Ship route real-time optimization method and system based on double starting mechanism and storage medium

Abstract

The application discloses a ship route real-time optimization method and system based on a double-start mechanism and a storage medium. The method comprises the following steps: collecting and monitoring multi-dimensional dynamic data in real time and classifying the data into hard indicators and soft indicators; calculating a sailing risk value according to a first prediction model when the hard indicators change; if the sailing risk value exceeds a first preset safety threshold, immediately starting a mandatory route adjustment and calculating an emergency adjustment route; calculating a pre-optimization adjustment route when the soft indicators change and calculating an adjustment benefit value of each route according to a second prediction model; starting a route optimization adjustment when the adjustment benefit value of a certain pre-optimization adjustment route is greater than a second preset benefit threshold and the sailing risk value thereof is less than a third preset safety threshold; and optimizing and adjusting the route according to the adjustment benefit value and the sailing risk value. The application solves the problem of low optimization efficiency and increased sailing risk caused by the lack of accurate calculation of adjustment risk and benefit in the existing intelligent ship real-time route optimization.

Description

Technical Field

[0001] This invention belongs to the field of intelligent ship technology, and in particular relates to a method, system and storage medium for real-time optimization of ship routes based on a dual-start mechanism. Background Technology

[0002] Ship route planning is a core aspect of shipping operations, and its optimization level directly affects navigation safety, operating costs, and transportation timeliness. Traditional route planning methods mainly rely on static electronic charts and historical meteorological data to determine a planned route before departure. This approach reveals significant shortcomings when faced with the complex and dynamically changing factors encountered during actual navigation.

[0003] Currently, some intelligent navigation systems have integrated some real-time data, such as meteorological information and automatic identification system data, and provide short-term collision avoidance or detour suggestions. However, these methods still have two key problems: (1) Existing technologies generally adopt a "passive response" or "single-objective optimization" mode. Usually, route adjustments are only triggered when the risk reaches an emergency level (such as an imminent collision or entering a typhoon range), lacking early warning and proactive avoidance mechanisms for potential risks. At the same time, adjustment decisions often only consider a single factor, such as the shortest path or immediate safety, failing to deeply integrate and synergistically optimize navigation safety, ship maneuverability, real-time economic costs, and commercial benefits. (2) The decision-making process lacks a quantitative and adaptive evaluation framework. Existing systems are unable to accurately and forward-lookingly calculate the risks and benefits of adjusting the route itself. Adjusting a route may introduce new navigation difficulties (such as entering complex waterways), different environmental risks (such as encountering another severe sea area), and fluctuating economic impacts (such as changes in port fees and market opportunity gains and losses). Currently, there is a lack of a comprehensive model that can uniformly quantify the difficulty of navigation, economic changes, and adjustment risks. As a result, the advice provided to crew members is often fragmented and contradictory, and ultimately still relies heavily on the personal experience of the crew members to make difficult trade-offs, resulting in low decision-making efficiency and poor consistency.

[0004] To address the problem that existing real-time route optimization methods for intelligent ships lack accurate and forward-looking calculations of the risks and benefits of route adjustments, resulting in low optimization efficiency and increased navigation risks, a real-time route optimization method, system, and storage medium based on a dual-start mechanism for ships is proposed. Summary of the Invention

[0005] This invention proposes a real-time optimization method, system, and storage medium for ship routes based on a dual-start mechanism, which aims to at least address the problem that existing intelligent ship real-time route optimization methods lack accurate and forward-looking calculations of the risks and benefits of route adjustments, resulting in low optimization efficiency and increased navigation risks.

[0006] According to an embodiment of the present invention, a real-time optimization method for ship routes based on a dual-start mechanism is provided, comprising:

[0007] Real-time collection and monitoring of multi-dimensional dynamic data related to ship navigation, and classification of the factors corresponding to the dynamic data into hard indicators and soft indicators;

[0008] When the hard indicators change, the navigation risk value is calculated based on the pre-trained first prediction model; the first prediction model uses the degree of impact of changes in different hard indicators of the ship on the ship's navigation safety as input features;

[0009] If the navigation risk value exceeds the first preset safety threshold, a mandatory route adjustment will be initiated immediately, and an emergency route adjustment will be calculated based on the current high-risk factors.

[0010] When soft indicators change, one or more pre-optimized adjustment routes are calculated based on the difficulty of route adjustment and / or economic impact and / or adjustment risk.

[0011] The route adjustment benefit value for each pre-optimized route is calculated based on the pre-trained second prediction model; the second prediction model uses the comprehensive impact of different routes on navigation difficulty and navigation economy as input features;

[0012] The route optimization adjustment will be initiated only when the route adjustment benefit value of a pre-optimized route is greater than the second preset benefit threshold and its estimated navigation risk value is less than the third preset safety threshold.

[0013] The pre-optimized adjustment routes are ranked based on the route adjustment benefit value and the estimated navigation risk value, and the optimal solution is selected as the recommended optimized adjustment route.

[0014] In a preferred embodiment, the multi-dimensional dynamic data includes at least:

[0015] Environmental and regional data: current and forecast information on ocean currents, wind, waves, visibility, reefs, and icebergs; location and status of pirate risk zones, emission control zones, and restricted areas;

[0016] Ship status data: the ship's position, speed, course, draft, and fuel oil level; the operating status of power equipment, cargo hold equipment, and navigation equipment;

[0017] Traffic and situational data: AIS information of nearby vessels; potential collision risks;

[0018] Personnel and cargo data: location and status of personnel on board; charges, supply prices, and congestion conditions at planned and alternative ports; cargo status and delivery deadlines;

[0019] Instructions and regulatory data: Instructions from ship owners or cargo owners; relevant maritime regulations.

[0020] In a preferred embodiment, classifying the factors corresponding to the dynamic data into hard indicators and soft indicators includes:

[0021] The risk index of each dynamic data point is calculated based on the correlation between each dynamic data point and the ship's risk level.

[0022] The economic indicators of each dynamic data point are calculated based on the correlation between each dynamic data point and the economic efficiency of ship navigation.

