A ship route planning method considering ship-wave coupling constraints

Abstract

The application relates to a ship route planning method considering ship-wave coupling constraints, which comprises the following steps: S1. constructing a ship-wave coupling data model, collecting ship basic parameters, navigation parameters and wave parameters, and training and evaluating the model after preprocessing by using a CNN and RNN hybrid architecture; S2. collecting and fusing multi-source real-time marine environment data; S3. determining and dynamically adjusting ship-wave coupling constraint conditions and safety thresholds based on the ship-wave coupling data model and real-time marine environment data; S4. constructing a multi-objective optimization function based on the ship-wave coupling constraint conditions, and performing route planning by using a genetic algorithm; and S5. real-time monitoring of ship navigation state and marine environment data, and re-planning of the route when the safety threshold is exceeded, to form a dynamic closed-loop adjustment. The application has the effects of improving navigation safety, optimizing navigation efficiency and economy, and real-time dynamic self-adaptation.

Description

Technical Field

[0001] This invention relates to the field of ship navigation technology, and in particular to a ship route planning method that takes into account ship-wave coupling constraints. Background Technology

[0002] In the context of a globalized economy, shipping is a core link in international trade, carrying approximately 90% of global cargo volume. Large ships, with their advantages of large carrying capacity and low transportation costs, have become the main force in maritime cargo transportation. However, the marine environment is complex and changeable, and waves are a key factor affecting the safety of ship navigation. High sea states and waves can easily cause cargo shifting, container loss, and even ship capsizing and structural damage, resulting not only in huge economic losses and threatening the lives of crew members, but also in damaging the marine ecosystem. Therefore, designing ship route planning schemes that take safety into account is an urgent need in the shipping industry.

[0003] Currently, ship route planning technologies are mostly optimized around navigation efficiency and economy. Some schemes incorporate conventional environmental factors such as wind direction and ocean currents, while others attempt to combine historical meteorological data and empirical models to predict wave conditions and incorporate them into planning considerations. Related technologies have formed a certain research foundation.

[0004] However, existing technologies still have significant shortcomings, all of which are problems that this invention can effectively solve: First, they do not adequately consider the ship-wave coupling effect, treating waves only as static interference factors and ignoring the dynamic coupling relationship between ships and waves, which involves mutual influence and constraint. They also fail to analyze the dynamic changes in the coupling effect by combining ship speed, heading, and wave characteristic parameters. Second, wave prediction has limitations. Relying on historical data and empirical models cannot reflect the dynamic changes in the marine environment in real time and accurately, and the models are poorly adaptable to complex high sea states. Third, the technical design lacks systematicity. Existing solutions are mostly optimized for single objectives and do not build a planning system that takes into account navigation safety, efficiency, and economy around ship-wave coupling constraints, making it difficult to meet the actual navigation safety requirements of ships under complex high sea states.

[0005] Therefore, developing a ship route planning method that considers ship-wave coupling constraints is of great practical significance for overcoming the shortcomings of existing technologies and improving the safety of ship navigation in complex sea conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide a ship route planning method that considers ship-wave coupling constraints, which has the effects of improving navigation safety, optimizing navigation efficiency and economy, and real-time dynamic self-adaptation.

[0007] The above-mentioned objective of this invention is achieved through the following technical solutions: A ship route planning method considering ship-wave coupling constraints includes the following steps: S1. Construct a ship-wave coupled data model, collect basic ship parameters, navigation parameters and wave parameters, and train and evaluate the model using a hybrid CNN and RNN architecture after preprocessing. S2. Collect and fuse multi-source real-time marine environmental data; S3. Based on the ship-wave coupling data model and real-time marine environment data, determine and dynamically adjust the ship-wave coupling constraints and safety thresholds; S4. Based on the ship-wave coupling constraints, a multi-objective optimization function is constructed, and a genetic algorithm is used for route planning; S5. Monitor ship navigation status and marine environment data in real time, and replan the route when the safety threshold is exceeded to form a dynamic closed-loop adjustment.

[0008] Through the above technical solutions, a complete process of ship-wave coupled route planning is established, realizing the full-dimensional application of coupling constraints, making up for the shortcomings of existing solutions in not considering ship-wave coupling, and improving the safety, real-time adaptability and comprehensive optimization of route planning.

[0009] As a further technical solution of the present invention, step S1 specifically includes the following steps: S11. Collect basic ship parameters, navigation parameters and wave parameters. The basic ship parameters include ship type, displacement and draft. The navigation parameters include speed and heading. The wave parameters include wave height, wavelength, wave period and wave direction. Normalize the data. S12. A hybrid architecture of convolutional neural network (CNN) and recurrent neural network (RNN) is used to build the model. Spatial features are extracted by CNN and temporal features are processed by RNN. S13. The model is trained using the mean squared error loss function, and the model parameters are optimized using the adaptive moment estimation (Adam) algorithm; S14. The root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²) are used as evaluation metrics to validate and evaluate the trained model.

