A ship route optimization method and system

Abstract

The application relates to the technical field of ship navigation, and discloses a ship navigation route optimization method and system. The method comprises meteorological change sequences, ocean current fluctuation sequences and channel obstacle distribution data; two-dimensional route topology construction processing is performed on the data, and route distribution characteristics are generated; a pre-trained path optimization model is called to perform nonlinear path analysis processing, and optimal route prediction values and key risk area identification are obtained; the optimal route prediction values are subjected to sailing environment compensation correction processing based on the correlation between the meteorological change sequences and ship control sensitivity, and corrected optimal route prediction values are generated; and finally, a route maintenance strategy set comprising a heading adjustment path optimization scheme and a sailing speed dynamic regulation and control processing scheme is generated according to the key risk area identification. The method can comprehensively process dynamic environmental factors, improve the accuracy and adaptability of route planning, effectively identify and avoid risks, and guarantee the efficiency and safety of ship sailing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship navigation, in particular to a ship navigation route optimization method and system. BACKGROUND

[0002] With the vigorous development of global shipping industry, ship navigation efficiency and safety have become the focus of the industry. In the complex and changeable marine environment, the rationality of ship navigation route directly affects the transportation cost, navigation time and the safety of crew and cargo. Currently, the traditional ship route planning method mainly relies on experience judgment or static channel data, which is difficult to cope with dynamic changes of environmental factors such as weather and ocean current.

[0003] The sudden change of weather conditions, such as sudden strong storm and thick fog, often leads to the deviation of ship from the planned route, increasing the risk of navigation. The irregular fluctuation of ocean current will significantly affect the speed and direction of the ship, so that the originally planned optimal route is no longer applicable in actual navigation. In addition, the distribution of obstacles in the channel has uncertainty, and temporary floating objects, shoals and other obstacles may pose a threat to ship navigation, and the traditional method often lags behind in response to these dynamic obstacles.

[0004] Most of the existing path optimization models are based on linear analysis method, which is difficult to handle the nonlinear relationship between environmental data and route characteristics. This leads to a large deviation between the predicted value of the generated route and the actual optimal route, which cannot meet the demand of high-precision navigation. At the same time, in the process of route planning, the identification of key risk areas is not accurate enough, so that the route maintenance strategy lacks pertinence and is difficult to effectively avoid potential risks.

[0005] The correlation between ship control sensitivity and weather changes is not fully considered, which leads to ineffective environmental compensation when correcting the route prediction value, further reducing the accuracy of route planning. The existence of these problems makes the ship often face the difficulties of frequent route adjustment, increased fuel consumption and prolonged navigation time during navigation, which not only affects the economic benefits of shipping enterprises, but also causes certain negative impact on the marine ecological environment, such as the increase of carbon emissions caused by unnecessary fuel consumption.

[0006] How to comprehensively process dynamic environmental data, build more accurate path optimization model, generate optimal route that can adapt to complex environmental changes, and develop effective route maintenance strategy has become a problem to be solved in the field of ship navigation route optimization. SUMMARY

[0007] The purpose of the present application is to provide a ship navigation route optimization method to solve the problems raised in the background.

[0008] To achieve the above object, the present invention provides a method for optimizing a ship's navigation route, the method comprising:

[0009] Collecting a dynamic environmental data set of the target ship in the navigation environment, wherein the dynamic environmental data set includes a weather change sequence, an ocean current fluctuation sequence, and a channel obstacle distribution data;

[0010] Performing a two-dimensional route topology construction process on the dynamic environment data set to generate a route distribution feature of the target ship, the route distribution feature including a heading deviation gradient, a speed matching cumulative amount, and an obstacle avoidance response fluctuation coefficient;

[0011] Calling a pre-trained path optimization model to perform nonlinear path analysis on the route distribution characteristics to generate an optimal route prediction value and a key risk area identifier for the target ship;

[0012] performing navigation environment compensation correction processing on the optimal route prediction value to generate a corrected optimal route prediction value, wherein the navigation environment compensation correction processing is implemented based on the correlation relationship between the meteorological change sequence and the ship control sensitivity;

[0013] A route maintenance strategy set is generated according to the key risk area identifier, and the route maintenance strategy set includes a heading adjustment path optimization solution and a speed dynamic control processing solution.

[0014] Preferably, the performing a two-dimensional route topology construction process on the dynamic environment data set to generate the route distribution characteristics of the target ship includes:

[0015] Dividing the weather change sequence into multiple environmental subsequences according to time windows, each environmental subsequence corresponding to a route planning cycle;

[0016] For each of the environment subsequences, the following processing is performed:

[0017] Constructing a two-dimensional navigation topology structure of the target ship according to the channel obstacle distribution data, wherein the two-dimensional navigation topology structure includes spatial distribution data of a heading deviation field, a speed adaptation field, and an obstacle avoidance response field;

[0018] Couple the two-dimensional navigation topology structure with the environmental subsequence for analysis and processing to generate a route topology result for the current time window, the route topology result including a spatial distribution matrix of a heading deviation component, a speed matching component, and an obstacle avoidance response component;

[0019] The route topology results of multiple consecutive time windows are cumulatively superimposed to calculate the heading deviation gradient, speed matching accumulation and obstacle avoidance response fluctuation coefficient; wherein,

[0020] The heading deviation gradient is the maximum rate of change of the heading deviation component along the route planning direction.

[0021] The speed matching cumulative amount is the integral of the speed matching component in the normal direction of the channel centerline.

[0022] The obstacle avoidance response fluctuation coefficient is the ratio of the standard deviation to the average value of the obstacle avoidance response component within a predetermined time interval.

[0023] Preferably, coupling analysis and processing the two-dimensional navigation topology structure with the environmental subsequence to generate a route topology result for the current time window includes:

[0024] Establishing a heading offset-correction mapping equation based on a correspondence between the heading offset field and the heading correction component in the environment subsequence, and obtaining a first distribution function of the heading deviation component by solving the heading offset-correction mapping equation;

[0025] According to the correlation characteristics between the speed adaptation field and the waterway speed limit, a speed matching calculation model is constructed, wherein the speed matching calculation model includes dynamic correction parameters of the ship load coefficient and propulsion power;

[0026] Combining the spatial change rate of the obstacle avoidance response field and the ship turning radius, an obstacle avoidance response iterative calculation process is established, wherein the iterative calculation process includes a feedback correction mechanism for position increment and obstacle avoidance response increment;

[0027] The output results of the first distribution function, the speed matching calculation model and the obstacle avoidance response iterative calculation process are subjected to spatial interpolation fusion processing to generate two-dimensional route topology distribution data including heading, speed and obstacle avoidance response components.

