Ship navigation control method and system, storage medium and electronic equipment

Abstract

The invention discloses a ship navigation control method and system, a storage medium and electronic equipment, and the method comprises the steps: predicting the predicted carbon emission data of a digital twin of a target ship in a current navigation environment based on the real-time target sensing data, navigation data and current environment data of the target ship; obtaining candidate navigation control parameters of the target ship based on the predicted carbon emission data; performing control strategy simulation on the digital twin based on the candidate navigation control parameters, preset reference navigation control parameters, the navigation data and a plurality of pieces of preset environment data to obtain a plurality of candidate performance index parameters, safety evaluation parameters and performance evaluation parameters of the candidate navigation control parameters, and reference performance index parameters; determining a control strategy verification result based on the candidate performance indicator parameters, the reference performance indicator parameters, the safety evaluation parameters and the performance evaluation parameters; target navigation control parameters are determined based on the control strategy verification result, and the target ship is controlled to operate according to the target navigation control parameters.

Description

Technical Field

[0001] This application relates to the field of computer control technology, specifically to a navigation control method, system, storage medium, and electronic device for ships. Background Technology

[0002] With increasingly stringent carbon emission regulations in the international shipping industry, ships need to pay attention to carbon emission control during voyages to avoid exceeding emission standards. This is generally achieved by optimizing speed and routes, using low-carbon fuels, and installing energy-saving devices.

[0003] However, existing methods mainly rely on rule-based strategies and static energy consumption models to control ship carbon emissions. However, these traditional control methods show significant rigidity and inadequacy when faced with the complex and ever-changing weather and sea conditions in actual navigation, resulting in inefficiency in ship navigation control. Summary of the Invention

[0004] A navigation control method for ships is provided, which aims to improve the accuracy of ship navigation control.

[0005] Firstly, a method for controlling the navigation of a ship is provided, comprising the following steps: Based on the target vessel's real-time target sensing data, navigation data, and current environmental data of the current navigation environment, predict the carbon emission data of the target vessel's digital twin in the current navigation environment. Based on the predicted carbon emission data, candidate navigation control parameters for the target vessel are obtained; Based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments, the control strategy simulation of the digital twin is performed to obtain candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to the candidate navigation control parameters under multiple preset navigation environments, as well as the baseline performance index parameters corresponding to the baseline navigation control parameters. Based on candidate performance index parameters, benchmark performance index parameters, safety assessment parameters, and performance assessment parameters, determine the control strategy verification results corresponding to the candidate navigation control parameters; Based on the verification results of the control strategy, the target navigation control parameters are determined from the candidate navigation control parameters, and the target ship is controlled to operate according to the target navigation control parameters.

[0006] In one exemplary embodiment, the ship's navigation control method further includes: Acquire real-time initial sensor data of the target vessel and initial environmental data of the current navigation environment; Spatiotemporal alignment of initial environmental data and initial sensor data is performed to obtain intermediate sensor data and current environmental data of the current navigation environment; State estimation is performed on intermediate sensing data to obtain target sensing data.

[0007] In one exemplary embodiment, the ship's navigation control method further includes: Obtain the initial digital twin of the target vessel; Based on the target vessel's real-time target sensing data and the current navigation environment data, the initial digital twin is updated with operational status to obtain the digital twin corresponding to the target vessel.

[0008] In one exemplary embodiment, based on real-time target sensing data, navigation data, and current environmental data of the target vessel's current navigation environment, the predicted carbon emission data of the target vessel's digital twin in the current navigation environment is predicted, including: Based on target sensing data and current environmental data, determine the target power parameters corresponding to the target vessel; Based on target sensing data, navigation data, current environmental data, and target dynamic parameters, predict the carbon emissions of the target ship's digital twin under the current navigation environment.

[0009] In an exemplary embodiment, the target dynamic parameters corresponding to the target vessel are determined based on target sensing data and current environmental data, including: Based on target sensing data and current environmental data, multiple reference dynamic parameters are determined in a preset dynamic database. By performing linear interpolation on multiple reference dynamic parameters, the target dynamic parameters corresponding to the target ship are obtained.

[0010] In one exemplary embodiment, candidate navigation control parameters for the target vessel are obtained based on predicted carbon emission data, including: The predicted carbon emission data is input into the navigation strategy optimization model, and the navigation strategy optimization model establishes the target state space based on the predicted carbon emission data and the digital twin. Based on the target state space and the preset action space, candidate navigation control parameters that meet the preset target constraints are generated.

[0011] In one exemplary embodiment, control strategy simulation is performed on the digital twin based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments, including: Based on the original simulation duration and the preset target simulation duration corresponding to the candidate navigation control parameters, the computational granularity of the navigation strategy simulation model in the time dimension is determined. Candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments are input into the navigation strategy simulation model. The navigation strategy simulation model then performs control strategy simulation on the digital twin according to the computational granularity.

[0012] In an exemplary embodiment, control strategy simulation is performed on the digital twin to obtain safety evaluation parameters and performance evaluation parameters corresponding to candidate navigation control parameters under multiple preset navigation environments, including: Obtain the candidate performance index parameters corresponding to the candidate navigation control parameters under each preset navigation environment; Based on the preset safety threshold and multiple candidate performance indicator parameters, the failure probability of the indicators is statistically analyzed, and the safety assessment parameters corresponding to the candidate navigation control parameters are obtained based on the failure probability of the indicators. The dispersion of multiple candidate performance index parameters is calculated to obtain the performance evaluation parameters corresponding to the candidate navigation control parameters.

[0013] In an exemplary embodiment, based on candidate performance index parameters, benchmark performance index parameters, safety assessment parameters, and performance assessment parameters, the control strategy verification result corresponding to the candidate navigation control parameters is determined, including: Compare the candidate performance indicator parameters with the benchmark performance indicator parameters to obtain the indicator gain corresponding to the candidate performance indicator parameters; When the index gain is greater than a preset threshold, the control strategy verification result corresponding to the candidate navigation control parameters is determined based on the safety assessment parameters and performance assessment parameters.

[0014] Secondly, a navigation control system for a ship is also provided, comprising: The data acquisition module is used to acquire real-time target sensor data, navigation data, and current environmental data of the target vessel's current navigation environment. The optimization module includes a carbon emission prediction model and a navigation strategy optimization model; The carbon emission prediction model is used to predict the carbon emission data of the digital twin of the target ship in the current navigation environment based on the target ship's real-time target sensing data, navigation data and current environmental data of the current navigation environment. A navigation strategy optimization model is used to obtain candidate navigation control parameters for a target vessel based on predicted carbon emission data. The strategy simulation module includes a navigation strategy simulation model; A navigation strategy simulation model is used to simulate the control strategy of a digital twin based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments. This model yields candidate performance index parameters, safety assessment parameters, and performance assessment parameters for the candidate navigation control parameters under multiple preset navigation environments, as well as the baseline performance index parameters for the baseline navigation control parameters. Based on these candidate performance index parameters, baseline performance index parameters, safety assessment parameters, and performance assessment parameters, the model determines the verification results of the control strategy corresponding to the candidate navigation control parameters. The strategy control module is used to determine the target navigation control parameters from the candidate navigation control parameters based on the control strategy verification results, and control the target ship's operation according to the target navigation control parameters.

[0015] Thirdly, an electronic device is also provided, including a memory and a processor, wherein computer instructions are stored in the memory, and when the computer instructions are executed by the processor, the method of any one of the above embodiments is implemented.