[0010] Through the above technical solutions, a ship-wave coupling data model is constructed in a refined manner, and the spatial and temporal features of the data are accurately extracted to ensure high model accuracy. This model can accurately reflect the ship-wave coupling characteristics of different ship types under various sea conditions, providing reliable model support for subsequent constraint determination and route planning.

[0011] As a further technical solution of the present invention, step S2 specifically includes the following steps: S21. Collect marine environmental data of ship navigation areas from multiple sources, including satellite remote sensing, marine monitoring buoys, and meteorological forecasting models; S22. Perform radiometric calibration and Lee filtering on satellite remote sensing data, and obtain wave parameters by inversion using the TVG model; S23. Kriging interpolation is used to fill in missing values ​​in marine monitoring buoy data, and Kalman filtering algorithm is used to fuse satellite remote sensing data with marine monitoring buoy data to obtain real-time observation data; S24. The meteorological prediction model data and the real-time observation data are fused using the Bayesian fusion method to obtain the final marine environmental data.

[0012] The above technical solutions enable the accurate acquisition and fusion of multi-source marine environmental data, solving problems such as noise, missing values, and large errors that exist in single data sources. They also enable multi-dimensional complementarity and accurate fusion of marine environmental data, providing real-time, accurate, and comprehensive marine environmental data support for determining ship-wave coupling constraints.

[0013] As a further technical solution of the present invention, step S3 specifically includes the following steps: S31. Based on the ship-wave coupled data model and real-time marine environmental data, establish the coupling relationship between ship motion state and wave parameters; S32. Based on ship dynamics and stability theory, derive and determine the safety thresholds for roll angle, pitch angle and acceleration; S33. Adjust the ship-wave coupling constraints in real time based on the dynamic changes in the ship's navigation status and the marine environment; S34. Verify the constraints through numerical simulation and actual ship experiments, and complete the optimization based on the verification results.

[0014] The above technical solutions clarify the process of determining and dynamically adjusting ship-wave coupling constraints, ensuring that the constraints conform to the actual navigation conditions of ships and adapt to the dynamic changes in the marine environment. Through numerical simulation and real ship experiments, the solutions are verified and optimized, thus defining a scientific and reasonable safety boundary for route planning.

[0015] As a further technical solution of the present invention: the method for determining the safety threshold in S3 is as follows: The roll angle safety threshold is derived from ship stability theory; The safe threshold for pitch angle is determined based on the principles of ship pitch dynamics. The acceleration safety threshold is determined based on the human body's acceleration limit, cargo securing safety standards, and a ship-wave coupling data model.

[0016] The above technical solutions allow for the differentiated derivation of safety thresholds for roll, pitch, and acceleration, taking into account ship engineering theory, human physiological limits, cargo safety standards, and ship-wave coupling data models, thus avoiding navigational safety hazards caused by unreasonable threshold settings.

[0017] As a further technical solution of the present invention: the roll angle safety threshold is: ;in For coefficients, The power exponent. For wave height, For speed, The direction of the wave. For the course, For drainage volume, This represents the initial stability height.

[0018] The above technical solution provides a quantitative formula for the safe threshold of roll angle, incorporating key influencing factors such as wave height, speed, and wave direction, to achieve dynamic and accurate calculation of the threshold, effectively limiting the ship's roll amplitude and reducing the probability of accidents such as lost containers and capsizing.

[0019] As a further technical solution of the present invention, step S4 specifically includes the following steps: S41. Construct a multi-objective optimization objective function that integrates navigation safety, navigation efficiency, and fuel economy; S42. Set constraints, including ship-wave coupling constraints, speed range constraints, and heading change angle constraints; S43. A genetic algorithm is used to perform a global route search. The route is encoded with real numbers. A fitness function is constructed based on the objective function. Roulette wheel selection, crossover, and mutation operations with set probabilities are performed in sequence. S44. Based on changes in real-time marine environmental data, reassess the ship-wave coupling constraints and update the objective function parameters, then rerun the genetic algorithm to dynamically adjust the route.

[0020] The above technical solutions clarify the route planning process of genetic algorithms, construct an objective function that integrates navigation safety, efficiency and fuel economy, adapts to dynamic changes in the marine environment, realizes dynamic route adjustment, and improves the practicality and adaptability of the algorithm.

