An offshore oil spill emergency dispatch method, system, device and medium

Abstract

The application discloses a kind of offshore oil spill emergency scheduling method, system, equipment and medium, it is related to emergency scheduling technical field, including steps: obtaining oil film diameter, containment material demand, determine demand point position change;Based on the optimized environmental sensitive index ESI, the sensitivity of sea area is classified, and the environmental pollution loss is quantified by combining the oil spill influence time, oil film area and value conversion coefficient;A double-objective model is established, and the double-objective model is solved by using the optimized genetic algorithm to obtain the scheduling scheme suitable for emergency accident scene.The application constructs a double-objective model, which can capture the change of ecological loss increment rate per unit time in real time, drive the emergency resources to be deployed in advance to the upstream of the high-sensitive area, avoid the redundant transportation of materials or insufficient containment capacity due to inaccurate demand prediction, and effectively reduce the total cost of emergency scheduling and ecological damage.

Description

Technical Field

[0001] This invention relates to the field of emergency dispatching technology, and in particular to a method, system, equipment and medium for emergency dispatching of marine oil spills. Background Technology

[0002] The continued growth in demand for maritime oil transportation and the significant increase in the frequency of tanker voyages have led to a corresponding rise in the risk of accidents such as tanker collisions and groundings. Simultaneously, the expansion of offshore oil and gas exploration and development activities has increased the likelihood of sudden oil spills due to equipment malfunctions and operational errors on drilling platforms. Oil slicks formed by leaked oil drift and spread under the influence of ocean currents, waves, and wind, posing a serious threat to marine ecosystems, human safety, and economic activities. Scientific and efficient emergency material deployment is crucial for improving response efficiency and minimizing losses.

[0003] The drift and diffusion of oil slicks is the core source of the "dynamic" nature of emergency dispatch. Regarding path dynamics, in 2014, a multi-node dispatch framework was first proposed for the first time, targeting drifting targets on the water, laying the foundation for dynamic path planning on the water and breaking through the limitations of static paths. However, it did not quantify the impact of oil slick drift speed and direction on the path. Existing technologies can further construct dynamic path planning models to reveal the mutual influence between oil slick drift and emergency dispatch, but these models focus on the single driving force of oil slick dynamics on the path, without considering the moderating effect of differences in marine environmental and ecological sensitivity on dispatch priorities. Considering the uncertainty of emergency resource demand and the time-varying characteristics of demand points, a dual-objective model with transportation costs and environmental losses as optimization objectives is established, and a genetic-simulated annealing hybrid algorithm is used for solution. However, its oil slick dynamics are only reflected in changes in navigation paths, without linkage with the priority of sensitive area protection. The above studies provide theoretical and methodological support for the dispatch of marine oil spill emergency supplies from the dimensions of water dispatch characteristics, dynamic path planning, and target characteristics, but lack priority processing for sensitive areas.

[0004] However, existing research often homogenizes the areas requiring rescue, neglecting the differences in marine ecological environments. In 2022, the study first introduced the differences in material demand for marine oil spill emergency dispatch into the research, constructing a two-layer programming model that considers rescue time windows and material priorities, and designing a hybrid heuristic algorithm for optimization. However, it only distinguishes priorities based on "rescue time windows" and does not take the sensitivity of marine ecology as the core basis for priority division, resulting in limitations in the dispatch scheme. Du Jian et al. used "pollution cleanup scenarios" to characterize sensitive resource areas along the coastline and constructed a robust dispatch model for oil booms, incorporating the type of sensitive area into the dispatch constraints for the first time. However, the model only focuses on "static sensitive area containment" and does not address the dynamic handling of oil spill slicks. In reality, oil slicks drift over a wide range, and the environments of different areas within the same sea area are different. There is an urgent need for a dispatch method that comprehensively considers the differences in regional environments and the time-varying characteristics of oil slicks within the same sea area.

[0005] While existing technologies have recognized the differences in marine ecological environments and attempted to prioritize them through rescue time windows or static marking of sensitive resource areas, and used the dynamics of oil slicks to drive adjustments in ship navigation paths, they essentially still homogenize or quasi-statically partition marine areas. The Environmental Sensitivity Index (ESI) is only used as a background layer and cannot calculate the differentiated ecological and environmental pollution losses under "specific accident scenarios." As a result, when oil slicks drift from low-sensitivity areas to high-sensitivity areas, they cannot be used as the core optimization target for pre-emptive intervention in emergency dispatch. Consequently, emergency resources are often passively deployed after pollution occurs, resulting in poor emergency dispatch effects and high dispatch costs and ecological damage. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the prior art by providing a method, system, equipment, and medium for emergency dispatching of marine oil spills, thereby solving the problems in the prior art.

