Ship management method and system based on digital twin scenario construction

Abstract

The present invention discloses a ship management method and system based on digital twin scene construction. The method includes: acquiring multi-source scene data of the target waters in real time and constructing a corresponding digital twin model for ship management; conducting an integrity assessment on the constructed digital twin model and compensating for defects based on the assessment results to obtain an optimized digital twin model; further collecting ship density and weather condition data for each sub-area of ​​the target waters, constructing a network transmission delay prediction model, analyzing relevant data based on the model, and generating a network delay coefficient distribution map for the target waters; determining the real-time update lag of each sub-unit of the optimization model based on the delay distribution map, and correcting the ship position information in the digital twin model accordingly. This method can significantly improve the real-time performance and accuracy of the digital twin model and enhance the efficiency and safety of ship operation management in complex water environments.

Description

Technical Field

[0001] The present invention relates to the technical field of ship management, and in particular to a ship management method and system based on digital twin scenarios. Background Art

[0002] With the continuous growth of ship traffic density and the increasing complexity of water environments, traditional ship management methods face significant challenges in information perception, risk prediction, and dispatch coordination. Currently, ship management mainly relies on methods such as automatic identification systems (AIS), radar monitoring, video surveillance, and manual intervention. These methods have significant limitations in data integration, model building, dynamic tracking, and intelligent analysis, and cannot meet the demand for high-precision and real-time ship management in complex and changing water environments.

[0003] Digital twin technology, an advanced approach that integrates physical entities with virtual models, has been gradually applied in various fields, including industrial manufacturing, smart cities, and transportation. By acquiring multi-source heterogeneous data in real time and constructing highly realistic digital models, it enables comprehensive perception, prediction, and control of physical systems, providing an effective means to improve system operational efficiency and safety. However, the application of digital twins in ship management is still in its infancy, with a lack of mature model construction methods and system architectures. In particular, there are technical gaps in key areas such as model integrity assessment, data lag compensation, network delay prediction, and dynamic position correction.

[0004] While some existing studies have proposed preliminary solutions based on 3D scene reconstruction or data fusion, they generally lack comprehensive consideration of data spatiotemporal consistency, model update lags, and the dynamic changes in complex hydrological environments. This results in insufficient accuracy, real-time performance, and robustness in the constructed ship management models, making it difficult to support the practical needs of large-scale, multi-vessel, and multi-environmental collaborative management. Therefore, there is an urgent need to propose a ship management method and system based on digital twin scenarios that can achieve dynamic reconstruction of water scenes, automatic compensation for model defects, intelligent prediction of network delays, and high-precision correction of ship positions, comprehensively improving the intelligent and refined level of ship management. Summary of the Invention

[0005] In order to solve at least one of the above technical problems, the present invention proposes a ship management method and system based on digital twin scenario construction.

[0006] The first aspect of the present invention provides a ship management method based on digital twin scenario construction, comprising:

[0007] Acquire multi-source scene data of the target waters in real time, and construct a digital twin model of ship management in the target waters based on the multi-source scene data;

[0008] Performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result;

[0009] Compensating for defects in the ship management digital twin model according to the evaluation results to obtain an optimized digital twin model;

[0010] Obtaining ship density and weather condition data for each sub-area in the target waters, as well as historical ship density and weather condition data, to construct a network transmission delay prediction model; analyzing the ship density and weather condition data based on the network transmission delay prediction model to construct a network delay coefficient distribution map for the target waters;

[0011] The real-time update lag of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update lag.

[0012] In this solution, the multi-source scene data of the target waters is acquired in real time, and a digital twin model of ship management in the target waters is constructed based on the multi-source scene data, specifically:

[0013] Acquire multi-source scene data of the target waters in real time, including geographical environment data, hydrological data, and ship AIS data of the target waters;

[0014] Constructing a three-dimensional geographic model of the target water area based on the geographic environment data, and mapping the hydrological data to the three-dimensional geographic model to generate an initial water area model integrating geographic and hydrological features;

[0015] A real-time motion model of the ship is constructed based on the ship AIS data, the real-time motion model of the ship is calibrated and matched with the spatial coordinate system of the initial water area model, and the calibrated real-time motion model of the ship is coupled with the initial water area model to construct a digital twin model of ship management in the target water area.