[0021] As a further technical solution of the present invention: the multi-objective optimization objective function in S4 is: The ship-wave coupling constraint in S42 includes: roll angle Pitch angle acceleration The speed range constraint is as follows: The heading change angle constraint is that the heading change does not exceed an angle threshold. By using the above technical solutions, the multi-objective optimization function and various constraints are quantified, making the route planning objectives clear and the constraints specific, effectively balancing ship navigation safety and efficiency, and avoiding safety risks or benefit losses caused by the planned route exceeding the constraints.

[0022] As a further technical solution of the present invention, step S5 specifically includes the following steps: S51. Collect marine environmental data and ship navigation data in real time at fixed time intervals using meteorological sensors, wave sensors and satellite communication equipment; S52. Compare the real-time collected data with the ship-wave coupling constraints, and trigger an early warning when the parameters exceed the corresponding safety threshold; S53. After the warning is triggered, the route is replanned based on the genetic algorithm to construct a multi-objective optimization function that integrates navigation safety, navigation efficiency and fuel economy; S54. Dynamically adjust and provide closed-loop feedback for the replanned routes, continuously monitor data, and trigger route replanning again based on changes in the objective function.

[0023] The above technical solution clarifies the closed-loop process of dynamic route adjustment. Data is collected in real time during ship navigation and compared with constraints. After triggering an early warning when the threshold is exceeded, the route is replanned, forming a closed-loop feedback system. This ensures that the ship is always in a safe and optimal navigation state, greatly improving the ship's adaptability and robustness in complex and ever-changing marine environments.

[0024] In summary, the present invention has at least one of the following beneficial technical effects: 1. This invention discloses a ship route planning method that considers ship-wave coupling constraints. By building a complete ship-wave coupled route planning process, integrating multi-source data and dynamic constraints, and using genetic algorithm planning and closed-loop adjustment, it achieves synergistic optimization of ship navigation safety, real-time adaptability and comprehensive benefits.

[0025] 2. This invention discloses a ship route planning method that considers ship-wave coupling constraints. It achieves high accuracy and high reliability of the ship-wave coupling data model by using a hybrid architecture of CNN and RNN to refine modeling and accurately extract spatiotemporal features of data, thus providing solid support for constraint determination and route planning.

[0026] 3. This invention discloses a ship route planning method that considers ship-wave coupling constraints. It achieves strong adaptability and robustness of ship navigation under complex sea conditions through targeted processing and fusion of multi-source data, dynamic optimization of constraints, dynamic replanning of routes, and closed-loop feedback. Attached Figure Description

[0027] Figure 1This is a flowchart illustrating a ship route planning method that considers ship-wave coupling constraints according to the present invention.

[0028] Figure 2 for Figure 1 A flowchart illustrating the sub-steps of S1.

[0029] Figure 3 for Figure 1 A flowchart illustrating the sub-steps of S2.

[0030] Figure 4 for Figure 1 A flowchart illustrating the sub-steps of S3.

[0031] Figure 5 for Figure 1 A flowchart illustrating the sub-steps of S4.

[0032] Figure 6 for Figure 1 A flowchart illustrating the sub-steps of S4. Detailed Implementation

[0033] 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. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0034] In the description of this application, it should be noted that the terms "upper," "lower," "inner," "outer," "top / bottom," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0035] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed," "equipped with," "sleeved / connected," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Example

[0036] Reference Figure 1 This invention discloses a ship route planning method considering ship-wave coupling constraints. The core of this method lies in fully considering the coupling effect between the ship's motion state and wave parameters. Through multi-stage technical means, it achieves safe, efficient, and economical route planning and real-time dynamic adjustment, including the following steps: S1. Construct a ship-wave coupled data model, collect basic ship parameters, navigation parameters and wave parameters, and train and evaluate the model using a hybrid architecture of CNN and RNN after preprocessing.

[0037] Reference Figure 2 Step S1 specifically includes the following sub-steps: S11. Collect and preprocess ship and marine environment data Collect a large amount of navigation data of different types of ships under various sea conditions. This data covers basic ship parameters. Navigation parameters and wave parameters Among them, the basic parameters of the ship , Indicates ship type, For drainage volume, Draft, etc.; Navigation parameters , It's the speed. For heading; Wave parameters , It’s the wave height, For wavelength, It is a wave cycle. The direction of the wave.

[0038] The collected data underwent preprocessing, including data cleaning and normalization. For data with missing values, interpolation was used to impute them. Let... For the first The first sample For each feature, if a missing value exists, linear interpolation is performed based on the values ​​of the preceding and following samples. Outliers are detected and processed using statistical methods. Then, all data are normalized using the following formula: in, This is the original data. The mean of the data. denoted as the standard deviation of the data.