[0023] Data with risk indicators exceeding preset risk indicator thresholds will be used as hard indicators;

[0024] Data whose economic indicators exceed the preset economic indicator threshold are considered soft indicators.

[0025] In a preferred embodiment, calculating the navigation risk value based on the pre-trained first prediction model includes:

[0026] Climate and environmental risk indicators are calculated based on the spatiotemporal overlap between the location and predicted path of severe weather conditions and the planned route of the ship, and / or the impact of sea state change gradient on navigation risks, and / or the impact of comprehensive environmental indicators on ship structure and equipment operation.

[0027] Geographic situation risk indicators are calculated based on the risk of intrusion into high-risk areas and / or the risk of navigation in complex waters and / or the risk of traffic conflicts between the vessel and its current position and planned route.

[0028] Ship condition risk indicators are calculated based on the impact of failures of critical equipment and / or the stress and stability risks of the hull structure and / or the watertightness risks.

[0029] Calculate response capability risk indicators based on the risk of disability of key personnel and / or sudden command and decision-making conflicts;

[0030] The first prediction model is constructed based on the positive correlation between climate and environmental risk indicators and / or geographical situation risk indicators and / or ship status risk indicators and / or response capability risk indicators and navigation risk values, and the navigation risk values ​​are calculated accordingly.

[0031] In a preferred embodiment, the step of calculating emergency route adjustments based on current high-risk factors includes:

[0032] Calculate the navigation area boundary based on the location and impact range of obstacles and / or dynamic collision avoidance zones and / or areas with severe weather and sea conditions on the electronic nautical chart;

[0033] Calculate the navigation margin width based on the ship's maneuverability and / or positioning accuracy and / or environmental uncertainties;

[0034] Calculate the safe drifting or anchoring area under restricted maneuvering based on the degree of impact of its own state-type risk factors on the ship's power and the expected repair time.

[0035] The navigation area is generated based on the navigation area boundary, navigation margin width, and safe drifting or anchoring area under restricted maneuvering, and the navigation area is discretized into a grid.

[0036] The navigation cost of each grid is calculated based on the wind and wave resistance and / or ocean current effects and / or water depth margin for each grid.

[0037] A real-time heuristic search algorithm is used to perform path search within a gridded navigation area to generate the optimal emergency adjustment route.

[0038] In a preferred embodiment, calculating one or more pre-optimized adjustment routes based on the difficulty and / or economic impact and / or adjustment risk of route adjustment includes:

[0039] The adjustment efficiency ratio is calculated based on the positive correlation between the vessel's maneuvering load and / or adjustment timing costs and / or coordination complexity and the difficulty of route adjustment;

[0040] Economic indicators are calculated based on the economic impact of fuel consumption differences due to route changes and / or port charges changes and / or expected time-of-arrival value gains and losses.

[0041] The net increase in risk is calculated based on the impact of the execution risk of the adjustment process and / or the estimated risk of new routes and / or operational risks on the adjustment risk;

[0042] Adjusting the energy efficiency ratio and / or economic indicators and / or net risk increment as the core parameters of multi-objective optimization, and using a constraint-based multi-objective genetic algorithm to generate one or more pre-optimized adjustment routes.

[0043] In a preferred embodiment, calculating the route adjustment benefit value for each pre-optimized route based on the pre-trained second prediction model includes:

[0044] The navigation difficulty index of the route is calculated based on the environmental resistance load and / or ship maneuvering load and / or navigation difficulty and / or personnel load of the pre-optimized route.

[0045] The route's sailing economy indicators are calculated based on the difference in fuel costs and / or the expected time of arrival value gains and losses and / or changes in process costs.

[0046] A second prediction model is constructed based on the correlation between the navigation difficulty index and / or the navigation economy index of the route and the route adjustment benefit value, and the route adjustment benefit value of each pre-optimized adjustment route is calculated accordingly.

[0047] In a preferred embodiment, the step of ranking the pre-optimized adjustment routes based on the route adjustment benefit value and the estimated navigation risk value, and selecting the optimal solution as the recommended optimized adjustment route, includes:

[0048] The weight of the adjustment index for each pre-optimized route is calculated based on the weighted sum of the route adjustment benefit value and the estimated navigation risk value.

[0049] The pre-optimized routes are sorted in descending order of the weight of the adjustment indicators for each route.

[0050] The pre-optimized adjustment route with the highest adjustment index weight is selected as the recommended optimized adjustment route.

[0051] According to another embodiment of the present invention, a computer-readable storage medium is provided that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the above-described real-time ship route optimization method based on a dual-start mechanism.

[0052] According to another embodiment of the present invention, a real-time ship route optimization system based on a dual-start mechanism is provided, comprising:

[0053] At least one processor;

[0054] and a memory communicatively connected to the at least one processor;

[0055] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute the above-described real-time ship route optimization method based on a dual-start mechanism.

[0056] The advantages of the real-time ship route optimization method, system, and storage medium based on the dual-start mechanism of the present invention are as follows:

[0057] (1) The present invention adopts a dual-start decision mechanism (hard start and soft start) for real-time route optimization of ships. Compared with the traditional single risk response or static economic planning technology, it can intelligently and proactively capture economic benefit optimization opportunities under the premise of ensuring absolute priority of navigation safety, and realize the automated and intelligent balance between safety bottom line and operational benefits, and transform from passive response to proactive optimization.

[0058] (2) Based on the pre-trained first prediction model, the present invention performs dynamic and multi-dimensional risk assessment of changes in hard indicators. Compared with traditional risk judgment techniques based on rules or simple thresholds, it can effectively quantify the real-time and forward-looking impact of complex and concurrent hard threats (such as the combined effects of severe weather, equipment failure, and regional risks) on ship safety, provide more accurate and timely risk warnings, and provide a reliable basis for mandatory risk avoidance decisions.

[0059] (3) Under hard start, the present invention constructs a dynamic safety corridor based on threat type and generates an emergency adjustment route. Compared with traditional manual experience-based risk avoidance or simple geographical detour technology, it can comprehensively consider ship maneuverability, environmental constraints and threat evolution to automatically generate the optimal safety path, which greatly improves the response speed and risk avoidance reliability in emergency situations.

