and meets the emergency needs of multiple scenarios.

[0007] Optionally, based on at least one possible scenario in the development process of an emergency at sea, multiple emergency tasks related to any of the possible scenarios and the task type to which each of the emergency tasks belongs are determined, including: analyzing at least one possible scenario that may occur according to the development trend of the emergency at sea; for any of the possible scenarios, determining multiple emergency tasks that need to be performed to handle the possible scenario, and separately determining the task type to which each of the emergency tasks belongs.

[0008] Optionally, for any of the possible scenarios, the candidate emergency subjects that satisfy the task parameter constraints are determined according to a plurality of emergency goals that need to be achieved in the possible scenario, including: for any of the possible scenarios, the task parameters of each emergency task are determined according to a plurality of emergency goals that need to be achieved in the possible scenario, and the candidate emergency subjects whose performance parameters satisfy the task parameter constraints are obtained.

[0009] Optionally, for any of the possible scenarios, according to the task type of each emergency task and the multiple emergency goals that need to be achieved in the possible scenario, a differential evolution algorithm is applied to assign corresponding emergency subjects to each emergency task from all candidate emergency subjects, so as to obtain the optimal task allocation plan for the possible scenario, including: for any of the possible scenarios, according to the multiple emergency goals that need to be achieved in the possible scenario, each emergency task and the task type to which each emergency task belongs, determining multiple objective functions related to the possible scenario; applying a differential evolution algorithm to select from all candidate emergency subjects an emergency subject that optimizes the multiple objective functions and is used to complete each emergency task, and at least one emergency subject is assigned to each emergency task.

[0010] Optionally, the application of the differential evolution algorithm to select an emergency subject that optimizes the multiple objective functions and is used to complete each of the emergency tasks from all the candidate emergency subjects includes: randomly constructing a first initial population based on all the candidate emergency subjects, wherein any individual in the first initial population represents an allocation scheme; applying the differential evolution algorithm to perform cyclic iterative operations of mutation and crossover recombination on the individuals in the first initial population, and selecting optimized individuals according to the multiple objective functions to form the first initial population of the next generation, until a preset first maximum number of iterations is reached, thereby obtaining a first target population; and selecting the allocation scheme corresponding to the individuals with the best multiple objective functions from the first target population as the optimal task allocation scheme.

[0011] Optionally, the selecting optimized individuals according to the multiple objective functions to form the first initial population of the next generation includes: calculating the nominal objective function value of any individual, combining to generate a decision matrix, and performing standardization; based on the standardized decision matrix, performing a weighted sum operation on the objective function values ​​of any individual to obtain the target optimization value of the individual; selecting the optimized individuals according to the target optimization value as the individuals of the next generation to form the first initial population of the next generation.

[0012] Optionally, the application of an intelligent optimization algorithm based on the optimal task allocation scheme under each of the possible scenarios obtains a flexible task allocation scheme that maximizes the flexible target and meets the emergency needs of multiple scenarios, including: calculating a flexible objective function based on the optimal task allocation scheme for each of the possible scenarios; randomly constructing a second initial population based on all of the candidate emergency subjects, wherein any individual in the second initial population represents an allocation scheme; applying a differential evolution algorithm according to the flexible objective function to perform cyclic iterative operations of mutation and crossover recombination on the individuals in the second initial population until a preset second maximum number of iterations is reached, to obtain a second target population, wherein the flexible objective function is used to characterize the matrix norm between the current allocation scheme and the optimal task allocation scheme for each of the possible scenarios; and selecting from the second target population an allocation scheme corresponding to the individual with the optimal flexible objective function as the flexible task allocation scheme that maximizes the flexible target and meets the emergency needs of multiple scenarios.

[0013] Based on the same inventive concept, the present invention also proposes a flexible solution acquisition device based on multi-objective emergency task allocation, including: a task setting unit, which is used to determine multiple emergency tasks related to the possible scenario and the task type to which each emergency task belongs based on at least one possible scenario in the development process of an emergency at sea; a subject candidate unit, which is used to determine each candidate emergency subject that meets the task parameter constraints for any of the possible scenarios according to the multiple emergency goals that need to be achieved in the possible scenario; an optimal solution allocation unit, which is used to apply a differential evolution algorithm to assign corresponding emergency subjects to each of the emergency tasks from all of the candidate emergency subjects for any of the possible scenarios according to the task type to which each emergency task belongs and the multiple emergency goals that need to be achieved in the possible scenario, so as to obtain the optimal task allocation solution under the possible scenario; a flexible allocation unit, which is used to apply an intelligent optimization algorithm based on the optimal task allocation solution under each of the possible scenarios to obtain a flexible task allocation solution that maximizes flexible goals and meets the emergency needs of multiple scenarios.

[0014] Based on the same inventive concept, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the flexible solution acquisition method based on multi-objective emergency task allocation as described above is implemented.

[0015] Based on the same inventive concept, the present invention also proposes a computer storage medium, in which at least one executable instruction is stored, and the executable instruction enables a processor to execute the flexible solution acquisition method based on multi-objective emergency task allocation as described above.

[0016] From the above, it can be seen that the beneficial effects of the technical solution provided by the present invention are: a method and device for obtaining a flexible solution based on multi-objective emergency task allocation provided by the present invention, the method comprising: based on at least one possible scenario in the development process of an emergency at sea, determining multiple emergency tasks related to the possible scenario and the task type to which each emergency task belongs; for any emergency task of any of the possible scenarios, determining each candidate emergency subject that meets the task parameter constraints according to the multiple emergency goals that need to be achieved in the possible scenario; for any of the possible scenarios, according to the task type to which each emergency task belongs and the multiple emergency goals that need to be achieved in the possible scenario, applying a differential evolution algorithm to assign corresponding emergency subjects to each of the emergency tasks from all the candidate emergency subjects, and obtaining the optimal task allocation solution for the possible scenario; based on the optimal task allocation solution for each of the possible scenarios, applying an intelligent optimization algorithm to obtain a flexible task allocation solution that maximizes flexible goals that meets the emergency needs of multiple scenarios, which can achieve flexible, fast, and low-cost solution adjustments for any scenario, and provide scientific and flexible decision-making support for various emergencies at sea. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flow chart of a flexible solution acquisition method based on multi-objective emergency task allocation according to an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a framework for generating a two-stage flexible task allocation solution according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of a linear search method in the second type of task in an embodiment of the present invention;

[0021] Figure 4 Schematic diagram of the algorithm framework of the improved differential evolution algorithm of the embodiment of the present invention;

[0022] Figure 5 An example diagram of the generation process of the optimal task allocation scheme and the flexible task allocation scheme according to an embodiment of the present invention;

[0023] Figure 6 An example diagram of adjustment costs between an optimal task allocation scheme and a flexible task allocation scheme according to an embodiment of the present invention;

[0024] Figure 7 This is an example diagram of contributions of various emergency entities in the flexible task allocation scheme of an embodiment of the present invention;

[0025] Figure 8 It is a structural schematic diagram of a flexible solution acquisition device based on multi-objective emergency task allocation according to an embodiment of the present invention;

[0026] Fig. 9 Schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0028] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0029] The task allocation scheme for maritime emergencies refers to an action plan formulated by decision makers based on the current situation and possible scenarios to implement various emergency tasks to specific emergency entities, with the purpose of performing various emergency tasks on specific emergency entities to reduce the number of deaths, injuries or material losses. In an embodiment of the present invention, the term "emergency entity" refers to a specific entity with emergency response capabilities, that is, a specific execution unit that can provide a certain emergency capability, and therefore includes both organizational departments and ships and aircraft. This is because ships and aircraft are equipment platforms, and generally emergency resources such as search and positioning equipment, oil absorbent felt, oil booms, life nets, etc. will be configured on equipment platforms. Flexibility refers to the ability to respond effectively to a changing environment. Combined with the actual characteristics of maritime emergencies, a flexible task allocation scheme is defined as taking into account the action requirements under multiple possible scenarios, so that the scheme can not only efficiently complete the action goals under specific scenarios, but also agilely respond to subsequent changing scenarios, thereby achieving a flexible, fast, and low-cost adjustment solution for the scheme.

[0030] The present invention implements a flexible solution acquisition method based on multi-objective emergency task allocation, such as Figure 1 As shown, including:

[0031] Step S11: Based on at least one possible scenario in the development process of an emergency at sea, a plurality of emergency tasks related to the possible scenario and a task type to which each of the emergency tasks belongs are determined.