[0039] S12. Building a hybrid CNN and RNN architecture model A hybrid architecture combining recurrent neural networks (RNNs) and convolutional neural networks (CNNs) from deep learning is used to construct a ship-wave coupling data model. The CNN is responsible for extracting the spatial features of the data, while the RNN is responsible for processing the temporal features. Together, they achieve accurate modeling of the ship-wave coupling relationship.

[0040] For the CNN part, the input layer receives preprocessed data. The data undergoes multiple convolutional layers, pooling layers, and activation layers to extract spatial features. The formula for calculating a convolutional layer is: in, For the first The first convolutional layer outputs the feature map of the 1st convolutional layer. One element, For the first The output of the layer, For the first The convolution kernel of the layer, For bias terms, This is the activation function.

[0041] The spatial feature data extracted by the CNN is directly input into the RNN part for deep mining of temporal features. In the RNN, the hidden state... The update formula is: in, , These are the weight matrices from hidden state to hidden state and from input to hidden state, respectively. For bias terms, For a moment Input. S13. Train and optimize the ship-wave coupling data model. The model is trained using the collected data, employing the mean squared error loss function. : in, For the true value, These are the model's predicted values. This represents the number of samples.

[0042] To optimize the model parameters, an improved algorithm of stochastic gradient descent (SGD), namely the Adaptive Moment Estimation (Adam) algorithm, is employed. The parameters of this algorithm... The updated formula is: in, For a moment gradient, These are the first-order moment estimate and the second-order moment estimate, respectively. The exponential decay rate, For learning rate, To prevent division by zero by a small constant.

[0043] During training, cross-validation is used to divide the dataset into training, validation, and test sets. Hyperparameters of the model, such as kernel size, number of hidden layers, and learning rate, are continuously adjusted to improve the model's performance and generalization ability.

[0044] S14. Verify and evaluate model performance. The trained model was validated and evaluated using a test set, employing root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination. The accuracy and reliability of the model are measured by indicators such as ( ). in, This represents the mean of the true values. By continuously optimizing the model structure and parameters, the RMSE and MAE of the model can be minimized. Get as close to 1 as possible.

[0045] S2. Collect and fuse multi-source real-time marine environmental data.

[0046] Reference Figure 3 Step S2 specifically includes the following sub-steps: S21. Acquire multi-source marine environmental data To acquire real-time marine environmental data, particularly wave-related parameters, within a ship's navigation area, various methods are employed, including satellite remote sensing, ocean monitoring buoys, and meteorological forecasting models. Satellite remote sensing uses sensors onboard satellites, such as synthetic aperture radar (SAR) and altimeters, to obtain information about the ocean surface over a wide area. Ocean monitoring buoys are monitoring devices deployed in specific sea areas that can directly measure marine environmental parameters. Meteorological forecasting models predict future weather and marine environmental conditions based on meteorological observation data and physical equations.

[0047] S22, Processing satellite remote sensing data Satellite remote sensing data typically involves large amounts of data and complex noise. Taking SAR data as an example, the specific processing flow is as follows: First, the SAR image is radiometrically calibrated to convert the image grayscale values ​​into backscattering coefficients. : in, It is the grayscale value of the image. and These are calibration parameters, which can be obtained from the calibration table provided by the satellite.

[0048] Then, the calibrated SAR image is filtered to remove noise. The Lee filtering algorithm is used, and its filtered pixel values... It can be represented as: in, These are the original pixel values. It is the mean within the window. These are adaptive weighting coefficients, defined as: It is the variance within the window. It is the equivalent apparent number of noise.

[0049] Finally, wave parameters are retrieved based on SAR data. An empirical model, such as the TVG (Terrain-VaryingGain) model, is established to fit the relationship between the backscattering coefficient and wave parameters, and the wave height is obtained. The estimated value is given by the following formula: in, and These are model parameters, which can be obtained by fitting a large amount of measured data.

[0050] S23, Processing marine monitoring buoy data The main problems with ocean monitoring buoy data are data gaps and errors. These problems are addressed using spatiotemporal interpolation and data fusion methods. Taking wave height data as an example, let... It is the first A buoy in For missing observations, the wave height at time [time] is estimated using Kriging interpolation. The estimated value from Kriging interpolation is [value]. It can be represented as: in, These are weighting coefficients. This represents the number of observations involved in the interpolation. The weighting coefficients are obtained by solving the variogram: in, It is the first The and the first The distance between observation points.

[0051] In terms of data fusion, satellite remote sensing data and ocean monitoring buoy data are combined to improve data accuracy and completeness. The Kalman filter algorithm is employed, with the following state equations and observation equations: in, It is the system state vector, which contains information such as wave parameters; It is the state transition matrix; It is a noise transfer matrix; It is process noise; It is an observation vector that includes satellite remote sensing data and ocean monitoring buoy data; It is the observation matrix; It is observation noise.