[0060] (4) The present invention uses a pre-trained second prediction model to quantify the benefits of route adjustment. Compared with the traditional economic assessment technology that only considers fuel or distance, it can effectively integrate navigation difficulty (operational load), changes in economic costs throughout the cycle (fuel, time, port fees) and new risks introduced by the adjustment, and output a comprehensive route adjustment benefit value, providing accurate and comparable data support for business decisions.

[0061] (5) Under soft start, the present invention uses the adjustment benefit being greater than the threshold and the adjustment risk being controllable as the trigger condition for route optimization. Compared with the traditional technology based on a single economic indicator or fixed rules, it can ensure that each route optimization suggestion simultaneously meets the requirements of economic significance and safety acceptability, avoids frequent and unnecessary adjustments or unprofitable optimizations, and improves the practicality and feasibility of the suggestions.

[0062] (6) The present invention constructs a real-time route optimization system that integrates real-time data from multiple dimensions such as environment, ships, transportation, and commerce. Compared with traditional technologies that rely on limited data sources (such as only AIS and weather), it can provide more comprehensive and more realistic inputs for navigation risk assessment and navigation revenue calculation, making optimization decisions more stable and accurate, and reducing decision-making biases caused by missing information. Attached Figure Description

[0063] Figure 1 This is a flowchart of a real-time ship route optimization method based on a dual-start mechanism according to an embodiment of the present invention;

[0064] Figure 2 This is a flowchart of step S01 in an embodiment of the present invention;

[0065] Figure 3 This is a flowchart of step S02 in an embodiment of the present invention;

[0066] Figure 4This is a flowchart of step S03 in an embodiment of the present invention;

[0067] Figure 5 This is a flowchart of step S04 in an embodiment of the present invention;

[0068] Figure 6 This is a flowchart of step S05 in an embodiment of the present invention;

[0069] Figure 7 This is a flowchart of step S07 in an embodiment of the present invention;

[0070] Figure 8 This is a schematic diagram of a real-time ship route optimization system based on a dual-start mechanism according to an embodiment of the present invention. Detailed Implementation

[0071] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0072] According to an embodiment of the present invention, a real-time optimization method for ship routes based on a dual-start mechanism is provided, the flowchart of which is shown below. Figure 1 As shown, it includes:

[0073] Step S01: Collect and monitor multi-dimensional dynamic data related to ship navigation in real time, and classify the factors corresponding to the dynamic data into hard indicators and soft indicators;

[0074] Step S02: When the hard indicators change, the navigation risk value is calculated according to the pre-trained first prediction model; the first prediction model uses the degree of impact of changes in different hard indicators of the ship on the ship's navigation safety as input features;

[0075] Step S03: If the navigation risk value exceeds the first preset safety threshold, a mandatory route adjustment is immediately initiated, and an emergency route adjustment is calculated based on the current high-risk factors.

[0076] Step S04: When soft indicators change, calculate one or more pre-optimized adjustment routes based on the difficulty of route adjustment and / or economic impact and / or adjustment risk;

[0077] Step S05: Calculate the route adjustment benefit value for each pre-optimized route based on the pre-trained second prediction model; the second prediction model uses the comprehensive impact of different routes on navigation difficulty and navigation economy as input features;

[0078] Step S06: The route optimization adjustment is initiated only when the route adjustment benefit value of a certain pre-optimized adjustment route is greater than the second preset benefit threshold and its estimated navigation risk value is less than the third preset safety threshold.

[0079] Step S07: Sort the pre-optimized adjustment routes according to the route adjustment benefit value and the estimated navigation risk value, and select the optimal solution as the recommended optimized adjustment route.

[0080] In a preferred embodiment, the multi-dimensional dynamic data in step S01 includes at least:

[0081] Environmental and regional data: current and forecast information on ocean currents, wind, waves, visibility, reefs, and icebergs; location and status of pirate risk zones, emission control zones, and restricted areas;

[0082] Ship status data: the ship's position, speed, course, draft, and fuel oil level; the operating status of power equipment, cargo hold equipment, and navigation equipment;

[0083] Traffic and situational data: AIS information of nearby vessels; potential collision risks;

[0084] Personnel and cargo data: location and status of personnel on board; charges, supply prices, and congestion conditions at planned and alternative ports; cargo status and delivery deadlines;

[0085] Instructions and regulatory data: Instructions from ship owners or cargo owners; relevant maritime regulations.

[0086] In a preferred embodiment, in step S01, the factors corresponding to the dynamic data are classified into hard indicators and soft indicators, as shown in the flowchart below. Figure 2 As shown, it includes:

[0087] Step S011: Calculate the hazard index of each dynamic data point based on the correlation between each dynamic data point and the ship's hazard level;

[0088] Step S012: Calculate the economic indicators of each dynamic data based on the correlation between each dynamic data and the ship's navigation economy;

[0089] Step S013: Data with hazard indicators greater than the preset hazard indicator threshold are used as hard indicators;

[0090] Step S014: Data with economic indicators greater than the preset economic indicator threshold are used as soft indicators.

[0091] In this embodiment, in step S011, the risk index of each dynamic data is quantified according to the risk, severity of risk, probability of occurrence and controllability of each dynamic data, as shown in equation (1).

[0092] (1)

[0093] Where R i S represents the hazard index value (dimensionless, range 0-100) of the i-th dynamic data item; j represents the j-th hazard factor under a certain dynamic data item (e.g., wave height and visibility under "Environmental and Regional Data"); ij It is a severity score, indicating the severity of the consequences when a hazard occurs (quantification standard: collision risk S). ij =90−100; Equipment failure S ij =50−80; slight environmental changes S ij =10−40); P ij It is the probability coefficient of occurrence, indicating the likelihood of the hazard occurring based on current data (quantification standard: extremely high risk (immediate occurrence) P). ij =1.0; Medium risk (within a few hours) P ij =0.5; Low-risk P ij =0.1); C ij The controllability coefficient indicates the effectiveness of human intervention under current conditions (the larger the value, the less controllable it is; quantification standard: completely uncontrollable C). ij =1.0; Partially controllable C ij =0.4; Fully controllable C ij =0.1).