[0016] Fourthly, a computer-readable storage medium is also provided, including computer instructions that, when executed on a device, cause the device to perform the method as described in any of the foregoing aspects and any of the embodiments thereof.

[0017] Fifthly, embodiments of this application provide a computer program product, including a computer program or instructions, which are executed by a processor to implement the steps of any of the methods described above.

[0018] Beneficial Effects: This application provides a navigation control method for ships. By predicting the carbon emission data of a digital twin of the target ship under the current navigation environment based on real-time target sensor data, navigation data, and current environmental data, the method can achieve accurate carbon emission prediction by combining the digital twin. Based on the accurate carbon emission prediction, candidate navigation control parameters for the target ship can be effectively output. Then, based on the candidate navigation control parameters, predicted baseline navigation control parameters, navigation data, and preset environmental data of multiple preset navigation environments, the method simulates the control strategy for the digital twin. Based on the candidate performance index parameters, baseline performance index parameters, safety assessment parameters, and performance assessment parameters obtained from the control strategy simulation, the method determines the control strategy verification results for the candidate navigation control parameters. By using a digital twin for control strategy simulation, the accuracy of the control strategy verification for the candidate navigation control parameters is ensured. The control strategy verification results effectively verify the control effectiveness of the candidate navigation control parameters, thereby obtaining accurate and effective target navigation control parameters. Accurate control of the target ship is achieved through these target navigation control parameters. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a ship navigation control method provided by an exemplary embodiment of this disclosure; Figure 2 This is another schematic flowchart of a ship navigation control method provided by an exemplary embodiment of this disclosure; Figure 3 This is another schematic flowchart of a ship navigation control method provided by an exemplary embodiment of this disclosure; Figure 4 This is another schematic flowchart of a ship navigation control method provided by an exemplary embodiment of this disclosure; Figure 5 This is a flowchart illustrating a digital twin-based ship carbon emission prediction and optimization system provided in an exemplary embodiment of this disclosure. Figure 6 This is a schematic diagram of a ship navigation control system provided by an exemplary embodiment of this disclosure; Figure 7 This is an internal structural diagram of an electronic device provided by an exemplary embodiment of this disclosure. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0023] This application provides a navigation control method, system, storage medium, and electronic device for a ship. The electronic device can be a server, a terminal, or other similar device, and can be deployed within the target ship. In an exemplary embodiment, the terminal predicts the predicted carbon emissions of the target ship's digital twin under the current navigation environment based on real-time target sensor data, navigation data, and current environmental data of the current navigation environment. The server obtains candidate navigation control parameters corresponding to the target ship based on the predicted carbon emission data. The terminal simulates the control strategy for the digital twin based on the candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environmental data of multiple preset navigation environments, obtaining candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to the candidate navigation control parameters under multiple preset navigation environments, as well as baseline performance index parameters corresponding to the baseline navigation control parameters. The terminal determines the control strategy verification result corresponding to the candidate navigation control parameters based on the candidate performance index parameters, baseline performance index parameters, safety assessment parameters, and performance assessment parameters. Based on the control strategy verification result, the terminal determines the target navigation control parameters from the candidate navigation control parameters and controls the target ship's operation according to the target navigation control parameters. The terminal may include, but is not limited to, computers, and the server may be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, big data, and artificial intelligence platforms. The terminal and the server may be connected directly or indirectly through wired or wireless communication, and this application does not impose any restrictions on this.

[0024] On the one hand, this embodiment provides a method for controlling the navigation of a ship, such as... Figure 1 As shown, it includes the following steps: S101, based on the target vessel's real-time target sensing data, navigation data, and current environmental data of the current navigation environment, predicts the predicted carbon emission data of the target vessel's digital twin in the current navigation environment.

[0025] The target vessel refers to the ship currently requiring navigation control during its voyage. Target sensor data refers to navigational condition data collected by sensors deployed on the target vessel, such as fuel flow, speed, heading, main engine power / speed, exhaust composition, draft, and trim. Navigation data refers to navigation plan data used to guide the target vessel's navigation, such as route and navigation data. A digital twin is a high-fidelity digital virtual ship model constructed in virtual space that perfectly corresponds to the physical entity of the target vessel. Predicted carbon emission data is carbon emission data predicted based on the digital twin; that is, carbon emission data predicted through simulated operation based on the digital twin in virtual space.

[0026] For example, in response to navigation control commands to the target vessel, the terminal acquires real-time target sensor data, navigation data, and current environmental data of the target vessel's current navigation environment. The current environmental data includes sea state data and external resource data, such as weather information and port traffic density. The terminal can acquire real-time sea state data through sensors deployed on the exterior of the target vessel, and obtain external resource data such as weather information and port traffic density from an external resource server.

[0027] An initial digital twin of the target vessel and its external environment model are pre-deployed in an integrated manner on a terminal or in the cloud. High-precision 3D modeling technology is used to virtually model key components of the target vessel, such as its hull structure, propulsion system, fuel system, and energy efficiency devices, creating an initial digital twin. An external environment model is then deployed in virtual space, constructing a virtual navigation environment corresponding to the current navigation environment based on current environmental data. The terminal sends real-time target sensor data of the target vessel to the initial digital twin, which updates its operational status based on the target sensor data, resulting in an updated digital twin whose operational status matches that of the target vessel. Finally, in the virtual navigation environment, the digital twin predicts carbon emissions based on the target sensor data, navigation data, and current environmental data, obtaining the predicted carbon emissions data for the digital twin in the corresponding virtual navigation environment.

[0028] In the process of carbon emission prediction using digital twins, the power parameters of the target ship, such as resistance data and propulsion efficiency, can be calculated based on the target sensor data. Carbon emission prediction is then performed based on the power parameters, target sensor data, navigation data, and current environmental data to obtain the predicted carbon emission data.

[0029] S102, based on predicted carbon emission data, obtain candidate navigation control parameters for the target vessel.

[0030] The candidate navigation control parameters represent the control parameters used to control the ship's navigation that are to be verified, including speed commands and heading commands. Speed ​​commands include a sequence of speeds at various time points, such as maintaining a speed of 13 knots for the first two hours and increasing to 15 knots for the next three hours. Heading commands are, for example, a heading of 5 degrees east.

[0031] For example, the terminal also deploys a navigation strategy optimization model. After obtaining the predicted carbon emission data, the terminal inputs the predicted carbon emission data, target sensor data, current environmental data, and preset navigation constraints into the navigation strategy optimization model for calculation, thereby obtaining candidate navigation control parameters. The candidate navigation control parameters can be understood as the optimal ship operation strategy that can reduce energy consumption and carbon emissions while meeting the preset navigation constraints.

[0032] In the process of model calculation by the navigation strategy optimization model, data interaction can be carried out with the digital twin. That is, the navigation strategy optimization model obtains the target virtual operating state of the corresponding ship virtual model from the digital twin after the operating state is updated based on the target sensing data. Then, the navigation strategy optimization model performs model calculation based on the target virtual operating state to obtain candidate navigation control parameters.

[0033] S103, based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data and preset environment data of multiple preset navigation environments, performs control strategy simulation on the digital twin to obtain candidate performance index parameters, safety assessment parameters and performance assessment parameters corresponding to candidate navigation control parameters under multiple preset navigation environments, as well as baseline performance index parameters corresponding to baseline navigation control parameters.