[0032] Maritime emergencies involve a complex emergency task system, and the emergency task system for maritime emergencies needs to be classified. Maritime emergencies refer to incidents in which ships and facilities have fires, explosions, collisions, groundings, sinkings, oil substances or dangerous chemicals leak, and civil aircraft are in distress at sea, causing or may cause casualties and property losses. Therefore, the two main types of accidents at sea can be divided into two categories: people and ships in distress (for example, ship fires, explosions, collisions, groundings, sinkings) and pollutant leaks (for example, accidental oil leaks, other navigation hazards that cause toxic chemicals, and plastic pollution). The emergency action goals of these two accidents are very different. The emergency action for people and ships in distress requires the rescue of human lives, and the emergency tasks involved are classified as life-saving types, while the emergency action for pollutant leaks requires the removal of pollutants, and the emergency tasks involved are classified as pollution removal types. The emergency tasks common to the two types of accidents, such as receiving alarms, monitoring, and commanding, are classified as basic task types. In the embodiment of the present invention, the classification of emergency tasks in marine emergencies is shown in Table 1 below. Emergency tasks can belong to any of three task types. The three task types include the first type of basic tasks T 1 , corresponding to the second type of life-saving task T 2 and the corresponding third type of pollution removal task T 3 There are differences in the allocation rules for different types of emergency tasks, which are closely related to the characteristics of emergency tasks. The subsequent objective function modeling will also vary according to different task types.

[0033] Table 1 Classification of emergency response tasks for maritime emergencies

[0034]

[0035] After an emergency occurs at sea, the maritime emergency command center can obtain information such as the marine environment, accident situation, emergency task requirements, and emergency subject performance parameters of the current accident scene. Table 2 describes the accident elements and environmental parameter information, among which the accident parameters mainly describe the accident disaster situation at the time, including the geographical location Q (lngQ, latQ) when the marine accident occurred, the amount of pollutant leakage A, the radius of the pollutant leakage area R, the number of people to be rescued Num, the search area S and other accident information. The marine environmental parameters include sea conditions B, sea temperature information T, etc., which can be obtained from various meteorological departments and maritime departments.

[0036] Table 2 Marine accident factors and environmental information

[0037]

[0038] Table 3 describes the task parameter information of emergency mission requirements in different scenarios. The task parameter information of emergency mission requirements is determined by the expert group of the maritime emergency command center. Different task types T = {T 1 ,T 2 ,T 3} indicates that the task weight θ i Represents the importance of the task. Decision makers will give basic and maximum capability requirements for different emergency tasks based on their experience and preferences.

[0039] Table 3 Task parameter information of emergency tasks

[0040]

[0041] Table 4 describes the surrounding emergency subjects that can participate in emergency tasks. j The performance parameter information of the emergency subject, in addition to the position coordinates G j (lng j ,lat j ), the distance from the incident point d j , the travel speed V of the emergency subject j , the maximum sea condition allowed for the emergency subject to go out to sea b j In addition, the model also considers the experience E of the emergency subject ji It also includes the performance parameters of the subject participating in different types of emergency tasks, and the performance parameters of the subject participating in basic support tasks. 1 The emergency response capabilities of the main Participate in search and rescue missionsT 2 The sea sweeping width W of the emergency subject j , route spacing S track , coverage C j , discovery probability POD j and the maximum number of people the ship can accommodate j . Participate in pollution cleanup tasks T 3 The weight of pollutants that can be cleaned and recovered per unit time by the emergency subject ef j , and the pollutant containment speed of the emergency subject Most of the parameters in Tables 2-4 can be obtained directly from the emergency departments and are assumed to be known.

[0042] Table 4 Performance parameters of emergency subjects

[0043]

[0044] In the embodiment of the present invention, it is assumed that the survival status of all victims follows a normal distribution, and the time it takes for all search and rescue aircraft to cover the designated search area is the time of the search phase, and the factor of aircraft returning for refueling is not considered here. It is assumed that the special equipment and resources on each ship, such as firefighting, lifesaving, and lifting equipment, are all working normally. The rescue operation will not be terminated due to bad weather conditions or no hope of rescue.

[0045] In step S11, optionally, at least one possible scenario that may occur is analyzed based on the development trend of maritime emergencies; for any of the possible scenarios, a plurality of emergency tasks that need to be performed to handle the possible scenario are determined, and the task type to which each of the emergency tasks belongs is determined respectively.

[0046] Step S12: for any of the possible scenarios, determine candidate emergency subjects that meet the task parameter constraints according to multiple emergency goals that need to be achieved in the possible scenario.

[0047] In an embodiment of the present invention, optionally, for any of the possible scenarios, task parameters of each of the emergency tasks are determined according to a plurality of emergency goals that need to be achieved in the possible scenario, and candidate emergency subjects whose performance parameters satisfy the task parameter constraints are obtained.

[0048] Let X be the decision variable, that is, the task allocation scheme, including the decision on which emergency subjects will perform each emergency task, which can be represented by a decision matrix, where the rows represent the emergency tasks T i , column represents the emergency subject A j , the matrix element x ij is an integer decision variable, and its value represents whether the emergency subject performs the emergency task. ij The value of x is 0 or 1. ij = 0, it means that the emergency subject j does not participate in the execution of emergency task i. ij =1, it means that emergency subject j participates in executing emergency task i.

[0049]

[0050] Generating a task allocation plan for maritime emergencies is a complex decision-making problem. The accident attributes and emergency tasks in different possible scenarios are different, and the locations, capabilities, speeds, cleaning efficiency, search and rescue efficiency, and handling experience of each emergency subject are also different. In order to make successful decisions, it is necessary to consider various flexible real-world needs, allocation rules, and constraints imposed by organizations and the environment. In an embodiment of the present invention, according to the comprehensive marine environmental conditions and the specific attributes of task allocation, there are some constraints in the task allocation plan, which are mainly divided into the following types:

[0051] 1) The number of emergency tasks that each emergency subject can perform is 1, that is, each emergency subject can perform one emergency task at the same time:

[0052]

[0053] 2) Response time constraint: The response time of the emergency subject to arrive at the emergency task location must be less than the response time required by the emergency task. Because when the response time of the emergency subject to arrive at the emergency task location exceeds the response time required by the emergency task, it means that the emergency task may have ended before the emergency subject arrives at the emergency task location.

[0054]

[0055] Among them, RT i represents the expected response time of emergency task i.

[0056] 3) T 1 Basic requirements and constraints of emergency tasks: Execution of emergency tasks T i 1 The capabilities of all emergency entities must meet the basic capability requirements of emergency tasks:

[0057]

[0058] Where n represents the number of emergency entities performing emergency tasks.

[0059] 4) T 2 Basic requirement constraints for emergency tasks: In order to ensure that everyone can be rescued, the number of people that can be accommodated by the rescue boat must be greater than the number of people to be rescued who have fallen into the water.

[0060]

[0061] Among them, u j It represents the upper limit of the number of people that emergency subject j can carry, and J is the total number of emergency subjects.

[0062] 5)T 3 Basic requirements and constraints for emergency tasks: T 3 This type of emergency task requires the removal of pollutants, so the total amount of recycling by all emergency entities should be greater than the current pollutant leakage amount A.

[0063]

[0064] Among them, ef j represents the amount of pollution recovered by the jth emergency subject per unit time, t3 is an intermediate variable, representing the emergency plan x ij Complete T 3 Time, emergency plan xij Refers to the emergency subject j performing emergency task i.

[0065] 6) Maximum sea condition constraints allowed for normal sea travel by emergency entities: In order to ensure that emergency entities perform emergency tasks under safe navigation conditions, the maximum sea condition allowed for operations is j It should be greater than the current sea condition level B. Sea condition level B refers to the degree of fluctuation of the sea surface caused by the strength of the wind. The higher the waves, the greater the level. When the wind and waves are too strong, the sea surface will be bumpy. Some ships will be dangerous to go out to sea in bad sea conditions. Generally speaking, sea conditions are divided into 9 levels according to the effective wave height, the corresponding wind level and the wind speed.