[0094] In step S012, the economic indicators of each dynamic data are calculated based on the cost impact of changes in each dynamic data on ship navigation and the time sensitivity of each dynamic data, as shown in equation (2).

[0095] (2)

[0096] Among them, E i ∆C represents the economic index value (dimensionless, range 0-100) of the i-th dynamic data item. i This indicates the cost fluctuation caused by the change in this data, such as the expected increase in voyage cost for every $10 / ton increase in fuel prices. C 基准 T represents the baseline total cost of this voyage; i The time sensitivity coefficient indicates the impact of data changes on delivery time (quantification standard: if the data change directly leads to a delay penalty, then T). i =1.5; If only costs are increased but delivery is not affected, T i =1.0).

[0097] In step S013, according to the preset hazard index threshold R TData with risk indicators exceeding preset risk indicator thresholds will be used as hard indicators. Hard indicators include, but are not limited to, severe weather conditions, navigation restrictions, serious equipment failures, collision risks, and serious illnesses of crew members.

[0098] In step S014, according to the preset economic indicator threshold E T Data with economic indicators exceeding the preset economic indicator threshold will be used as soft indicators. Soft indicators include, but are not limited to, changes in port fees, fluctuations in fuel prices, and adjustments to cargo delivery deadlines.

[0099] When a dynamic data point meets both the hard indicator classification criteria and the soft indicator classification criteria, the data point is classified as a hard indicator (i.e., hard indicators have higher priority than soft indicators).

[0100] In a preferred embodiment, step S02 involves calculating the navigation risk value based on a pre-trained first prediction model, as shown in the flowchart below. Figure 3 As shown, it includes:

[0101] Step S021: Calculate the climate and environmental risk indicators based on the spatiotemporal intersection rate between the location and predicted path of severe weather and the planned route of the ship and / or the impact of sea state change gradient on navigation risk and / or the impact of comprehensive environmental indicators on ship structure and equipment operation.

[0102] Step S022: Calculate the geographical situation risk index based on the risk of intrusion into high-risk areas and / or the risk of navigation in complex waters and / or the risk of traffic conflicts between the vessel and the planned route.

[0103] Step S023: Calculate the ship's condition risk index based on the impact of failures of key ship equipment and / or the stress and stability risk of the hull structure and / or the watertightness integrity risk.

[0104] Step S024: Calculate response capability risk indicators based on the disability risk of key personnel and / or sudden command and decision-making conflicts;

[0105] Step S025: Construct a first prediction model based on the positive correlation between climate and environmental risk indicators and / or geographical situation risk indicators and / or ship status risk indicators and / or response capability risk indicators and navigation risk values, and calculate the navigation risk value accordingly.

[0106] In this embodiment, the climate and environmental risk index R is calculated in step S021. env This indicator reflects the direct physical threat posed by the external natural environment to navigation safety, as shown in equation (3).

[0107] (3)

[0108] Among them, I stormThe severe weather spatiotemporal intersection rate is calculated as the ratio of the intersection mileage between the planned route and the severe weather area to the total mileage of the planned route; in one implementation method... L overlap L represents the intersection distance between the planned route and the severe weather zone. total α represents the total planned route mileage. severity Indicates the severity level coefficient of the weather. I gradient The gradient of sea state change is represented by the wave height variation and the distance to the meteorological fault; in one implementation method... ∆H s This represents the change in significant wave height per unit distance, and ∆D represents the distance to the meteorological fault (the larger this value, the higher the risk of encountering a "meteorological fault"). structure The influence of the overall environment on the structure is represented by calculations based on temperature and / or air pressure and / or sea ice concentration; in one embodiment... T low / T design P represents the ratio of the lowest temperature to the design minimum temperature. diff / P std Indicates the pressure difference, C ice β1, β2, and β3 represent the sea ice concentration (value range 0-1), and β1, β2, and β3 represent the corresponding weighting coefficients.

[0109] In step S022, the geographical situation risk index R is calculated. geo This indicator reflects the risk of interaction between the ship and the surrounding geographical environment and traffic flow, as shown in Equation (4).

[0110] (4)

[0111] Among them, I proximity The intrusion risk value for a high-risk area is calculated based on the distance to the high-risk area and / or the frequency of recent security incidents in the area; in one implementation... D current D represents the current distance from high-risk areas (such as pirate zones or restricted areas). safe It is the preset recommended safe distance threshold, F incident This indicates the frequency of recent security incidents in the area. waterway The navigation risk value in complex waters is calculated based on the real-time ratio of the vessel's draft to the available water depth and / or the ratio of the track width to the channel width and / or the influence of tidal current speed and direction on the vessel's position deviation; in one embodiment... T draft / D epth B represents the draft-to-depth ratio (indicating the risk of bottoming out). path / W idthV represents the ratio of track width to channel width (indicating the risk of maneuvering in restricted waters). current ·sinθ / V ship This represents the ratio of the crossflow component to the ship's speed (indicating the risk of drift). traffic The traffic conflict risk value is calculated based on AIS data; in one implementation method... DCPA represents the nearest encounter distance for each target vessel (calculated based on AIS data), and TCPA represents the time to nearest encounter point (calculated based on AIS data). k The encounter coefficient is obtained through prior training.

[0112] In step S023, the ship condition risk index R is calculated. ship This indicator reflects the physical integrity of the ship and the health of its equipment functions, as shown in Equation (5).

[0113] (5)

[0114] Among them, I machinery The impact of critical equipment failure is calculated based on the expected percentage decrease in the ship's power and / or maneuverability; I hull The stress and stability risk values ​​for the hull structure are calculated based on the loading status and real-time sea state, and then compared with safety thresholds. tightness The watertight integrity risk value is calculated based on factors such as the compartment liquid level, water ingress location, and rate. In one implementation method... Q inflow / Q pump V represents the ratio of inflow rate to drainage capacity (a value greater than 1 indicates that the water is sinking). flooded / V comp This indicates the ratio of the amount of water that has entered to the volume of the compartment.

