A multimodal multi-objective-based sea route planning method, device and medium

Abstract

The application discloses a kind of based on multimodal multi-objective sea route planning method, equipment and medium, belong to route planning technical field, for solving in existing ship meteorological route planning, single target optimization cannot meet complex shipping decision-making needs, lack of consideration of the dynamic characteristics of ship-specific technical problems.The method comprises: obtaining the route planning basic information of target ship and the meteorological ocean environment forecast data required for route planning in target sea area;Multi-objective function target optimization processing is carried out on the sea route planning of target ship, and a multi-objective optimization function system is generated;The route planning constraint strategy is obtained by carrying out route planning constraint processing on the sea route of target ship under multidimension;The generated several candidate sea routes are subjected to multimodal and multi-objective iterative optimization calculation, and a set of Pareto optimal routes of target ship in target sea area is obtained;The Pareto optimal route set is smoothed and output visually.

Description

Technical Field

[0001] This application relates to the field of route planning technology, and in particular to a method, equipment and medium for maritime route planning based on multimodal and multi-objective approaches. Background Technology

[0002] Ship meteorological route planning is a key technology for the shipping industry to achieve energy conservation, emission reduction, and ensure navigation safety. Existing route planning methods typically involve: using a given optimization objective as the primary focus (such as minimizing distance or shortest sailing time), and then, based on meteorological forecast data and ship performance parameters, employing A... Algorithms, genetic algorithms, dynamic programming, and other methods are used to find the "optimal" route for a single destination. For example, existing solutions typically include the following steps: obtaining the coordinates of the starting and destination points; loading meteorological and marine environmental forecast data; establishing an optimization model based on a single objective (such as the shortest sailing time); using intelligent optimization algorithms to find the optimal route for a single destination; and providing the optimal route to the captain as a decision-making reference.

[0003] However, existing ship meteorological route planning methods still have certain objective shortcomings, including: 1) There are fundamental differences between maritime and air transport scenarios. Existing air route planning methods are mainly designed for aircraft, and the factors they consider (such as cloud density and aircraft yaw angle constraints) differ significantly from the actual needs of maritime scenarios. Maritime route planning needs to focus on: ocean currents, waves, tides, and other marine hydrological elements (rather than clouds); propeller propulsion efficiency and ship stall models; and complex voyage structures with multiple port calls (rather than simple point-to-point flights).

[0004] 2) Single-objective optimization cannot meet the needs of complex shipping decisions. Most existing methods are guided by a single objective (such as the shortest voyage) and output an "optimal" route. However, in actual shipping, captains need to make trade-offs between multiple objectives such as fuel saving, time saving, comfort and safety. A single-objective output cannot provide decision-makers with a basis for comparing multiple options.

[0005] 3) Lack of consideration for the unique dynamic characteristics of ships. Existing methods rarely incorporate shipping-specific mathematical models, such as ship stall models, fuel consumption models under the influence of wind, and the impact of ocean currents on ground speed, into a unified framework for route optimization. This leads to discrepancies between optimization results and actual navigation performance.

[0006] 4) Insufficient handling of conflicts between multiple objectives. The shortest flight time route and the lowest fuel consumption route often conflict (e.g., high-speed navigation saves time but consumes more fuel). Existing methods usually use a weighted combination to transform multiple objectives into a single objective, but the weight setting depends on experience and cannot fully display the Pareto front between objectives. Summary of the Invention

[0007] This application provides a method, device, and medium for multimodal and multi-objective maritime route planning to solve the following technical problems: in existing ship meteorological route planning, single-objective optimization cannot meet the needs of complex shipping decisions and lacks consideration of the unique dynamic characteristics of ships.

[0008] The embodiments of this application adopt the following technical solutions: On one hand, this application provides a multimodal, multi-objective maritime route planning method, comprising: acquiring basic route planning information of the target vessel and meteorological and marine environmental forecast data required for route planning in the target sea area, based on the target vessel and the target sea area; performing multi-objective function optimization processing on the maritime route planning of the target vessel according to the basic route planning information and the meteorological and marine environmental forecast data, generating a multi-objective optimization function system; wherein, the multi-objective function includes: a shortest sailing time objective function, a minimum fuel consumption objective function, and an optimal sailing comfort function; performing multi-dimensional route planning constraint processing on the maritime route of the target vessel to obtain a route planning constraint strategy; performing multimodal and multi-objective iterative optimization calculation on several generated candidate maritime routes through the multi-objective optimization function system and the route planning constraint strategy to obtain a Pareto optimal route set of the target vessel in the target sea area; and smoothing and visualizing the Pareto optimal route set.

[0009] This application provides multiple Pareto-optimal routes through multi-objective optimization, achieving the shortest voyage time, lowest fuel consumption, and best comfort. The captain can make flexible decisions based on real-time sea conditions and mission priorities. In the event of unexpected situations such as sudden severe weather or changes in restricted areas during voyages, multiple pre-planned alternative routes can be quickly switched without recalculating from scratch, significantly reducing emergency decision-making time. Furthermore, by incorporating factors such as ocean currents, wind and waves, and stall effects into the fuel consumption model, the optimization results more closely reflect actual navigation conditions. Moreover, this application is specifically designed for maritime scenarios, fully considering factors unique to the shipping industry such as ship stall, ocean current utilization, and wave drag, filling a technological gap in this field. It is also applicable to transoceanic route planning tasks for various commercial vessels such as bulk carriers, container ships, and tankers, and is compatible with existing ship navigation systems.

[0010] In one feasible implementation, based on the target vessel and the target sea area, the basic information for the route planning of the target vessel and the meteorological and marine environmental forecast data required for the route planning in the target sea area are obtained respectively. Specifically, this includes: obtaining the basic information for route planning based on the basic design attribute data and navigation task constraint information of the target vessel; wherein, the basic information for route planning includes at least: the location of the departure port, the location of the destination port, the design parameters of the target vessel, and the task constraint information; the task constraint information includes at least: departure time, planned arrival time window, maximum still water set speed and minimum still water set speed of the vessel; the design parameters of the target vessel include at least: length, beam, design draft, design speed, displacement, main engine power, and fuel consumption characteristics; and obtaining the meteorological and marine environmental forecast data for the planned navigation area in the target sea area within the planned navigation time period from the meteorological and marine data service platform via HTTP network communication protocol or FTP protocol; wherein, the meteorological and marine environmental forecast data includes at least: wave-related data, ocean current-related data, sea surface wind field data, and air pressure and temperature data.