[0066] B≤b j ,b j ∈{0,1,……,9}

[0067] 7) The maximum number of emergency entities that can be accommodated in the sea area: When the area of ​​the sea area to be searched is certain, only a certain number of search and rescue ships and search and rescue aircraft can be allowed to participate in the search and rescue operation at the same time. The main reasons are, first, if there are too many ships and aircraft participating in the search operation at the same time, it will be more difficult for the commander to deploy ships and aircraft to work together; second, it is easy to cause the routes of different search and rescue entities to cross, which is prone to collision accidents.

[0068]

[0069] Where Q s Represents different possible scenarios S i The maximum number of emergency entities involved.

[0070] In step S12, for any possible scenario, an emergency subject whose performance parameters meet the above constraints is selected from all emergency subjects as a candidate emergency subject.

[0071] Step S13: For any of the possible scenarios, according to the task type of each emergency task and the multiple emergency goals that need to be achieved in the possible scenario, a differential evolution algorithm is applied to assign corresponding emergency subjects to each emergency task from all the candidate emergency subjects, so as to obtain the optimal task allocation plan under the possible scenario.

[0072] In the embodiment of the present invention, the generation of a task allocation plan for maritime emergencies is a complex decision-making problem. The accident attributes and emergency tasks in different scenarios are different, and the locations, capabilities, speeds, cleaning efficiency, search and rescue efficiency, and handling experience of each emergency subject are also different. In order to make successful decisions, it is necessary to consider various flexible real-world needs, allocation rules, and constraints imposed by organizations and environments. In order to generate a flexible task allocation plan, the embodiment of the present invention constructs a two-stage system based on multi-objective programming and intelligent optimization algorithms. Figure 2 The flexible task allocation scheme generation framework shown in the figure, in the first stage, generates the optimal task allocation scheme for each possible scenario based on multi-objective planning, and uses the intelligent optimization algorithm to solve it. The embodiment of the present invention incorporates some real-world characteristics, organizational and environmental constraints, considers basic response time, basic capability requirements, safety sea conditions restrictions, the maximum number of emergency entities, etc., and designs three objective functions including minimizing response time, maximizing subject enthusiasm, and maximizing task completion efficiency to measure the allocation efficiency of the scheme. The task completion efficiency is described by constructing capability satisfaction Sa, search and rescue success rate POR, and pollution removal time CT according to the characteristics of the three types of tasks. In the second stage, a maximum flexible goal that meets the emergency needs of multiple scenarios is constructed to optimize the multiple optimal task allocation schemes generated in the first stage, and a flexible task allocation scheme is quickly generated using an intelligent optimization algorithm.

[0073] In step S13, optionally, for any of the possible scenarios, first determine multiple objective functions related to the possible scenario according to multiple emergency goals that need to be achieved in the possible scenario, each of the emergency tasks, and the task type to which each of the emergency tasks belongs. Then, apply the differential evolution algorithm to select the emergency subject that optimizes the multiple objective functions from all the candidate emergency subjects for completing each of the emergency tasks, and at least one emergency subject is assigned to each of the emergency tasks.

[0074] In the first stage, the task allocation of maritime emergencies is mostly decided by the maritime command center, and the emergency tasks are generally allocated based on the emergency action objectives. The embodiment of the present invention determines three action objectives of capability satisfaction Sa, search and rescue success rate POR, and pollution removal time CT by analyzing the actual needs of emergency actions. The weak economy of emergency actions means that the lowest cost is not the goal in the embodiment of the present invention.

[0075] The first goal is to minimize the response time (RT). Since the accident usually occurs at sea, it takes a certain amount of time for each emergency subject to reach the incident center. Due to the urgency of the incident, the principle of proximity and timeliness will be considered when allocating emergency tasks, requiring the emergency subject to arrive at the scene as soon as possible to carry out emergency actions. The geographic coordinates G of each emergency subject are known. j (lng j ,latj ) and the incident center (lng Q ,lat Q ), the distance d between the emergency subject and the scene at the time of the incident can be calculated j , and then calculate the arrival time of emergency subject j to emergency task i, that is, the response time. Since the severity of the marine environment has a great impact on the navigation of the emergency subject unit, a weather penalty parameter ε is added to act on the navigation speed of the emergency subject. Minimizing the response time (RT) specifically refers to the minimum average response time O1 of each emergency subject in a possible scenario. The corresponding calculation formula of the first objective function is as follows:

[0076]

[0077] Among them, R e is the radius of the Earth.

[0078] The second goal is to maximize the subject's enthusiasm (UE). Currently, emergency tasks are mostly assigned directly without considering the subjective wishes of different emergency subjects. In current practice, emergency subjects will have different handling experiences and risk preferences when performing different emergency tasks. If the enthusiasm of the emergency subject is very low, it will affect the execution effect of the emergency task. If the subjective factors are fully considered, the emergency task allocation plan will be more reasonable. The embodiment of the present invention considers that the less difficult the emergency task is and the more experienced the emergency subject is, the more enthusiastic the emergency subject will be to participate in the emergency task. Taking into account each emergency task There may be multiple emergency entities A j If you participate, then the emergency task The enthusiasm of the emergency subject is UE ij , represents the enthusiasm of emergency subject j to participate in emergency task i. The enthusiasm of the emergency subject in any possible scenario of task allocation is obtained by aggregating UE ij The priority of different emergency tasks is mainly considered. To aggregate, The higher the value, the higher the enthusiasm of the emergency subject. ij The subject's experience in participating in the emergency task E ji And the emergency task difficulty dif i k Interactive composition, among which, the treatment experience E ji Positive impact, emergency task difficulty dif i k Inversely related, dif i k The larger the reciprocal of E is, the more willing the emergency subject is to participate in the emergency task. jiThe larger the value, the more willing the emergency subject is to participate in the emergency task. Taking the experience value of 0.5 as the threshold, when the experience value is greater than 0.5, Greater than 1, when the experience value is less than 0.5, If it is less than 1, the enthusiasm of the emergency subject will be even smaller. i k The basic requirement value of emergency task capability and the maximum demand value Decision, it can be seen and The bigger the harder it is, and and The larger the difference, the greater the uncertainty risk and difficulty of the emergency task. Among them, α is a hyperparameter, which is mainly used to prevent division by zero and takes values ​​between 0.01 and 0.1. The specific calculation formula of the second objective function corresponding to maximizing the subject's enthusiasm (UE) is as follows:

[0079]

[0080]

[0081]

[0082] The third goal is to maximize the task completion efficiency (TCE), which requires the emergency subjects to give full play to their respective capabilities so as to achieve the highest efficiency of emergency actions. The intrinsic meaning and measurement criteria of the completion function O3 of the emergency task will vary according to the different task types, which is related to the characteristics of the emergency task itself. Therefore, it is necessary to model the task completion efficiency function for different task types, which is specifically expressed as follows:

[0083]

[0084] Among them, Sa(T 1 ) represents the first type of task T 1 Completion efficiency capability satisfaction, POR (T 2 ) represents the second type of task T 2 The search and rescue success rate of completion efficiency, CT(T 3 ) represents the third type of task T 3 Complete efficient cleaning time.

[0085] Basic Tasks 1Generally, only emergency entities within their scope of responsibility can undertake such tasks. For example, only forecast centers can perform ocean forecasting tasks, and the greater the forecasting capability, the more it can meet the needs of emergency tasks. Therefore, the task completion efficiency TCE of such emergency tasks can be measured by capability satisfaction Sa, which refers to the degree to which the capability level provided by the emergency entity performing the emergency task can meet the needs of the emergency task. According to the calculation process of aggregation from the bottom up, it is divided into a single emergency task TCE. i 1 Satisfaction of ability i And the ability satisfaction of the total action Sa, the first type of task T 1 The calculation formula of the third objective function is as follows:

[0086]

[0087]

[0088]

[0089]

[0090] in, represents the priority level of emergency task i, Represents the emergency subject A participating in the first task type j The k-th guarantee capability value of Indicates emergency task T i 1 The ability value obtained in the kth ability, By participating in emergency tasks i 1 The emergency entities have the ability k The sum of the above ability values. They represent the emergency tasks of the first task type T i 1 The basic and maximum capability requirements, Indicates emergency task T i 1 The degree of competence satisfaction on the kth competence. For a single emergency task T i 1 Satisfaction of ability i In terms of emergency tasks, i 1 May require multiple abilities k For example, atmospheric monitoring capabilities require the joint efforts of upstream monitoring capabilities and downstream monitoring capabilities, neither of which can be missing. iThe modeling principle can be expressed by the wooden barrel principle, that is, if any one of the capabilities required by the emergency task is missing, the emergency task will not be completed. The embodiment of the present invention adopts the method of taking the minimum value to express the actual capability satisfaction of the emergency task, which is expressed as At the same time, for For multiple emergency entities A j Possess the same ability k Can participate in emergency tasks at the same time i 1 For example, both the coast guard ship and the firefighting ship can provide firefighting capabilities and can participate in firefighting tasks at the same time. It can be modeled as an additive coordination rule with diminishing marginal returns. This is because when multiple emergency subjects participate in the emergency task at the same time, when the capabilities are improved to a certain level, the capability satisfaction of the additional emergency subjects will no longer increase.