[0028] Preferably, the calling of a pre-trained path optimization model to perform nonlinear path analysis on the route distribution characteristics to generate an optimal route prediction value and a key risk area identifier for the target ship includes:

[0029] Inputting the heading deviation gradient into the first feature analysis layer of the path optimization model, and determining the distribution coordinates of the heading deviation area and the heading adjustment amplitude change curve through the heading deviation factor calculation module;

[0030] Inputting the speed matching cumulative amount into the second feature analysis layer of the path optimization model, performing speed adaptation damage cumulative calculation, and generating a navigation delay probability and adjustment rate prediction value for the speed matching section;

[0031] Inputting the obstacle avoidance response fluctuation coefficient into the third characteristic analysis layer of the path optimization model, and calculating the number of ship turns and channel traffic efficiency evolution data of the obstacle avoidance response surface based on the channel obstacle avoidance degradation model;

[0032] fusing the course adjustment amplitude change curve, the probability of navigation delay, and the number of ship turns to generate a comprehensive path index for the target ship, and determining an optimal route prediction value based on a comparison result of the comprehensive path index with a preset route threshold;

[0033] Based on the spatial superposition results of the distribution coordinates, the adjustment rate prediction value and the waterway traffic efficiency evolution data, the geometric positions of the heading deviation area, the delay extension path and the obstacle avoidance high-risk area are identified.

[0034] Preferably, performing navigation environment compensation correction processing on the optimal route prediction value to generate a corrected optimal route prediction value includes:

[0035] Extracting extreme weather values ​​and weather change frequencies from the weather change sequence, and calculating the dynamic adjustment amount of the ship's control sensitivity as the weather changes;

[0036] Performing meteorological influence compensation calculation on the heading deviation gradient according to the dynamic adjustment amount to generate a corrected heading deviation gradient;

[0037] Based on the correlation between the weather change frequency and the ship's rolling characteristics, performing roll effect correction processing on the speed matching cumulative amount to generate a corrected speed matching cumulative amount;

[0038] According to the ship braking distance change data under extreme weather conditions, the obstacle avoidance response fluctuation coefficient is adaptively adjusted to adjust the control capability to generate a corrected obstacle avoidance response fluctuation coefficient;

[0039] The corrected heading deviation gradient, speed matching accumulation and obstacle avoidance response fluctuation coefficient are input into the path optimization model for recalculation to generate an optimal route prediction value after compensating for environmental factors.

[0040] Preferably, performing meteorological influence compensation calculation on the heading deviation gradient according to the dynamic adjustment amount to generate a corrected heading deviation gradient includes:

[0041] Obtaining the initial maneuvering sensitivity of the target ship under reference weather conditions and the dynamic adjustment amount, and establishing a maneuvering sensitivity-weather correlation function;

[0042] Calculating a heading correction increment based on the control sensitivity-weather correlation function, wherein the heading correction increment is the product of the weather change and the control sensitivity change;

[0043] Adding the heading correction increment to the calculation process of the heading deviation gradient to generate a heading deviation gradient correction value including meteorological influence;

[0044] The heading deviation gradient correction value is subjected to a control delay effect compensation process, which is based on a product factor of a ship turning delay curve and a weather retention time.

[0045] Preferably, the generating a route maintenance strategy set according to the key risk area identification includes:

[0046] For the identification of the heading deviation area, an optimal heading adjustment path is calculated, which is achieved by adjusting the heading distribution ratio of adjacent segments;

[0047] According to the identification of the delay expansion path, a dynamic speed regulation processing scheme is constructed, which includes the selection of acceleration intervals and the optimization of propulsion power parameters;

[0048] Based on the identification of the obstacle avoidance high-risk area, a channel bypass strategy is generated, which dynamically adjusts the bypass distance and turning angle according to the obstacle avoidance response rate prediction value;

[0049] The optimal heading adjustment path, the dynamic speed regulation processing scheme and the channel bypass strategy are subjected to priority sorting processing to generate a maintenance strategy set containing execution timing and implementation parameters.

[0050] Preferably, the constructing a dynamic speed regulation processing scheme includes:

[0051] Extract the geometric features of the delay expansion path, calculate the path curvature radius and the expansion direction angle;

[0052] According to the curvature radius, the coverage density of the acceleration interval is selected, which is inversely proportional to the curvature radius;

[0053] Based on the expansion direction angle, the application direction of the propulsion power is adjusted, so that the propulsion direction forms a predetermined angle with the delay expansion direction;

[0054] According to the ship load test data, the propulsion power duration is dynamically adjusted to ensure that the propulsion power is below the ship rated power critical value;

[0055] A regulation parameter configuration table containing coverage density, propulsion direction and power duration is generated.

[0056] Preferably, the method further includes:

[0057] In a preset verification period, the actual navigation deviation amount and the delay expansion length of the target ship are collected;

[0058] The actual navigation deviation amount and the predicted heading deviation gradient are subjected to deviation analysis processing to generate a first error correction coefficient;

[0059] Performing a time domain comparison process on the delay extension length and the predicted adjustment rate to generate a second error correction coefficient;

[0060] adjusting the weight parameters of the path optimization model according to the first error correction coefficient and the second error correction coefficient to generate an optimized path optimization model;

[0061] The optimized path optimization model is applied to subsequent batches of ship route optimization tasks.

[0062] Preferably, the present invention also includes a ship navigation route optimization system, including a processor and a memory, the memory and the processor are connected, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the above-mentioned ship navigation route optimization method.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] This ship route optimization method effectively addresses the complex and ever-changing ocean navigation environment. By collecting dynamic environmental data on the target ship's navigation environment, including weather patterns, ocean current fluctuations, and the distribution of obstacles in the waterway, it provides comprehensive and real-time foundational information for route planning. Compared to traditional methods that rely on empirical or static data, this comprehensive collection of dynamic data allows route planning to move beyond fixed conditions and adapt to changing circumstances.