[0052] The estimation steps of Kalman filtering include prediction and updating. The estimated value in the prediction step... Covariance Matrix They are respectively: in, It is the covariance matrix of the process noise.

[0053] Update step estimate Covariance Matrix They are respectively: in, It is the covariance matrix of the observation noise. It is the Kalman gain matrix. It is an identity matrix.

[0054] S24. Integrating real-time observation data with meteorological forecast model data Meteorological forecasting models can provide marine environmental data for a future period, but they have a certain degree of error. Fusing data from meteorological forecasting models with real-time observation data improves data accuracy. A Bayesian fusion method is employed, assuming... It is a weather forecasting model. It is a probability distribution of real-time observation data after fusion. It can be represented as: in, It is the likelihood function of the observed data. It is the prior probability distribution of the weather forecasting model.

[0055] By employing the above multi-source data acquisition and processing methods, marine environmental data within the ship's navigation area can be obtained in real time and accurately, providing reliable data support for determining ship-wave coupling constraints and route planning.

[0056] S3. Based on the ship-wave coupling data model and real-time marine environment data, determine and dynamically adjust the ship-wave coupling constraints and safety thresholds.

[0057] Reference Figure 4 Step S3 specifically includes the following sub-steps: S31. Establishing the Foundation for Coupled Analysis Determining ship-wave coupling constraints relies on ship-wave coupling data models and real-time ocean environmental data. On one hand, ship-wave coupling data models reflect the characteristics and patterns of wave interaction between different ship types and waves under different sea states; on the other hand, real-time ocean environmental data provides specific wave parameters under the current ocean environment. Based on these two factors, constraints can be determined from a ship dynamics perspective.

[0058] Let the motion state vector of the ship be... ,in For ship speed, For the course, The roll angle, For pitch angle, These are the angular velocities of the roll and pitch, respectively. These represent the ship's acceleration in the three coordinate axes.

[0059] The wave parameter vector is ,in For wave height, For wavelength, For wave cycles, The direction of the wave.

[0060] Ship-wave interaction can be expressed by a functional relationship This function embodies the coupling relationship between the ship's motion state and wave parameters.

[0061] S32, Deriving the safety threshold for motion parameters To ensure safe navigation of a ship under different speeds, headings, and wave parameter combinations, it is necessary to determine the safety thresholds for each motion parameter. Taking the roll angle as an example, its safety threshold must consider the ship's stability. According to ship stability theory, the ship's restoring moment... With roll angle It is relevant and can be expressed as ,in For ship displacement, This is the initial stability height of the ship.

[0062] When a ship rolls, it experiences wave disturbance torque. It is related to wave parameters, speed, and heading. Through theoretical analysis and fitting of experimental data, we can obtain... ,in For coefficients, It is a power exponent and can be determined through a ship-wave coupling data model.

[0063] To ensure that the ship's rolling does not exceed the safe range, the following must be met: ,Right now From this, the safe threshold for the roll angle can be derived. : Similarly, for the pitch angle The safety threshold can be derived based on the principles of ship pitch dynamics. Ship pitch restoring force. With pitch angle Related to the pitching disturbance caused by waves. Related to wave parameters and ship motion state, the safe threshold for the pitch angle can be obtained through dynamic analysis. Regarding acceleration Because crew comfort and cargo stability require a certain level of acceleration, and based on the limits of human acceleration tolerance and cargo securing safety standards, combined with a ship-wave coupling data model, safe acceleration thresholds can be determined for different navigation conditions.

[0064] S33, Dynamically Adjusting Constraints The ship's navigation status and the marine environment are dynamically changing; therefore, the ship-wave coupling constraints also need to be dynamically adjusted. When the ship's speed... ,course Wave parameters in the marine environment When changes occur, the constraints need to be updated in real time.

[0065] Let the initial constraints be During the ship's voyage, as Changes, new constraints It can be done through a mapping function get.

[0066] For example, when wave parameters change, the safety thresholds for each motion parameter can be recalculated using the formulas derived from the above theory. Meanwhile, considering the specific requirements of ships under different transport missions, constraints can be appropriately tightened for ships transporting dangerous goods to improve navigation safety.

[0067] S34. Verify and optimize the constraints. The established constraints were verified through numerical simulation and full-scale ship experiments. In the numerical simulation, computer software was used to simulate the ship's motion under different sea states, and the degree of agreement between the simulation results and the constraints was compared. In the full-scale ship experiments, the ship's motion parameters and marine environmental data were recorded during actual navigation, and the safety of the ship's navigation within the constraints was analyzed.