[0091] Finally, by polymerizing Sa i The total capability satisfaction Sa can be obtained. In the embodiment of the present invention, considering that the execution importance of different emergency tasks in the emergency response action of marine emergencies is different, for example, the life-saving task has a higher priority than the pollution cleaning task, the single capability satisfaction of all emergency tasks is weighted and aggregated with the importance of the emergency task as the weight to obtain the total capability satisfaction Sa (T 1 ).

[0092] Maximizing the search and rescue success rate (POR) is the second task T 2 The completion efficiency of life-saving mission T 2 It is the primary task of emergency action, emphasizing the urgency of the golden rescue time and the survival rate of the victim. Therefore, the measurement of this type of emergency task cannot be simply based on the ability to meet the emergency task. The embodiment of the present invention measures the task completion efficiency TCE by maximizing POR. POR is defined as the product of the search success rate and the survival probability. That is, in emergency actions, the emergency task goal is not only to ensure that the search and rescue target can be successfully searched with a higher probability, but also to take into account the survival status of the search and rescue target within the golden search and rescue time. The second type of task TCE 2 The third objective function is as follows:

[0093] maxPOR(T 2 )=POS×P survival (t)

[0094]

[0095] S j =(T s -t j )·V j S track ·xij

[0096]

[0097] P survival (t)=exp(-0.349t×exp(-0.094×Temp)),t≤T s

[0098] POS stands for search success rate, P survival (t) represents the survival probability. Based on the search theory, the embodiment of the present invention improves the search success rate POS model, which is the product of the inclusion probability POC and the discovery probability POD. Assuming that the search targets are all within the search area, the inclusion probability POC is 1, and the search area size S covered by the search subjects with different performances in the total search time is j Different from POS, POS can be modeled as the probability of discovery POD for each emergency subject covering its assigned search area. j According to the definition of modern search theory, the probability of finding the target to be rescued POD j It can be regarded as coverage C j An exponential function of A measure of how thoroughly the search area has been scanned. j W is the sea sweep width, which is strongly related to different search equipment, ocean environment and the type of target to be rescued. j It is obtained by statistical analysis of a large amount of experimental data and actual data. track is the distance between routes, which depends on the search mode and path, such as Figure 3 As shown, the linear search method is divided into an extended square search mode and a parallel line search mode.

[0099] The success rate of search and rescue activities cannot only be measured by the fastest search and positioning, but also the life survival status of the search and rescue object. The survival time of people in water refers to the longest time that the human body can survive in seawater under normal conditions. Therefore, if the search and rescue at sea exceeds a certain time, the search and rescue object will lose its life characteristics and the search and rescue operation will lose its meaning. According to analysis, considering the duration of the plan execution T s The survival time of the target to be rescued will be affected. The embodiment of the present invention can calculate the survival probability P of the person who falls into the water according to the time and the change of the ambient temperature Temp (0-25°C). survival (t).

[0100] Minimizing the cleaning time (CT) is the third task T 3The completion efficiency. Considering that the main criterion for pollution cleaning tasks is to clean up faster, the embodiment of the present invention measures the task completion efficiency TCE by minimizing the pollution cleaning time CT. When an emergency occurs at sea and pollutants are leaked, the pollutants will drift and spread irregularly due to the influence of waves and sea breeze. If the pollutants are not contained and controlled in time, they will endanger ecological protection areas, breeding areas, and shores. In addition, most pollutants are basically harmful, so it is necessary to quickly dispatch relevant pollution cleaning entities to clean and recycle them. There are many ways to recycle them, which can be professional equipment or manual salvage by fishermen. Therefore, this type of emergency task has two main pollution cleaning activities, namely pollutant containment and pollution cleaning and recycling. The embodiment of the present invention will minimize the disposal time T c The modeling is expressed as minimizing the recovery time under the influence of drift diffusion, using the pollutant diffusion factor β co To represent the diffusion effect of pollutants, the larger the diffusion factor, the lower the disposal efficiency. 3 The third objective function is as follows:

[0101]

[0102] Diffusing factor β co The modeling mainly takes into account that in actual actions, emergency entities will contain pollutant areas. Therefore, the pollutant diffusion process is formed by the simultaneous action of artificial containment and natural diffusion by emergency entities. Assuming that the initial pollutant leakage area is circular and its radius R is known, referring to the end flow diffusion theory, π(R+E) 2 is the area of ​​pollutants after diffusion, E is the diffusion distance, Where K α is the horizontal diffusion coefficient, which can be obtained through specific experiments and numerical simulations under different time scales and ocean currents. α When the time scale is greater than 48h, K α tends to be stable. For time scales less than 24h, K α It increases with the increase of time scale. Δt represents the time for pollutants to spread at sea after leakage, that is, the interval from the occurrence of the accident to the start of containment mission. Indicates the speed at which pollutants diffuse within a certain period of time. The containment speed V of the known containment subject j c , so the ratio of diffusion speed to confinement speed can be expressed as β co The larger the ratio, the larger the diffusion range. re It represents the total time required to recover the pollutants, where A is the total amount of pollutants spilled, ef j It refers to the amount of pollution that can be recovered per unit time.

[0103] In order to reduce the computational complexity and improve the quality of the solution, the algorithm framework of the improved differential evolution algorithm of the embodiment of the present invention is as follows: Figure 4 As shown in the figure, due to the large number of emergency subjects for maritime emergencies, the emergency subjects are first preliminarily screened according to 7 hard constraints, and the candidate emergency subjects that meet the constraints are obtained, and the first initial population that meets the constraints is randomly generated. Then the first stage of solution is carried out, and after mutation and cross-recombination of all feasible solutions, the optimal task allocation scheme under different possible scenarios is generated after selection. In the selection step, each objective function will be forward-oriented and standardized, and finally, each objective function will be weighted and aggregated according to expert experience to obtain the optimal task allocation scheme under each possible scenario. Finally, the second stage of solution is carried out based on the optimal task allocation scheme under each possible scenario, and finally a flexible task allocation scheme is generated.

[0104] In the embodiment of the present invention, when solving the first stage, the first initial population is randomly constructed based on all the candidate emergency subjects, and any individual in the first initial population represents an allocation scheme. Since the constraints are relatively complex, the first initial population is randomly constructed according to the constraints in the first t generations, and the encoding form of the initial individuals is shown in the following formula:

[0105]

[0106] in, represents the pth individual in the 0th generation population, np represents the total number of individuals in the population, H represents the dimension of the decision variable, and its value is the product of the number of candidate emergency subjects and the number of emergency tasks in this possible scenario.

[0107] Then, the differential evolution algorithm is applied to perform cyclic iterative operations of mutation and crossover recombination on the individuals in the first initial population, and the optimized individuals are selected according to the multiple objective functions to form the first initial population of the next generation, until the preset first maximum number of iterations is reached, and the first target population is obtained. The encoding form of the pth individual of the gth generation chromosome population is as follows:

[0108]

[0109] The mutation operation is to randomly select three different individuals in the g-th generation population and Generate new variant individuals using the following formula

[0110]

[0111] Where α is a scaling factor that controls the influence of the difference vector.

[0112] Crossover recombination can enrich population diversity. Parent individual coding and Crossover recombination to generate offspring individuals The specific method is shown in the following formula:

[0113]

[0114] Where cr∈[0,1] represents the crossover probability. rand(0,1) represents a random number that follows a uniform distribution in the interval [0,1].