[0011] In one feasible implementation, based on the route planning basic information and the meteorological and marine environmental forecast data, the maritime route planning of the target vessel is subjected to multi-objective function optimization processing to generate a multi-objective optimization function system, specifically including: based on... The objective function for the shortest travel time is obtained. ;in, The total sailing time in the basic route planning information is calculated by dividing the distance of each segment in the basic route planning information by the ship's actual speed in wind and waves. The sum is obtained by accumulation; The cumulative time a vessel spends navigating within a severe sea state warning zone, and The coefficients represented are used to penalize routes crossing dangerous sea areas; according to The objective function for minimizing fuel consumption is obtained. Where Q represents the total fuel consumption of the target shipping route, obtained by summing the fuel consumption of each segment after correction using the aforementioned meteorological and marine environmental forecast data; according to The optimal flight comfort function is obtained. ;in, The motion response of the target vessel at each waypoint along the shipping route. The roll of the target vessel in the i-th segment, The pitch of the target vessel in the i-th segment, The heave of the target vessel in the i-th segment. The temperature, humidity, and pressure of the target vessel in the i-th segment are given; the shortest sailing time objective function, the lowest fuel consumption objective function, and the optimal sailing comfort function are associated and stored to construct the multi-objective optimization function system.

[0012] In one feasible implementation, the actual speed for: ; The ship speed is set for still water, h is the meaningful wave height, q is the relative angle between the ship's heading and the wave direction, and D is the ship's displacement.

[0013] In one feasible implementation, the total fuel consumption of the target shipping route is obtained by summing the fuel consumption of each segment after correction using the meteorological and marine environmental forecast data, specifically including: according to The fuel consumption for a single flight segment under conditions unaffected by weather factors was obtained. ;in, The fuel consumption of the target vessel on the i-th segment; Fuel consumption rate; P is the rated power of the main engine; V represents the actual speed of the target vessel on the i-th segment; V is the set speed in still water corresponding to the rated power of the target vessel. The distance of the target vessel on the i-th segment; according to The fuel consumption impact coefficient based on the meteorological and marine environmental forecast data is obtained. Where AD represents relative wind direction and AS represents wind force level. c1 is the wind direction and wind speed penalty factor; c2 is the speed-related exponential parameter; c3 is the displacement-related exponential parameter. It refers to the amount of water discharged. The target vessel's current speed is used as an example. The fuel consumption impact coefficient is used to correct the fuel consumption of a single segment under the influence of weather factors, resulting in the fuel consumption of a single segment after correction using the meteorological and marine environmental forecast data, and the total fuel consumption after adding up the fuel consumption of a single segment.

[0014] In one feasible implementation, the shipping route of the target vessel is subjected to multi-dimensional route planning constraint processing to obtain a route planning constraint strategy. Specifically, this includes: locating and identifying static no-navigation zones along the target vessel's shipping route; applying route constraints to the fixed obstacle areas corresponding to the static no-navigation zones to generate a static no-navigation zone constraint strategy; wherein the static no-navigation zones include at least: land, islands, shoals, and military restricted areas; and using meteorological and marine environmental forecast data, locating and identifying dynamic weather warning zones along the target vessel's shipping route; applying real-time route avoidance constraints to the dynamic weather warning zones to obtain dynamic weather warning zone constraints. The strategy involves: imposing spherical rectangular boundary constraints on the navigation area of ​​the target vessel's shipping route based on a predetermined route, generating a navigation area boundary constraint strategy; imposing speed constraints on each segment of the target vessel's shipping route, generating a speed constraint strategy; and imposing route elimination constraints on the shipping route based on a pre-set comprehensive route safety assessment model, obtaining a route safety assessment constraint strategy. The comprehensive route safety assessment model is a fuzzy hierarchical route safety assessment model. The route planning constraint strategies include: the static no-navigation zone constraint strategy, the dynamic weather warning zone constraint strategy, the navigation area boundary constraint strategy, the speed constraint strategy, and the route safety assessment constraint strategy.

[0015] In one feasible implementation, the multi-objective optimization function system and the route planning constraint strategy are used to perform multimodal and multi-objective iterative optimization calculations on several generated candidate shipping routes to obtain the Pareto optimal route set for the target vessel in the target sea area. Specifically, this includes: dividing each optimization objective in the multi-objective optimization function system into sub-populations and performing independent iterative optimization on each sub-population to obtain candidate shipping routes for each sub-population; wherein the optimization objectives include: shortest sailing time objective, lowest fuel consumption objective, and best sailing comfort objective; merging the candidate shipping routes of all sub-populations and performing Pareto non-dominated sorting on all merged candidate shipping routes, while retaining the Pareto front solution set; updating the pheromone concentration or particle position for each optimization objective based on the Pareto front solution set, and performing multimodal constraint processing on the iteratively selected candidate shipping routes through the route planning constraint strategy; and outputting the candidate shipping routes after multimodal and multi-objective iterative optimization calculations when the maximum number of iterations or convergence condition is reached, to obtain the Pareto optimal route set.

[0016] In one feasible implementation, the Pareto optimal route set is smoothed and visualized, specifically including: smoothing each shipping route in the Pareto optimal route set using a preset spline interpolation or Dubins curve smoothing algorithm, and visually outputting several recommended shipping routes; visualizing the multi-dimensional indicators corresponding to the several recommended shipping routes; wherein the multi-dimensional indicators include at least: sailing time, fuel consumption, comfort score, route planning constraints, and meteorological and marine environmental forecasts.

[0017] Secondly, embodiments of this application also provide a multimodal, multi-objective maritime route planning device, the device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, enabling the at least one processor to execute a multimodal, multi-objective maritime route planning method as described in any of the above embodiments.