[0065] The system constructs a two-dimensional route topology from dynamic environmental data sets, generating route distribution characteristics such as heading deviation gradient, speed matching accumulation, and obstacle avoidance response fluctuation coefficient. This transforms the originally complex environmental data into route characteristic parameters that are easier to analyze and apply. These characteristic parameters reflect the relationship between the ship's navigation and the environment from multiple dimensions, providing precise input for subsequent path analysis and avoiding the difficulties caused by complex data.

[0066] A pre-trained route optimization model is used to perform nonlinear route analysis on route distribution characteristics, overcoming the limitations of traditional linear analysis methods. Nonlinear analysis better captures the complex relationships between environmental data and routes, resulting in more accurate optimal route predictions and identification of critical risk areas. This ensures that predicted routes not only meet efficiency requirements but also proactively identify potential risk areas, providing a stronger guarantee for navigation safety.

[0067] The optimal route prediction value is subjected to a sailing environment compensation correction process based on the correlation between the meteorological change sequence and the ship maneuvering sensitivity, and the accuracy of the route prediction is further improved. The ship maneuvering sensitivity will change with the change of the meteorological condition, and this factor is taken into account in the correction process, so that the corrected route can better adapt to the actual sailing environment and reduce the route deviation caused by environmental changes.

[0068] According to the key risk area identification, a route maintenance strategy set containing a heading adjustment path optimization scheme and a sailing speed dynamic regulation processing scheme is generated, which provides specific and feasible operation guidance for the ship to cope with risks during sailing. When the ship approaches the key risk area, the heading and sailing speed can be adjusted in time according to these strategies to effectively avoid risks and ensure the smooth sailing. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 The working principle diagram of the ship sailing route optimization method is provided.

[0070] Figure 2 The flowchart of the sailing environment compensation correction process is provided.

[0071] Figure 3 The flowchart of the meteorological influence compensation calculation is provided.

[0072] Figure 4 The flowchart of the path optimization model verification and optimization is provided. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0074] Please refer to Figures 1-4 The present application provides a ship sailing route optimization method, and the specific implementation steps are as follows:

[0075] The dynamic environmental data set of the target ship in the navigation environment is collected. The dynamic environmental data set includes meteorological change series, ocean current fluctuation series, and channel obstacle distribution data. The meteorological change series can be obtained in real time through the meteorological sensors on board the ship, covering parameters that change over time, such as wind speed, wind direction, precipitation, and visibility. The ocean current fluctuation series is collected by ocean monitoring buoys or ocean current detection equipment on the ship, including time series data on ocean current speed and direction. The channel obstacle distribution data is combined with electronic nautical charts, radar detection, and satellite remote sensing information to form a spatial distribution data set that includes fixed obstacles (such as islands and reefs) and mobile obstacles (such as other ships).

[0076] The dynamic environment data set is processed through a two-dimensional route topology construction process to generate the target ship's route distribution characteristics, including heading deviation gradient, speed matching accumulation, and obstacle avoidance response fluctuation coefficient. This process combines dynamic environment data with the spatial characteristics of the ship's navigation to construct characteristic indicators that reflect the changes in the ship's navigation state under different environments, providing basic data support for subsequent path optimization.

[0077] A pre-trained route optimization model is used to perform nonlinear route analysis on the route distribution characteristics, generating an optimal route prediction for the target vessel and identifying key risk areas. This model, built using a deep learning framework and trained on extensive historical navigation data, is capable of processing nonlinear relationships within route distribution characteristics, predicting the optimal route after comprehensively considering environmental impacts and risk factors, and identifying potential key risk areas during navigation.

[0078] The optimal route prediction value is then subjected to navigation environment compensation and correction processing to generate a revised optimal route prediction value. This navigation environment compensation and correction processing is performed based on the correlation between the weather change sequence and the ship's maneuverability. Since weather conditions directly affect the ship's maneuverability, a correlation model between weather changes and ship maneuverability is established to compensate and correct the initially predicted optimal route, thereby improving the accuracy and adaptability of the route.

[0079] Based on the key risk area identification, a set of route maintenance strategies is generated. These strategies include a course adjustment path optimization solution and a dynamic speed control solution. A corresponding route maintenance strategy is formulated for the identified key risk areas. By adjusting course and speed, the vessel can safely and efficiently avoid the risk areas, ensuring safe navigation.

[0080] Example 1:

[0081] When constructing a two-dimensional route topology for a dynamic environmental data set to generate the target ship's route distribution characteristics, the meteorological change sequence is divided into multiple environmental subsequences based on time windows, with each environmental subsequence corresponding to a route planning period. The division of time windows requires a comprehensive consideration of the environmental stability of the navigation area and the timeliness of route planning. For example, in busy nearshore waters, where the environment changes rapidly, the time window can be set to 1 to 2 hours to capture short-term meteorological fluctuations. In open waters, where the environment is relatively stable, the time window can be extended to 4 to 6 hours, reducing data processing while ensuring data representativeness. Each environmental subsequence contains a continuous record of meteorological parameters such as wind speed, wind direction, precipitation, and visibility during that time period, providing basic data for subsequent time-segmented route analysis.

[0082] For each environmental subsequence, a two-dimensional navigation topology structure for the target ship is first constructed based on the channel obstacle distribution data. This structure includes the spatial distribution data of the heading deviation field, the speed adaptation field, and the obstacle avoidance response field. The heading deviation field is gridded across the channel area, with each grid point indicating the possible heading deviation angle of the ship at that location. Data for this field is derived from historical navigation records of course deviations in that area and the estimated impact of current currents and wind direction on the course. The speed adaptation field is also presented in a grid format, with the value at each grid point representing the adaptive speed that the ship can achieve at that location, taking into account the channel speed limit, water depth, and the ship's power performance parameters. The obstacle avoidance response field reflects the ship's response sensitivity to obstacles at different locations. The closer the obstacle, the higher the response sensitivity. The value distribution is also adjusted based on the obstacle type (e.g., fixed reef or moving ship). The response sensitivity distribution is more concentrated around fixed obstacles, while the sensitivity distribution around moving obstacles changes dynamically based on the ship's movement.