[0068] Based on the verification results, the constraints are optimized. If some constraints are found to be too lenient or too strict, the coefficients and exponents in the relevant formulas can be adjusted to make the constraints more reasonable. Simultaneously, the ship-wave coupling data model is continuously updated to improve the accuracy of constraint determination.

[0069] S4. Based on the ship-wave coupling constraints, construct a multi-objective optimization function and use a genetic algorithm for route planning.

[0070] Reference Figure 5 Step S4 specifically includes the following sub-steps: This route planning algorithm comprehensively considers ship navigation safety, efficiency, and fuel economy. Based on ship-wave coupling constraints, it employs a heuristic search algorithm to find the optimal route and introduces a dynamic adjustment mechanism to optimize the route based on real-time changes in marine environmental data. Specifically, it utilizes multi-objective programming theory to construct an objective function incorporating multiple factors, and uses a genetic algorithm for global search while satisfying ship-wave coupling constraints.

[0071] S41. Constructing a multi-objective optimization objective function Assume the ship starts from the starting point sail to the destination Discretize the navigation path into There are 1 flight segments, and the length of each flight segment is 1. ,in Navigation safety objective: Roll angle under ship-wave coupling constraints. Pitch angle and acceleration These are the core security indicators. A security penalty function is defined. as follows: in, For security penalty weights, These are the safety thresholds for roll angle, pitch angle, and acceleration, respectively.

[0072] Navigation efficiency target: Navigation efficiency is measured by navigation time, assuming the ship is on the [missing information - likely a date or time]. The actual speed of each segment is The total sailing time for: To incorporate this into the objective function, we define an efficiency objective function. for: in, Efficiency weight.

[0073] Fuel economy target: A ship's fuel consumption is related to factors such as speed and seakeeping. Let the ship be at a certain point in time... Fuel consumption rate per segment Then the total fuel consumption for: Define fuel economy objective function for: in, As a weight for fuel economy, and Combining the above three objectives, a multi-objective optimization objective function is constructed. as follows:

[0074] S42. Setting route planning constraints First, ship-wave coupling constraints: for each segment Roll angle Pitch angle and acceleration Must meet: Second, ship handling constraints: ship speed. Within the permitted range within, that is The change in heading cannot exceed a certain angle threshold. This is to ensure the ship's maneuverability.

[0075] S43. Use a genetic algorithm to perform a global route search. Encoding initialization: Using real number encoding, each possible route is mapped to a chromosome. Each gene in the chromosome corresponds to the core parameters of the route, such as the start point, end point, and speed, thus completing the encoding construction of the initial population.

[0076] Fitness Evaluation: Multi-objective Optimization Objective Function Based on Route Planning Define the fitness function as This function calculates the fitness value of each individual (route plan) in the initial population, quantifying the merits of different routes.

[0077] Selection operation: The roulette wheel selection method is used to determine the probability of an individual being selected based on its fitness value (the higher the fitness value, the greater the probability of selection), and to select the best individuals.

[0078] Crossover operation: Perform a two-point crossover operation on the selected superior individuals, randomly select two crossover points, exchange gene segments of the two chromosomes between the crossover points, and generate a new individual with the superior characteristics of both parents.

[0079] Mutation operation: According to the preset mutation probability, some genes of the new individual are randomly mutated to introduce a new search space, avoid the algorithm from getting stuck in local optima, and enrich the diversity of the population.

[0080] Iteration Termination: Repeat the fitness evaluation, selection, crossover, and mutation steps above until the preset number of iterations or the fitness value converges, and finally output the optimal route plan.

[0081] S44. Dynamically adjust the route according to changes in the marine environment. During ship navigation, real-time marine environmental data is acquired. When the marine environmental data changes, the ship-wave coupling constraints are reassessed, and the parameters in the objective function are updated. For example, when wave height increases, the impact of ship-wave coupling on ship safety intensifies. The value may need to be increased. Then, the genetic algorithm is rerun to adjust the ship's route to adapt to the new marine environment.

[0082] S5. Real-time monitoring of ship navigation status and marine environment data; replanning routes when safety thresholds are exceeded, forming a dynamic closed-loop adjustment.

[0083] Reference Figure 6 Step S5 specifically includes the following sub-steps: S51, Data Monitoring and Acquisition Various real-time monitoring devices are installed on the ship, mainly including meteorological sensors (which can measure wind speed, wind direction, temperature, etc.), wave sensors (used to acquire parameters such as wave height, wave period, and wave direction), and satellite communication equipment (used to transmit and receive marine environmental information and ship position data from different data sources).