[0034] The preset baseline navigation control parameters refer to the navigation strategy currently actually adopted by the target vessel. The preset navigation environment is the navigation environment used for verifying the control strategy. Control strategy verification includes safety verification and performance verification. Preset environment data is used to construct the corresponding virtual navigation environment in virtual space. Candidate performance index parameters are the navigation index parameters of the digital twin during its navigation process according to the candidate navigation control parameters. Baseline performance index parameters are the navigation index parameters of the digital twin during its navigation process according to the candidate navigation control parameters. Navigation index parameters include, for example, total carbon emissions, fuel consumption rate, and navigation time. Safety assessment parameters are parameters obtained by conducting a safety assessment of the candidate performance index parameters, characterizing the navigation safety level of the digital twin when navigating according to the candidate index parameters. Performance assessment parameters are parameters obtained by evaluating and calculating the candidate performance index parameters, characterizing the performance fluctuation level of the digital twin when navigating according to the candidate index parameters.

[0035] For example, the terminal also deploys a navigation strategy simulation model. Candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments are input into the navigation strategy simulation model. The navigation strategy simulation model constructs the corresponding preset navigation environment in the virtual space through the external environment model. In each preset navigation environment, the digital twin is driven to perform control strategy simulation based on navigation data according to the candidate navigation control parameters and the preset baseline navigation control parameters, to obtain the candidate performance index parameters corresponding to the candidate navigation control parameters in multiple preset navigation environments, as well as the baseline performance index parameters corresponding to the baseline navigation control parameters.

[0036] Then, based on the candidate performance index parameters corresponding to each preset navigation environment, the safety assessment parameters and performance assessment parameters corresponding to the candidate navigation control parameters are determined.

[0037] S104. Based on candidate performance index parameters, benchmark performance index parameters, safety assessment parameters, and performance assessment parameters, determine the control strategy verification results corresponding to the candidate navigation control parameters.

[0038] For example, after the terminal obtains candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to candidate navigation control parameters under multiple preset navigation environments, as well as benchmark performance index parameters corresponding to benchmark navigation control parameters, it compares the candidate performance index parameters with the benchmark performance index parameters, and determines the index gain corresponding to the candidate performance index parameters based on the comparison results, such as a 5% reduction in fuel consumption rate or a reduction in carbon emissions of less than 3%.

[0039] Then, the candidate performance indicator parameters are compared according to a preset safety threshold, and the failure probability of the indicators is statistically analyzed based on the comparison results to obtain the safety assessment parameters corresponding to the candidate navigation control parameters. Then, the dispersion of the multiple candidate performance indicator parameters is calculated to obtain the performance assessment parameters corresponding to the candidate navigation control parameters.

[0040] The terminal determines the control strategy verification result corresponding to the candidate navigation control parameters based on the index gain, safety assessment parameters, and performance assessment parameters. The control strategy verification result includes verification pass and verification fail. Specifically, the control strategy verification result is determined to be verification pass when all three parameters (index gain, safety assessment parameters, and performance assessment parameters) meet the corresponding parameter threshold conditions, or verification fail when at least one of the three parameters fails to meet the corresponding parameter threshold conditions.

[0041] S105, based on the control strategy verification results, determine the target navigation control parameters from the candidate navigation control parameters, and control the target ship's operation according to the target navigation control parameters.

[0042] For example, when the control strategy verification result is "verification passed," the candidate navigation control parameters are used as the target navigation control parameters, and the target vessel is controlled to operate according to the target navigation control parameters. When the control strategy verification result is "verification failed," the terminal instructs the navigation strategy optimization model to generate new candidate navigation control parameters, and verifies the control strategy using the new candidate navigation control parameters through the navigation strategy simulation model until the verification result is "verification passed," thus obtaining the target navigation control parameters.

[0043] In this embodiment, by using real-time target sensor data, navigation data, and current environmental data of the target vessel, the predicted carbon emission data of the target vessel's digital twin in the current navigation environment is predicted. This allows for accurate prediction of carbon emission data by combining the digital twin, and then, based on the accurate predicted carbon emissions, candidate navigation control parameters for the target vessel can be effectively output. Then, based on the candidate navigation control parameters, predicted baseline navigation control parameters, navigation data, and preset environmental data of multiple preset navigation environments, the control strategy simulation of the digital twin is performed. Based on the candidate performance index parameters, baseline performance index parameters, safety assessment parameters, and performance assessment parameters obtained from the control strategy simulation, the control strategy verification results of the candidate navigation control parameters are determined. By using the digital twin for control strategy simulation, the accuracy of the control strategy verification of the candidate navigation control parameters is ensured, and the control effectiveness of the candidate navigation control parameters can be effectively verified based on the control strategy verification results. This leads to accurate and effective target navigation control parameters, and accurate control of the target vessel is achieved through these target navigation control parameters.

[0044] In one exemplary embodiment, the ship's navigation control method further includes: Acquire real-time initial sensor data of the target vessel and initial environmental data of the current navigation environment; Spatiotemporal alignment of initial environmental data and initial sensor data is performed to obtain intermediate sensor data and current environmental data of the current navigation environment; State estimation is performed on intermediate sensing data to obtain target sensing data.

[0045] Initial sensor data refers to sensor data that has not undergone data processing. Initial environmental data refers to environmental data that has not undergone data processing.

[0046] For example, the terminal acquires real-time initial sensing data and initial environmental data of the current navigation environment of the target vessel through sensors deployed on the target vessel. The initial environmental data and initial sensing data are then spatiotemporally aligned to obtain intermediate sensing data and current environmental data of the current navigation environment. Specifically, this can be achieved by unifying the coordinates of the initial environmental data in the spatial dimension to obtain the current environmental data of the current navigation environment; and by unifying the time of the current environmental data and initial sensing data in the temporal dimension. This can be done by acquiring the timestamps of the current environmental data and the initial sensing data, and then performing time matching in the initial sensing data based on the timestamp of the current environmental data to obtain intermediate sensing data that matches the time of the current environmental data. Then, an extended Kalman filter is used to perform dynamic state estimation on the intermediate sensing data to obtain the target sensing data.

[0047] In this embodiment, by performing spatiotemporal alignment and state estimation on the initial environmental data and initial sensor data, the data accuracy of the current environmental data and target sensor data can be guaranteed, thereby improving the accuracy of predicted carbon emission data.

[0048] In one exemplary embodiment, the ship's navigation control method further includes: Obtain the initial digital twin of the target vessel; Based on the target vessel's real-time target sensing data and the current navigation environment data, the initial digital twin is updated with operational status to obtain the digital twin corresponding to the target vessel.

[0049] The initial digital twin refers to a virtual ship model that is pre-built but not aligned with the current operating status of the target ship.

[0050] For example, after the terminal obtains the target vessel's real-time target sensor data, navigation data, and current environmental data of the current navigation environment, it constructs a virtual navigation environment consistent with the current navigation environment in the virtual space where the initial digital twin resides, using an external environment model, and sends the target sensor data to the initial digital twin. The initial digital twin updates its operating status based on the target sensor data, resulting in a digital twin with an updated status corresponding to the target vessel. The operating status of the digital twin is consistent with the operating status of the target vessel.

[0051] Then, in a virtual navigation environment, the digital twin performs carbon emission prediction based on target sensor data, navigation data, and current environmental data, obtaining the predicted carbon emission data of the digital twin in the virtual navigation environment corresponding to the current navigation environment.