[0007] The present invention specifically provides the following technical solution: A method for emergency dispatch of marine oil spills, comprising the following steps: The diameter of the oil film in the oil spill area obtained by the diffusion of the oil film under the action of wind, waves and fluid is obtained. The amount of containment materials required at different times is obtained by the positive correlation between the oil film diameter and the amount of containment materials required. The location change of the demand point is obtained according to the drift speed of the oil film under the action of wind and ocean currents. A geographic information system was introduced into the Environmental Sensitivity Index (ESI) classification model to construct an optimized ESI classification model. The marine area was classified by sensitivity index using the optimized ESI, and the ecological and environmental pollution loss was quantified by combining the sensitivity index classification results, oil spill impact time, oil film area and value conversion coefficient. A dual-objective model is established with the objectives of minimizing the added value of ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch. The total cost of emergency dispatch is based on the configuration of emergency bases and ships, and is obtained in combination with the demand for containment materials at the demand points and the changes in the location of the demand points. By inputting the configuration of emergency bases and vessels and the location data of demand points into the dual-objective model, a dispatching scheme adapted to the marine oil spill emergency scenario can be obtained.

[0008] Preferably, the step of obtaining the change in the location of the demand point based on the drift velocity of the oil film under the influence of wind and ocean currents specifically involves: Obtain the ocean surface current velocity vector and the wind speed vector above the sea surface, and obtain the oil film drift velocity by weighted sum of the ocean surface current velocity vector and the wind speed vector above the sea surface; The location of the required point after drifting is determined by the oil film drift speed, drift angle, and ship sailing time; wherein the location of the required point serves as a condition for updating the ship's sailing path.

[0009] Preferably, the Environmental Sensitivity Index (ESI) grading model is introduced into a geographic information system to construct an optimized ESI grading model, specifically as follows: By weighting ecological vulnerability (E), socioeconomic value (S), and pollution recovery capacity (H), the optimized Environmental Sensitivity Index (ESI) classification model is obtained, and its specific expression is as follows: ; in, , , Ecological vulnerability E Socioeconomic value S and pollution recovery capacity H The weighted calculation weights.

[0010] Preferably, the dual-objective model is established with the objectives of minimizing the increased ecological and environmental pollution loss before and after emergency dispatch, and minimizing the total cost of emergency dispatch, specifically as follows: Minimize the increase in ecological and environmental pollution losses before and after emergency dispatch F 1; The specific expression is: ; in, The ecological and environmental pollution losses at the end of the dispatch period, This refers to the ecological and environmental pollution losses at the start of the dispatch process; Minimize the total cost of emergency dispatch; the specific expression is: ; in, To minimize the total cost of emergency dispatch, including the cost of emergency base deployment, vessel deployment, and vessel navigation, The cost of deploying a single emergency base, The cost of deploying a single emergency vessel, The cost of navigation per unit distance for emergency vessels. As decision variables, For the fleet k The number of ships equipped, As decision variables, V For the ship's speed, For the fleet k From node i To the node j The sailing time.

[0011] Preferably, after establishing the dual-objective model, the method further includes: By introducing 2-opt local optimization and an adaptive mutation strategy based on population diversity into the genetic algorithm, an optimized genetic algorithm is obtained. The optimized genetic algorithm is then used to solve the bi-objective model, and the optimal scheduling scheme of the bi-objective model is output.

[0012] Preferably, the step of solving the bi-objective model using the optimized genetic algorithm specifically involves: A chromosome is constructed using a two-layer integer encoding method. During decoding, the nearest emergency base is selected as the starting point based on the centroid of the sub-path demand point cluster. The path is updated by combining the drift dynamics of the demand points, and the number of ships is allocated according to demand and ship capacity. The chromosome consists of a demand point sequence and implicit split points. The demand point sequence represents the fleet's access order, and the split points divide the sequence into sub-paths. The scheduling cost and environmental pollution loss are obtained by analyzing the number of ships and routes. The scheduling cost and environmental pollution loss are normalized and then weighted to obtain the fitness, and a fitness assessment is performed. Based on the fitness, a hybrid strategy of elite retention and roulette wheel selection is used for selection, and 2-opt local optimization is performed on elite individuals; the elite individuals refer to individuals in the population with fitness higher than a threshold. A sequential crossover operation is adopted, and the mutation rate is dynamically adjusted based on the standard deviation of the population fitness after the crossover operation. A multi-strategy mutation operation is performed, and individuals are inserted into the mutated population. The optimal individual output is used to obtain a scheduling scheme adapted to the marine oil spill emergency scenario.

[0013] Preferably, the adaptive mutation strategy is to dynamically adjust the actual mutation rate between the basic mutation rate and the maximum mutation rate based on the ratio of the current population fitness standard deviation to the initial population fitness standard deviation.