[0084] During the ship's voyage, at fixed time intervals Collect data, The vector of marine environmental data collected at any given time can be represented as: ,in for High wave height at any moment for Wave cycles at any moment for The direction of the waves at any moment. for Wind speed at all times for Constant wind direction. Simultaneously, collect ship navigation data vectors. ,in for Ship speed at any time for Ship's course at all times for Ship position at any time.

[0085] S52, Threshold Judgment and Early Warning The real-time acquired data is compared with the pre-determined ship-wave coupling constraints. Let the maximum allowable roll angle in the ship-wave coupling constraints be... , The roll angle at any given time is measured by sensors or calculated based on a model. .when At that time, an early warning signal is triggered.

[0086] Similarly, for the maximum permissible pitch angle Maximum permissible acceleration Similar judgments can be made for constraints. Taking acceleration as an example, if... Acceleration at all times If a certain condition is detected, a corresponding safety warning will be issued. The warning level can be categorized based on the degree to which the condition exceeds a threshold; for example, the percentage exceeding the threshold may be [not specified]. Different settings are available. Different warning levels correspond to different intervals.

[0087] S53, Route replanning based on early warning Once a warning signal is triggered, the route needs to be replanned. Heuristic search algorithms (such as genetic algorithms) are used to ensure that ship-wave coupling constraints are met in the new marine environment.

[0088] Let the objective function be... It takes into account navigation safety. Navigation efficiency and fuel economy , can be represented as ,in These are the weighting coefficients, and Navigation safety It can be represented as ,in For the first Individual ship-wave coupling constraint parameters (such as roll angle, pitch angle, etc.). For its corresponding security threshold, To constrain the number of indicators, the smaller the value, the safer the navigation.

[0089] sailing efficiency ,in To plan the total distance of the flight route, This is the estimated sailing time.

[0090] fuel economy ,in For estimated fuel consumption.

[0091] During the planning process, the population (i.e., the set of route options) is iteratively updated through operations such as selection, crossover, and mutation to continuously optimize the objective function. .set up Population of individuals ,in For the first There are several route options, each consisting of a series of discrete waypoints. The selection operation can use a roulette wheel selection based on the objective function value to retain the better individual. The crossover operation can exchange some waypoints between two selected individuals to generate a new individual, while the mutation operation randomly changes the positions of some waypoints in an individual.

[0092] S54, Route Dynamic Adjustment and Closed-Loop Feedback The replanned route information is transmitted to the ship's bridge in a timely manner via satellite communication equipment. During the implementation of the new route, marine environmental and ship status data are continuously monitored, forming a closed-loop feedback system. If data exceeds constraints again, the replanning process is repeated. Let the time step be... Every time Evaluate the objective function, if ( If a pre-set threshold is reached, a new route planning process will be triggered to ensure that the ship is always in an optimal and safe navigation state.

[0093] The implementation principle of this invention is as follows: Through a closed-loop process of "data support - constraint construction - algorithm planning - dynamic adjustment," safe, efficient, and economical ship route planning considering ship-wave coupling constraints is achieved. First, basic ship parameters, navigation parameters, and wave parameters are collected. After preprocessing, a ship-wave coupling data model is constructed and trained using a hybrid CNN and RNN architecture to accurately capture the dynamic coupling relationship between ship motion and wave parameters. Second, marine environmental data is collected from multiple sources, including satellite remote sensing, ocean monitoring buoys, and meteorological forecasting models. After targeted processing and multi-source fusion, real-time, high-precision marine environmental parameters are obtained. Next, based on the trained ship-wave coupling data model and real-time marine environmental data, the coupling relationship between ship motion state and wave parameters is established. Based on ship dynamics and stability theory, safety thresholds for motion parameters such as roll angle, pitch angle, and acceleration are derived to determine the ship-wave coupling relationship. The system incorporates wave coupling constraints that can be dynamically adjusted based on navigation status and the marine environment. Then, a multi-objective optimization function is constructed, integrating navigation safety, efficiency, and fuel economy. Constraints such as ship-wave coupling, speed range, and heading change angle are set. A genetic algorithm is used for global route search to obtain the optimal route scheme that satisfies the multi-objective optimization. Finally, by real-time monitoring of ship navigation status and marine environment data and comparing them with the constraints, an early warning is triggered and the route is replanned when safety thresholds are exceeded, forming a closed-loop feedback loop. This ensures that the ship remains in a safe and efficient navigation state in the complex and ever-changing marine environment, ultimately solving the problems of insufficient consideration of ship-wave coupling constraints and poor adaptability in existing technologies.