[0018] Thirdly, embodiments of this application also provide a non-volatile computer storage medium, which is a non-volatile computer-readable storage medium storing at least one program, each program including instructions, which, when executed by a terminal, cause the terminal to execute a multimodal, multi-objective maritime route planning method as described in any of the above embodiments.

[0019] This application provides a method, device, and medium for multimodal and multi-objective maritime route planning. Compared with the prior art, the embodiments of this application have the following beneficial technical effects: 1. Provides multi-option decision support. Unlike existing technologies that only output a single optimal route, this method can simultaneously provide multiple Pareto optimal routes, such as the shortest voyage time, lowest fuel consumption, and best comfort, allowing the captain to make flexible decisions based on real-time sea conditions and mission priorities.

[0020] 2. Enhance the robustness of route planning. When unexpected situations such as severe weather or changes in restricted areas occur during the voyage, multiple pre-planned alternative routes can be quickly switched without having to recalculate from scratch, significantly reducing emergency decision-making time.

[0021] 3. Improve ship energy efficiency and economic benefits. By incorporating factors such as ocean currents, wind and waves, and stall effects into the fuel consumption model, the optimization results more closely reflect actual navigation conditions. Real-world testing has shown that the optimized routes with the lowest fuel consumption can save approximately 5%-8% of fuel compared to traditional empirical routes.

[0022] 4. Filling the gap in multi-objective planning for maritime scenarios. Compared with existing aircraft route planning methods, this application is specifically designed for maritime scenarios, fully considering factors unique to the shipping industry such as ship stall, ocean current utilization, and wave drag, thus filling a technological gap in this field.

[0023] 5. Wide applicability. This method is applicable to transoceanic route planning for various commercial vessels such as bulk carriers, container ships, and tankers, and is compatible with existing ship navigation systems. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of a multimodal, multi-objective maritime route planning method provided for embodiments of this application; Figure 2 This is a schematic diagram of the structure of a multimodal, multi-objective maritime route planning device provided in an embodiment of this application. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0026] This application provides a multimodal, multi-objective maritime route planning method, such as... Figure 1 As shown, the multimodal, multi-objective shipping route planning method specifically includes steps S101-S105: S101. Based on the target vessel and the target sea area, obtain the basic information on the route planning of the target vessel and the meteorological and marine environmental forecast data required for the route planning in the target sea area.

[0027] Specifically, the basic information for route planning is first obtained based on the target vessel's fundamental design attribute data and navigation task constraints. This basic information includes at least: the location of the departure port, the location of the destination port, the target vessel's design parameters, and task constraints. The task constraints include at least: departure time, planned arrival time window, and the vessel's maximum and minimum still water speeds. The target vessel's design parameters include at least: length, beam, design draft, design speed, displacement, main engine power, and fuel consumption characteristics.

[0028] Furthermore, meteorological and marine environmental forecast data for the planned navigation area within the target sea area during the planned navigation period are obtained from the meteorological and marine data service platform via HTTP or FTP protocols. This meteorological and marine environmental forecast data includes at least: wave-related data, ocean current-related data, sea surface wind field data, and air pressure and temperature data.

[0029] In one embodiment, meteorological and marine environmental forecast data for the planned navigation area during the planned navigation period are obtained from the meteorological and marine data service platform via HTTP network communication protocol or FTP protocol: ocean waves (which may include: significant wave height, wave direction, period); ocean currents (which may include: current speed, current direction); sea surface wind field (which may include: wind speed, wind direction); air pressure, temperature, and other necessary meteorological elements.

[0030] S102. Based on the basic information of the route planning and meteorological and marine environmental forecast data, perform multi-objective function optimization processing on the target vessel's maritime route planning to generate a multi-objective optimization function system. The multi-objective functions include: the shortest sailing time objective function, the lowest fuel consumption objective function, and the optimal sailing comfort function.

[0031] It should be noted that a shipping route for a target vessel is described using the number of segments, waypoint coordinates, heading variables, and speed variables, among which: (1) Number of segments: A shipping route consists of multiple segments. The number of segments on a route can be represented by n. This parameter can be adjusted according to certain steps of the optimization algorithm. It should generally satisfy the inequality n≥2.

[0032] (2) Waypoint coordinates: Waypoint coordinates can be represented by vectors. It is represented by the coordinates of each waypoint as follows: , where λ i Let l represent the latitude of the i-th waypoint. i This represents the longitude of the coordinates of the i-th waypoint.

[0033] (3) Heading variable: As mentioned above, a ship's route consists of multiple route segments. The ship navigates between each route segment using a constant heading line. The heading of the constant heading line in each route segment can be represented by a vector φ, which constitutes the heading variable. ; where φ i This represents the heading of the i-th route segment, which can be determined by the latitude and longitude coordinates of the i-th waypoints and the (i+1)-th waypoint. φ i ∈[0°, 360°], with due north as 0° and increasing clockwise.

[0034] (4) Speed ​​variable: The dynamic behavior of a ship on its course depends on the frequency of the propeller rotation. This frequency depends entirely on the desired speed setpoint in still water. Therefore, the ship speed is represented by the vector v: ; where each component of the speed vector represents the desired still water speed for the corresponding route segment.

[0035] Specifically, according to The objective function for the shortest travel time is obtained. .in, The total sailing time for the route, as defined in the basic route planning information, is calculated by dividing the distance of each segment in the basic route planning information by the ship's actual speed in wind and waves. The sum is obtained by accumulation. The cumulative time a vessel spends navigating within a severe sea state warning zone, and The coefficients represent the penalties imposed on routes crossing dangerous sea areas.

[0036] As a feasible implementation method, actual speed for: .in, The ship speed is set for still water, h is the meaningful wave height, q is the relative angle between the ship's heading and the wave direction, and D is the ship's displacement.