[0083] The two-dimensional navigation topology is coupled with the environmental subsequence for analysis and processing to generate the route topology for the current time window. This process unfolds across three dimensions: In the heading dimension, a correlation model is established based on the correspondence between the heading offset field and the heading correction components in the environmental subsequence. The heading correction components are calculated from the wind direction and ocean current data in the environmental subsequence. For example, headwinds or headwinds produce specific heading correction values. These correction values ​​are matched to the grid data in the heading offset field through the model to obtain the heading deviation component for each grid point, forming a first distribution function. This function mathematically describes the spatial variation of the heading deviation.

[0084] In the speed dimension, a speed matching calculation model is constructed based on the correlation between the speed adaptation field and the channel speed limit. The model introduces dynamic correction parameters for the ship's load factor and propulsion power. The ship's load factor is the ratio of the actual cargo capacity to the maximum cargo capacity, while the propulsion power is derived from the real-time monitoring data of the ship's engine. When the ship's load is heavy, the load factor increases, and the model will lower the calculated value of the adaptive speed accordingly. When the propulsion power is high, the model will appropriately increase the adaptive speed while complying with the channel speed limit. Through the dynamic adjustment of these two parameters, the calculated speed matching component is made more consistent with actual navigation conditions, ultimately forming the spatial distribution data of the speed matching component.

[0085] In the obstacle avoidance dimension, an iterative calculation process for obstacle avoidance response is established by combining the spatial rate of change of the obstacle avoidance response field with the ship's turning radius. The spatial rate of change reflects how quickly the values ​​in the obstacle avoidance response field change with position. A greater rate of change indicates a more dramatic change in response sensitivity when approaching an obstacle. The ship's turning radius is determined by the ship's length, draft, and current speed. The smaller the turning radius, the greater the ship's obstacle avoidance flexibility. During the iterative calculation, the position increment is first determined based on the spatial rate of change of the current position. The obstacle avoidance response increment is then calculated based on the turning radius. This increment is fed back into the next calculation, and the result is gradually adjusted until the difference between two consecutive calculations is less than the set threshold, resulting in stable spatial distribution data for the obstacle avoidance response component.

[0086] The output results of the first distribution function, speed matching calculation model, and obstacle avoidance response iterative calculation process obtained above are subjected to spatial interpolation fusion processing. Spatial interpolation methods are used to process the discrete grid data of each component, filling the blank areas between grids and making the data continuously distributed in space. During the fusion process, the consistency of the data of the three components in spatial coordinates must be ensured. For example, the heading deviation component, speed matching component, and obstacle avoidance response component at the same grid point must correspond to the same geographic location. Ultimately, a spatial distribution matrix containing these three components is generated, which is the route topology result for the current time window.

[0087] The route topology results for multiple consecutive time windows are cumulatively superimposed to calculate three indicators of route distribution characteristics. The heading deviation gradient is calculated by performing a differential operation on the heading deviation components of consecutive time windows along the planned route direction (usually the main direction from the starting point to the end point). The value with the largest rate of change is found, which is the heading deviation gradient. The speed matching cumulative value is calculated by integrating the speed matching component along the normal direction of the channel centerline (perpendicular to the main route direction). The integral interval covers the effective width of the channel. The resulting integral value is the speed matching cumulative value, which reflects the cumulative effect of speed matching in the lateral space. The obstacle avoidance response fluctuation coefficient is calculated by selecting a predetermined time interval (which can be equal to the route planning period or an integer multiple thereof). The standard deviation and average value of the obstacle avoidance response components within this interval are calculated. The ratio of the two is the obstacle avoidance response fluctuation coefficient. A larger standard deviation and a smaller average value indicate a less stable obstacle avoidance response and a higher fluctuation coefficient. Through this series of processes, the heading deviation gradient, speed matching cumulative value, and obstacle avoidance response fluctuation coefficient of the target ship are ultimately obtained, forming a complete route distribution characteristic.

[0088] Example 2:

[0089] When calling the pre-trained path optimization model to perform nonlinear path analysis on the route distribution characteristics and generate the optimal route prediction value and key risk area identification of the target ship, the heading deviation gradient is first input into the first feature analysis layer of the path optimization model. The first feature analysis layer has a built-in heading deviation factor calculation module, which defines the area where heading deviation may occur by analyzing the numerical distribution and spatial change trend of the heading deviation gradient. Specifically, the module will set multiple gradient thresholds. When the heading deviation gradient exceeds a certain threshold, the corresponding area will be marked as a potential heading deviation area, and the specific distribution coordinates of these areas will be determined through the coordinate system. At the same time, according to the change amplitude of the heading deviation gradient at different positions, a heading adjustment amplitude change curve is generated. The horizontal axis of the curve is the route position coordinate, and the vertical axis is the required heading adjustment angle, which intuitively reflects the change of the amplitude of the ship's heading adjustment during navigation with the position.

[0090] After the speed matching cumulative amount is input into the second feature analysis layer of the path optimization model, this layer will perform the speed adaptation damage accumulation calculation. This calculation process is based on the change trajectory of the speed matching cumulative amount over time, combined with the power system loss characteristics of the ship at different speeds, to analyze the impact of the speed matching status on the navigation progress. By comparing the navigation delay records corresponding to similar speed matching cumulative amounts in historical data, the navigation delay probability of the current speed matching segment is generated. This probability value reflects the possibility of delay under the current speed configuration. At the same time, based on the rate of change of the speed matching cumulative amount, the speed adjustment rate required to reduce the delay probability is predicted, that is, the adjustment rate prediction value, which provides a quantitative reference for subsequent speed regulation.

[0091] After the obstacle avoidance response fluctuation coefficient is input into the third feature analysis layer of the path optimization model, this layer performs calculations based on the channel obstacle avoidance degradation model. The channel obstacle avoidance degradation model simulates how obstacle avoidance response capability changes with the obstacle avoidance response fluctuation coefficient. When the obstacle avoidance response fluctuation coefficient is large, the model determines that the obstacle avoidance response capability has decreased. This model calculates the number of turns required on the obstacle avoidance response surface under the current obstacle avoidance response fluctuation coefficient to ensure effective obstacle avoidance. The model also analyzes the impact of these turns on channel efficiency, generating data on the evolution of channel efficiency over time. This data reflects the trend of efficiency changes caused by obstacle avoidance operations.