[0052] In this embodiment, a high-fidelity digital twin is used in conjunction with real-time environmental data, navigation status and other multi-source data to achieve dynamic prediction of ship carbon emissions and improve the accuracy of carbon emission prediction.

[0053] In one exemplary embodiment, such as Figure 2 As shown, based on the target vessel's real-time target sensor data, navigation data, and current environmental data of the current navigation environment, the predicted carbon emission data of the target vessel's digital twin in the current navigation environment is predicted, including: S201, based on target sensing data and current environmental data, determine the target dynamic parameters corresponding to the target vessel; S202, based on target sensing data, navigation data, current environmental data, and target dynamic parameters, predicts the carbon emissions of a digital twin of a target ship under the current navigation environment.

[0054] Among them, target dynamic parameters are parameters that characterize the impact on the sailing power of the target ship, such as resistance data and propulsion efficiency.

[0055] For example, the terminal inputs the target vessel's real-time target sensing data, navigation data, and current environmental data of the current navigation environment into the digital twin. The digital twin estimates the dynamic parameters based on the target sensing data and the current environmental data to obtain the target dynamic parameters corresponding to the target vessel.

[0056] The digital twin includes a carbon emission prediction model, which predicts carbon emissions based on target sensor data, navigation data, current environmental data, and target power parameters, thus obtaining the predicted carbon emission data of the digital twin under the current navigation environment.

[0057] In an exemplary embodiment, during the state estimation process of intermediate sensor data, the drag coefficient of the target vessel under the current navigation environment is also obtained. The drag coefficient includes appendage drag increment coefficients, etc. Specifically, the drag coefficient of the target vessel under the current navigation environment can be determined based on a hull-environment coupled dynamic feature library. The hull-environment coupled dynamic feature library is constructed based on real-time data calculation of wind and wave drag increase and historical data learning of appendage drag changes. The terminal learns appendage drag changes in advance by comparing theoretical and actual energy consumption. When it detects that the actual energy consumption is continuously high, it uses a parameter identification algorithm to reverse-calculate an appendage drag increment coefficient, constructing the hull-environment coupled dynamic feature library, thereby realizing online learning of drag changes caused by conditions such as hull fouling.

[0058] The target sensor data, navigation data, current environmental data, and drag coefficient are input into the digital twin. The digital twin includes a ship drag model. Based on the target sensor data, current environmental data, and drag coefficient, the target dynamic parameters corresponding to the target ship are obtained through the ship drag model.

[0059] For example, the ship resistance model incorporates a built-in CFD (Computational Fluid Dynamics) algorithm. It weights and fuses the base resistance curve calculated by CFD based on target sensor data, current environmental data, and resistance coefficients with environmental correction factors identified in real-time by machine learning to obtain the target resistance curve. Then, based on the target resistance curve, the corresponding resistance value and propulsion efficiency are determined to obtain the target dynamic parameters. This embodiment enables the ship resistance model to adapt to changes in ship fouling and sea state.

[0060] In one exemplary embodiment, the carbon emission prediction model in the digital twin also includes a model correction factor, which can dynamically adjust the weight of the correction factor according to the reliability of the current data and the characteristics of the operating conditions, thereby achieving dynamic optimization of the carbon emission data prediction model in the digital twin.

[0061] In this embodiment, the target dynamic parameters are calculated using a digital twin based on the target sensing data and current environmental data, and carbon emission prediction is performed by combining the target dynamic parameters, which can improve the accuracy of carbon emission prediction data.

[0062] In an exemplary embodiment, S201, based on target sensing data and current environmental data, the target dynamic parameters corresponding to the target vessel are determined, including: Based on target sensing data and current environmental data, multiple reference dynamic parameters are determined in a preset dynamic database. By performing linear interpolation on multiple reference dynamic parameters, the target dynamic parameters corresponding to the target ship are obtained.

[0063] For example, the terminal obtains the resistance coefficient of the target ship in the current navigation environment, inputs the target sensor data, navigation data, current environment data and resistance coefficient into the digital twin, and obtains the target resistance curve based on the target sensor data, current environment data and resistance coefficient through the ship resistance model in the digital twin.

[0064] Based on target sensor data, current environmental data, and the target drag curve, multiple reference dynamic parameters are determined in a pre-defined dynamic database. Specifically, a pre-defined dynamic database is pre-deployed within the digital twin, containing the target ship's pre-defined dynamic parameters for various pre-defined operating conditions, such as combinations of different speeds, drafts, trim, wave heights, wave directions, wind speeds, and wind directions.

[0065] The digital twin determines the current operating condition based on target sensor data, current environmental data, and the target drag curve. It then performs operating condition matching against a preset power database. If a matching preset operating condition is found, the preset power parameters corresponding to that preset operating condition are used as the target power parameters for the current operating condition.

[0066] If no matching preset operating condition is found, the similarity between the current operating condition and each preset operating condition is calculated in the preset power database. Multiple reference preset operating conditions are determined based on the similarity; for example, preset operating conditions with a similarity greater than a preset similarity threshold can be used as reference preset operating conditions. The preset power parameters of these multiple reference preset operating conditions are obtained as reference power parameters. Linear interpolation is then performed on these reference power parameters to obtain the target power parameters corresponding to the target ship. The target power parameters include the target ship's resistance value and propulsion efficiency under the current navigation environment. The linear interpolation calculation may be a high-dimensional interpolation algorithm.

[0067] In this embodiment, by performing linear interpolation in a preset power database based on the current operating conditions, the target power parameters corresponding to the current operating conditions can be obtained quickly and accurately, thereby improving the prediction accuracy of carbon emission data.

[0068] In an exemplary example, a parameter space is defined to influence the dynamic parameters of the target ship, including parameters in various dimensions such as speed, draft, trim, wave height, and wave direction. A range and step size are set for each dimension parameter; for example, for speed, a range of 5 knots to 20 knots, with each point representing a 0.5-knot interval. All parameters in each dimension are combined to generate a grid. Each grid point contains a set of preset operating conditions, such as (speed i, draft j, trim k, wave height l, wave direction m). For each grid point, a CFD (Computational Fluid Dynamics) algorithm is run to calculate parameters such as drag and propulsion efficiency, obtaining the dynamic parameters corresponding to the preset operating conditions for each grid point. These parameters are then stored in a database, resulting in a preset dynamic database.

[0069] For example, a high-fidelity CFD model can be reduced in order to obtain a reduced-order model. This reduced-order model is then deployed to a pre-defined dynamic database, using the dynamic parameters of a set of pre-defined operating conditions corresponding to each grid point in the database as CFD samples. When retrieving target dynamic parameters from the pre-defined dynamic database based on the current operating condition, the reduced-order model, combined with a high-dimensional interpolation algorithm, searches for and calculates the corresponding target dynamic parameters within the database. This transforms the computation from "calculation" to "querying," reducing computation time from days / hours to seconds while maintaining CFD-level accuracy.

[0070] In one exemplary embodiment, such as Figure 3As shown, based on predicted carbon emission data, candidate navigation control parameters for the target vessel are obtained, including: S301, input the predicted carbon emission data into the navigation strategy optimization model, and establish the target state space based on the predicted carbon emission data and the digital twin through the navigation strategy optimization model; S302, based on the target state space and the preset action space, generates candidate navigation control parameters that meet the preset target constraints.