[0115] In step S024, the response capability risk index R is calculated. crew This indicator reflects the risk caused by human factors, that is, whether the crew can respond effectively when danger occurs, as shown in Equation (6).

[0116] (6)

[0117] Among them, I manning This represents the disability risk value for key personnel, calculated based on the proportion of on-duty personnel below the safety staffing requirement and the correlation between the expected response time delay of critical operations and the risk value; I command The risk value representing a sudden command and decision-making conflict is calculated based on the severity level of the conflict and the expected duration of the decision vacuum. In one implementation method... δlevel This indicates the severity of the command conflict (e.g., a value of 0.8 if the shipowner's command conflicts with the collision avoidance rules), t vacuum t represents the duration of the vacuum period during which no one makes a decision. critical This indicates the maximum permissible delay in decision-making under critical circumstances.

[0118] In step S025, a first prediction model is constructed based on the risk indicators of the above four dimensions, and the navigation risk value R is calculated. total As shown in equation (7).

[0119] (7)

[0120] Among them, w env w geo w ship w crew These represent the dynamic risk weights of each dimension of the risk indicators, satisfying w env +w geo +w ship +w crew =1, the risk weight values ​​can be adjusted according to the ship's current dominant risk source, such as increasing w when sailing near the shore. geo (Geographical situation is more important), increasing w during ocean voyages. env (Climate and environment are more important), increase w when entering and leaving the port. ship (Low-speed maneuverability) and w crew (Lookout and command).

[0121] In a preferred embodiment, step S03 involves calculating an emergency route adjustment based on current high-risk factors, as shown in the flowchart below. Figure 4 As shown, it includes:

[0122] Step S031: Calculate the navigation area boundary based on the location and impact range of obstacles and / or dynamic collision avoidance zones and / or areas with severe weather and sea conditions on the electronic nautical chart;

[0123] Step S032: Calculate the navigation margin width based on the ship's maneuverability and / or positioning accuracy and / or environmental uncertainties;

[0124] Step S033: Calculate the safe drifting or anchoring area under restricted maneuvering based on the degree of impact of its own state-type risk factors on the ship's power and the expected repair time.

[0125] Step S034: Generate a navigation area based on the navigation area boundary, navigation margin width, and safe drifting or anchoring area under restricted maneuvering, and discretize the navigation area into a grid.

[0126] Step S035: Calculate the navigation cost of each grid cell based on the wind and wave resistance and / or ocean current influence and / or water depth margin.

[0127] Step S036: Use a real-time heuristic search algorithm to perform path search within the gridded navigation area to generate the optimal emergency adjustment route.

[0128] In this embodiment, in step S031, calculating the navigation area boundary based on the location and influence range of obstacles and / or dynamic collision avoidance zones and / or areas with severe weather and sea conditions on the electronic nautical chart involves delineating static obstacles, dynamic targets, and areas with severe weather conditions on the electronic nautical chart as areas inaccessible to ships, and thereby obtaining the navigable area boundary, denoted as Area. 可行 (t).

[0129] In step S032, the calculation of the navigation margin width based on ship maneuvering performance and / or positioning accuracy and / or environmental uncertainty involves reserving a safety buffer distance within the feasible domain boundary based on ship maneuvering uncertainty and environmental errors. The navigation margin width is denoted by B. 余量 The calculation is shown in equation (8).

[0130] (8)

[0131] Among them, R turn η represents the minimum turning radius of a ship (a core parameter of maneuverability). 定位 η represents the positioning error of GNSS / compass. 流 Δt represents the maximum error in ocean current prediction, and ∆t represents the expected time window for the current to pass through the region.

[0132] At this point, the navigation area boundary is corrected to Area. 航行 (t)=Area 可行 (t)−B 余量 .

[0133] In step S033, the safe drifting or anchoring area under restricted maneuvering calculated based on the degree of impact of its own state-type risk factors on the ship's power and the estimated repair time is a safe area pre-selected for emergency rescue or repair when the ship loses partial power due to equipment failure (such as main engine failure), denoted as Area. 应急 A set of discrete emergency rescue locations were selected based on drifting or anchoring conditions such as being far from traffic flow, having sufficient water depth, and having suitable bottom sediment for anchoring.

[0134] In step S034, a navigation area Grid is generated based on the navigation area boundary, navigation margin width, and safe drifting or anchoring area under restricted maneuvering. 总 =Area 航行 (t)∪Area 应急(Incorporate the emergency rescue area into the feasible grid set and mark it as a special node), discretize the navigation area into a grid, and ensure that the grid size can accommodate ship turning and meet the decision refresh frequency.

[0135] In step S035, calculating the navigation cost of a grid based on the wave resistance and / or ocean current influence and / or water depth margin of each grid involves assigning a weight value to each feasible grid, representing the difficulty or risk cost of a ship passing through that grid. The greater the wave resistance, the greater the ocean current influence, and the smaller the water depth margin, the greater the navigation cost. In a preferred embodiment, the navigation cost of a grid is calculated as shown in equation (9).

[0136] (9)

[0137] Where, ω wave H represents the cost weighting of wind and wave resistance. s (i,j) represents the effective wave height within the grid, H base This indicates the preset reference wave height. ω current V represents the cost weighting of ocean currents. c (i,j)·cosθ represents the forward and reverse components of the current direction and the planned course (reverse current increases cost when the current is positive), V ship Indicates the ship's speed. ω depth The weight of the water depth surplus cost is represented by D(i,j), which represents the water depth within the grid, and T. max Indicates the ship's maximum draft. ∆D safe D(i,j)-T represents the preset safety margin depth. max <∆D safe At that time, the risks increase dramatically.

[0138] In step S036, on the established grid cost map, a real-time heuristic search algorithm is used to search for the minimum cost path from the current location to the target port (or emergency rescue point) as the optimal emergency adjustment route.