[0014] This invention provides a method and system for emergency dispatching of marine oil spills, comprising: The demand acquisition module is used to acquire the diameter of the oil film that has spread to the oil spill area under the action of wind, waves and fluid. It obtains the demand for containment materials at different times by the positive correlation between the oil film diameter and the demand for containment materials, and obtains the change in the location of the demand point based on the drift speed of the oil film under the action of wind and ocean currents. The loss quantification module is used to introduce the Environmental Sensitivity Index (ESI) classification model into the geographic information system, construct an optimized ESI classification model, classify the sea area based on the optimized ESI, and quantify the ecological and environmental pollution loss by combining the sensitivity index classification results, oil spill impact time, oil film area and value conversion coefficient. The objective construction module is used to establish a dual-objective model with the objectives of minimizing the added value of ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch; the total cost of emergency dispatch is based on the configuration of emergency bases and ships, and is obtained in combination with the demand for containment materials at the demand points and the changes in the location of the demand points; The scheduling module is used to input the configuration of emergency bases and ships and the location data of demand points into the dual-objective model to obtain a scheduling scheme adapted to the marine oil spill emergency scenario.

[0015] The present invention provides a computer device, including a memory and a processor. The memory stores a program, and when the program is executed by the processor, the processor performs the steps of the above-described method for emergency dispatching of marine oil spills.

[0016] The present invention provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described method for emergency dispatching of marine oil spills.

[0017] Compared with the prior art, the present invention has the following significant advantages: This invention utilizes a geographic information system (GIS) to spatially optimize and classify the ESI (Effective Environmental Pollution Index), and couples the oil spill impact time, instantaneous oil slick coverage area, and value conversion coefficient to achieve dynamic quantification of differentiated ecological and environmental pollution losses under specific accident scenarios. It constructs a dual-objective model based on the total emergency dispatch cost, the configuration of emergency bases and vessels, and the ecological and environmental pollution loss increment resulting from changes in the location of demand points, as well as the total emergency dispatch cost. As the oil slick drifts from a low-sensitivity area to a high-sensitivity area, the model can capture the real-time change in the ecological loss increment rate per unit time and use it as the core optimization objective. This drives emergency resources to be deployed upstream of high-sensitivity areas in advance, enabling the types, loads, and navigation routes of vessels dispatched from emergency bases to dynamically adapt to the entire process of oil slick expansion and drift. This avoids redundant material transportation or insufficient containment capacity due to inaccurate demand prediction, effectively reducing the total emergency dispatch cost and ecological damage. Attached Figure Description

[0018] Figure 1 The present invention provides an emergency dispatch diagram for marine oil spill accidents; Figure 2 This is a schematic diagram of the changes in the navigation route provided by the present invention; Figure 3 The flowchart of the improved genetic algorithm provided by this invention; Figure 4 This is a decoding diagram provided by the present invention; Figure 5 The 2-opt optimization diagram provided by this invention; Figure 6 This is a cross diagram provided by the present invention; Figure 7 This is a schematic diagram of the multi-strategy variation provided by the present invention; Figure 8 This is a schematic diagram of simulated sea area sensitivity division provided by the present invention; Figure 9 This is a performance comparison chart of the algorithm provided by the present invention; wherein Figure 9 (a) represents a performance comparison chart of the C101_25 algorithm. Figure 9 (b) represents a performance comparison chart of the C101_50 algorithm. Figure 9 (c) represents the performance comparison chart of the C101_75 algorithm. Figure 9 (d) represents the performance comparison chart of the C101_100 algorithm; Figure 10 The convergence curve of the algorithm provided by this invention; Figure 11 The flowchart of an emergency dispatch method for marine oil spills provided by the present invention is shown. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] This paper integrates the ecological environment sensitivity index classification with the time-varying nature of oil films, and introduces an environmental sensitivity index calculation method based on ecological vulnerability, socio-economic value, and pollution recovery capacity. Furthermore, it establishes an emergency dispatch model that combines the time-varying characteristics of oil films under the influence of wind and waves at sea, taking into account the time-varying nature of oil films and the environmental sensitivity classification. With the optimization objective of minimizing dispatch costs and the added value of environmental pollution losses, an improved genetic algorithm with embedded 2-opt local optimization and adaptive mutation strategies is designed to solve the problem based on local optimization and avoiding premature convergence. The effectiveness of the method is verified through examples and control experiments.

[0021] The marine oil spill emergency response network consists of emergency bases, points of need, and rescue vessels. After an accident occurs, it is necessary to first analyze the characteristics of the scene, such as weather, sea conditions, oil spill characteristics, and the environmental sensitivity index of the accident area. Then, coordinate with the various emergency bases to form a rescue fleet and plan the dispatch scheme: the fleet departs from the emergency base, proceeds to each point of need according to the route, and returns to the emergency base (the process is as follows). Figure 1 (As shown).

[0022] This study focuses on two core issues: first, the dynamic changes in demand over time (location drift and range diffusion); and second, the significant differences in pollution sensitivity among different areas within the same sea area. The study aims to minimize emergency rescue transportation costs and environmental pollution losses in the early stages of an accident through the scientific allocation of rescue resources.