[0092] After obtaining the course adjustment curve, the probability of navigation delay, and the number of ship turns, these three factors must be fused to generate a comprehensive path index. During this fusion process, different weights are assigned to each factor based on their impact on route optimization. For example, in narrow waterways, the number of ship turns related to obstacle avoidance may be given a higher weight, while in open waters, the course adjustment range may be given a higher weight. The weighted values ​​are added together to form a comprehensive path index, which is then compared with the preset route threshold. The preset route threshold is set based on factors such as the route's safety standards and efficiency requirements. When the comprehensive path index exceeds this threshold, the corresponding route is the optimal route prediction value.

[0093] The geometric locations of critical risk areas can be identified by spatially overlaying the coordinates of the deviation areas, predicted speed adjustment rates, and the evolution of waterway efficiency data. This spatial overlay maps these data into the same spatial coordinate system. By analyzing the characteristics of the overlapping areas, the specific boundaries of the deviation areas, the paths along which delays may spread (i.e., delay expansion paths), and areas where obstacle avoidance is particularly challenging (i.e., high-risk obstacle avoidance areas) are determined. The locations and extents of these areas are then geometrically annotated on electronic nautical charts.

[0094] When generating a route maintenance strategy set based on the identification of critical risk areas, the optimal course adjustment path must be calculated for the identified deviation areas. This path is calculated by adjusting the course allocation ratios of adjacent segments. For example, in segments close to the deviation area, the course adjustment ratio is appropriately increased to allow the ship to adjust its course in advance to avoid the deviation area. In segments farther away from the area, the course adjustment ratio can be kept smaller to reduce unnecessary operations.

[0095] In view of the identification of the delay extension path, when constructing a dynamic speed control solution, it is necessary to select a suitable acceleration range based on the length and distribution of the delay extension path, and optimize the configuration of propulsion power parameters to shorten the navigation time and offset possible delays while ensuring safety.

[0096] When generating a detour strategy based on high-risk obstacle avoidance area identification, the detour distance and steering angle are dynamically adjusted based on the predicted obstacle avoidance response rate. A high predicted obstacle avoidance response rate indicates greater agility in obstacle avoidance, allowing for a smaller detour distance and steering angle. A low predicted value requires a larger detour distance and steering angle to ensure safe obstacle avoidance.

[0097] Prioritize the optimal course adjustment paths, dynamic speed control solutions, and route detour strategies. This ranking is based on the urgency of the risk, the difficulty of implementing the strategy, and its impact on navigation efficiency. For example, detours around high-risk obstacle avoidance areas are typically prioritized to ensure navigation safety. Based on the ranking results, the execution order and specific implementation parameters of each strategy are determined to form a complete set of route maintenance strategies.

[0098] Example 3:

[0099] When applying navigation environment compensation to the optimal route prediction to generate a revised optimal route prediction, the team first extracts extreme weather values ​​and the frequency of weather changes from the weather change sequence. Extreme weather values ​​include meteorological parameters that exceed the normal range, such as maximum wind speed, maximum precipitation, and minimum visibility. The frequency of weather changes is determined by counting the number of significant changes in meteorological parameters per unit time, such as the number of wind speed changes exceeding 5m / s per hour. Based on these parameters and in conjunction with the ship's maneuverability manual, the dynamic adjustment of the ship's maneuverability sensitivity as it changes with weather conditions is calculated. This adjustment reflects the degree of change in the ship's maneuverability under different weather conditions. For example, as wind speed increases, the ship's steering sensitivity decreases accordingly, resulting in a negative dynamic adjustment.

[0100] When calculating weather compensation for the heading deviation gradient based on the dynamic adjustment, the initial maneuvering sensitivity of the target vessel under baseline weather conditions and the aforementioned dynamic adjustment are first obtained to establish a maneuvering sensitivity-weather correlation function. The baseline weather conditions are defined as wind speeds less than 3 m / s, no precipitation, and visibility greater than 10 nautical miles. The initial maneuvering sensitivity is the inherent steering response parameter of the vessel under these conditions, such as the ratio of the steering angle to the rudder angle. The correlation function combines the initial maneuvering sensitivity with the dynamic adjustment to describe the maneuvering sensitivity under different weather parameter combinations. For example, when the wind speed is 10 m / s, the correlation function outputs a maneuvering sensitivity that is 0.7 times the initial value.

[0101] The heading correction increment is calculated based on the control sensitivity-weather correlation function. The calculation formula is:

[0102]

[0103] in, Indicates the heading correction increment, Indicates the meteorological change (the difference between the actual meteorological parameters and the reference meteorological parameters). Indicates the change in control sensitivity (the difference between the actual control sensitivity calculated by the control sensitivity-weather correlation function and the initial control sensitivity).

[0104] The heading correction increment is added to the calculation of the heading deviation gradient to generate a heading deviation gradient correction value that incorporates weather effects. For example, if the original heading deviation gradient at a certain location is 0.5 degrees / km and the heading correction increment is 0.2 degrees / km, the corrected heading deviation gradient is 0.7 degrees / km. The heading deviation gradient correction value is then compensated for maneuvering delay effects using a multiplication factor based on the ship's steering delay curve and the weather holdover time. The ship's steering delay curve, derived from field measurements, describes the time difference between rudder angle adjustment and actual steering. For example, a 10-degree rudder angle adjustment results in a 2-second delay. The weather holdover time is the duration of the current weather conditions. For example, if the current wind speed is 10 m / s for 30 minutes, the multiplication factor is 2 seconds x 30 minutes. This factor is multiplied by the heading deviation gradient correction value to obtain the final delay compensation value, which is then added to the correction value to complete the compensation.

[0105] Based on the correlation between the frequency of weather changes and the ship's roll characteristics, the speed matching cumulative value is corrected for roll effects. The ship's roll characteristics include roll period and roll angle. Analysis of historical data shows that higher weather change frequency leads to larger roll angles and shorter periods. Based on this correlation, a roll influence coefficient table is established. For example, if the weather change frequency is three times per hour, the roll influence coefficient is 1.2. This coefficient is multiplied by the original speed matching cumulative value to obtain a corrected speed matching cumulative value, eliminating the interference of roll on the speed matching calculation.