[0139] In a preferred embodiment, step S04 involves calculating one or more pre-optimized adjustment routes based on the difficulty and / or economic impact and / or adjustment risk of route adjustment, as shown in the flowchart below. Figure 5 As shown, it includes:

[0140] Step S041: Calculate the adjustment efficiency ratio based on the positive correlation between the ship's maneuvering load and / or adjustment timing cost and / or coordination complexity and the difficulty of route adjustment;

[0141] Step S042: Calculate the economic indicators based on the economic impact of fuel consumption differences due to route changes and / or port charges changes and / or expected time of arrival value gains and losses.

[0142] Step S043: Calculate the net risk increment based on the impact of the execution risk of the adjustment process and / or the estimated risk of the new route and / or the operational risk on the adjustment risk;

[0143] Step S044: Using the adjusted energy efficiency ratio and / or economic indicators and / or net risk increment as the core parameters of multi-objective optimization, one or more pre-optimized adjustment routes are generated by using a constraint-based multi-objective genetic algorithm.

[0144] In this embodiment, the adjusted energy efficiency ratio E is calculated in step S041. eff The cost of implementing route adjustments is quantified, including operational burden, timing cost, and coordination complexity, as shown in Equation (10).

[0145] (10)

[0146] Among them, I man To calculate the maneuvering load, the time and space required for steering or changing speed, as well as the mechanical load on the main engine and steering gear, are calculated based on the ship's current speed, load condition, and real-time sea state. ω m This represents the pre-trained manipulation load weight values; in a preferred embodiment... R req / R max P represents the ratio of the required rudder angle to the maximum rudder angle (steering load). req / P max ∆t represents the ratio of required main engine power to rated power (variable speed load). maneuver / ∆t safe This represents the ratio of the time required to perform the operation to the safe allowable time. cost To adjust timing costs, and to evaluate the merits of making adjustments at the current location, ω is calculated based on the current water area's openness and / or the difference between the optimal adjustment window and the current time. t This represents the adjustment timing cost weight value obtained through pre-training; in a preferred embodiment... A open / A req Indicates the current openness of the water area (maneuverable water area / minimum required area), |t now -t window | represents the time difference between the current moment and the optimal adjustment window. C plex The coordination complexity is represented by ω, calculated based on the complexity of communication or coordination with the traffic management center, other vessels, or port authorities involved in the adjustment. c This represents the coordination complexity weight values ​​obtained through pre-training; in a preferred embodiment, C plex =α1·NVTS+α2·N ship +α3·N portNVTS represents the number of traffic management centers that need to be coordinated, N ship N represents the number of surrounding vessels that may be affected. port This indicates the number of ports or agents that need to be communicated, and α1, α2, and α3 are calculated coefficients obtained through pre-training.

[0147] In step S042, the economic index E is calculated. eco The formula is used to quantify the economic differences between new and old routes, including fuel, port fees and time value, as shown in Equation (11).

[0148] (11)

[0149] Where, ∆C fuel This represents the difference in fuel consumption, calculated based on the difference in fuel requirements between the old and new routes. ∆C port This indicates changes in port charges, calculated based on the difference in charges arising from changing ports of call on new and old shipping routes. Port charges include, but are not limited to, pilotage fees, tugboat fees, terminal fees, and agency fees. ∆C time The estimated time-of-arrival value gain or loss is calculated based on the financial gains or losses associated with the difference in estimated arrival times between the old and new routes; in a preferred embodiment, ∆C time =(T new -T old )×V time +∆C demurrage T new -T old V represents the difference in arrival times between the old and new routes. time ∆C represents the value per unit of time (such as rent, late payment rate). demurrage This indicates the contractual rewards or penalties resulting from early / delayed arrival.

[0150] In step S043, the net risk increment ∆R is calculated to quantify the increased risk value of the new route compared to the original route, as shown in equation (12).

[0151] (12)

[0152] Among them, R proc To adjust for process execution risks, ω is calculated based on the duration or magnitude of risks such as decreased maneuverability and new encounters with other vessels. p This represents the risk weight value for the adjustment process obtained through pre-training; in a preferred embodiment... I man (t) represents the descent exponent of controllability at time t (e.g., loss of rudder effectiveness), N newCPA (t) / N total (t) represents the proportion of newly formed meeting situations. Rroute Risk assessment for new shipping routes is calculated based on the probability of potential severe weather and / or the level of piracy or political risk in the new navigation area and / or the complexity of the waterways in the new navigation area. ω r R represents the risk weight value for the new route obtained through pre-training. ops Representing operational risk, ω is calculated based on supply chain disruption risk and / or uncertainty regarding new port operations. o R represents the pre-trained operational risk weight value; in a preferred embodiment, R ops =λ1·P disrupt +λ2·(1−R port,rel ), P disrupt R represents the probability of supply chain disruption (such as strikes or congestion at new ports). port,rel λ1 and λ2 represent the operational reliability index of the new port, and are calculated coefficients obtained through pre-training.

[0153] In step S044, the energy efficiency ratio and / or economic indicators and / or net risk increment are used as core parameters for multi-objective optimization. A constraint-based multi-objective genetic algorithm is used to generate one or more pre-optimized adjustment routes, i.e., adjusting the energy efficiency ratio E eff Economic indicators E eco The optimal trade-off between the three objectives—risk, net risk increment ∆R, and risk itself—is sought to generate a Pareto front solution set. The optimization problem is defined as follows:

[0154] Minimize: f1 = -E eff (Maximize the adjustment of energy efficiency ratio)

[0155] Minimize: f2 = -E eco (Maximize economic benefits)

[0156] Minimize: f3 = ∆R (Minimize the risk increment)

[0157] Constraints: The flight path is within the feasible grid; the segment length is less than or equal to the maximum range; the estimated arrival time meets the time window agreed upon in the contract.

[0158] Based on the above optimization problem, a constraint-based multi-objective genetic algorithm is used to generate a Pareto front solution set, which is one or more pre-optimized adjustment routes.

[0159] In a preferred embodiment, step S05 calculates the route adjustment benefit value for each pre-optimized route based on the pre-trained second prediction model, as shown in the flowchart below. Figure 6 As shown, it includes:

[0160] Step S051: Calculate the navigation difficulty index of the route based on the environmental resistance load and / or ship maneuvering load and / or navigation difficulty and / or personnel load of the pre-optimized route.