[0023] The basic assumptions are as follows: (1) Basic information such as the location of the oil spill point can be obtained before dispatching, and no new oil film will be generated after the rescue begins; (2) The wind and wave environment is stable, and the oil film and the ship speed are constant; (3) Each demand point is rescued by only one fleet; (4) The rescue begins immediately after the fleet arrives at the demand point, and the demand point stops drifting and spreading.

[0024] like Figure 11 As shown in the figure, a method for emergency dispatch of marine oil spills in this embodiment includes the following steps: Step 1: Based on the diffusion model of oil film under the action of wind, waves and fluid, obtain the diameter of the oil film in the oil spill area obtained by the diffusion under the action of wind, waves and fluid. Obtain the demand for containment materials at different times by the positive correlation between the oil film diameter and the demand for containment materials. And obtain the change of the position of the demand point according to the drift speed of the oil film under the action of wind and ocean current.

[0025] Mathematical model: The parameters, variables, and symbols in the model are explained below: Table 1 Symbol Definitions 1. Dynamic demand calculation: Oil films diffuse under the influence of wind, waves, and fluids at sea, and their diameter increases non-linearly over time. ; in, For demand points i The diameter of the oil film at time T; The density of the spilled oil. The density of seawater; g It is the acceleration due to gravity; For demand points i The amount of oil spilled; The viscosity coefficient; This is the net surface tension coefficient.

[0026] The demand for containment materials is positively correlated with the diameter of the oil film: ; in, For demand points exist The required amount of oil booms at any given time (rounded up). The standard containment length for a single oil boom. To reserve redundancy.

[0027] 2. Analysis of changes in the location of demand points: The oil slick drifts under the influence of wind and ocean currents, allowing us to obtain the ocean surface current velocity vector. wind speed vector over the sea surface The drift velocity of the oil film is obtained by weighting the ocean surface current velocity vector with the wind speed vector above the sea surface. The specific expression is: ; The ship's navigation path will dynamically change as the demand point shifts. Figure 2 This describes the changes in the navigation path after the demand point drifts.

[0028] The location of the demand point after drift is determined by the oil film drift velocity, drift angle, and ship sailing time; the location of the demand point serves as a condition for updating the ship's sailing path, and the ship's actual sailing satisfies dynamic geometric constraints: (4); Step 2: Introduce the Environmental Sensitivity Index (ESI) grading model into a geographic information system to construct an optimized ESI grading model. Use the optimized ESI to classify the sea area based on its sensitivity index, and combine the sensitivity index grading results, oil spill impact time, oil film area, and value conversion coefficient to quantify the ecological and environmental pollution losses.

[0029] Environmental classification principles and mathematical models: The Environmental Sensitivity Index (ESI) directly reflects the importance and protection priority of sensitive marine resources. This study introduces an improved ESI classification model based on Geographic Information System (GIS), weighted by ecological vulnerability (E), socio-economic value (S), and pollution recovery capacity (H), to obtain the optimized ESI classification model, expressed as:

[0030] ; Sensitivity grading standards: Based on ESI values, different areas within the same sea area are divided into 5 sensitivity levels, as shown in Table 2: Table 2 Sensitivity Grading Standards Environmental pollution loss assessment: To accurately quantify the environmental pollution losses caused by oil spills, the Florida formula is introduced and modified according to the research content of this invention, integrating ESI level p, oil spill impact time t, oil film area S, and value conversion factor E: (6); Step 3: Establish a dual-objective model with the goals of minimizing the added value of ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch. The total cost of emergency dispatch is based on the configuration of emergency bases and ships, combined with the demand for containment materials at the demand points and the changes in the location of the demand points, and establishes constraints on the fleet's carrying capacity to meet the total demand.

[0031] Based on the above analysis, this study constructs a dual-objective model: ; in, The ecological and environmental pollution losses at the end of the dispatch period, To account for the ecological and environmental pollution losses at the start of scheduling, the function To minimize the increase in ecological and environmental pollution losses before and after emergency dispatch.

[0032] Minimize the total cost of emergency dispatch; the specific expression is: ; function To minimize the total cost of emergency dispatch, including the cost of emergency base deployment, vessel deployment, and vessel navigation. Among these, To minimize the total cost of emergency dispatch, including the cost of emergency base deployment, vessel deployment, and vessel navigation, The cost of deploying a single emergency base, The cost of deploying a single emergency vessel, The cost of navigation per unit distance for emergency vessels. As decision variables, For the fleet k The number of ships equipped, As decision variables, V For the ship's speed, For the fleet k From node i To the node j The sailing time.