[0106] Based on data on ship braking distance changes under extreme weather conditions, the obstacle avoidance response fluctuation coefficient is adaptively adjusted to optimize maneuverability. This data was obtained through simulation experiments. For example, in a wind speed of 15m / s, the distance a ship needs to brake from a speed of 10 knots to a standstill increases by 200 meters compared to baseline weather conditions. Based on this data, an adjustment relationship between the braking distance change rate and the obstacle avoidance response fluctuation coefficient is established. When the braking distance increases by 20%, the obstacle avoidance response fluctuation coefficient is adjusted upward by 15%, generating a revised obstacle avoidance response fluctuation coefficient.

[0107] The corrected heading deviation gradient, speed matching accumulation, and obstacle avoidance response fluctuation coefficient are input into the path optimization model for recalculation, generating an optimal route prediction after compensating for environmental factors. The path optimization model then performs a nonlinear analysis on these corrected characteristic parameters, adjusting parameters such as the turning point location and speed configuration. For example, in areas with high corrected obstacle avoidance response fluctuation coefficients, the model increases the safe distance between the route and obstacles. The resulting optimal route prediction is more accurately aligned with the impact of actual weather conditions on the ship's navigation.

[0108] Example 4:

[0109] When constructing a dynamic speed control solution, the geometric features of the delay extension path are first extracted, and the path curvature radius and extension direction angle are calculated. The delay extension path is a route where delays may spread, predicted based on historical navigation data and the cumulative amount of current speed matching. Its geometric features are obtained by fitting the coordinate points on the electronic nautical chart. For example, in a certain offshore channel, the delay extension path presents an arc curve. By mathematically fitting this curve, the curvature radius of this curve can be calculated. If the curve equation obtained after fitting is a circular arc with a radius of 5 nautical miles, then the path curvature radius is 5 nautical miles. The extension direction angle is obtained by determining the angle between the tangent direction at the starting point of the path and the north direction. If the tangent direction points to the northeast and forms a 30-degree angle with the north direction, then the extension direction angle is 30 degrees.

[0110] The coverage density of the acceleration interval is selected according to the curvature radius, and the coverage density is inversely proportional to the curvature radius. When the curvature radius is small, it means that the delay extension path is more curved, and more dense acceleration intervals are required to cope with complex path changes. For example, if the curvature radius is 2 nautical miles, which is a small radius, the coverage density of the acceleration interval can be set to one acceleration point every 0.5 nautical miles; when the curvature radius is 10 nautical miles and the path is relatively flat, the coverage density of the acceleration interval can be adjusted to one acceleration point every 2 nautical miles to reduce unnecessary acceleration operations. The demarcation of the acceleration interval must avoid restricted areas such as shoals in the waterway and no-fly zones to ensure the safety of acceleration operations.

[0111] The direction of propulsion power application is adjusted based on the expansion angle so that the propulsion direction forms a predetermined angle with the delay expansion direction. The setting of this predetermined angle must take into account the ship's dynamic characteristics and the direction of the water flow. For example, if the aforementioned expansion angle is 30 degrees and the water flow is due east, the propulsion direction can be adjusted to form a 45-degree angle with the delay expansion direction, i.e., a propulsion direction of 75 degrees (30 degrees + 45 degrees). This angle setting allows the ship to gradually deviate from the delay expansion path while accelerating, reducing the possibility of delay accumulation. The propulsion direction is adjusted using the ship's power system's vector propulsion device, ensuring precise control of the power application direction.

[0112] The propulsion power duration is dynamically adjusted based on the ship's load test data to ensure that the propulsion power remains below the critical value of the ship's rated power. The ship's load test data contains the corresponding relationship between power output and duration under different load conditions. For example, at a load of 5,000 tons, the maximum duration of continuous output of 80% of rated power is 30 minutes; at a load of 8,000 tons, the maximum duration of the same power output is 20 minutes. Based on the actual load of the current ship (for example, 6,000 tons), the corresponding power duration (for example, 25 minutes) is interpolated from the test data. This is used as the basis for setting the propulsion power duration to avoid engine overload due to excessive power or excessive duration.

[0113] Generate a control parameter configuration table that includes coverage density, propulsion direction, and power duration. The configuration table arranges the positions of each acceleration zone in the order of the route, and each position corresponds to a row of parameters, including the coverage density of the interval (such as one acceleration point every 0.5 nautical miles), the angle of the propulsion direction (such as 75 degrees), the percentage of propulsion power (such as 80% of rated power), and the duration (such as 25 minutes). This configuration table must be linked with the ship's navigation system. When the ship enters the preset acceleration zone, the corresponding control parameters are automatically called to achieve dynamic speed control. At the same time, the configuration table must also include an emergency adjustment mechanism. For example, when encountering sudden obstacles, the power duration is automatically shortened or the power output is reduced to ensure navigation safety.

[0114] Example 5:

[0115] The target vessel's actual navigation deviation and extended delay are collected within a pre-set verification period. The preset verification period should be determined based on the actual route length and voyage time. For example, for a short voyage between two ports, with a one-way trip time of approximately six hours, the verification period can be set to two hours, meaning relevant data is collected every two hours. For long-distance transoceanic routes, with one-way trip times exceeding 72 hours, the verification period can be extended to six hours to balance data collection frequency and processing efficiency. The actual navigation deviation is obtained using fused data from the vessel's GPS and inertial navigation system. Specifically, it is the straight-line distance between the vessel's real-time position coordinates and the planned route coordinates, expressed in nautical miles. For example, at a given moment, the deviation between the vessel's actual position and the planned route is 0.3 nautical miles. The extended delay is calculated by recording the difference between the vessel's actual arrival time at the preset waypoint and the planned arrival time, combined with the average speed for that leg. For example, if a leg is scheduled to arrive at 10:00 a.m. and arrives at 10:15 a.m., with an average speed of 12 knots, the extended delay is 12 knots x 0.25 hours = 3 nautical miles.