[0115] Perform a cyclic iterative operation of mutation and crossover recombination on each individual, calculate the objective function values ​​of any individual, combine to generate a decision matrix, and perform a standardized process; based on the standardized decision matrix, perform a weighted sum operation on the objective function values ​​of any individual to obtain the target optimization value of the individual; select the optimized individual as the individual of the next generation according to the target optimization value to form the first initial population of the next generation. In an embodiment of the present invention, the aforementioned three objective function values ​​of any individual are calculated, and the minimization of the response time objective function is converted into maximization and forward, and the calculation is as follows:

[0116] max(r 1. )-r 1.

[0117] Then the three objective functions are standardized, and the calculation formula is shown as follows:

[0118]

[0119]

[0120]

[0121] Among them, R represents the set of three objective functions, r ip is the i-th target value of the p-th individual (scheme), np is the number of individuals in the population, Obj i is the ith objective function value, is the individual in the new generation population, z ij is the standardized objective function value, and Z is the standardized decision matrix.

[0122] Based on the standardized decision matrix, the three objective function values ​​of any individual are weighted and summed to obtain the target optimization value of the individual. The calculation formula is as follows:

[0123] Obj = w i *Z

[0124] Among them, w i are weights of different objective functions.

[0125] After the iteration is completed, the allocation scheme corresponding to the individuals with the best objective functions is selected from the first target population as the optimal task allocation scheme. Optionally, the objective function value of each individual in the first target population is calculated, combined to generate a decision matrix, and standardized; based on the standardized decision matrix, the objective function value of each individual in the first target population is weighted summed to obtain the target optimization value of each individual; the allocation scheme corresponding to the individual with the largest target optimization value is selected from the first target population as the optimal task allocation scheme. In this way, the optimal task allocation scheme under different possible scenarios can be obtained. The optimal task allocation scheme under a possible scenario is calculated as a complete operation. Generally, there are multiple possible scenarios, so it is necessary to call the algorithm for multiple operations.

[0126] At this point, the first stage of solution calculation is completed, and the optimal task allocation plan for each possible scenario is obtained.

[0127] Step S14: Based on the optimal task allocation scheme under each of the possible scenarios, an intelligent optimization algorithm is applied to obtain a flexible task allocation scheme that maximizes the flexible target and meets the emergency needs of multiple scenarios.

[0128] In the embodiment of the present invention, the fourth objective function of the second stage is to maximize flexibility (F). Considering that the maritime emergency command center needs to adjust the current task allocation plan after receiving the updated forward accident information and scene evolution information, but the maritime accident site is far from the shore and the transportation is inconvenient, the plan adjustment is time-consuming and difficult. Therefore, the embodiment of the present invention measures the flexibility target as the minimum change of the plan, so as to flexibly adapt to the emergency needs of maritime emergencies.

[0129] In general real-life emergency actions, the adjustment of the plan generally includes adding, changing, or reducing the allocation of emergency entities. Considering that the decision variables in the embodiment of the present invention are matrices, and the matrix norm can well represent the size of the change caused by the matrix, it is a common basic concept in the fields of matrix theory, linear algebra, and functional analysis in mathematics. Therefore, the embodiment of the present invention uses the matrix norm to describe the changes between plans. Let the plan for the second stage be X1, and the task allocation plan for the first stage be X i , evaluate the flexible min diff1(X1,X1) of the second-stage task allocation scheme X1 i ) can be expressed as matrices X1 and X i The norm of the difference is as follows:

[0130]

[0131] Among them, a ij , b ijThey are task allocation scheme X1 and task allocation scheme X i The smaller the norm, the greater the flexibility of the task allocation scheme X1 in the second stage, and the faster it can adapt to changes in various possible scenarios.

[0132] The adjustment costs between different plans are different. The greater the change, the greater the adjustment cost, resulting in a waste of manpower and material resources. In order to measure the adjustment cost of the plan, the embodiment of the present invention simply models it as an adjustment fee, which refers to the loss caused by the emergency subject's participation in the action. Considering that the emergency subject's participation in the action generally requires a long period of sailing and the call for resources and personnel, the adjustment cost includes transportation fees, fuel fees, insurance agency fees, port fees, repair fees, operating risk fees, and management fees. According to surveys, based on the cost of one day's sea voyage, in general, a small ship costs 30×104 yuan (￥), and a large ship costs 40×104 ￥. The cost of an airplane going to sea is even higher. The embodiment of the present invention knows the cost of each emergency subject performing a single emergency task, and the plan adjustment cost (ΔC) can be calculated by changing the cost of the emergency subject to perform the emergency task (cos j ) aggregation can be obtained:

[0133]

[0134] In step S14, a flexible objective function is calculated based on the optimal task allocation scheme of each possible scenario. For the specific calculation method, please refer to the description of the aforementioned fourth objective function. A second initial population is randomly constructed based on all the candidate emergency subjects, and any individual in the second initial population represents an allocation scheme; according to the flexible objective function, a differential evolution algorithm is applied to the individuals in the second initial population for cyclic iteration operations of mutation and crossover recombination until the preset second maximum number of iterations is reached to obtain a second target population, wherein the flexible objective function is used to characterize the matrix norm between the current allocation scheme and the optimal task allocation scheme of each possible scenario; the allocation scheme corresponding to the individual with the optimal flexible objective function is selected from the second target population as the flexible task allocation scheme that satisfies the maximum flexible objective. That is, the method for obtaining a flexible task allocation scheme is the same as the method in step 3, and the only difference is that the objective function is different.

[0135] The following example is used to illustrate the background of the collision and explosion accident of the Sanchi ship, and multiple emergency tasks involved in the emergency action are allocated to verify the rationality of the flexible solution acquisition method based on multi-objective emergency task allocation and the effectiveness of the algorithm solution of the embodiment of the present invention. The bulk carrier Changfeng Crystal collided with the tanker Sanchi. The condensate oil carried by the Sanchi ship began to leak into the sea, and the maritime search and rescue center received a distress alarm signal. After determining that the location of the collision was 28.21 degrees north latitude and 125.57 degrees east longitude, the emergency action was initiated. The accident was about 160 nautical miles east of the Yangtze River estuary. It was initially estimated that the amount of oil spilled was 300-800 tons, and the radius of the oil spill belt was 5-8 kilometers, causing large-scale oil pollution in the sea. As the sea area where the accident occurred was far from the land, the cargo condensate oil of the Sanchi was flammable and explosive, and the ships were at risk of explosion. A total of 53 crew members on the two ships were in urgent need of rescue. Some of them jumped into the sea, and some were trapped in the cabin. The preliminary estimate was that the search area was about 500 square kilometers, and the average sea surface temperature was 17.4℃. According to measurements, the northerly wind was between 4-5 levels, the wave height was 2.5-3.5 meters, and the visibility was 3-5 nautical miles, which would have an adverse impact on the on-site pollution removal, monitoring and search and rescue. According to the wind level and wave height, according to the parameter query, the current sea condition (B) is level 4. Due to the uncertainty of information in the early stage of the accident, there may be three possible scenarios, and the detailed information is shown in Table 5. The scale of this accident was so large that the national first-level warning was activated, and the Ministry of Transport took the lead in establishing an emergency leadership group to command and coordinate. According to the relevant plan, it is necessary to monitor, evaluate, search and rescue, recover and remove pollution, marine forecast, and environmental restoration in the surrounding sea areas. According to the expert judgment, the emergency task demand information under different possible scenarios is given, as shown in Tables 6-8.

[0136] Table 5 Possible scenarios of the Sanchi accident

[0137]

[0138] Table 6 Basic information of emergency tasks under possible scenario S1

[0139]

[0140]

[0141] Table 7 Basic information of emergency tasks under possible scenario S2

[0142]

[0143] Table 8 Basic information of emergency tasks under possible scenario S3

[0144]

[0145] First, the probability P of possible scenarios S1, S2, and S3 is obtained through random simulationk , 0.2, 0.5, and 0.3 respectively. The decision maker sets the weights w of each objective function under each possible scenario according to the on-site situation. i . Due to the different casualties in different possible scenarios, w i There will be differences. Possible scenarios S1 and S2 belong to possible initial possible scenarios. The initial accident information is incomplete, and emergency subjects need to be dispatched to understand the accident situation first. Therefore, the response time O1 and task completion efficiency O3 are 0.4, and the subject enthusiasm O2 is 0.2. Possible scenario S3 is a spreading scenario. At this time, the accident situation has been basically understood, so the most important goal should be to better complete the emergency task, so the weights of objective functions O1, O2, and O3 are 0.3, 0.2, and 0.5 respectively. In addition, the total number of emergency subjects for possible scenarios S1, S2, and S3 is set to 22, 32, and 36 respectively.