[0033] Constraints: ; ; ; ; ; ; ; ; ; ; ; Equation (9) represents the total demand of the service points that the fleet's carrying capacity meets; Equation (10) represents that each demand point is rescued by only one fleet; Equation (11) represents that the departing fleet must go to the demand point; Equations (12) and (13) represent that the fleet departs from the only emergency base and cannot travel between emergency bases; Equation (14) represents the flow conservation between nodes; Equation (15) represents that the access order of path nodes strictly follows the chronological order; Equation (16) represents the arrival time relationship between successive demand nodes; Equation (17) represents that the fleet must return to port; Equations (18)-(19) represent the logical relationship between the fleet's departure, the existence of the path, and the responsible demand point.

[0034] Step S4: Input the configuration of emergency bases and ships and the location data of demand points into the dual-objective model to obtain a dispatching scheme adapted to the marine oil spill emergency scenario.

[0035] After establishing the dual-objective model, the following steps are also taken: introducing 2-opt local optimization and an adaptive mutation strategy based on population diversity into the genetic algorithm to obtain an optimized genetic algorithm, and then solving the dual-objective model through the optimized genetic algorithm to obtain a scheduling scheme adapted to the marine oil spill emergency scenario.

[0036] The marine oil spill emergency dispatching problem studied is an NP-hard problem of multi-node dynamic path planning, which is difficult to solve quickly with exact algorithms. While genetic algorithms have the advantage of global search, they suffer from premature convergence and insufficient local optimization. Therefore, an improved genetic algorithm is designed, introducing 2-opt local optimization and an adaptive mutation strategy based on population diversity to enhance local optimization capabilities and avoid premature convergence. The population represents multiple possible ship dispatching schemes in each generation. The algorithm flow is as follows: Figure 3 .

[0037] Encoding and Decoding: A chromosome is constructed using a two-layer integer encoding method: the chromosome consists of a "demand point sequence + implicit split points". The demand point sequence represents the fleet's access order, and the split points divide the sequence into sub-paths (each corresponding to a fleet task). During decoding, the nearest emergency base is selected as the starting point based on the centroid of the demand point clusters along the sub-paths. The path is updated dynamically based on the demand point drift, and the number of ships is allocated according to demand and ship capacity. Decoding results are then provided. Figure 4 .

[0038] The population is initialized and preprocessed. Scheduling costs and environmental pollution losses (EVL) are obtained using ship counts and routes. Scheduling costs and EVL are normalized to eliminate dimensional differences between the two, and then a bi-objective weighted average is applied to obtain fitness. Fitness is then evaluated. ; in, =0.5、 =0.5, C is the scheduling cost, and EVL is the environmental pollution loss.

[0039] The selection operation employs a hybrid strategy of elite retention and roulette wheel betting, combined with a 2-opt strategy to optimize elite individuals.

[0040] Based on fitness, a selection operation is performed using a hybrid strategy of elite retention and roulette wheel selection, and 2-opt local optimization is applied to elite individuals; elite individuals are those in the population whose fitness is higher than a threshold.

[0041] Choose between an elite retention strategy and a roulette-style hybrid strategy: The elite retains the top 20% of the fitness individuals in each generation for direct inheritance to avoid the loss of good genes; the remaining individuals are selected by roulette using normalized fitness (Equation (21)).

[0042] (twenty one); Where N is the population size. The worst adaptability in modern times. It serves as a smoothing factor to avoid zero probability.

[0043] 2-opt local optimization for elite individuals: Improved genetic algorithms that fit with other strategies have been shown to enhance the algorithm's optimization performance. To optimize the local performance of elite individuals, a 2-opt path correction is introduced: for elite individuals with intersecting paths, the sub-paths are traversed and the breakpoints are combined. Intersecting edges are eliminated by reversing the sub-paths, and the corrected fitness is calculated. If it is better, the individual is updated. (e.g.) Figure 5 (As shown).

[0044] Crossover operation: Sequential crossover is used: the parent crossover interval is randomly selected. The gene sequence within the specified interval is preserved, and the remaining genes are filled in according to another parental sequence to ensure that the offspring inherit the gene sequence characteristics of the parent. (e.g.) Figure 6 (As shown).

[0045] Adaptive Mutation: After the crossover operation, the mutation rate is dynamically adjusted based on the population diversity (fitness standard deviation), and a multi-strategy mutation operation is performed. When the maximum number of iterations is reached, the optimal individual output is used to obtain a scheduling scheme adapted to the marine oil spill emergency scenario. The diversity level is dynamically measured by calculating the population fitness standard deviation (Equation (22)). When the diversity decreases, the mutation rate is adjusted from the base value. =0.05 adaptive boost to the upper limit =0.2 (Equation (23)), enhancing the ability to explore the solution space; at the same time, multi-strategy mutation operations such as gene exchange, sub-path reversal and gene insertion are introduced ( Figure 7 ).

[0046] ; in, Indicators representing population diversity For population size, Indicates the individual's fitness value. This represents the average fitness of the population.