[0037] In one embodiment, due to the additional drag exerted on the hull by wind and waves, the actual speed of a ship in wind and waves is generally lower than its speed in calm water, assuming the ship's propulsion power remains constant. This stall phenomenon is called natural stall. The degree of speed loss affects the calculation of speed in weather conditions, as well as the calculated sailing time or fuel consumption cost of the ship's route. Speed ​​loss is the most important factor among the various factors involved in ship weather route design. Speed ​​loss has a significant impact on the accuracy of ship positioning and the feasibility of route optimization through weather routing. Therefore, this application calculates the actual speed of the ship after correcting for stall. for: Meanwhile, this actual speed The formula used in this calculation method uses fixed coefficients, making it simpler to perform. It should be noted that this formula works best when the significant wave height is no more than 5m. In this case, for a given still-water speed, the actual speed of the ship has a nearly linear relationship with the significant wave height. Furthermore, when the significant wave height is between 0 and 5m, the error between the formula and the actual result is less than 1 kN. However, when the significant wave height is between 5 and 8m, the error increases to 1 to 3 kN, and the calculated speed is often overestimated. Therefore, in practice, the formula needs to be corrected based on the significant wave height. Finally, the actual speed is obtained using the speed loss function. The total travel time of the route is obtained by summing them up. Thus, the objective function for the shortest travel time is finally obtained. .

[0038] Furthermore, according to The objective function for minimizing fuel consumption is obtained. Where Q represents the total fuel consumption of the target shipping route, which is obtained by summing the fuel consumption of each segment after correction using meteorological and marine environmental forecast data.

[0039] As a possible implementation method, according to The fuel consumption for a single flight segment under conditions unaffected by weather factors was obtained. .in, The target vessel's fuel consumption for the i-th segment; Fuel consumption rate; P is the rated power of the main engine; V represents the actual speed of the target vessel on the i-th segment; V is the set speed in still water corresponding to the rated power of the target vessel. Let i be the distance of the target vessel on the i-th segment. Then, based on... The fuel consumption impact coefficient based on meteorological and marine environmental forecast data was obtained. Where AD represents relative wind direction and AS represents wind speed rating. c1 is the wind direction and wind speed penalty factor; c2 is the speed-related exponential parameter; c3 is the displacement-related exponential parameter. It refers to the amount of water discharged. The target vessel's current speed is given. Finally, the fuel consumption for a single leg of the journey is corrected for weather factors using the fuel consumption impact coefficient. This yields the corrected fuel consumption for a single leg, based on meteorological and marine environmental forecast data, as well as the total fuel consumption after adding up the fuel consumption for each leg.

[0040] In one embodiment, regarding the fuel consumption model applied in the objective function to minimize fuel consumption, 1) the formula for calculating the total voyage cost is: Where C represents the total voyage cost, f represents the ship's fixed costs, and k represents the ship's performance coefficient. This coefficient is related to factors such as ship performance (propeller, draft, hull form, sailing conditions, etc.), main engine and propulsion system type, and fuel type and price. In this application... The values ​​were obtained from experimental measurements in the actual ship test waters: Where U is the fuel price, g is the effective fuel consumption rate of the main engine, Ne is the effective power of the main engine, and V is the ship's still water speed.

[0041] 2) The formula for calculating operating costs per unit distance is: Where U is the water flow velocity, positive for downstream and negative for upstream. Taking its derivative and setting it equal to zero yields the speed per hour that minimizes the cost per unit distance traveled, i.e., the economic speed. Then the optimal fuel cost The calculation formula is: The relationship between ship speed and main engine power is as follows: Among them, P e Where D is the main unit power, and C is the displacement. e V is a constant. s This refers to the ship's actual speed.

[0042] 3) The formula for combining ship fuel consumption (without considering other factors) is: Where G is any cruising speed V s Let g be the fuel consumption for a voyage of s nautical miles, S be the voyage distance, P be the rated power of the main engine, and V0 be the ship's speed at the rated power. Therefore, based on the ship's fuel consumption formula (ignoring other factors), the fuel consumption for a single voyage segment under ideal weather conditions can be obtained. Furthermore, the fuel consumption rate is relatively high when the main engine is running at low load. As the power increases, the fuel consumption rate gradually decreases. The fuel consumption rate is lowest when the main engine power reaches about 80%-85% of the rated power. The corresponding ship speed at this point is the minimum fuel consumption rate speed. If the ship operates under high load conditions for a long time, it is more economical to use the minimum fuel consumption rate speed.

[0043] 4) Ship fuel consumption model that takes weather into account The fuel consumption model used in this application considers weather factors, specifically wind force causing the ship to sail outside its minimum fuel consumption rate speed, leading to increased fuel consumption of the ship's main engine. Its fuel consumption impact... The calculation formula is as follows: Where p(AD, AS) is the velocity penalty factor, defined as: Where AD is wind direction, AS is wind speed, c1 is a linear parameter, c2 is a speed-related exponential parameter, and c3 is a displacement-related exponential parameter. This refers to the water displacement. The calculation method for μ(AD, AS) is as follows:

[0044] The wind direction AD is represented by vectors ABCDEFGH.

[0045] In the ship fuel consumption model, only historical voyage records are needed to perform ridge regression analysis to obtain the parameters of the ship's speed-fuel consumption model, thus constructing the ship's speed-fuel consumption model per unit time under weather influence. Then, through the ship speed-fuel consumption model, the total fuel consumption Q of the target shipping route is obtained, and finally, the objective function for minimizing fuel consumption is derived. .

[0046] Furthermore, according to To obtain the optimal flight comfort function .in, The motion response of the target vessel at various points along the shipping route. The roll of the target vessel in segment i, For the pitch of the target vessel in segment i, The heave of the target vessel in segment i, The temperature, humidity, and pressure of the target vessel in segment i.

[0047] In one embodiment, the optimal flight comfort function It can comprehensively evaluate the ship's motion response, including roll, pitch, and heave, at various points along the route. Specifically, a ship seakeeping assessment model can be used, taking the effective wave height, wave period, and encounter angle at each point along the route as inputs to evaluate the ship's motion amplitude. Comfort targets are generally used for passenger ships or special ship types transporting precision instruments, and their activation can be selected based on the mission type.

[0048] Furthermore, the objective functions for shortest flight time, lowest fuel consumption, and best flight comfort are linked and stored together to construct a multi-objective optimization function system.

[0049] As a feasible implementation method, a comprehensive objective function can also be constructed, namely: Wherein, α, β, and γ are weighting coefficients, set by the shipping company according to the voyage mission requirements, and α + β + γ = 1. When using multi-objective optimization, this application can directly solve for the Pareto optimal route set without pre-setting the weights.