[0116] The actual navigation deviation and the predicted heading deviation gradient are subjected to deviation analysis processing to generate the first error correction coefficient. The deviation analysis processing first calculates the difference between the actual navigation deviation and the predicted heading deviation gradient at the same time point to obtain a series of deviation values. For example, if the actual deviation is 0.3 nautical miles, the theoretical deviation calculated corresponding to the predicted heading deviation gradient is 0.25 nautical miles, and the deviation value is 0.05 nautical miles. These deviation values ​​are statistically analyzed, and the sliding window method is used to calculate the average and standard deviation of the deviation values ​​in multiple consecutive verification cycles. The deviation model is constructed based on these statistics. The first error correction coefficient is determined based on the output results of the deviation model. When the deviation value is positive as a whole, it means that the predicted heading deviation gradient is too small, and the first error correction coefficient is greater than 1 to increase the predicted value; when the deviation value is negative as a whole, the correction coefficient is less than 1 to reduce the predicted value.

[0117] The delay extension length and the predicted adjustment rate are compared in the time domain to generate a second error correction coefficient. This time domain comparison requires aligning the delay extension length and the predicted adjustment rate on the same time axis and analyzing their correlation in the time series. For example, during a certain verification cycle, the delay extension length was 3 nautical miles, while the theoretical delay length calculated based on the predicted adjustment rate was 2.5 nautical miles, with a difference of 0.5 nautical miles. By cumulatively analyzing the differences over multiple verification cycles, an error transfer function is established between the delay extension length and the predicted adjustment rate. This function describes the deviation pattern between the predicted adjustment rate and the actual delay. The second error correction coefficient is determined based on the output of this function. If the predicted adjustment rate is generally lower than the actual required value, resulting in a larger delay extension length, the second error correction coefficient will be greater than 1 to increase the predicted adjustment rate; otherwise, the second error correction coefficient will be less than 1.

[0118] The weight parameters of the path optimization model are adjusted based on the first and second error correction coefficients to generate an optimized path optimization model. The weight parameters of the path optimization model include the heading feature weight, the speed feature weight, and the obstacle avoidance feature weight, which respectively correspond to the degree of influence of the heading deviation gradient, the speed matching accumulation, and the obstacle avoidance response fluctuation coefficient in the route distribution characteristics on the model calculation. For example, when the first error correction coefficient is 1.2, it means that the prediction error of the heading-related features is large, and the heading feature weight needs to be increased from the original 0.3 to 0.36; when the second error correction coefficient is 0.9, it means that the prediction of the speed-related features is overestimated, and the speed feature weight needs to be reduced from the original 0.4 to 0.36. The adjustment of the weight parameters is achieved through the gradient descent method to ensure that the adjusted model has a better fitting effect on the training data set.

[0119] The optimized path optimization model is applied to subsequent batches of ship route optimization tasks. Subsequent batches of route optimization tasks include repeated voyages of the same ship on the same route, or route planning of different ships in similar waters. During the application process, it is necessary to record the deviation between the predicted results of the optimized model and the actual navigation data, and continuously accumulate new verification data to provide a basis for the next model optimization. For example, after the optimized model is applied to a certain route, the deviation between the newly collected actual navigation deviation and the predicted value is lower than before, indicating that the model optimization is effective and the model can continue to be used; if the deviation increases, it is necessary to re-examine the deviation analysis process, adjust the calculation method of the error correction coefficient, and optimize the model again. Through this cyclic optimization mechanism, the path optimization model can continuously adapt to different navigation environments and ship conditions to maintain the accuracy of the prediction.

[0120] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other inventors can develop based on the same general inventive concepts embodied by the described embodiments. That is, although the present application is described in terms of particular embodiments and illustrative figures, it should be apparent that the scope of the present application is not limited to these specific embodiments.

[0121] While the embodiments of the application have been shown and described herein, it is to be understood that the scope of the application, jointly pointed out in the appended claims, is not to be limited to the above-described embodiments but can be otherwise variously changed, modified, replaced, and altered within the principles and spirit of the present application.

Claims

1. A method for optimizing a ship's navigation route, characterized in that: The method comprises: Collecting a dynamic environmental data set of the target ship in the navigation environment, wherein the dynamic environmental data set includes a weather change sequence, an ocean current fluctuation sequence, and a channel obstacle distribution data; Performing a two-dimensional route topology construction process on the dynamic environment data set to generate a route distribution feature of the target ship, the route distribution feature including a heading deviation gradient, a speed matching cumulative amount, and an obstacle avoidance response fluctuation coefficient; Calling a pre-trained path optimization model to perform nonlinear path analysis on the route distribution characteristics to generate an optimal route prediction value and a key risk area identifier for the target ship; performing navigation environment compensation correction processing on the optimal route prediction value to generate a corrected optimal route prediction value, wherein the navigation environment compensation correction processing is implemented based on the correlation relationship between the meteorological change sequence and the ship control sensitivity; Generate a route maintenance strategy set based on the key risk area identifier, wherein the route maintenance strategy set includes a course adjustment path optimization solution and a speed dynamic control processing solution; The performing a two-dimensional route topology construction process on the dynamic environment data set to generate the route distribution characteristics of the target ship includes: Dividing the weather change sequence into multiple environmental subsequences according to time windows, each environmental subsequence corresponding to a route planning cycle; For each of the environment subsequences, the following processing is performed: Constructing a two-dimensional navigation topology structure of the target ship according to the channel obstacle distribution data, wherein the two-dimensional navigation topology structure includes spatial distribution data of a heading deviation field, a speed adaptation field, and an obstacle avoidance response field; Couple the two-dimensional navigation topology structure with the environmental subsequence for analysis and processing to generate a route topology result for the current time window, wherein the route topology result includes a spatial distribution matrix of a heading deviation component, a speed matching component, and an obstacle avoidance response component; The route topology results of multiple consecutive time windows are cumulatively superimposed to calculate the heading deviation gradient, speed matching accumulation and obstacle avoidance response fluctuation coefficient; wherein, The heading deviation gradient is the maximum rate of change of the heading deviation component along the route planning direction. The speed matching cumulative amount is the integral of the speed matching component in the normal direction of the channel centerline. The obstacle avoidance response fluctuation coefficient is the ratio of the standard deviation to the average value of the obstacle avoidance response component within a predetermined time interval.