[0047] ; in, Indicates the adaptive mutation rate. , The distribution represents the basic and maximum variability. , The distribution represents the current and initial population diversity.

[0048] Example Background: This study selects a major oil spill incident in the Bohai and Yellow Seas as the simulation object. The Bohai and Yellow Seas are typical representatives of the coastal ecosystems of northern my country, combining the characteristics of "ecological protection areas" and "conventional development areas," and these sea areas are also high-incidence areas for marine oil spill accidents in my country. The relevant parameters are set as follows: Within 6 hours of the accident response, 24 discrete oil slicks formed in the sea area, drifting in directions of 37° south of east and 24° north of west, respectively. There are three emergency bases in the vicinity: Dalian, Yantai, and Weihai. The cost of deploying each emergency base is 50,000 yuan. The average ship speed is 22 km / h, the maximum carrying capacity is 10 inflatable oil booms of 30m each, the ship deployment cost is 5,000 yuan per vessel, and the unit transportation cost is 450 yuan / km.

[0049] The simulation area is a 100km × 100km sea area surrounding the emergency base, divided into 10 × 10 cells. The ESI value of each cell is calculated using ESI = 0.5E + 0.3S + 0.2H. Ecological vulnerability (E) is quantified based on the distribution of marine ecosystem types: nature reserves (E=8-10), important fishing areas (E=4-8), and ordinary and open sea areas (E=0-4). Socioeconomic value (S) is quantified based on wind farm distribution and nighttime light radiation: high economic value areas (S=8-10), medium economic value areas (E=4-8), and low economic value and no economic activity areas (S=0-4). Pollution recovery capacity (H) is quantified based on marine quality status and sediment type: low recovery capacity (H=8-10), medium recovery capacity (H=4-8), and high recovery capacity (H=0-4). Specific data are sourced from "Priority Areas for Biodiversity Conservation in China," "Atlas of China," and "Bulletin on the Status of China's Marine Ecological Environment." In summary, based on the ESI classification standard, the simulated sea area is divided into environmentally sensitive areas of levels 1-5. Figure 8 ).

[0050] Comparative analysis of the options: The algorithm parameters are set as follows: population size 200, maximum iterations 500, crossover probability 0.9, basic mutation probability 0.05 (maximum 0.2), and the exchange / reverse / insertion probabilities in the mutation operator are 0.2 / 0.5 / 0.3 respectively.

[0051] To verify the effectiveness of the environmental sensitivity grading strategy, a control experiment was conducted. Enabling ESI grading involved using the grading results from a simulated sea area; disabling grading involved adjusting the ESI values ​​of each region in the simulated sea area to the average ESI value under the enabled grading condition, thus eliminating the influence of grading. Control group 1 consisted of 24 actual oil film data points, while control groups 2-4 had 25, 50, and 100 randomized demand points respectively (oil spill volume 50-100t, drift velocity 1-5m / s, angle 0-2π, fixed random seed). The experiment was repeated 10 times, and the average value was taken.

[0052] The results (Table 3) show that in small-scale scenarios (control groups 1-2), the tiered strategy reduced pollution losses by 24.48% and 21.91%, respectively; in medium-to-large-scale scenarios (control groups 3-4), the losses increased with the increase in demand points, but still achieved a reduction of 5.06%-7.42%.

[0053] Table 3. Comparison of Experimental Results of Different Schemes To evaluate the performance of the hybrid genetic algorithm of this invention, performance comparison experiments were conducted with the traditional genetic algorithm (GA), simulated annealing algorithm (SA), and improved particle swarm optimization (PSO). Based on the Solomon dataset C101, the demand points were expanded to include oil spill volume, drift velocity (1-5 m / s), and angle (0-2π), and datasets of different sizes were constructed with 25 / 50 / 75 / 100 demand points.

[0054] like Figure 9 As shown, for small-to-medium scale scenarios (C101_25, C101_50), the algorithm of this invention is superior to the comparison algorithm in terms of both optimization ability and stability. For large-scale scenarios (C101_75, C101_100), the median fitness of the algorithm of this invention is significantly lower than that of GA and SA, and the convergence interval is more compact. Although PSO shows good optimization ability, the expansion of the fitness range indicates that there are fluctuations.

[0055] The algorithm's convergence curve is as follows Figure 10 GA is prone to local optima, SA has low convergence efficiency, and PSO converges prematurely without completely breaking through the local optimum limitation. The algorithm in this invention combines optimization for marine oil spill scenarios, possessing both convergence efficiency and global search capability, resulting in significant overall performance advantages.