[0050] S103. Perform multi-dimensional route planning constraint processing on the target vessel's shipping route to obtain the route planning constraint strategy.

[0051] Specifically, it is necessary to perform area location and identification processing of static no-navigation zones for the target vessel's shipping route, and to apply route constraint processing to the fixed obstacle areas corresponding to the static no-navigation zones to generate static no-navigation zone constraint strategies. These static no-navigation zones include at least: land, islands, shallow waters, and military restricted areas.

[0052] In one embodiment, under static no-navigation zone constraints, routes may not cross areas with fixed obstacles such as land, islands, shallow waters, or military restricted areas.

[0053] Furthermore, by combining meteorological and marine environmental forecast data, the target vessel's shipping route is located and identified in a dynamic weather warning area, and the dynamic weather warning area is constrained in real time to obtain a dynamic weather warning area constraint strategy.

[0054] In one embodiment, under dynamic weather warning zone constraints, flight routes should avoid dynamic weather warning zones caused by severe weather conditions such as typhoons and giant waves as much as possible.

[0055] Furthermore, the target vessel's shipping route must be constrained by a spherical rectangular boundary of the navigation area based on the predetermined route, generating a navigation area boundary constraint strategy. Additionally, speed constraints must be applied to each segment of the target vessel's shipping route, generating a speed constraint strategy.

[0056] In one embodiment, under the navigation area boundary constraints, the route should be located within a pre-selected spherical rectangle of the navigation area, which can be defined as: Ω0 {(λ,l)|λmin≤λ≤λmax,lmin≤l≤lmax}. Furthermore, in the speed constraints, the still water speed for each segment should satisfy V. min ≤V i ≤V max .

[0057] Furthermore, a pre-defined comprehensive route safety assessment model is needed to apply route elimination constraints based on safety assessment to the shipping routes, resulting in a route safety assessment constraint strategy. The comprehensive route safety assessment model is a fuzzy hierarchical model. That is, in the safety assessment constraints, the fuzzy hierarchical comprehensive safety assessment model can be used to score the safety of candidate routes, and routes with scores below a threshold are directly eliminated.

[0058] In one embodiment, regarding the route safety assessment model: based on the route safety assessment method under route planning constraints, this application constructs a fuzzy hierarchical comprehensive safety assessment model (route safety assessment model). Fuzzy hierarchical comprehensive safety assessment is a safety assessment method that integrates the analytic hierarchy process (AHP) and fuzzy assessment methods. Guided by the basic theory of fuzzy mathematics, it uses the AHP to obtain the weights of evaluation factors and utilizes the principle of fuzzy relation transformation to quantify the fuzzy critical objects in the evaluation indicators. This maximizes the combination of data on system safety influencing factors, forming a highly reliable set of judgments on the system safety membership degree. The final comprehensive assessment result can be obtained according to the principle of maximum membership degree. Furthermore, this fuzzy hierarchical method can accurately characterize the fuzziness of the factors themselves and evaluate the safety status of the system by assessing the membership degree of each factor to danger (or safety), making the evaluation results more objective, practical, and reasonable.

[0059] In one embodiment, the steps for constructing the route safety assessment model of this application may include: establishing a factor set, establishing a weight set, establishing an evaluation set, establishing an evaluation matrix, and comprehensive evaluation.

[0060] 1) The factors to be established include human factors (maneuvering and response requirements), ship factors (stability, hull structure, heel, ship age, ship equipment, tonnage) and weather and environmental factors (wind, waves).

[0061] 2) When establishing a weight set, the AHP method can be used to obtain the indicator weights through experts, process the survey data using the geometric mean method to obtain the judgment matrix, and then check the consistency indicators.

[0062] 3) In establishing the evaluation set, the evaluation level set in this application can be divided into 5 levels (low risk, lower risk, moderate risk, higher risk and high risk).

[0063] 4) Establish an evaluation matrix: Expert evaluation and scoring methods can be used; for example: if there are N... k Personally, I think factor U i It should be categorized as comment V. j Then its evaluation value for: Therefore, the evaluation matrix R of the factor set U={U1,U2,U3} is: .

[0064] The route planning constraint strategies include: static no-navigation zone constraint strategy, dynamic weather warning zone constraint strategy, navigation area boundary constraint strategy, speed constraint strategy, and route safety assessment constraint strategy.

[0065] In one embodiment, the optimal weather route design for a ship is generally based on predicted weather, sea conditions, and customer needs. Simultaneously, certain constraints must be considered in route planning, such as static constraints arising from coastlines, shallow waters, islands, and military restricted areas, which can be obtained in advance; and dynamic constraints, such as severe weather like strong winds and large waves, which can be predicted using weather forecasting technology. Furthermore, utilizing favorable weather conditions such as downwinds and currents can increase sailing speed, reduce sailing time, and improve economic efficiency. Therefore, this application addresses the route design problem from one port to another, primarily considering the following constraints: 1) The outer boundary of the navigation area; 2) Safety boundaries around shallow waters, restricted areas such as land and islands; 3) Weather warning zones caused by severe weather conditions; 4) The maximum and minimum permissible still water speed of the vessel.

[0066] The first three constraints define the feasible domain of the ship's route position information, while the fourth constraint restricts the speed variable.

[0067] Constraint 1) represents an area determined by the initial parameters of the optimization algorithm, denoted as Ω0. The starting and target points of the problem should be contained within this area. Typically, when determining the optimal route, the entire Earth's surface is not considered; instead, a specific sea area is selected based on the chosen starting and target points, and the route optimization is performed within this local area. Therefore, assume the navigation area Ω0 is contained within the following spherical rectangle, as shown below: Where λ represents the latitude coordinates of the navigation area, l represents the longitude coordinates of the navigation area, and λ min and λ max These represent the minimum and maximum latitude values ​​of the selected marine environment, respectively, while lmin and lmax represent the minimum and maximum longitude values ​​of the selected marine environment.

[0068] Constraint 2) is a static constraint area that remains unchanged during navigation, including the safety outline around shallow waters, and obstacle areas such as land and islands. These obstacles are generally fixed during navigation. After selecting the navigation area Ω0, the static obstacles within that area can be determined, thus obtaining the static restricted area.