[0146] In addition to the above input data, the longest expected survival time at a sea temperature of 17.4°C is 25 hours. The visibility at sea is 6-9 nautical miles, and the sweep width W of the ship to search for rescued people on the sea surface is j Can be set to 0.7 and 0.9. The aircraft is at 1000 feet below W j The distance between the parallel line search routes is set to 0.1 nautical miles. The diffusion coefficient K of the pollutant α It is related to the time scale. The ship that arrives the fastest after the accident is taken as the start time of the containment mission. The possible scenarios S1 and S2 are 12 hours, and K α The value is 101000cm 3 / s(0.0365km / h), the possible time of scenario S3 is 24h, K α The value is 115000cm 3 / s(0.0414km / h). Based on all the above data, the flexible solution acquisition method based on multi-objective emergency task allocation of the embodiment of the present invention is applied to help decision makers select emergency subjects for each emergency task.

[0147] In the first stage, the optimal task allocation schemes under possible scenarios S1, S2, and S3 are solved to be X1, X2, and X3 respectively. The solution process is performed in PyCharm2020. The specific parameter settings of the improved differential evolution algorithm are as follows: the initial population size np is 200, the number of iterations L1, L2, and L3 under the three possible scenarios S1, S2, and S3 are 1000, the mutation probability Pm is 0.6, and the crossover probability Xovr is 0.7. For each possible scenario, the algorithm is run three times under the same parameter settings, and then the best evolution curve data for each possible scenario is saved. There are a total of 2764 constraints and 576 decision variables. The average time for generating the optimal task allocation schemes for the possible scenarios S1, S2, and S3 does not exceed 3 minutes. After the first stage is completed, the optimal task allocation scheme change curves for the possible scenarios S1, S2, and S3 are as follows: Figure 5 As shown in (abc), the gray curve represents the evolution trend of the optimal task allocation solution, and the black curve represents the evolution trend of the population as a whole. The horizontal axis is the evolutionary generation, which represents the iteration in the process of running the algorithm. The vertical axis is the value of the total objective function Obj of the task allocation solution, which is obtained by aggregating the objective functions O1, O2, and O3 according to different weights, where the objective function O1 is processed by normalization and standardization. Figure 5 (b) It can be seen that the possible scenario S2 can reach the optimal solution of the solution in the 210th generation, and the total target value is 0.832. Combined with the trend of the black line, it can be seen that the entire population conducts a wide-area search before the 180th generation. When the population diversity reaches the maximum at the 180th generation, the algorithm enters the local search, and then the curve gradually rises, indicating that the objective function value of the entire population is getting better and better. At the same time, it can be seen from the gray line that the optimal solution has stabilized. The solution results of the optimal task allocation solutions X1, X2, and X3 under the final three possible scenarios are shown in Table 9.

[0148] Table 9 Optimal task allocation schemes under possible scenarios S1, S2, and S3

[0149]

[0150]

[0151] In the second stage, based on the optimal task allocation schemes X1, X2, and X3, the flexible task allocation scheme X0 is further solved and the specific optimization results are as follows: Figure 5 As shown in (d), since the scheme optimization only involves one emergency target, the change curve is relatively stable. It can be seen from the figure that the optimal solution of the 300th generation scheme gradually approaches stability, and the average norm value of the flexible task allocation scheme is 9.2. The final solution results of the flexible task allocation scheme are shown in Table 10.

[0152] Table 10 Results of flexible task allocation scheme (X0)

[0153]

[0154] In the first stage, the optimal task allocation schemes X1, X2, and X3 were generated for possible scenarios S1, S2, and S3, respectively. The specific task allocation is shown in Table 11. The specific composition of the flexible task allocation scheme generated in the second stage is shown in Table 12. From the input information, it can be seen that the degree of casualties, task priorities, and target weights in different possible scenarios are different. In reality, emergency subjects generally have the ability to perform multiple emergency tasks. For example, the Shanghai Maritime Safety Administration can perform drift prediction tasks and accident assessment tasks. However, considering the workload, one emergency subject can only undertake one emergency task. Therefore, there are great differences in the task allocation under the three possible scenarios.

[0155] Table 11 Detailed display of the optimal task allocation scheme under possible scenarios S1, S2, and S3

[0156]

[0157]

[0158] Table 12 Detailed display of flexible task allocation scheme (X0)

[0159]

[0160]

[0161] In scenario S2, there may be oil spills and explosions from ships, accompanied by casualties, and 12 emergency tasks need to be assigned. First, in order to ensure the safety of emergency entities going out to sea, it is necessary to exclude emergency entities that do not meet the safety requirements for going out to sea. The current sea condition is level 4. Under the restrictions of level 4 sea conditions, the A8 deep-diving salvage ship and the A19 drone cannot go out to sea normally. Then exclude emergency entities that cannot meet basic capabilities. The atmospheric emergency monitoring task requires the emergency entity's capability to reach at least 0.2, but the capability of the A16 China Environmental Monitoring Center is 0.1, which is not competent for this emergency task. Therefore, emergency entities that do not meet the constraints will not be included in the allocation range.

[0162] For basic tasks of basic support type 1, take the atmospheric emergency monitoring task as an example to illustrate, the capacity requirement of this emergency task is within the range of (0.2, 0.7), and only three emergency subjects, A11 Shanghai Pudong Super Air Automatic Monitoring Station, A12 Jiangsu Nantong Super Air Automatic Monitoring Station and A14 Ningbo Super Air Automatic Monitoring Station, were dispatched through calculation. Among them, the dispatchable A13 Hangzhou Bay Wetland Super Air Automatic Monitoring Station was not selected because its response time is longer, and the A14 Ningbo Super Air Automatic Monitoring Station is better in experience. In order to avoid waste of resources, the more emergency subjects participating in the emergency task, the longer the response time O1 will be. Therefore, O1 and O3 become constraints in this embodiment, and it can be realized to consider using as few emergency subjects as possible to efficiently realize the emergency task. In the optimal task allocation scheme, the Sa of 3 emergency subjects participating in the atmospheric monitoring task is 0.945, and if all 4 emergency subjects participate, the Sa is 0.983. It can be seen that adding an emergency subject is not efficient for task completion.

[0163] For life-saving missions such as saving lives 2 Considering the limited survival time of people, the emergency subject is required to arrive at the scene within 12 hours to carry out rescue. However, the A46 fishing boat 1 takes 13 hours to reach the scene due to the long distance, so it does not meet the requirements of this emergency task. It is known that the longest survival time of a person in 17.4℃ water temperature is 25 hours, so the action needs to be completed within 25 hours. Finally, 7 emergency subjects including A28 Huaji "ship 1, A4 Donghai Rescue 101, A17 Coast Guard Ship 31240, A43 South Korean Coast Guard Ship 1, A44 Donghai First Rescue Flight Team, A45 Merchant Ship 1, and A48 Fixed-wing Aircraft were deployed to perform life-saving tasks. At this time, the POR of the plan is 0.22, it takes 7.51 hours to cover the 500km search area, and Psurvival(t) is 0.558. From the input data, it can be seen that the selected emergency subjects are close to the scene and have a fast speed. High, the A33 volunteer ship was not selected because although it was close, it lacked experience. From the results, it can be seen that the joint sea and air search and rescue of aircraft and ships is very necessary. This is because the aircraft has a fast speed and can achieve efficient search, but the aircraft can carry a limited number of people and is not convenient for rescue, while ships can carry more victims and medical equipment. Compared with possible scenario S1, due to casualties in possible scenario S2, A4 Donghai Rescue 101, which was originally performing the on-site command task in possible scenario S1, was adjusted to the personnel search and rescue task in possible scenario S2. This is because the personnel search and rescue task in possible scenario S2 has the highest priority.