[0016] In this solution, the ship management digital twin model is evaluated for its integrity, and the evaluation results are as follows:

[0017] Discretize the surface of the ship management digital twin model according to a preset grid size, construct N model subunits, tile the subunits, and map them onto a two-dimensional plane to construct a two-dimensional display diagram of the ship management digital twin;

[0018] Acquire historical multi-source scenario data of a preset time length in the target waters, build a test data connection channel for the ship management digital twin model, and import the historical multi-source scenario data into the ship management digital twin model for simulation operation according to the test data connection channel;

[0019] Mapping the simulated operation of the ship management digital twin simulation to the two-dimensional display diagram, integrating the two-dimensional display diagrams during the simulation operation according to a preset frame rate, and constructing a two-dimensional dynamic display diagram of the simulation operation;

[0020] The ViBe algorithm is introduced, and a static background model is established based on the initial frame image of the two-dimensional dynamic display image according to the ViBe algorithm. The subsequent frame sequence of the two-dimensional dynamic display image is input into the static background model frame by frame. The difference between the pixel value of the current frame and the pixel value of the initial frame image is calculated for each pixel point, and whether the pixel belongs to the dynamic foreground is determined according to a preset difference threshold, thereby forming a dynamic foreground pixel set;

[0021] Decomposing the two-dimensional dynamic image frame by frame into a static background layer and a dynamic foreground layer according to the dynamic foreground pixel set;

[0022] Performing image segmentation processing on the static background layer to extract the defect area outline, calculating the static background vacancy rate based on the mapping relationship between the defect area outline and the model sub-unit, performing multi-frame motion trajectory tracking on the dynamic foreground layer to generate a dynamic connectivity topology map, and calculating the dynamic trajectory fracture rate based on the dynamic connectivity topology map, wherein the motion trajectory includes the ship navigation trajectory and the hydrological flow trajectory;

[0023] The construction integrity of the ship management digital twin model is evaluated based on the static background vacancy rate and the dynamic trajectory fracture rate to obtain an evaluation result. The evaluation result includes the construction completeness rate, the model construction vacancy position and the model defect items. The model defect items include the vacancy in the model three-dimensional scene construction and the inconsistent data update frequency of each sub-unit of the model.

[0024] In this solution, the ship management digital twin model is compensated for defects based on the evaluation results to obtain an optimized digital twin model, specifically:

[0025] According to the evaluation results, subunits with gaps in three-dimensional scene construction in the ship management digital twin model are calibrated as scene-missing subunits, and subunits with inconsistent data update frequencies are calibrated as subunits with update frequencies to be optimized;

[0026] For the vacant sub-units of the scene, historical multi-source scene data of the target waters and environmental characteristic data of adjacent sub-units are obtained, and the geographical environmental data of the vacant sub-units are supplemented based on a spatiotemporal interpolation algorithm to generate supplementary data. The vacant sub-units of the scene are filled according to the supplementary data, which is the first defect compensation method;

[0027] For the subunit whose update frequency is to be optimized, extract the average of the data update frequencies of adjacent subunits, set the update frequency reference value of the subunit to be optimized according to the average, and analyze the data update delay of the subunit within a preset time window using a sliding window algorithm;

[0028] When the update delay exceeds the allowable fluctuation range of the reference value, the timing characteristics of the ship position data and hydrological data in the sub-unit are aligned and compensated based on the dynamic time warping algorithm, and a data update synchronization compensation channel with the adjacent sub-unit is established, which is the second defect compensation method;

[0029] Defect compensation is performed on the ship management digital twin model according to the first defect compensation method and the second defect compensation method to obtain an optimized digital twin model.