[0071] Here, the state space represents the set of all possible operational states of the target ship at a given moment. The action space represents the set of all possible actions that the model can perform at each decision point. The objective constraints are hard restrictions on the states and actions to ensure that the generated strategy is safe and feasible.

[0072] For example, predicted carbon emission data is input into a navigation strategy optimization model. The model extracts the current operational status from the digital twin, such as speed, heading, draft, trim, main engine speed / power, cumulative fuel consumption, and cumulative carbon emissions. It also extracts the environmental status from the virtual navigation environment in which the digital twin resides, such as wind speed, wind direction, current speed, current direction, significant wave height, and wave period. Finally, it extracts the mission status from the navigation data, such as the remaining distance to the destination and the planned arrival time. Based on the extracted current operational status, environmental status, and mission status, a basic state space is established. Then, the predicted carbon emission data is encoded, and the encoded predicted carbon emission data is merged into the basic state space to obtain the target state space.

[0073] Then, a preset action space and preset target constraints are obtained. The preset action space includes executable control parameters such as speed setting, heading adjustment, and route selection. The preset target constraints include ship stability, strength limits, latest arrival time, and maximum / minimum engine speed. The navigation strategy optimization model generates initial navigation control parameters based on the target state space and the preset action space. If the initial navigation control parameters do not meet the preset target constraints, the strategy iterates through the initial navigation control parameters until candidate navigation control parameters that satisfy the preset target constraints are obtained.

[0074] In an exemplary embodiment, during the training and deployment of the navigation strategy optimization model, a deep reinforcement learning algorithm is used to construct the model. Historical navigation data is used to pre-train the value network and policy network within the model offline. This historical navigation data includes ship motion time-series data (historical speed, course, trajectory), engine room monitoring data (main engine speed, power, fuel consumption rate), environmental records (wind and wave data), and ship configuration data (draft, load status, etc.). The state space of the navigation strategy optimization model is defined to include ship speed, course, main engine load, sea state, and load factor. The action space includes speed adjustments, course deviations, and the activation / deactivation of energy efficiency devices. The reward function is weighted by carbon emission reduction, fuel consumption reduction, navigation time, and safety factor.

[0075] The navigation strategy optimization model interacts with the digital twin in real time for online training, continuously improving its decision-making capabilities under different sea states, load conditions, and weather conditions. For example, the navigation strategy optimization model obtains real-time ship status data from the digital twin, such as speed, position, and environmental data, and outputs action commands, i.e., navigation control parameters, such as speed adjustments. The digital twin then infers the multi-dimensional results of this action, such as carbon emissions and fuel consumption, and feeds these results back to the navigation strategy optimization model as reward signals. The navigation strategy optimization model updates its value network and policy network parameters online, achieving real-time evolution of the strategy.

[0076] In this embodiment, by using a navigation strategy optimization model to generate candidate navigation control parameters based on predicted carbon emission data, the effectiveness of the candidate navigation control parameters can be improved.

[0077] In one exemplary embodiment, control strategy simulation is performed on the digital twin based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments, including: Based on the original simulation duration and the preset target simulation duration corresponding to the candidate navigation control parameters, the computational granularity of the navigation strategy simulation model in the time dimension is determined. Candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments are input into the navigation strategy simulation model. The navigation strategy simulation model then performs control strategy simulation on the digital twin according to the computational granularity.

[0078] Here, the original simulation duration refers to the initial time required for simulation of the control strategy based on the candidate navigation control parameters. The preset target simulation duration refers to the desired duration of the control strategy simulation. Calculation granularity indicates the time granularity at which the navigation strategy simulation model performs calculations, such as the time step.

[0079] For example, while maintaining the key dynamic characteristics of the system, the high-fidelity CFD model is downgraded to a real-time computation model based on transfer functions and deployed into the navigation strategy simulation model. The navigation strategy simulation model obtains the original simulation duration and the preset target simulation duration corresponding to the candidate navigation control parameters, and uses an adaptive algorithm to dynamically adjust the computational granularity according to the original simulation duration and the preset target simulation duration.

[0080] Candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments are input into the navigation strategy simulation model. The navigation strategy simulation model performs control strategy simulation on the digital twin according to the calculation granularity, and obtains candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to the candidate navigation control parameters under multiple preset navigation environments, as well as the baseline performance index parameters corresponding to the baseline navigation control parameters.

[0081] In this embodiment, by using an adaptive algorithm to dynamically adjust the calculation granularity, the flight simulation that originally required several days can be completed within minutes, thus improving the efficiency of control strategy simulation.

[0082] In one exemplary embodiment, such as Figure 4 As shown, control strategy simulation is performed on the digital twin to obtain safety evaluation parameters and performance evaluation parameters corresponding to candidate navigation control parameters under multiple preset navigation environments, including: S401, Obtain the candidate performance index parameters corresponding to the candidate navigation control parameters under each preset navigation environment; S402, based on the preset indicator safety threshold and multiple candidate performance indicator parameters, the failure probability of the indicators is statistically calculated, and the safety assessment parameters corresponding to the candidate navigation control parameters are obtained based on the failure probability of the indicators. S403, calculate the dispersion of multiple candidate performance index parameters to obtain the performance evaluation parameters corresponding to the candidate navigation control parameters.

[0083] Among them, the probability of indicator failure refers to the probability that one of the multiple candidate performance indicator parameters fails to reach the preset indicator safety threshold.

[0084] For example, the terminal acquires candidate performance index parameters corresponding to the candidate navigation control parameters under each preset navigation environment. These candidate performance index parameters include performance indicators such as total carbon emissions, fuel consumption rate, and navigation time. Then, it acquires the preset safety threshold for each performance index. Performance indicators that exceed the corresponding preset safety threshold are identified as failure indicators. The proportion of failure indicators among multiple candidate performance index parameters is calculated to obtain the indicator failure probability. This indicator failure probability is used as the safety assessment parameter corresponding to the candidate navigation control parameter.

[0085] The dispersion of multiple candidate performance index parameters is calculated to obtain the performance variance of multiple candidate performance index parameters, and the performance variance is used as the performance evaluation parameter corresponding to the candidate navigation control parameter.

[0086] In this embodiment, the safety assessment parameters are obtained by statistically analyzing the failure probabilities of multiple candidate performance indicator parameters, and the performance assessment parameters are obtained by calculating the dispersion of multiple candidate performance indicator parameters. This ensures the accuracy and effectiveness of the safety assessment parameters and the performance assessment parameters.

[0087] In an exemplary embodiment, based on candidate performance index parameters, benchmark performance index parameters, safety assessment parameters, and performance assessment parameters, the control strategy verification result corresponding to the candidate navigation control parameters is determined, including: Compare the candidate performance indicator parameters with the benchmark performance indicator parameters to obtain the indicator gain corresponding to the candidate performance indicator parameters; When the index gain is greater than a preset threshold, the control strategy verification result corresponding to the candidate navigation control parameters is determined based on the safety assessment parameters and performance assessment parameters.

[0088] Among them, indicator gain refers to the quantitative improvement or change in a specific performance indicator compared to the benchmark performance indicator parameter.

[0089] For example, the terminal acquires candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to multiple candidate navigation control parameters under the preset navigation environment, as well as benchmark performance index parameters corresponding to benchmark navigation control parameters. The average values ​​of the multiple candidate performance index parameters, multiple benchmark performance index parameters, multiple safety assessment parameters, and multiple performance assessment parameters are calculated respectively to obtain target candidate performance index parameters, target benchmark performance index parameters, target safety assessment parameters, and target performance assessment parameters.