[0069] Constraint 3) defines dynamic weather warning zones as those caused by weather conditions, primarily severe weather events such as hurricanes and large waves. Unlike static restricted areas, these zones change over time based on weather forecasts. These dynamic weather warning zones should also be avoided during ship navigation, as they increase the risk of accidents and reduce the safety and comfort of navigation.

[0070] When considering constraints 1), 2), and 3), time should also be taken into account when analyzing the selection of path variables. Allowed candidate routes should fully comply with the constraints of the navigable area. It should be noted that not only do these waypoints need to fall within the navigable area, but the path trajectory formed by these waypoints should also be within the navigable area; that is, the path trajectory formed by these waypoints should not intersect with either static or dynamic restricted areas.

[0071] Constraint 4) imposes a speed limit. The vessel travels at the same set speed in still water for each segment of the route. However, due to the vessel's own structural parameters and the manually set minimum speed, the vessel's set speed in still water is limited to a certain range [V]. min V max [Inside, where V] min V is the minimum still water speed set for the vessel during navigation. max Set the maximum still water speed for the ship's navigation area.

[0072] S104. Through a multi-objective optimization function system and route planning constraint strategy, iterative optimization calculations under multi-modal and multi-objective conditions are performed on several candidate shipping routes to obtain the Pareto optimal route set for the target vessel in the target sea area.

[0073] Specifically, the optimization objectives in the multi-objective optimization function system are first divided into subpopulations, and each subpopulation undergoes independent iterative optimization to obtain candidate shipping routes for each subpopulation. The optimization objectives include: shortest sailing time, lowest fuel consumption, and optimal sailing comfort. Then, the candidate shipping routes from all subpopulations are merged, and the merged candidate shipping routes are subjected to Pareto non-dominated sorting, while retaining the Pareto front solution set.

[0074] Furthermore, based on the Pareto front solution set, the pheromone concentration or particle position of each optimization objective is updated, and the iterative candidate shipping routes are subjected to multimodal constraints through a route planning constraint strategy.

[0075] Furthermore, when the maximum number of iterations or the convergence condition is reached, the output processing of several candidate shipping routes after iterative optimization calculation under multimodal and multi-objective conditions is performed to obtain the Pareto optimal route set.

[0076] As a feasible implementation method, an improved multi-swarm ant colony algorithm or particle swarm optimization algorithm can be used to find the optimal route for multiple objectives. In the execution of the multi-modal multi-objective optimization strategy: 1) Parallel optimization of multiple populations: The total population is divided into 3 sub-populations, corresponding to the three optimization objectives of sailing time, fuel consumption and comfort, respectively. Each sub-population iterates independently; 2) Pareto non-dominated sorting: After each iteration, the candidate routes generated by all sub-populations are merged and non-dominated sorted, and the Pareto front solution set is retained; 3) Pheromone / position update: The pheromone concentration (ant colony algorithm) or particle position (particle swarm optimization algorithm) is updated for each objective; 4) Iteration termination: The route planning constraint strategy is first used to perform multi-modal constraint processing on several candidate shipping routes after iteration. The iteration is terminated when the maximum number of iterations or the convergence condition is reached, and finally the Pareto optimal route set is output.

[0077] In one embodiment, ant colony optimization (ACO) is a biomimetic algorithm that simulates the pathfinding behavior of ants in nature. It is a probabilistic algorithm used to find optimal paths in a graph. Due to the diversity of pathfinding problems and the presence of various influencing factors, some optimizations are made to the pheromone update and termination rules during model building. First, the optimal path for each generation is calculated, its maximum benefit is determined, and pheromone is updated. During the optimization process of the next generation, a certain proportion of the previous generation's pheromone is evaporated. Further optimization is then performed to obtain the best path, and its maximum benefit is calculated. If this benefit ratio is less than the previous one, it is directly evaporated; if this benefit is greater than the previous one, it is selected as the optimal route, and pheromone is updated. This process continues until the maximum number of iterations is reached, at which point the iteration terminates. Therefore, the complete optimal path planning method of this application can be: (1) Based on the voyage plan, select the correct starting point for the route design and obtain the position coordinates; (2) Select an area on a nautical chart of appropriate scale based on the starting point; (3) Based on the gridding method, the nautical chart is preprocessed to discretize the obstacle grid and the navigable sea area grid, and the center of all navigable sea area grids is used as the navigable point for route design; (4) Initialize the parameters of the ant colony algorithm search path, set the initial distribution of pheromones, and place the ants at the starting position (i.e., the starting grid). (5) Based on the objective function, determine the algorithm transfer strategy, pheromone update method, and combine the constraints in the route planning constraint strategy to iterate out the optimal path; (6) Perform feasibility processing on the optimal path and output the optimal path.

[0078] S105. Smooth and visualize the Pareto optimal route set.

[0079] Specifically, the system first uses a pre-defined spline interpolation or Dubins curve smoothing algorithm to smooth each shipping route in the Pareto optimal route set, and then visualizes and outputs several recommended shipping routes.

[0080] Furthermore, the multi-dimensional indicators corresponding to several recommended shipping routes can be visualized and output. These multi-dimensional indicators include at least: sailing time, fuel consumption, comfort score, route planning constraints, and meteorological and marine environmental forecasts.

[0081] As a feasible approach, Pareto optimal routes can be smoothed (e.g., using spline interpolation or Dubins curve smoothing) to ensure the routes meet ship maneuverability requirements. Finally, multiple recommended routes are output along with their corresponding sailing times, fuel consumption, comfort scores, and other indicators for captains or shipping dispatchers to use as a reference for decision-making.

[0082] In one embodiment, this application conducted multi-objective route planning based on meteorological and marine environmental data based on the demonstration route of the demonstration vessel Zheng Tong, and developed route optimization system software.

[0083] First, this application collected relevant parameters of the demonstration vessel Zhengtong (see Table 1) and input them into the route optimization system.

[0084] Table 1 Design Parameters of Zhengtong Wheel

[0085] Secondly, the route optimization system used the MMSI of the Zhengtong vessel as a search parameter and, using the HTTP network communication protocol jointly built with the previous system service platform, obtained the ID of the Zhengtong vessel's demonstration voyage, as well as related information such as name, departure and arrival times.