[0030] In this solution, the ship density and weather condition data of each sub-area in the target waters, as well as the historical ship density and weather condition data, are obtained to construct a network transmission delay prediction model. The ship density and weather condition data are analyzed based on the network transmission delay prediction model to construct a network delay coefficient distribution map for the target waters, specifically:

[0031] Divide the target waters into N sub-areas, obtain historical ship density and historical weather data for each sub-area over a preset time period, the historical weather data including wind speed, rainfall intensity, and visibility parameters, and obtain historical network transmission delay data for each sub-area over a preset time period;

[0032] Aligning the historical network transmission delay data, historical ship density, and historical weather condition data in time series to construct a data analysis matrix;

[0033] performing a correlation analysis on the data analysis matrix based on the Pearson correlation coefficient to determine the impact of different ship densities and different weather conditions on network transmission delay, and obtaining impact data, the impact data including network delay coefficients corresponding to different ship densities and weather conditions;

[0034] constructing an impact feature vector based on the ship density and weather conditions according to the impact data, using the impact feature vector as an input feature of the prediction model, and constructing a training sample set with the network delay coefficient corresponding to the impact feature vector as an output target;

[0035] A decision tree algorithm is used to select feature splitting nodes for the training sample set. The weight of the input feature's impact on network delay is calculated based on the information gain rate. Feature splitting rules are constructed based on the weight ranking results. The branch structure of the decision tree model is generated through recursive splitting. A pre-pruning strategy is used to optimize the model complexity to form a network transmission delay prediction model.

[0036] Obtain ship density and weather condition data for each sub-area in the target waters, import the ship density and weather condition data into the network transmission delay prediction model to predict the network delay of each sub-area, and construct a network delay coefficient distribution map for the target waters.

[0037] In this solution, the real-time update hysteresis of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update hysteresis, specifically:

[0038] Mapping the network delay coefficient distribution map to each subunit in the optimized digital twin model, determining the network delay coefficient of each subunit, and obtaining data transmission volume information of each subunit;

[0039] Calculate the real-time update lag of each sub-unit during the data transmission process according to the network delay coefficient and data transmission volume information of each sub-unit;

[0040] Acquire the real-time position data of the ship in each subunit of the optimized digital twin model, compare the timestamp of the real-time position data with the current system time, and obtain the actual delay duration of the ship position data; when the real-time update lag is greater than the actual delay duration, determine that the subunit has a ship position update lag;

[0041] If a ship update lag occurs, a ship motion prediction model is constructed based on historical ship navigation trajectory data, the real-time position data of the ship is input into the ship motion prediction model, a predicted navigation trajectory of the ship within the time span of the real-time update lag is generated, and the predicted position of the ship is obtained based on the predicted navigation trajectory;

[0042] The ship position of the optimized digital twin model is corrected according to the predicted position, and the corrected ship position is monitored and managed in real time.

[0043] A second aspect of the present invention further provides a ship management system based on a digital twin scenario, the system comprising: a memory and a processor, wherein the memory includes a ship management method program based on the digital twin scenario, and when the ship management method program based on the digital twin scenario is executed by the processor, the following steps are implemented:

[0044] Acquire multi-source scene data of the target waters in real time, and construct a digital twin model of ship management in the target waters based on the multi-source scene data;

[0045] Performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result;

[0046] Compensating for defects in the ship management digital twin model according to the evaluation results to obtain an optimized digital twin model;

[0047] Obtaining ship density and weather condition data for each sub-area in the target waters, as well as historical ship density and weather condition data, to construct a network transmission delay prediction model; analyzing the ship density and weather condition data based on the network transmission delay prediction model to construct a network delay coefficient distribution map for the target waters;

[0048] The real-time update lag of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update lag.