[0090] The target candidate performance indicator parameters are compared with the target benchmark performance indicator parameters. For example, the difference between the target candidate performance indicator parameters and the target benchmark performance indicator parameters is calculated to obtain the indicator gain corresponding to the candidate performance indicator parameters.

[0091] When the index gain is greater than a preset threshold, the target safety assessment parameter is compared with the preset safety assessment parameter threshold, and the target performance assessment parameter is compared with the preset performance assessment parameter threshold. When both the target safety assessment parameter and the target performance assessment parameter are less than the preset performance assessment parameter threshold, the verification result of the control strategy corresponding to the candidate navigation control parameter is determined to be verified successfully. When either the target safety assessment parameter is greater than or equal to the preset safety assessment parameter threshold, or the target performance assessment parameter is greater than or equal to the preset performance assessment parameter threshold, the verification result of the control strategy corresponding to the candidate navigation control parameter is determined to be verified unsuccessfully.

[0092] In this embodiment, by determining the control strategy verification results based on index gain, safety assessment parameters, and performance assessment parameters, the accuracy and effectiveness of the control strategy verification results can be guaranteed.

[0093] In an exemplary embodiment, the terminal can set multiple extreme operating conditions under preset navigation environments through a parameter perturbation algorithm, obtain candidate performance index parameters of the digital twin based on candidate navigation control parameters, simulate operation under multiple extreme operating conditions, and then perform sensitivity analysis on multiple candidate performance index parameters to statistically analyze the failure probability and performance variance of the strategy under extreme operating conditions.

[0094] In this embodiment, the verification method based on digital twins can not only quantify the emission reduction benefits of the strategy, but also comprehensively evaluate its robustness through safety boundary detection (such as main engine overload alarm and ship stability loss early warning), and finally select the optimal solution that balances efficiency and safety.

[0095] In one exemplary embodiment, such as Figure 5 The diagram illustrates a flowchart of a ship carbon emission prediction and optimization system based on digital twins. This system includes a high-fidelity digital twin module, a real-time data acquisition and transmission module, a multi-source data fusion module, a reinforcement learning optimization module, a virtual simulation testing platform, and a visualization decision support module. The modules interact via wired / wireless communication networks, and are coordinated and calculated by a central control unit. The ship's data center and shore-based data center maintain a continuous connection via satellite communication links, enabling two-way data synchronization.

[0096] High-fidelity digital twin module: Utilizing high-precision 3D modeling technology, this module virtually models key components of the ship, including its hull structure, propulsion system, fuel system, and energy efficiency devices. Computational fluid dynamics and finite element analysis methods are incorporated to ensure the model's accuracy in hydrodynamics and structural mechanics. External operating environment modeling is integrated, including real-time sea conditions, weather conditions, and port traffic density, ensuring the digital twin accurately replicates both the ship itself and its operational scenarios. At the hardware level, a distributed sensor network acquires real-time data such as fuel flow, main engine speed, main engine load, exhaust composition, ship speed, and ship position, maintaining second-level synchronization with the twin's parameters to achieve bidirectional data interaction between the physical system and the virtual model.

[0097] The digital twin is a virtual representation of a physical entity, achieving a two-way mapping between the real and virtual worlds through high-precision modeling and real-time data synchronization. Changes in the actual ship's status are reflected in the twin model in real time, and the predictions and simulation results from the virtual model are also fed back to the actual ship system for adjustments. Specifically, parameters such as fuel flow, emission gas concentration, ship speed, and heading, collected in real time by the ship's sensors, are preprocessed by the ship's local processor and transmitted to the cloud via a communication module to update the digital twin's status parameters in real time. When the digital twin simulation yields a new optimization strategy, instructions are sent to the actual ship via the same communication link. Optimization strategies include speed-course joint optimization, achieving a balance between carbon emissions and navigation efficiency through dynamic speed profile planning and heading fine-tuning. Rolling optimization is performed using model predictive control methods. After full-condition safety verification on the virtual simulation platform, the accuracy of the prediction results is determined through confidence intervals, ultimately forming a closed-loop control command that combines safety, economy, and environmental friendliness.

[0098] Real-time data acquisition and transmission module: Hardware configuration: The fuel flow sensor is installed on the fuel delivery pipeline, the exhaust gas analyzer is installed at the exhaust manifold, the GPS / BeiDou dual-mode positioning module is used to obtain the real-time ship position and speed, and the meteorological sensor is arranged on the top of the bridge.

[0099] Data Acquisition and Preprocessing: The acquisition frequency is set to 1Hz-5Hz to ensure that valid data is acquired within seconds. Data filtering and outlier removal are performed locally, using moving average filtering or Kalman filtering to remove high-frequency noise.

[0100] Data transmission: Data is transmitted in real time to a shore-based server via a ship's satellite communication terminal, while some data is cached in the ship's onboard storage to prevent communication interruption.

[0101] The real-time data acquisition and transmission module employs a multi-channel data acquisition architecture to integrate internal shipboard sensors and external data sources onto a unified data bus. Data is transmitted in real-time to the cloud data center via a ship-to-shore communication link, ensuring that optimization algorithms can make decisions based on the latest data. Data compression and outlier detection mechanisms are used to reduce bandwidth consumption and improve data validity, while interpolation is performed to complete missing data, ensuring data continuity and integrity.

[0102] The multi-source data fusion module, based on front-end sensor data synchronization, utilizes extended Kalman filtering for dynamic state estimation. It deeply integrates real-time data from fuel consumption meters, GPS, anemometers, and ship attitude sensors, and employs a spatiotemporal alignment algorithm to unify coordinates and compensate for delays in external weather and sea state forecasts. Specifically, the multi-source data fusion module uses data fusion algorithms to unify data from different sources and accuracy levels, improving the reliability and timeliness of carbon emission prediction.

[0103] The reinforcement learning optimization module includes a navigation strategy optimization model: a deep reinforcement learning framework is deployed on a cloud or local high-performance computing platform, using a digital twin as the environment, ship operating parameters as the state space, and carbon emissions, fuel consumption, sailing time, and safety factor as multi-objective reward functions. Through policy iteration and value function updates, key parameters are optimized to achieve dynamic adaptive optimization.

[0104] During model training, historical navigation data is used for pre-training, followed by online training through real-time interaction with the digital twin. This continuously improves the decision-making ability under different sea states, loads, and weather conditions. Further online training through repeated interactions with the digital twin environment allows the reinforcement learning agent to continuously adjust its decisions based on the reward function, maximizing carbon reduction benefits while balancing fuel economy and navigation safety. Compared to static rules, this approach possesses adaptive and self-learning capabilities, enabling it to handle complex nonlinear multi-objective optimization problems.

[0105] After the model training is completed, the navigation strategy optimization model is deployed on a cloud-based high-performance computing platform. Upon receiving real-time ship operational status data, the optimal decision-making strategy is generated based on predicted carbon emission data over a certain future timeframe, such as carbon footprint change curves. The optimized strategy is first transmitted to a virtual simulation platform for rapid testing. Only after verifying the absence of safety risks is the strategy issued to the ship for execution.

[0106] The virtual simulation testing platform includes a navigation strategy simulation model: Multiple operating scenarios are preset within a digital twin environment, including a library of typical operating scenarios such as stable sea states, severe weather, and port congestion, to conduct full lifecycle simulation tests on different optimization strategies. Through simulation and analysis, the impact of fluctuations in various parameters on carbon emission prediction results is evaluated, and emission reduction strategies that demonstrate stable performance under various operating conditions are selected.