[0086] Then, the route optimization system used the Zhengtong vessel's demonstration voyage ID as a parameter and continued to send network requests to the system service platform using the HTTP network communication protocol to obtain the planned route information of the demonstration voyage.

[0087] Next, the route optimization system uses the FTP network transmission protocol to download the meteorological and marine environmental forecast data corresponding to the spatiotemporal coverage provided by this application from the platform server based on the departure and arrival times of the demonstration voyage and the area covered by the planned route, and loads it into the route optimization system of this application.

[0088] Finally, the route optimization system aggregates all data related to route optimization calculations. Utilizing the aforementioned multi-objective route planning mathematical model and intelligent optimization algorithm, it iteratively solves for the optimal route using an intelligent route optimization algorithm modified from the ant colony algorithm, with the shortest flight time, lowest fuel consumption, and best comfort as objective functions. The optimized routes are then pushed to the system service platform via a shared HTTP network communication protocol. Simultaneously, the route optimization system can also display the optimal routes corresponding to the shortest flight time (red), lowest fuel consumption (green), and best comfort (blue), as well as the planned routes (black). To clearly distinguish between the optimized and planned routes, the three optimized routes can be displayed separately.

[0089] The route optimization results show that, with the shortest travel time as the objective function, the optimized route is closer to the great circle route than the planned route; with the lowest fuel consumption as the objective function, the optimized route favors routes with stronger current fields, thus choosing the eastern side of the island instead of passing through the strait; and with optimal comfort as the objective function, the optimized route favors routes with lower wave height, wind speed, and current speed. In summary, the routes calculated using the route optimization algorithm proposed in this application have shorter travel distances compared to the planned routes.

[0090] In addition, embodiments of this application also provide a multimodal, multi-objective maritime route planning device, such as... Figure 2 As shown, the multimodal, multi-objective maritime route planning equipment 200 specifically includes: At least one processor 201; and a memory 202 communicatively connected to the at least one processor 201; wherein the memory 202 stores instructions executable by the at least one processor 201 to enable the at least one processor 201 to execute: Based on the target vessel and the target sea area, the basic information on the route planning of the target vessel and the meteorological and marine environmental forecast data required for the route planning in the target sea area are obtained respectively. Based on the basic information of the route planning and meteorological and marine environmental forecast data, the maritime route planning of the target vessel is optimized using a multi-objective function to generate a multi-objective optimization function system. Among them, the multi-objective functions include: the shortest sailing time objective function, the lowest fuel consumption objective function, and the best sailing comfort function. The shipping routes of the target vessel are subjected to multi-dimensional route planning constraints to obtain the route planning constraint strategy; By using a multi-objective optimization function system and route planning constraint strategy, iterative optimization calculations under multi-modal and multi-objective conditions are performed on several candidate shipping routes to obtain the Pareto optimal route set for the target vessel in the target sea area. The Pareto optimal route set is smoothed and visualized.

[0091] This application provides multiple Pareto optimal routes through multi-objective optimization, achieving the shortest voyage time, lowest fuel consumption, and best comfort. The captain can make flexible decisions based on real-time sea conditions and mission priorities. In the event of unexpected situations such as sudden severe weather or changes in restricted areas during voyages, multiple pre-planned alternative routes can be quickly switched without recalculating from scratch, significantly reducing emergency decision-making time. Furthermore, by incorporating factors such as ocean currents, wind and waves, and stall effects into the fuel consumption model, the optimization results more closely reflect actual navigation conditions. Moreover, this application is specifically designed for maritime scenarios, fully considering factors unique to the shipping industry such as ship stall, ocean current utilization, and wave drag, filling a technological gap in this field. It is also applicable to transoceanic route planning tasks for various commercial vessels such as bulk carriers, container ships, and tankers, and is compatible with existing ship navigation systems.

[0092] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0093] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0094] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of this specification.

Claims

1. A multimodal, multi-objective maritime route planning method, characterized in that, The method includes: Based on the target vessel and the target sea area, the basic information on the route planning of the target vessel and the meteorological and marine environmental forecast data required for the route planning in the target sea area are obtained respectively. Based on the route planning information and the meteorological and marine environmental forecast data, the maritime route planning of the target vessel is optimized using a multi-objective function to generate a multi-objective optimization function system. The multi-objective functions include: the shortest sailing time objective function, the lowest fuel consumption objective function, and the optimal sailing comfort function. The shipping route of the target vessel is subjected to multi-dimensional route planning constraints to obtain a route planning constraint strategy. Using the multi-objective optimization function system and the route planning constraint strategy, iterative optimization calculations under multi-modal and multi-objective conditions are performed on several candidate sea routes to obtain the Pareto optimal route set of the target vessel in the target sea area. The Pareto optimal route set is smoothed and visualized.

2. The maritime route planning method based on multimodal and multi-objective modes according to claim 1, characterized in that, Based on the target vessel and the target sea area, the basic route planning information of the target vessel and the meteorological and marine environmental forecast data required for route planning in the target sea area are obtained, specifically including: Based on the basic design attribute data of the target vessel and the navigation task constraint information, the basic information for route planning is obtained; wherein, the basic information for route planning includes at least: the location of the departure port, the location of the destination port, the design parameters of the target vessel, and the task constraint information; the task constraint information includes at least: the departure time, the planned arrival time window, the maximum and minimum set speeds in still water; the design parameters of the target vessel include at least: the length, beam, design draft, design speed, displacement, main engine power, and fuel consumption characteristics; Meteorological and marine environmental forecast data for the planned navigation area in the target sea area during the planned navigation period are obtained from the meteorological and marine data service platform via HTTP network communication protocol or FTP protocol; wherein, the meteorological and marine environmental forecast data includes at least: wave-related data, ocean current-related data, sea surface wind field data, and air pressure and temperature data.