[0049] The present invention discloses a ship management method and system based on digital twin scene construction. The method includes: acquiring multi-source scene data of the target waters in real time and constructing a corresponding digital twin model for ship management; conducting an integrity assessment on the constructed digital twin model and compensating for defects based on the assessment results to obtain an optimized digital twin model; further collecting ship density and weather condition data for each sub-area of ​​the target waters, constructing a network transmission delay prediction model, analyzing relevant data based on the model, and generating a network delay coefficient distribution map for the target waters; determining the real-time update lag of each sub-unit of the optimization model based on the delay distribution map, and correcting the ship position information in the digital twin model accordingly. This method can significantly improve the real-time performance and accuracy of the digital twin model and enhance the efficiency and safety of ship operation management in complex water environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flow chart of a ship management method based on digital twin scenario construction according to the present invention is shown;

[0051] Figure 2 The flowchart of the present invention for constructing a digital twin model for ship management is shown;

[0052] Figure 3 A flow chart of the present invention for correcting the ship position of the optimized digital twin model is shown;

[0053] Figure 4 A block diagram of a ship management system constructed based on a digital twin scenario of the present invention is shown. DETAILED DESCRIPTION

[0054] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0056] Figure 1 A flow chart of a ship management method based on a digital twin scenario is shown in the present invention.

[0057] like Figure 1 As shown, the first aspect of the present invention provides a ship management method based on digital twin scene construction, including:

[0058] S102, acquiring multi-source scene data of the target waters in real time, and constructing a digital twin model of ship management in the target waters based on the multi-source scene data;

[0059] S104, performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result;

[0060] S106, performing defect compensation on the ship management digital twin model according to the evaluation result to obtain an optimized digital twin model;

[0061] S108, obtaining ship density and weather condition data and historical ship density and weather condition data for each sub-area in the target waters, constructing a network transmission delay prediction model, analyzing the ship density and weather condition data based on the network transmission delay prediction model, and constructing a network delay coefficient distribution map for the target waters;

[0062] S110, determining the real-time update hysteresis of each subunit of the optimized digital twin model according to the network delay coefficient distribution map, and correcting the ship position of the optimized digital twin model according to the update hysteresis.

[0063] It should be noted that by acquiring multi-source scene data of the target waters in real time, it is possible to fully perceive the water environment and the operating status of the ship, realize accurate fusion modeling of geographical, hydrological and ship dynamic information, and enhance the true restoration of the model; by conducting integrity assessment on the constructed digital twin model, the structural defects and data update inconsistencies in the model can be discovered in time, and the model quality monitoring capability can be improved; further defect compensation based on the assessment results can effectively fill the vacant areas of the three-dimensional scene and unify the model data update frequency, thereby building a more accurate and stable optimized digital twin model; using ship density and weather data to construct a network transmission delay prediction model, and based on this, generate a network delay coefficient distribution map, which can realize the prediction of network environment changes and ensure that each subunit of the model still has high data synchronization and responsiveness under complex communication conditions; finally, the real-time update lag of each subunit is corrected according to the delay distribution map to ensure the accurate repair and real-time update of the ship position information, thereby significantly improving the accuracy of ship monitoring and scheduling.

[0064] Figure 2 A flow chart of the present invention for constructing a digital twin model for ship management is shown.

[0065] According to an embodiment of the present invention, the real-time acquisition of multi-source scene data of the target waters and the construction of a digital twin model of ship management in the target waters based on the multi-source scene data are specifically as follows:

[0066] S202, acquiring multi-source scene data of the target waters in real time, wherein the multi-source scene data includes geographical environment data, hydrological data, and ship AIS data of the target waters;

[0067] S204, constructing a three-dimensional geographic model of the target water area based on the geographic environment data, and mapping the hydrological data to the three-dimensional geographic model to generate an initial water area model integrating geographic and hydrological features;

[0068] S206: Construct a real-time motion model of the ship based on the ship AIS data, calibrate and match the real-time motion model of the ship with the spatial coordinate system of the initial water area model, couple the calibrated real-time motion model of the ship with the initial water area model, and construct a digital twin model of ship management in the target water area.