[0107] Before implementing the actual ship verification, the optimization strategy is verified for safety and evaluated for performance using a simulation platform to reduce test risks and economic costs. In the digital twin environment, multiple rounds of simulation tests and sensitivity analysis are conducted by setting different operating conditions and optimization strategies. Sensitivity analysis is used to evaluate the robustness of the strategy under different environmental fluctuations, screen out the lowest-risk solution, and evaluate the impact of each parameter on carbon emissions, thereby providing a scientific basis for actual ship application.

[0108] Visualized Decision Support Module: Through a visual interface on the shore-based monitoring center and the ship's bridge, it displays the current carbon emission level, predicted trends, optimization suggestions, and strategy simulation results in real time. It supports historical data retrospective analysis and report export, facilitating long-term carbon emission management and iterative optimization strategies.

[0109] In an exemplary embodiment, when the ship carbon emission prediction and optimization system is in operation, the modules interact collaboratively through a closed-loop pipeline: the real-time data acquisition module collects initial sensor data and initial environmental data from multiple sources, performs preprocessing such as difference compensation, and then transmits the data to the cloud via ship-to-shore communication; the multi-source data fusion module uses extended Kalman filtering to perform spatiotemporal alignment and state estimation on the initial sensor data and initial environmental data, outputting accurate ship dynamic parameters (i.e., target sensor data) and current environmental data to the digital twin module; the digital twin drives carbon emission prediction based on a reduced-order CFD model and real-time current environmental data, obtains predicted carbon emission data, and synchronizes the predicted carbon emission data to the reinforcement learning module; the reinforcement learning module iterates the strategy using the twin as the environment, generates an optimized strategy (i.e., candidate navigation control parameters), and sends it to the virtual simulation platform; the simulation platform completes the strategy verification of the candidate navigation control parameters through accelerated simulation and sensitivity analysis, and finally feeds back the safe and compliant optimization instructions to both ship and shore ends for execution through the visualization module, while simultaneously feeding the execution data back to the database to form a complete closed loop for continuous optimization.

[0110] In this embodiment, by combining a high-fidelity digital twin with multi-source data, dynamic prediction of ship carbon emissions can be made within seconds, with a prediction error rate lower than that of traditional models, thus improving prediction accuracy. The reinforcement learning algorithm can find the optimal strategy in real time under complex navigation conditions, achieving a comprehensive balance between multiple objectives such as carbon emissions, fuel economy, and navigation time, thereby improving optimization results. The new strategy is first verified through multiple rounds of simulation in a virtual twin environment, avoiding the safety risks and high costs of direct testing on the actual ship, thus improving safety and economy. Through multi-source data fusion and a unified interface, data silos are broken down, enabling ship-shore collaborative optimization and transparent management, thereby improving data utilization. The combination of reinforcement learning and digital twins allows the system to continuously accumulate experience and optimize strategies during operation, sustainably improving emission reduction effects in the long term, and achieving rapid strategy iteration and continuous optimization.

[0111] Secondly, Figure 6 This is a schematic diagram of a ship navigation control system according to an embodiment of this application, such as... Figure 6 As shown, the navigation control system 600 of the above-mentioned ship includes: a data acquisition module 601, an optimization module 602, a strategy simulation module 603, and a strategy control module 604. The device will be described below.

[0112] The data acquisition module 601 is used to acquire real-time target sensor data, navigation data, and current environmental data of the target vessel's current navigation environment. Optimization module 602 includes a carbon emission prediction model and a navigation strategy optimization model; The carbon emission prediction model is used to predict the carbon emission data of the digital twin of the target ship in the current navigation environment based on the target ship's real-time target sensing data, navigation data and current environmental data of the current navigation environment. A navigation strategy optimization model is used to obtain candidate navigation control parameters for a target vessel based on predicted carbon emission data. The strategy simulation module 603 includes a navigation strategy simulation model, used to perform control strategy simulation on a digital twin based on candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments. This model obtains candidate performance index parameters, safety assessment parameters, and performance assessment parameters for the candidate navigation control parameters under multiple preset navigation environments, as well as baseline performance index parameters for the baseline navigation control parameters. Based on the candidate performance index parameters, baseline performance index parameters, safety assessment parameters, and performance assessment parameters, the module determines the control strategy verification results corresponding to the candidate navigation control parameters. The strategy control module 604 is used to determine the target navigation control parameters from the candidate navigation control parameters based on the control strategy verification results, and control the target ship to operate according to the target navigation control parameters.

[0113] In an exemplary embodiment, the ship's navigation control system 600 is further configured to acquire real-time initial sensing data of the target ship and initial environmental data of the current navigation environment; perform spatiotemporal alignment of the initial environmental data and the initial sensing data to obtain intermediate sensing data and current environmental data of the current navigation environment; and perform state estimation on the intermediate sensing data to obtain target sensing data.

[0114] In an exemplary embodiment, the ship's navigation control system 600 is further configured to acquire an initial digital twin corresponding to the target ship; based on the target ship's real-time target sensing data and the current environmental data of the current navigation environment, the initial digital twin is updated with an operational status to obtain a digital twin corresponding to the target ship.

[0115] In an exemplary embodiment, the optimization module 602 is further configured to determine the target power parameters corresponding to the target vessel based on the target sensing data and the current environmental data; and to predict the predicted carbon emission data of the digital twin of the target vessel in the current navigation environment based on the target sensing data, navigation data, current environmental data and target power parameters.

[0116] In an exemplary embodiment, the optimization module 602 is further configured to determine multiple reference power parameters in a preset power database based on target sensing data and current environmental data; and to perform linear interpolation on the multiple reference power parameters to obtain the target power parameters corresponding to the target ship.

[0117] In an exemplary embodiment, the optimization module 602 is further configured to input the predicted carbon emission data into the navigation strategy optimization model, and establish a target state space based on the predicted carbon emission data and the digital twin through the navigation strategy optimization model; and generate candidate navigation control parameters that meet the preset target constraints based on the target state space and the preset action space.

[0118] In an exemplary embodiment, the strategy simulation module 603 is further configured to determine the computational granularity of the navigation strategy simulation model in the time dimension based on the original simulation duration corresponding to the candidate navigation control parameters and the preset target simulation duration; input the candidate navigation control parameters, preset baseline navigation control parameters, navigation data and preset environment data of multiple preset navigation environments into the navigation strategy simulation model, and perform control strategy simulation on the digital twin according to the computational granularity through the navigation strategy simulation model.

[0119] In an exemplary embodiment, the strategy simulation module 603 is further configured to obtain candidate performance index parameters corresponding to candidate navigation control parameters under each preset navigation environment; based on the preset index safety threshold and multiple candidate performance index parameters, calculate the index failure probability, obtain the safety assessment parameters corresponding to the candidate navigation control parameters based on the index failure probability; and calculate the dispersion of multiple candidate performance index parameters to obtain the performance assessment parameters corresponding to the candidate navigation control parameters.

[0120] In an exemplary embodiment, the strategy simulation module 603 is further configured to compare the candidate performance indicator parameters with the benchmark performance indicator parameters to obtain the indicator gain corresponding to the candidate performance indicator parameters; when the indicator gain is greater than a preset threshold, the control strategy verification result corresponding to the candidate navigation control parameters is determined based on the safety assessment parameters and performance assessment parameters.