3. The maritime route planning method based on multimodal and multi-objective modes according to claim 1, characterized in that, Based on the aforementioned route planning information and the aforementioned meteorological and marine environmental forecast data, the maritime route planning of the target vessel is subjected to multi-objective function optimization processing to generate a multi-objective optimization function system, specifically including: according to The objective function for the shortest travel time is obtained. ;in, The total sailing time in the basic route planning information is calculated by dividing the distance of each segment in the basic route planning information by the ship's actual speed in wind and waves. The sum is obtained by accumulation; This refers to the cumulative time a vessel spends navigating within a severe sea state warning zone, and The coefficients represented are used to penalize routes that cross dangerous sea areas; according to The objective function for minimizing fuel consumption is obtained. Where Q represents the total fuel consumption of the target shipping route, which is obtained by summing the fuel consumption of a single segment after correction using the meteorological and marine environmental forecast data. according to The optimal flight comfort function is obtained. ;in, The motion response of the target vessel at each waypoint along the shipping route. The roll of the target vessel in the i-th segment, The pitch of the target vessel in the i-th segment, The heave of the target vessel in the i-th segment. The temperature, humidity, and pressure of the target vessel in the i-th segment; The shortest flight time objective function, the lowest fuel consumption objective function, and the optimal flight comfort function are associated and stored to construct the multi-objective optimization function system.

4. The maritime route planning method based on multimodal and multi-objective modes according to claim 3, characterized in that, The actual speed for: ; Set the ship speed for still water, h is the meaningful wave height, q is the relative angle between the ship's heading and the wave direction, and D is the ship's displacement.

5. The maritime route planning method based on multimodal and multi-objective modes according to claim 3, characterized in that, The total fuel consumption of the target shipping route is obtained by summing the fuel consumption of each segment after correction using the aforementioned meteorological and marine environmental forecast data, specifically including: according to The fuel consumption for a single flight segment under conditions unaffected by weather factors was obtained. ;in, The fuel consumption of the target vessel on the i-th segment; Fuel consumption rate; P is the rated power of the main engine; V represents the actual speed of the target vessel on the i-th segment; V is the set speed in still water corresponding to the rated power of the target vessel. The distance of the target vessel on the i-th segment; according to The fuel consumption impact coefficient based on the meteorological and marine environmental forecast data is obtained. Where AD represents relative wind direction and AS represents wind force level. c1 is the wind direction and wind speed penalty factor; c2 is the speed-related exponential parameter; c3 is the displacement-related exponential parameter. It refers to the amount of water discharged. The current speed of the target vessel; The fuel consumption impact coefficient is used to correct the fuel consumption of a single flight segment under the influence of weather factors, resulting in the fuel consumption of a single flight segment after correction based on the meteorological and marine environmental forecast data, and the total fuel consumption after adding up the fuel consumption of a single flight segment.

6. The maritime route planning method based on multimodal and multi-objective modes according to claim 1, characterized in that, The shipping route of the target vessel is subjected to multi-dimensional route planning constraints to obtain a route planning constraint strategy, which specifically includes: The target vessel's shipping route is subjected to static no-navigation zone location and identification processing, and the fixed obstacle area corresponding to the static no-navigation zone is subjected to route constraint processing to generate a static no-navigation zone constraint strategy; wherein, the static no-navigation zone includes at least: land, islands, shallow waters and military restricted areas; Using the meteorological and marine environmental forecast data, the target vessel's shipping route is located and identified in a dynamic weather warning area, and the dynamic weather warning area is constrained in real time to obtain a dynamic weather warning area constraint strategy. The target vessel's sea route is constrained by a spherical rectangular boundary of the navigation area based on the predetermined route, thereby generating a navigation area boundary constraint strategy. For each segment of the target vessel's shipping route, a speed constraint process is applied to generate a speed constraint strategy. By using a pre-defined comprehensive route safety assessment model, route elimination constraints are applied to shipping routes based on safety assessment, resulting in a route safety assessment constraint strategy; wherein, the comprehensive route safety assessment model is a route safety assessment model based on fuzzy hierarchy. The route planning constraint strategies include: the static no-navigation zone constraint strategy, the dynamic weather warning zone constraint strategy, the navigation area boundary constraint strategy, the speed constraint strategy, and the route safety assessment constraint strategy.

7. The maritime route planning method based on multimodal and multi-objective modes according to claim 1, characterized in that, Using the multi-objective optimization function system and the route planning constraint strategy, iterative optimization calculations under multimodal and multi-objective conditions are performed on several generated candidate shipping routes to obtain the Pareto optimal route set for the target vessel in the target sea area, specifically including: Each optimization objective in the multi-objective optimization function system is divided into sub-populations, and each sub-population is subjected to independent iterative optimization to obtain candidate shipping routes for each sub-population; wherein, the optimization objectives include: shortest sailing time objective, lowest fuel consumption objective, and best sailing comfort objective; The candidate shipping routes of all subpopulations are merged, and the merged candidate shipping routes are sorted by Pareto non-dominated order, while retaining the Pareto front solution set. Based on the Pareto front solution set, the pheromone concentration or particle position is updated for each optimization objective, and the route planning constraint strategy is used to perform multimodal constraint processing on several candidate shipping routes after iteration. When the maximum number of iterations or the convergence condition is reached, the algorithm terminates by outputting several candidate shipping routes after iterative optimization calculations under multimodal and multi-objective conditions, and obtains the Pareto optimal route set.

8. The maritime route planning method based on multimodal and multi-objective modes according to claim 1, characterized in that, The Pareto optimal route set is smoothed and visualized, specifically including: By using a preset spline interpolation or Dubins curve smoothing algorithm, each shipping route in the Pareto optimal route set is smoothed, and several recommended shipping routes are visualized and output. The multi-dimensional indicators corresponding to the recommended shipping routes are visualized and output; wherein, the multi-dimensional indicators include at least: sailing time, fuel consumption, comfort score, route planning constraints, and meteorological and marine environmental forecasts.

9. A multimodal, multi-objective maritime route planning device, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, enabling the at least one processor to execute a multimodal, multi-objective maritime route planning method according to any one of claims 1-8.

10. A non-volatile computer storage medium, characterized in that, The storage medium is a non-volatile computer-readable storage medium that stores at least one program, each program including instructions that, when executed by a terminal, cause the terminal to perform a multimodal, multi-objective maritime route planning method according to any one of claims 1-8.

Images