[0161] Step S052: Calculate the route's sailing economy index based on the difference in fuel costs and / or the expected time of arrival value gains and losses and / or changes in process costs.

[0162] Step S053: Construct a second prediction model based on the correlation between the navigation difficulty index and / or the navigation economy index of the route and the route adjustment benefit value, and use this model to calculate the route adjustment benefit value for each pre-optimized adjustment route.

[0163] In this embodiment, the navigation difficulty index D of the route is calculated in step S051. route The cost of executing the route is quantified, including environmental resistance, maneuvering burden, navigation complexity and personnel load, as shown in Equation (13).

[0164] (13)

[0165] Among them, I env Environmental drag load is represented by ω, which is calculated based on the positive correlation between the proportion of additional propulsion power required for navigation against environmental conditions such as headwinds and currents, and / or the duration and proportion of severe weather and / or the proportion of low visibility time, and environmental drag load. d 1 represents the environmental resistance load weight value obtained through pre-training. man The ship's maneuvering load is calculated according to the calculation method in step S04. ω d 2 represents the pre-trained ship maneuvering load weight value. I nav The navigation difficulty is indicated by ω, calculated based on the proportion of complex waters in the route and / or the expected number of encounters with other vessels and / or the number of new or changed ports of call. d 3 represents the navigation difficulty weight value obtained through pre-training. In a preferred embodiment... , where L complex / L 总 N represents the percentage of navigation segments in complex waterways (narrow waterways, restricted waterways). encounter / N encounter,ref N represents the ratio of the expected number of encounters with other vessels to a reference value. port,change Indicates the number of newly added or changed ports of call. crew ω represents personnel load, calculated based on the positive correlation between the bridge team's operating time and / or decision frequency under the flight conditions of the route and personnel load. d 4 represents the pre-trained personnel load weight value;

[0166] In step S052, the navigation economy index E of the route is calculated. routeThis is used to quantify the financial gains or losses from executing the route, including fuel, time value, and process costs, and is calculated based on the economic indicator E in step S042. eco The calculation method for the route's sailing economy index E route .

[0167] In step S053, a second prediction model is constructed based on the correlation between the navigation difficulty index and / or the navigation economy index of the route and the route adjustment benefit value, and the route adjustment benefit value of each pre-optimized adjustment route is calculated. The larger the value, the more worthwhile the route is to adopt. The route adjustment benefit value is denoted as V. adj As shown in equation (14).

[0168] (14)

[0169] Among them, w econ For economic weighting (dynamically adjusted according to the degree of emphasis on economic benefits), w diff Difficulty weight (dynamically adjusted based on the emphasis on operational burden and safety margin).

[0170] In step S06, only when the route adjustment benefit value V of a certain pre-optimized route is... adj Greater than the second preset revenue threshold V T And its estimated navigation risk value R total Less than the third preset safety threshold R T’ At that time, route optimization and adjustments will be initiated.

[0171] In a preferred embodiment, in step S07, the pre-optimized adjustment routes are sorted according to the route adjustment benefit value and the estimated navigation risk value, and the optimal solution is selected as the recommended optimized adjustment route. The flowchart is as follows. Figure 7 As shown, it includes:

[0172] Step S071: Calculate the adjustment index weight of each pre-optimized adjustment route based on the weighted sum of the route adjustment benefit value and the estimated navigation risk value.

[0173] Step S072: Sort the pre-optimized adjustment routes in descending order of the adjustment index weights of each pre-optimized adjustment route;

[0174] Step S073: Select the pre-optimized adjustment route with the largest adjustment index weight as the recommended optimized adjustment route.

[0175] In this embodiment, the adjustment index weight W for each pre-optimized route is calculated by combining the combined route adjustment benefit value (calculated according to step S05) and the estimated navigation risk value (calculated according to step S02). routeAs shown in equation (15).

[0176] (15)

[0177] Among them, V adj R represents the revenue value from route adjustment (calculated in step S053; a larger value indicates better economic efficiency and lower difficulty). total μ1 represents the estimated navigation risk value (the result of step S025, the larger the value, the higher the risk), μ2 represents the benefit weight coefficient (dynamically adjusted according to the importance attached to economy and operational convenience), and μ2 represents the risk weight coefficient (dynamically adjusted according to the importance attached to navigation safety), satisfying μ1+μ2=1 and μ1, μ2∈[0,1].

[0178] The pre-optimized adjustment route set is sorted from largest to smallest according to the adjustment index weight (if the adjustment index weights of the routes are equal, they are arranged in order of navigation risk value from smallest to largest), and the pre-optimized adjustment route with the largest adjustment index weight is selected as the recommended optimized adjustment route.

[0179] According to another embodiment of the present invention, a computer-readable storage medium is provided that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the above-described real-time ship route optimization method based on a dual-start mechanism.

[0180] According to another embodiment of the present invention, a real-time ship route optimization system based on a dual-start mechanism is provided, the structural schematic diagram of which is shown below. Figure 8 As shown, it includes:

[0181] At least one processor;

[0182] and a memory communicatively connected to the at least one processor;

[0183] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute the above-described real-time ship route optimization method based on a dual-start mechanism.

[0184] Of course, those skilled in the art should recognize that the above embodiments are only used to illustrate the present invention and are not intended to limit the present invention. Any changes or modifications to the above embodiments that are within the scope of the present invention will fall within the protection scope of the present invention.