[0056] This study focuses on optimizing emergency dispatch for marine oil spills and yields the following results: An environmental sensitivity grading strategy effectively reduces environmental pollution losses. Based on the Environmental Sensitivity Index (ESI), constructed using ecological vulnerability, socio-economic value, and resilience, the sea area is divided into five sensitivity levels, providing a quantitative basis for differentiated rescue efforts. Experimental verification shows that in small-scale studies (24-25 demand points), this strategy reduces environmental pollution losses by 21.91%-24.48%; in medium-to-large-scale studies (50-100 demand points), although the overall loss increases with the number of demand points, a reduction of 5.06%-7.42% is still achieved, confirming the significant role of prioritizing rescue in highly sensitive areas in mitigating ecological damage.

[0057] The improved genetic algorithm exhibits superior optimization performance. By embedding 2-opt local optimization and an adaptive mutation strategy, the algorithm demonstrates outstanding performance across various scales of computational examples. In small-to-medium scale scenarios (25-50 requirement points), its optimization capability and stability are comprehensively superior to the comparative algorithms. In large-scale scenarios (75-100 requirement points), although the optimization difficulty increases, its optimization accuracy and convergence speed remain significantly superior, achieving stable convergence within 150 iterations.

[0058] Sensitivity analysis revealed the impact mechanism of key factors on scheduling effectiveness. Increased ship speed can reduce environmental damage by shortening arrival time, but has a limited impact on cost; the drift speed of demand points is positively correlated with scheduling costs and environmental damage, faster near-shore oil slick drift increases pollution accumulation, while longer voyages further offshore increase costs; ship capacity mainly affects scheduling costs, with a negligible effect on environmental damage.

[0059] Based on the Environmental Sensitivity Index (ESI) constructed from ecological vulnerability, socio-economic value, and resilience, the sea area is divided into five levels of sensitive zones, providing a quantitative basis for differentiated rescue. By embedding 2-opt local optimization and adaptive mutation strategies, the algorithm performs outstandingly in cases of different scales. In small and medium-scale scenarios (25-50 demand points), its optimization ability and stability are comprehensively superior to the comparison algorithms.

[0060] This invention proposes a method and system for emergency dispatching of marine oil spills, comprising: The system comprises several modules: a demand acquisition module to obtain the diameter of the oil slick in the spill area under the influence of wind, waves, and fluids; a loss quantification module to introduce the Environmental Sensitivity Index (ESI) grading model into a geographic information system to construct an optimized ESI grading model; a target construction module to establish a dual-objective model with the objectives of minimizing the increase in ecological and environmental pollution loss before and after emergency dispatch, and minimizing the total cost of emergency dispatch; a dispatch module to input the configuration of emergency bases and vessels, combined with the demand for containment materials and the changes in the location of the demand points; and a dispatch module to input the configuration of emergency bases and vessels and the location data of the demand points into the dual-objective model to obtain a dispatch plan suitable for marine oil spill emergency scenarios.

[0061] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a program, and when the program is executed by the processor, the processor performs the steps of a marine oil spill emergency dispatch method.

[0062] According to the disclosed embodiments, the computer device can communicate with one or more external devices (e.g., keyboard, pointing device, Bluetooth communication, etc.) or with any device that enables the computing device to communicate with one or more other computing devices (e.g., router, demodulator, etc.).

[0063] The present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of a marine oil spill emergency dispatch method.

[0064] According to the disclosed embodiments, the storage medium can be a non-volatile computer-readable storage medium, such as, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, the storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0065] The above description, in conjunction with specific preferred embodiments, provides a more detailed explanation of the present invention. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for emergency dispatching of marine oil spills, characterized in that, Including the following steps: The diameter of the oil film in the oil spill area obtained by the diffusion of the oil film under the action of wind, waves and fluid is obtained. The amount of containment materials required at different times is obtained by the positive correlation between the oil film diameter and the amount of containment materials required. The location change of the demand point is obtained according to the drift speed of the oil film under the action of wind and ocean currents. A geographic information system was introduced into the Environmental Sensitivity Index (ESI) classification model to construct an optimized ESI classification model. The marine area was classified by sensitivity index using the optimized ESI, and the ecological and environmental pollution loss was quantified by combining the sensitivity index classification results, oil spill impact time, oil film area and value conversion coefficient. A dual-objective model is established with the objectives of minimizing the added value of ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch. The total cost of emergency dispatch is based on the configuration of emergency bases and ships, and is obtained in combination with the demand for containment materials at the demand points and the changes in the location of the demand points. By inputting the configuration of emergency bases and vessels and the location data of demand points into the dual-objective model, a dispatching scheme adapted to the marine oil spill emergency scenario can be obtained.

2. The marine oil spill emergency dispatch method as described in claim 1, characterized in that, The method of obtaining the change in the location of the demand point based on the drift velocity of the oil film under the influence of wind and ocean currents is as follows: Obtain the ocean surface current velocity vector and the wind speed vector above the sea surface, and obtain the oil film drift velocity by weighted sum of the ocean surface current velocity vector and the wind speed vector above the sea surface; The location of the required point after drifting is determined by the oil film drift speed, drift angle, and ship sailing time; wherein the location of the required point serves as a condition for updating the ship's sailing path.