[0164] For pollutant cleaning tasks T 3, by calculating the CT of the optimal task allocation plan is 17.8h. If there is no diffusion, it can be calculated that it only takes 14.3h to complete the pollution cleaning. It can be seen that the diffusion of oil pollution has a great impact on the cleaning of marine pollutants. Compared with possible scenario S1, the capacity requirement of the pollutant cleaning task in possible scenario S2 has changed from cleaning 300t of oil to 500t, and the capacity requirement of pollutant containment has changed from a radius of 5km to 6km. By calculating possible scenario S2, 3 A38 oil shipping companies and 2 A40 "Huaji" ships have been added to the recovery and cleaning tasks, and 2 oil shipping companies have been added to the pollution containment tasks. Compared with possible scenario S3, possible scenario S2, A38 oil shipping company 3 performs the recovery and cleaning tasks in possible scenario S2, and is adjusted to the containment task in possible scenario S3. This is because the containment task priority of possible scenario S3 is higher than that of possible scenario S2, which affects the change of the plan.

[0165] In this embodiment, the possible scenarios are S1, S2, and S3. The changes between the possible scenarios will cause the solution to be adjusted. Although there are 6 possible changes between scenarios (S1→S2, S2→S1 are both considered one), there are only 3 differences in the adjustment between the solutions. The adjusted differences are equal), including Adjustment, Adjustment, Adjustment. The following verifies the advantages of the flexible task allocation scheme by comparing the adjustment differences between different schemes. Table 13 shows the change in the total target value after the scheme is adjusted. The adjustment difference between X1→X2 and X0→X1, X2 is analyzed as an example. The practical meaning is that if the decision maker responds according to the optimal task allocation plan X1 of the possible scenario S1 when he is not clear about the specific situation ahead, but as the first batch of rescue personnel arrive at the scene to implement the rescue, they find that there are still crew members falling into the water and missing. At this time, the possible scenario is S2, so it is necessary to adjust from the optimal task allocation plan X1 to the optimal task allocation plan X2. The calculated change in the total objective function value Obj is 0.09. If the flexible task allocation plan X0 is used to allocate at the beginning of the accident, and it is necessary to adjust from the flexible task allocation plan X0 to X2, the change in Obj is 0.05. Conversely, if it is adjusted from X0 to X1, the change in Obj is 0.08. It can be seen that the difference in Obj when X0 is adjusted to X1 or X2 is smaller, which means that the change in X0 is smaller. Similarly, it can be seen from the data in the table as well as The changing trend is the same.

[0166] Table 13 Changes in total target value after program adjustment

[0167]

[0168] The flexible solution acquisition method based on multi-objective emergency task allocation according to the embodiment of the present invention can calculate the cost of solution adjustment under different possible scenarios, such as Figure 6 As shown in the figure, it can be seen that the cost of adjusting the flexible task allocation scheme X0 to any possible scenario is greater than the adjustment cost between different possible scenarios. The adjustment cost between X0 and the optimal solution under possible scenario S2 is the smallest, which is 163x104￥. This is because the probability of occurrence of possible scenario S2 is 0.5, which is the highest. Therefore, the flexible task allocation scheme will be closer to the scheme of possible scenario S2. The adjustment cost of adjusting possible scenario S1 to possible scenario S3 is the highest, which is 544x104￥. This is because the emergency tasks involved in possible scenarios S1 and S3 are the most different, resulting in a large difference between the schemes, so the adjustment is the most difficult. From the analysis, it can be seen that the flexible task allocation scheme can be adjusted to any possible scenario more flexibly and at a low cost, thereby meeting the emergency needs of maritime emergencies.

[0169] In addition, the contribution of different emergency subjects to the action in the flexible task allocation scheme X0 is analyzed to provide decision-making basis for decision makers. The contribution of the emergency subject is defined as the difference between the objective function of the scheme after removing the emergency subject and the objective function of the original scheme. Figure 7 The contributions of all emergency subjects in the flexible task allocation scheme are listed. It can be seen that the contribution of different emergency subjects to the action is different. Among them, the contribution rates of emergency subjects A15 and A21 are the lowest, which shows that their participation is not very meaningful in improving the efficiency of the action. The contribution values ​​of A37 Haixun 22, A38 Oil Transport Company 3, and A42 Donglei 6 Salvage Ship are the largest, indicating that these emergency subjects will play a very important role in the action. This is because A37 Haixun 22 and A38 Oil Transport Company 3 have the strongest containment and recovery capabilities, while A42 Donglei 6 Salvage Ship has the highest experience and is closest to the scene and can respond quickly. Therefore, decision makers need to focus on these emergency subjects when making plan decisions and consider not giving priority to adjustments.

[0170] The embodiment of the present invention first analyzes the differences among the different emergency subjects involved, and can effectively utilize the emergency subjects of all parties to maximize the benefits of the emergency process. Then, by comparing the flexible task allocation scheme with the optimal task allocation scheme under each possible scenario, it can be seen that although the optimal task allocation scheme under the current possible scenario can meet the current needs, it cannot flexibly and low-costly respond to changes in possible scenarios, and the adjustment cost of the scheme is huge. On the basis of being able to meet the needs of emergency tasks, the flexible task allocation scheme can adjust changes faster and at a lower cost, thereby meeting the emergency needs of rapid adjustment. In addition, by analyzing the contribution of different emergency subjects to the action, it can provide reference suggestions for decision makers when making program decisions.

[0171] The embodiment of the present invention is used to solve large-scale maritime emergencies. The characteristics of maritime emergencies are analyzed and the classification of maritime emergency tasks and corresponding emergency goals are given. Then, a two-stage flexible task allocation scheme generation framework is proposed, and the flexible task allocation scheme is defined as being able to achieve flexible, fast and low-cost adjustments. The embodiment of the present invention also develops a multi-objective planning framework to generate a flexible task allocation scheme in two stages, taking into account more comprehensive realistic factors and organizational goals. In the first stage, three objective functions are designed to minimize response time, maximize subject enthusiasm, and maximize task completion efficiency, wherein the task completion efficiency is modeled according to three types of tasks respectively; the second stage is optimized according to the constructed maximum flexible objective function. In order to solve the combinatorial explosion problem, the embodiment of the present invention proposes an improved differential evolution algorithm based on the two-stage flexible task allocation scheme generation problem to quickly and accurately generate a flexible task allocation scheme. Through analysis, it is found that the flexible task allocation scheme can be adjusted to any scenario flexibly, quickly and at low cost, which not only ensures the superiority of the scheme, but also is more suitable for the emergency needs of maritime emergencies.

[0172] In summary, the flexible solution acquisition method based on multi-objective emergency task allocation of the embodiment of the present invention determines multiple emergency tasks related to the possible scenario and the task type to which each emergency task belongs based on at least one possible scenario in the development process of an emergency at sea; for any of the possible scenarios, determine the candidate emergency subjects that meet the task parameter constraints according to the multiple emergency goals that need to be achieved in the possible scenario; for any of the possible scenarios, apply the differential evolution algorithm to assign corresponding emergency subjects to each of the emergency tasks from all the candidate emergency subjects according to the task type to which each emergency task belongs and the multiple emergency goals that need to be achieved in the possible scenario, and obtain the optimal task allocation scheme under the possible scenario; based on the optimal task allocation scheme under each of the possible scenarios, apply the intelligent optimization algorithm to obtain a flexible task allocation scheme that maximizes flexible goals and meets the emergency needs of multiple scenarios, which can realize flexible, fast and low-cost scheme adjustment for any scenario, and provide scientific and flexible decision-making support for various emergencies at sea.

[0173] The above specific embodiments of the present invention are described. In some cases, the actions or steps recorded in the embodiments of the present invention can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0174] The embodiment of the present invention also provides a flexible solution acquisition device based on multi-objective emergency task allocation, such as Figure 8As shown, the flexible solution acquisition device based on multi-objective emergency task allocation includes: a task setting unit, a subject candidate unit, an optimal solution allocation unit and a flexible allocation unit.

[0175] A task setting unit, configured to determine, based on at least one possible scenario in the development of an emergency at sea, a plurality of emergency tasks associated with any of the possible scenarios and a task type to which each of the emergency tasks belongs;

[0176] A subject candidate unit is used to determine, for any of the possible scenarios, candidate emergency subjects that meet the task parameter constraints according to multiple emergency goals that need to be achieved in the possible scenario;

[0177] An optimal solution allocation unit is used to allocate corresponding emergency subjects to each of the emergency tasks from all the candidate emergency subjects according to the task type of each emergency task and the multiple emergency goals that need to be achieved in the possible scenario, and obtain the optimal task allocation solution under the possible scenario by applying a differential evolution algorithm for any of the possible scenarios;

[0178] The flexible allocation unit is used to apply an intelligent optimization algorithm based on the optimal task allocation plan under each possible scenario to obtain a flexible task allocation plan that maximizes the flexible target and meets the emergency needs of multiple scenarios.