Claims

1. A real-time optimization method for ship routes based on a dual-start mechanism, characterized in that, include: Real-time collection and monitoring of multi-dimensional dynamic data related to ship navigation, and classification of the factors corresponding to the dynamic data into hard indicators and soft indicators; When hard indicators change, the navigation risk value is calculated based on the pre-trained first prediction model; The first prediction model uses the degree of impact of changes in different hard indicators of a ship on the safety of ship navigation as input features; The calculation of navigation risk value based on the pre-trained first prediction model includes: calculating climate environment risk indicators based on the spatiotemporal intersection rate of the location of severe weather and the predicted path with the ship's planned route and / or the impact of sea state change gradient on navigation risk and / or the impact of comprehensive environmental indicators on ship structure and equipment operation; calculating geographical situation risk indicators based on the risk of intrusion into high-risk areas based on the ship's current position and planned route and / or the ship's navigation risk in complex waters and / or the ship's traffic conflict risk; calculating ship status risk indicators based on the impact of failure of key ship equipment and / or the stress and stability risk of the hull structure and / or the watertight integrity risk; calculating response capability risk indicators based on the risk of disability of key personnel and / or sudden command and decision-making conflicts; and constructing a first prediction model based on the positive correlation between climate environment risk indicators and / or geographical situation risk indicators and / or ship status risk indicators and / or response capability risk indicators and navigation risk value, and calculating the navigation risk value accordingly. If the navigation risk value exceeds the first preset safety threshold, a mandatory route adjustment will be initiated immediately, and an emergency route adjustment will be calculated based on the current high-risk factors. When soft indicators change, one or more pre-optimized adjustment routes are calculated based on the difficulty of route adjustment and / or economic impact and / or adjustment risk. The route adjustment benefit value for each pre-optimized adjusted route is calculated based on the pre-trained second prediction model. The second prediction model uses the comprehensive impact of different routes on navigation difficulty and navigation economy as input features. The calculation of the route adjustment benefit value for each pre-optimized adjusted route based on the pre-trained second prediction model includes: calculating the navigation difficulty index of the route based on the environmental resistance load and / or ship maneuvering load and / or navigation difficulty and / or personnel load of the pre-optimized adjusted route; calculating the navigation economy index of the route based on the fuel cost difference and / or expected time of arrival value gain and / or process cost changes; constructing a second prediction model based on the correlation between the navigation difficulty index and / or the navigation economy index of the route and the route adjustment benefit value, and calculating the route adjustment benefit value for each pre-optimized adjusted route accordingly. The route optimization adjustment will be initiated only when the route adjustment benefit value of a pre-optimized route is greater than the second preset benefit threshold and its estimated navigation risk value is less than the third preset safety threshold. The pre-optimized adjustment routes are ranked based on the route adjustment benefit value and the estimated navigation risk value, and the optimal solution is selected as the recommended optimized adjustment route.

2. The real-time optimization method for ship routes based on a dual-start mechanism according to claim 1, characterized in that, The multi-dimensional dynamic data includes at least: Environmental and regional data: current and forecast information on ocean currents, wind, waves, visibility, reefs, and icebergs; location and status of pirate risk zones, emission control zones, and restricted areas; Ship status data: the ship's position, speed, course, draft, and fuel oil level; the operating status of power equipment, cargo hold equipment, and navigation equipment; Traffic and situational data: AIS information of nearby vessels; potential collision risks; Personnel and cargo data: location and status of personnel on board; charges, supply prices, and congestion conditions at planned and alternative ports; cargo status and delivery deadlines; Instructions and regulatory data: Instructions from ship owners or cargo owners; relevant maritime regulations.

3. The real-time optimization method for ship routes based on a dual-start mechanism according to claim 1, characterized in that, The classification of factors corresponding to the dynamic data into hard indicators and soft indicators includes: The risk index of each dynamic data point is calculated based on the correlation between each dynamic data point and the ship's risk level. The economic indicators of each dynamic data point are calculated based on the correlation between each dynamic data point and the economic efficiency of ship navigation. Data with risk indicators exceeding preset risk indicator thresholds will be used as hard indicators; Data whose economic indicators exceed the preset economic indicator threshold are considered soft indicators.

4. The real-time optimization method for ship routes based on a dual-start mechanism according to claim 1, characterized in that, The calculation of emergency route adjustments based on current high-risk factors includes: Calculate the navigation area boundary based on the location and impact range of obstacles and / or dynamic collision avoidance zones and / or areas with severe weather and sea conditions on the electronic nautical chart; Calculate the navigation margin width based on the ship's maneuverability and / or positioning accuracy and / or environmental uncertainties; Calculate the safe drifting or anchoring area under restricted maneuvering based on the degree of impact of its own state-type risk factors on the ship's power and the expected repair time. The navigation area is generated based on the navigation area boundary, navigation margin width, and safe drifting or anchoring area under restricted maneuvering, and the navigation area is discretized into a grid. The navigation cost of each grid is calculated based on the wind and wave resistance and / or ocean current effects and / or water depth margin for each grid. A real-time heuristic search algorithm is used to perform path search within a gridded navigation area to generate the optimal emergency adjustment route.

5. The real-time optimization method for ship routes based on a dual-start mechanism according to claim 1, characterized in that, The calculation of one or more pre-optimized adjustment routes based on the difficulty and / or economic impact and / or adjustment risk of route adjustment includes: The adjustment efficiency ratio is calculated based on the positive correlation between the vessel's maneuvering load and / or adjustment timing costs and / or coordination complexity and the difficulty of route adjustment; Economic indicators are calculated based on the economic impact of fuel consumption differences due to route changes and / or port usage fee changes and / or expected time of arrival value gains and losses. The net increase in risk is calculated based on the impact of the execution risk of the adjustment process and / or the estimated risk of new routes and / or operational risks on the adjustment risk; Adjusting the energy efficiency ratio and / or economic indicators and / or net risk increment as the core parameters of multi-objective optimization, and using a constraint-based multi-objective genetic algorithm to generate one or more pre-optimized adjustment routes.

6. The real-time optimization method for ship routes based on a dual-start mechanism according to claim 1, characterized in that, The process of ranking pre-optimized adjustment routes based on route adjustment benefit value and estimated navigation risk value, and selecting the optimal solution as the recommended optimized adjustment route, includes: The weight of the adjustment index for each pre-optimized route is calculated based on the weighted sum of the route adjustment benefit value and the estimated navigation risk value. The pre-optimized routes are sorted in descending order of the weight of the adjustment indicators for each route. The pre-optimized adjustment route with the highest adjustment index weight is selected as the recommended optimized adjustment route.

7. A computer-readable storage medium storing a computer program for electronic data interchange, wherein, The computer program causes the computer to perform the method as described in any one of claims 1-6.

8. A real-time ship route optimization system based on a dual-start mechanism, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-6.

Images