2. The ship navigation route optimization method according to claim 1, characterized in that: The coupling analysis and processing of the two-dimensional navigation topology structure and the environment subsequence to generate a route topology result of the current time window includes: Establishing a heading offset-correction mapping equation based on a correspondence between the heading offset field and the heading correction component in the environment subsequence, and obtaining a first distribution function of the heading deviation component by solving the heading offset-correction mapping equation; According to the correlation characteristics between the speed adaptation field and the waterway speed limit, a speed matching calculation model is constructed, wherein the speed matching calculation model includes dynamic correction parameters of the ship load coefficient and propulsion power; Combining the spatial change rate of the obstacle avoidance response field and the ship turning radius, an obstacle avoidance response iterative calculation process is established, wherein the iterative calculation process includes a feedback correction mechanism for position increment and obstacle avoidance response increment; The output results of the first distribution function, the speed matching calculation model and the obstacle avoidance response iterative calculation process are subjected to spatial interpolation fusion processing to generate two-dimensional route topology distribution data including heading, speed and obstacle avoidance response components.

3. The ship navigation route optimization method according to claim 1, characterized in that: The calling of the pre-trained path optimization model to perform nonlinear path analysis on the route distribution characteristics to generate the optimal route prediction value and key risk area identification of the target ship includes: Inputting the heading deviation gradient into the first feature analysis layer of the path optimization model, and determining the distribution coordinates of the heading deviation area and the heading adjustment amplitude change curve through the heading deviation factor calculation module; Inputting the speed matching cumulative amount into the second feature analysis layer of the path optimization model, performing speed adaptation damage cumulative calculation, and generating a navigation delay probability and adjustment rate prediction value for the speed matching section; Inputting the obstacle avoidance response fluctuation coefficient into the third characteristic analysis layer of the path optimization model, and calculating the number of ship turns and channel traffic efficiency evolution data of the obstacle avoidance response surface based on the channel obstacle avoidance degradation model; fusing the course adjustment amplitude change curve, the probability of navigation delay, and the number of ship turns to generate a comprehensive path index for the target ship, and determining an optimal route prediction value based on a comparison result of the comprehensive path index with a preset route threshold; Based on the spatial superposition results of the distribution coordinates, the adjustment rate prediction value and the waterway traffic efficiency evolution data, the geometric positions of the heading deviation area, the delay extension path and the obstacle avoidance high-risk area are identified.

4. The ship navigation route optimization method according to claim 1, characterized in that: The performing navigation environment compensation correction processing on the optimal route prediction value to generate a corrected optimal route prediction value includes: Extracting extreme weather values ​​and weather change frequencies from the weather change sequence, and calculating the dynamic adjustment amount of the ship's control sensitivity as the weather changes; Performing meteorological influence compensation calculation on the heading deviation gradient according to the dynamic adjustment amount to generate a corrected heading deviation gradient; Based on the correlation between the weather change frequency and the ship's rolling characteristics, performing roll effect correction processing on the speed matching cumulative amount to generate a corrected speed matching cumulative amount; According to the ship braking distance change data under extreme weather conditions, the obstacle avoidance response fluctuation coefficient is adaptively adjusted to adjust the control capability to generate a corrected obstacle avoidance response fluctuation coefficient; The corrected heading deviation gradient, speed matching accumulation and obstacle avoidance response fluctuation coefficient are input into the path optimization model for recalculation to generate an optimal route prediction value after compensating for environmental factors.

5. The ship navigation route optimization method according to claim 4, characterized in that: The performing weather influence compensation calculation on the heading deviation gradient according to the dynamic adjustment amount to generate a corrected heading deviation gradient includes: Obtaining the initial maneuvering sensitivity of the target ship under reference weather conditions and the dynamic adjustment amount, and establishing a maneuvering sensitivity-weather correlation function; Calculating a heading correction increment based on the control sensitivity-weather correlation function, wherein the heading correction increment is the product of the weather change and the control sensitivity change; Adding the heading correction increment to the calculation process of the heading deviation gradient to generate a heading deviation gradient correction value including meteorological influence; A control delay effect compensation process is performed on the heading deviation gradient correction value, and the control delay effect compensation process is implemented based on a multiplication factor of a ship steering delay curve and a weather holding time.

6. The ship navigation route optimization method according to claim 3, characterized in that: Generating a route maintenance strategy set according to the key risk area identifier includes: Calculating an optimal heading adjustment path based on the identifier of the heading deviation area, wherein the optimal heading adjustment path is achieved by adjusting the heading allocation ratio of adjacent flight segments; Constructing a dynamic speed control solution based on the identifier of the delay extension path, wherein the dynamic speed control solution includes selecting an acceleration range and optimizing the configuration of propulsion power parameters; Based on the identification of the high-risk obstacle avoidance area, a channel detour strategy is generated, wherein the channel detour strategy dynamically adjusts the detour distance and steering angle according to the predicted value of the obstacle avoidance response rate; The optimal heading adjustment path, the dynamic speed control processing scheme and the channel detour strategy are prioritized to generate a maintenance strategy set including execution sequence and implementation parameters.

7. The ship navigation route optimization method according to claim 6, characterized in that: The construction of the dynamic speed control processing solution includes: Extracting geometric features of the delay extension path, and calculating the path curvature radius and extension direction angle; Selecting a coverage density of the acceleration interval according to the curvature radius, wherein the coverage density is inversely proportional to the curvature radius; adjusting the direction of application of propulsion power based on the extension direction angle so that the propulsion direction forms a predetermined angle with the delay extension direction; Dynamically adjust the propulsion power duration based on the ship's load test data to ensure that the propulsion power is below the critical value of the ship's rated power; Generate a control parameter configuration table including coverage density, propulsion direction and power duration.

8. The method for optimizing a ship's navigation route according to claim 1, wherein: The method further comprises: Collecting the actual navigation deviation and delay extension length of the target ship within a preset verification period; Performing deviation analysis on the actual navigation deviation and the predicted heading deviation gradient to generate a first error correction coefficient; Performing a time domain comparison process on the delay extension length and the predicted adjustment rate to generate a second error correction coefficient; adjusting the weight parameters of the path optimization model according to the first error correction coefficient and the second error correction coefficient to generate an optimized path optimization model; The optimized path optimization model is applied to subsequent batches of ship route optimization tasks.

9. A ship navigation route optimization system, characterized in that: It includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the ship navigation route optimization method described in any one of claims 1 to 8.

Images