[0094] The embodiments described herein are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A ship route planning method considering ship-wave coupling constraints, characterized in that, Includes the following steps: S1. Construct a ship-wave coupled data model, collect basic ship parameters, navigation parameters and wave parameters, and train and evaluate the model using a hybrid CNN and RNN architecture after preprocessing. S2. Collect and fuse multi-source real-time marine environmental data; S3. Based on the ship-wave coupling data model and real-time marine environment data, determine and dynamically adjust the ship-wave coupling constraints and safety thresholds; S4. Based on the ship-wave coupling constraints, a multi-objective optimization function is constructed, and a genetic algorithm is used for route planning; S5. Monitor ship navigation status and marine environment data in real time, and replan the route when the safety threshold is exceeded to form a dynamic closed-loop adjustment.

2. The ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, S1 specifically includes the following steps: S11. Collect basic ship parameters, navigation parameters and wave parameters. The basic ship parameters include ship type, displacement and draft. The navigation parameters include speed and heading. The wave parameters include wave height, wavelength, wave period and wave direction. Normalize the data. S12. A hybrid architecture of convolutional neural network (CNN) and recurrent neural network (RNN) is used to build the model. Spatial features are extracted by CNN and temporal features are processed by RNN. S13. The model is trained using the mean squared error loss function, and the model parameters are optimized using the adaptive moment estimation (Adam) algorithm; S14. The root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²) are used as evaluation metrics to validate and evaluate the trained model.

3. The ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, S2 specifically includes the following steps: S21. Collect marine environmental data of ship navigation areas from multiple sources, including satellite remote sensing, marine monitoring buoys, and meteorological forecasting models; S22. Perform radiometric calibration and Lee filtering on satellite remote sensing data, and obtain wave parameters by inversion using the TVG model; S23. Kriging interpolation is used to fill in missing values ​​in marine monitoring buoy data, and Kalman filtering algorithm is used to fuse satellite remote sensing data with marine monitoring buoy data to obtain real-time observation data; S24. The meteorological prediction model data and the real-time observation data are fused using the Bayesian fusion method to obtain the final marine environmental data.

4. The ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, S3 specifically includes the following steps: S31. Based on the ship-wave coupled data model and real-time marine environmental data, establish the coupling relationship between ship motion state and wave parameters; S32. Based on ship dynamics and stability theory, derive and determine the safety thresholds for roll angle, pitch angle and acceleration; S33. Adjust the ship-wave coupling constraints in real time based on the dynamic changes in the ship's navigation status and the marine environment; S34. Verify the constraints through numerical simulation and actual ship experiments, and complete the optimization based on the verification results.

5. A ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, The method for determining the security threshold in S3 is as follows: The roll angle safety threshold is derived from ship stability theory; The safe threshold for pitch angle is determined based on the principles of ship pitch dynamics. The acceleration safety threshold is determined based on the human body's acceleration limit, cargo securing safety standards, and a ship-wave coupling data model.

6. A ship route planning method considering ship-wave coupling constraints according to claim 5, characterized in that, The safe threshold for the roll angle is: ;in For coefficients, The power exponent. For wave height, For speed, The direction of the wave. For the course, For drainage volume, This represents the initial stability height.

7. A ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, S4 specifically includes the following steps: S41. Construct a multi-objective optimization objective function that integrates navigation safety, navigation efficiency, and fuel economy; S42. Set constraints, including ship-wave coupling constraints, speed range constraints, and heading change angle constraints; S43. A genetic algorithm is used to perform a global route search. The route is encoded with real numbers. A fitness function is constructed based on the objective function. Roulette wheel selection, crossover, and mutation operations with set probabilities are performed in sequence. S44. Based on changes in real-time marine environmental data, reassess the ship-wave coupling constraints and update the objective function parameters, then rerun the genetic algorithm to dynamically adjust the route.

8. A ship route planning method considering ship-wave coupling constraints according to claim 7, characterized in that, The multi-objective optimization objective function in S4 is: The ship-wave coupling constraint in S42 includes: roll angle Pitch angle acceleration The speed range constraint is as follows: The heading change angle constraint is that the heading change does not exceed an angle threshold. .

9. A ship route planning method considering ship-wave coupling constraints according to claim 1, characterized in that, S5 specifically includes the following steps: S51. Collect marine environmental data and ship navigation data in real time at fixed time intervals using meteorological sensors, wave sensors and satellite communication equipment; S52. Compare the real-time collected data with the ship-wave coupling constraints, and trigger an early warning when the parameters exceed the corresponding safety threshold; S53. After the warning is triggered, the route is replanned based on the genetic algorithm to construct a multi-objective optimization function that integrates navigation safety, navigation efficiency and fuel economy; S54. Dynamically adjust and provide closed-loop feedback for the replanned routes, continuously monitor data, and trigger route replanning again based on changes in the objective function.

Images