[0121] The various modules in the navigation control of the aforementioned ships can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0122] Thirdly, this embodiment provides an electronic device, including a memory and a processor. The memory stores computer instructions, and when the computer instructions are executed by the processor, they implement the method of any of the above embodiments.

[0123] In one embodiment, this embodiment also provides an electronic device, which may be a server, and its internal structure diagram may be as follows. Figure 7 As shown, this electronic device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer instructions, and a database. The internal memory provides the environment for the operation of the operating system and computer instructions stored in the non-volatile storage media. The database stores data involved in business data processing methods. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer instructions are executed by the processor, a navigation control method for a ship is implemented.

[0124] Those skilled in the art will understand that Figure 7The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0125] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions thereon, which are loaded by a processor to execute the arrangements in any of the methods described above. In embodiments of this application, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0126] Fifthly, embodiments of this application provide a computer program product, including a computer program or instructions, which are executed by a processor to implement the steps of any of the methods described above.

[0127] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0128] The foregoing has provided a detailed description of a ship navigation control, electronic device, and computer-readable storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for controlling the navigation of a ship, characterized in that, include: Based on the target vessel's real-time target sensing data, navigation data, and current environmental data of the current navigation environment, predict the predicted carbon emission data of the target vessel's digital twin in the current navigation environment. Based on the predicted carbon emission data, candidate navigation control parameters corresponding to the target vessel are obtained; Based on the candidate navigation control parameters, the preset baseline navigation control parameters, the navigation data, and the preset environment data of multiple preset navigation environments, the control strategy simulation is performed on the digital twin to obtain the candidate performance index parameters, safety assessment parameters, and performance assessment parameters corresponding to the candidate navigation control parameters under multiple preset navigation environments, as well as the baseline performance index parameters corresponding to the baseline navigation control parameters. Based on the candidate performance indicator parameters, the benchmark performance indicator parameters, the safety assessment parameters, and the performance assessment parameters, the control strategy verification results corresponding to the candidate navigation control parameters are determined; Based on the verification results of the control strategy, the target navigation control parameters are determined from the candidate navigation control parameters, and the target vessel is controlled to operate according to the target navigation control parameters.

2. The method according to claim 1, characterized in that, The method further includes: Acquire the target vessel's real-time initial sensor data and the initial environmental data of the current navigation environment; The initial environmental data and the initial sensor data are spatiotemporally aligned to obtain intermediate sensor data and the current environmental data of the current navigation environment; The intermediate sensing data is used to perform state estimation to obtain the target sensing data.

3. The method according to claim 1, characterized in that, The method further includes: Obtain the initial digital twin of the target vessel; Based on the real-time target sensing data of the target vessel and the current environmental data of the current navigation environment, the initial digital twin is updated in terms of its operating status to obtain the digital twin corresponding to the target vessel.

4. The method according to claim 1, characterized in that, The method of predicting the predicted carbon emissions of the digital twin of the target vessel under the current navigation environment, based on real-time target sensing data, navigation data, and current environmental data of the target vessel, includes: Based on the target sensing data and the current environmental data, the target dynamic parameters corresponding to the target vessel are determined; Based on the target sensing data, the navigation data, the current environmental data, and the target dynamic parameters, the predicted carbon emission data of the digital twin of the target ship under the current navigation environment is predicted.

5. The method according to claim 4, characterized in that, The step of determining the target dynamic parameters corresponding to the target vessel based on the target sensing data and the current environmental data includes: Based on the target sensing data and the current environment data, multiple reference power parameters are determined in a preset power database; Linear interpolation is performed on the multiple reference dynamic parameters to obtain the target dynamic parameters corresponding to the target ship.

6. The method according to claim 1, characterized in that, The process of obtaining candidate navigation control parameters for the target vessel based on the predicted carbon emission data includes: The predicted carbon emission data is input into the navigation strategy optimization model, and the navigation strategy optimization model establishes a target state space based on the predicted carbon emission data and the digital twin. Based on the target state space and the preset action space, candidate navigation control parameters that satisfy the preset target constraints are generated.

7. The method according to claim 1, characterized in that, Based on the candidate navigation control parameters, preset baseline navigation control parameters, the navigation data, and preset environment data of multiple preset navigation environments, the control strategy simulation of the digital twin is performed, including: Based on the original simulation duration and the preset target simulation duration corresponding to the candidate navigation control parameters, the computational granularity of the navigation strategy simulation model in the time dimension is determined. The candidate navigation control parameters, preset baseline navigation control parameters, navigation data, and preset environment data of multiple preset navigation environments are input into the navigation strategy simulation model. The navigation strategy simulation model then performs control strategy simulation on the digital twin according to the calculated granularity.

8. The method according to claim 1, characterized in that, The control strategy is simulated on the digital twin to obtain safety evaluation parameters and performance evaluation parameters corresponding to the candidate navigation control parameters under multiple preset navigation environments, including: Obtain the candidate performance index parameters corresponding to the candidate navigation control parameters under each preset navigation environment; Based on the preset safety threshold and multiple candidate performance indicator parameters, the failure probability of the indicators is statistically calculated, and the safety assessment parameters corresponding to the candidate navigation control parameters are obtained based on the failure probability of the indicators. The dispersion of multiple candidate performance index parameters is calculated to obtain the performance evaluation parameters corresponding to the candidate navigation control parameters.

9. The method according to claim 1, characterized in that, The step of determining the control strategy verification result corresponding to the candidate navigation control parameters based on the candidate performance indicator parameters, the benchmark performance indicator parameters, the safety assessment parameters, and the performance assessment parameters includes: The candidate performance indicator parameters are compared with the benchmark performance indicator parameters to obtain the indicator gain corresponding to the candidate performance indicator parameters; When the index gain is greater than a preset threshold, the control strategy verification result corresponding to the candidate navigation control parameter is determined based on the safety assessment parameter and the performance assessment parameter.

10. A navigation control system for a ship, characterized in that, include: The data acquisition module is used to acquire real-time target sensor data, navigation data, and current environmental data of the target vessel's current navigation environment. The optimization module includes a carbon emission prediction model and a navigation strategy optimization model; The carbon emission prediction model is used to predict the predicted carbon emission data of the digital twin of the target ship in the current navigation environment based on the target ship's real-time target sensing data, navigation data and current environmental data of the current navigation environment. The navigation strategy optimization model is used to obtain candidate navigation control parameters for the target vessel based on the predicted carbon emission data. The strategy simulation module includes a navigation strategy simulation model; The navigation strategy simulation model is used to simulate the control strategy of the digital twin based on the candidate navigation control parameters, preset benchmark navigation control parameters, the navigation data, and preset environment data of multiple preset navigation environments. This simulation yields candidate performance index parameters, safety assessment parameters, and performance assessment parameters for the candidate navigation control parameters under multiple preset navigation environments, as well as benchmark performance index parameters for the benchmark navigation control parameters. Based on the candidate performance index parameters, the benchmark performance index parameters, the safety assessment parameters, and the performance assessment parameters, the model determines the control strategy verification result corresponding to the candidate navigation control parameters. The strategy control module is used to determine the target navigation control parameters from the candidate navigation control parameters based on the control strategy verification results, and control the target vessel to operate according to the target navigation control parameters.

11. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 9.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method of any one of claims 1 to 9.

Images