[0179] For the convenience of description, the above device is described in terms of functions and is divided into various units and described separately. Of course, when implementing the embodiment of the present invention, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0180] The device of the above embodiment is applied to the corresponding method in the above embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0181] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of the above embodiments is implemented.

[0182] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the method described in any of the above embodiments.

[0183] Fig. 9A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 901, a memory 902, an input / output interface 903, a communication interface 904, and a bus 905. The processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are connected to each other in communication within the device through the bus 905.

[0184] The processor 901 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided by the method embodiment of the present invention.

[0185] The memory 902 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 902 can store an operating system and other application programs. When the technical solution provided by the method embodiment of the present invention is implemented by software or firmware, the relevant program code is stored in the memory 902 and called and executed by the processor 901.

[0186] The input / output interface 903 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0187] The communication interface 904 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0188] The bus 905 includes a path that transmits information between various components of the device (eg, the processor 901 , the memory 902 , the input / output interface 903 , and the communication interface 904 ).

[0189] It should be noted that, although the above device only shows the processor 901, the memory 902, the input / output interface 903, the communication interface 904 and the bus 905, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiment of the present invention, and does not necessarily include all the components shown in the figure.

[0190] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0191] The embodiments of the present invention are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of this disclosure.

Claims

1. A flexible solution acquisition method based on multi-objective emergency task allocation, Its characteristics are: The method comprises: Based on at least one possible scenario in the development process of an emergency at sea, determining a plurality of emergency tasks related to the possible scenario and a task type to which each of the emergency tasks belongs; For any of the possible scenarios, determine candidate emergency subjects that meet the task parameter constraints according to multiple emergency goals that need to be achieved in the possible scenario; For any of the possible scenarios, according to the task type to which each of the emergency tasks belongs and the multiple emergency goals that need to be achieved in the possible scenario, a differential evolution algorithm is applied to assign corresponding emergency subjects to each of the emergency tasks from all of the candidate emergency subjects, so as to obtain the optimal task allocation scheme under the possible scenario; wherein, for any of the possible scenarios, multiple objective functions related to the possible scenario are determined according to the multiple emergency goals that need to be achieved in the possible scenario, each of the emergency tasks and the task type to which each of the emergency tasks belongs; the differential evolution algorithm is applied to select emergency subjects for completing each of the emergency tasks that optimize the multiple objective functions from all of the candidate emergency subjects, and each of the emergency tasks is assigned at least one emergency subject, including: randomly constructing a first initial population based on all of the candidate emergency subjects, and any individual in the first initial population represents an allocation scheme; applying the differential evolution algorithm to perform cyclic iterative operations of mutation and crossover recombination on the individuals in the first initial population, and selecting optimized individuals according to the multiple objective functions to form the first initial population of the next generation, until a preset first maximum number of iterations is reached, and a first target population is obtained; and the allocation scheme corresponding to the individuals with the optimal multiple objective functions is selected from the first target population as the optimal task allocation scheme; The step of selecting optimized individuals according to the multiple objective functions to form a first initial population of the next generation includes: calculating each objective function value of any individual, combining to generate a decision matrix, and performing a standardized process; performing a weighted sum operation on each objective function value of any individual based on the standardized decision matrix to obtain a target optimized value of the individual; selecting optimized individuals according to the target optimized value as individuals of the next generation to form a first initial population of the next generation; Based on the optimal task allocation scheme under each possible scenario, an intelligent optimization algorithm is applied to obtain a flexible task allocation scheme that maximizes flexible objectives and meets the emergency needs of multiple scenarios.

2. The flexible solution acquisition method based on multi-objective emergency task allocation according to claim 1, Its characteristics are: The at least one possible scenario in the development process of the marine emergency is determined based on the possible scenario and multiple emergency tasks related to the possible scenario and the task type to which each emergency task belongs, including: Analyze at least one possible scenario that may occur based on the development trend of maritime emergencies; For any of the possible scenarios, multiple emergency tasks that need to be performed to handle the possible scenario are determined, and the task type to which each of the emergency tasks belongs is determined respectively.

3. The flexible solution acquisition method based on multi-objective emergency task allocation according to claim 1, Its characteristics are: For any of the possible scenarios, determining candidate emergency subjects that meet the task parameter constraints according to multiple emergency goals that need to be achieved in the possible scenarios includes: For any of the possible scenarios, task parameters of each of the emergency tasks are determined according to a plurality of emergency goals that need to be achieved in the possible scenario, and candidate emergency subjects whose performance parameters satisfy the task parameter constraints are obtained.

4. The flexible solution acquisition method based on multi-objective emergency task allocation according to claim 1, Its characteristics are: The method of applying an intelligent optimization algorithm based on the optimal task allocation scheme under each possible scenario to obtain a flexible task allocation scheme that maximizes the flexible target and meets the emergency needs of multiple scenarios includes: Calculating a flexible objective function based on an optimal task allocation solution for each of the possible scenarios; Randomly constructing a second initial population based on all the candidate emergency subjects, wherein any individual in the second initial population represents an allocation scheme; Applying a differential evolution algorithm according to a flexible objective function to perform a cyclic iterative operation of mutation and crossover recombination on individuals in the second initial population until a preset second maximum number of iterations is reached, thereby obtaining a second target population, wherein the flexible objective function is used to characterize a matrix norm between a current allocation scheme and an optimal task allocation scheme for each of the possible scenarios; The allocation scheme corresponding to the individual with the best flexible objective function is selected from the second target population as the flexible task allocation scheme that maximizes the flexible objective and meets the emergency needs of multiple scenarios.

5. A flexible solution acquisition device based on multi-objective emergency task allocation, Its characteristics are: The flexible solution acquisition device based on multi-objective emergency task allocation includes: A task setting unit, configured to determine, based on at least one possible scenario in the development process of an emergency at sea, a plurality of emergency tasks related to the possible scenario and a task type to which each of the emergency tasks belongs; A subject candidate unit is used to determine, for any of the possible scenarios, candidate emergency subjects that meet the task parameter constraints according to multiple emergency goals that need to be achieved in the possible scenario; The optimal solution allocation unit is used for allocating corresponding emergency subjects to each of the emergency tasks from all of the candidate emergency subjects according to the task type to which each of the emergency tasks belongs and the multiple emergency goals to be achieved in the possible scenario, and obtaining the optimal task allocation solution under the possible scenario by applying the differential evolution algorithm for any of the possible scenarios; wherein, for any of the possible scenarios, multiple objective functions related to the possible scenario are determined according to the multiple emergency goals to be achieved in the possible scenario, the emergency tasks and the task type to which each of the emergency tasks belongs; the differential evolution algorithm is applied to select the emergency subjects for completing each of the emergency tasks that optimize the multiple objective functions from all of the candidate emergency subjects, and each of the emergency tasks corresponds to at least one emergency subject, including: randomly constructing a first initial population based on all of the candidate emergency subjects, and the first initial population Any individual in the group represents an allocation scheme; applying the differential evolution algorithm to perform cyclic iterative operations of mutation and crossover recombination on the individuals in the first initial population, and selecting optimized individuals according to the multiple objective functions to form the first initial population of the next generation, until the preset first maximum number of iterations is reached, and the first target population is obtained; the allocation scheme corresponding to the individuals with the best multiple objective functions is selected from the first target population as the optimal task allocation scheme; the selecting optimized individuals according to the multiple objective functions to form the first initial population of the next generation includes: calculating the objective function values ​​of any individual, combining to generate a decision matrix, and performing standardization processing; based on the standardized decision matrix, performing a weighted sum operation on the objective function values ​​of any individual to obtain the target optimization value of the individual; selecting the optimized individuals as the individuals of the next generation according to the target optimization value to form the first initial population of the next generation; The flexible allocation unit is used to apply an intelligent optimization algorithm based on the optimal task allocation plan under each possible scenario to obtain a flexible task allocation plan that maximizes the flexible target and meets the emergency needs of multiple scenarios.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the method for obtaining a flexible solution based on multi-objective emergency task allocation according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the method for obtaining a flexible solution based on multi-objective emergency task allocation according to any one of claims 1 to 4 is implemented.

Images