3. The method for emergency dispatching of marine oil spills as described in claim 1, characterized in that, The Environmental Sensitivity Index (ESI) grading model is introduced into a geographic information system to construct an optimized ESI grading model, specifically as follows: By weighting ecological vulnerability (E), socioeconomic value (S), and pollution recovery capacity (H), the optimized Environmental Sensitivity Index (ESI) classification model is obtained, and its specific expression is as follows: ； in, , , Ecological vulnerability E Socioeconomic value S and pollution recovery capacity H The weighted calculation weights.

4. The marine oil spill emergency dispatch method as described in claim 1, characterized in that, The aforementioned dual-objective model is established with the objectives of minimizing the increased ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch. Specifically: Minimize the increase in ecological and environmental pollution losses before and after emergency dispatch F 1; The specific expression is: ； in, The ecological and environmental pollution losses at the end of the dispatch period, This refers to the ecological and environmental pollution losses at the start of the dispatch process; Minimize the total cost of emergency dispatch; the specific expression is: ； in, To minimize the total cost of emergency dispatch, including the cost of emergency base deployment, vessel deployment, and vessel navigation, The cost of deploying a single emergency base, The cost of deploying a single emergency vessel, The cost of navigation per unit distance for emergency vessels. As decision variables, For the fleet k The number of ships equipped, As decision variables, V For the ship's speed, For the fleet k From node i To the node j The sailing time.

5. A method for emergency dispatching of marine oil spills as described in claim 1, characterized in that, After establishing the dual-objective model, it also includes: By introducing 2-opt local optimization and an adaptive mutation strategy based on population diversity into the genetic algorithm, an optimized genetic algorithm is obtained. The optimized genetic algorithm is then used to solve the bi-objective model, and the optimal scheduling scheme of the bi-objective model is output.

6. The marine oil spill emergency dispatch method as described in claim 5, characterized in that, The optimization of the genetic algorithm to solve the bi-objective model is as follows: A chromosome is constructed using a two-layer integer encoding method. During decoding, the nearest emergency base is selected as the starting point based on the centroid of the sub-path demand point cluster. The path is updated by combining the drift dynamics of the demand points, and the number of ships is allocated according to demand and ship capacity. The chromosome consists of a demand point sequence and implicit split points. The demand point sequence represents the fleet's access order, and the split points divide the sequence into sub-paths. The scheduling cost and environmental pollution loss are obtained by analyzing the number of ships and routes. The scheduling cost and environmental pollution loss are normalized and then weighted to obtain the fitness, and a fitness assessment is performed. Based on the fitness, a hybrid strategy of elite retention and roulette wheel selection is used for selection, and 2-opt local optimization is performed on elite individuals; the elite individuals refer to individuals in the population with fitness higher than a threshold. A sequential crossover operation is adopted, and the mutation rate is dynamically adjusted based on the standard deviation of the population fitness after the crossover operation. A multi-strategy mutation operation is performed, and individuals are inserted into the mutated population. The optimal individual output is used to obtain a scheduling scheme adapted to the marine oil spill emergency scenario.

7. A method for emergency dispatching of marine oil spills as described in claim 6, characterized in that, The adaptive mutation strategy is as follows: based on the ratio of the current population fitness standard deviation to the initial population fitness standard deviation, the actual mutation rate is dynamically adjusted between the basic mutation rate and the maximum mutation rate.

8. A method and system for emergency dispatching of marine oil spills, characterized in that, include: The demand acquisition module is used to acquire the diameter of the oil film that has spread to the oil spill area under the action of wind, waves and fluid. It obtains the demand for containment materials at different times by the positive correlation between the oil film diameter and the demand for containment materials, and obtains the change in the location of the demand point based on the drift speed of the oil film under the action of wind and ocean currents. The loss quantification module is used to introduce the Environmental Sensitivity Index (ESI) classification model into the geographic information system, construct an optimized ESI classification model, classify the sea area based on the optimized ESI, and quantify the ecological and environmental pollution loss by combining the sensitivity index classification results, oil spill impact time, oil film area and value conversion coefficient. The objective construction module is used to establish a dual-objective model with the objectives of minimizing the added value of ecological and environmental pollution losses before and after emergency dispatch, and minimizing the total cost of emergency dispatch; the total cost of emergency dispatch is based on the configuration of emergency bases and ships, and is obtained in combination with the demand for containment materials at the demand points and the changes in the location of the demand points; The scheduling module is used to input the configuration of emergency bases and ships and the location data of demand points into the dual-objective model to obtain a scheduling scheme adapted to the marine oil spill emergency scenario.

9. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores a program that, when executed by the processor, causes the processor to perform the steps of a marine oil spill emergency dispatch method as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of any one of the marine oil spill emergency dispatch methods according to claims 1 to 7.

Images