[0069] It should be noted that by acquiring multi-source scene data of the target waters in real time, including geographic environment data, hydrological data, and ship AIS data, comprehensive coverage and timely perception of the static and dynamic information of the waters are ensured. On this basis, by constructing a three-dimensional geographic model and accurately mapping the hydrological data to the model, the fusion expression of topography and flow characteristics is achieved, effectively restoring the real physical environment of the target waters. Then, a real-time ship motion model is constructed based on the ship AIS data, and it is calibrated and matched with the spatial coordinate system of the initial water model with high precision, so that the ship's motion trajectory can be accurately located and synchronously displayed in the digital model. Finally, through model coupling, the static environment and dynamic targets are seamlessly integrated, and a highly realistic, dynamic, and interactive digital twin model for ship management is constructed. The geographic environment data includes waterway boundaries, three-dimensional port structures, and seabed topography grid data; the hydrological data includes velocity fields, tidal cycles, and water temperature stratification information; and the ship AIS data includes the ship's dynamic position, speed, heading, and ship size parameters.

[0070] According to an embodiment of the present invention, the ship management digital twin model is subjected to a construction integrity assessment to obtain an assessment result, specifically:

[0071] Discretize the surface of the ship management digital twin model according to a preset grid size, construct N model subunits, tile the subunits, and map them onto a two-dimensional plane to construct a two-dimensional display diagram of the ship management digital twin;

[0072] Acquire historical multi-source scenario data of a preset time length in the target waters, build a test data connection channel for the ship management digital twin model, and import the historical multi-source scenario data into the ship management digital twin model for simulation operation according to the test data connection channel;

[0073] Mapping the simulated operation of the ship management digital twin simulation to the two-dimensional display diagram, integrating the two-dimensional display diagrams during the simulation operation according to a preset frame rate, and constructing a two-dimensional dynamic display diagram of the simulation operation;

[0074] The ViBe algorithm is introduced, and a static background model is established based on the initial frame image of the two-dimensional dynamic display image according to the ViBe algorithm. The subsequent frame sequence of the two-dimensional dynamic display image is input into the static background model frame by frame. The difference between the pixel value of the current frame and the pixel value of the initial frame image is calculated for each pixel point, and whether the pixel belongs to the dynamic foreground is determined according to a preset difference threshold, thereby forming a dynamic foreground pixel set;

[0075] Decomposing the two-dimensional dynamic image frame by frame into a static background layer and a dynamic foreground layer according to the dynamic foreground pixel set;

[0076] Performing image segmentation processing on the static background layer to extract the defect area outline, calculating the static background vacancy rate based on the mapping relationship between the defect area outline and the model sub-unit, performing multi-frame motion trajectory tracking on the dynamic foreground layer to generate a dynamic connectivity topology map, and calculating the dynamic trajectory fracture rate based on the dynamic connectivity topology map, wherein the motion trajectory includes the ship navigation trajectory and the hydrological flow trajectory;

[0077] The construction integrity of the ship management digital twin model is evaluated based on the static background vacancy rate and the dynamic trajectory fracture rate to obtain an evaluation result. The evaluation result includes the construction completeness rate, the model construction vacancy position and the model defect items. The model defect items include the vacancy in the model three-dimensional scene construction and the inconsistent data update frequency of each sub-unit of the model.

[0078] It should be noted that in digital twin models for ship management, due to the highly heterogeneous nature of data sources, frequent dynamic scene changes, and the high real-time requirements of system operation, the construction process of digital twin models is prone to problems such as incomplete scene restoration, missing local model data, and discontinuous dynamic information representation. Therefore, by simulating the ship management digital twin model and extracting the corresponding 2D dynamic graphics, the dynamic evolution of ships and hydrological elements in the target waters can be fully reproduced without affecting the actual operating environment, facilitating a systematic analysis of the model's performance. The 2D dynamic graphics generated during the simulation are processed using the ViBe algorithm to accurately extract the static and dynamic components of the dynamic graphics. Segmenting the dynamic graphics into a static background layer and a dynamic foreground layer not only helps clearly identify areas of missing scene construction in the model but also precisely locates gaps and loopholes in the model's 3D construction. Furthermore, multi-frame motion trajectory tracking of the dynamic foreground layer and the construction of a dynamic connectivity topology map enable the identification of the continuity of ship navigation and hydrological flow trajectories. Trajectory connectivity analysis then calculates the dynamic trajectory fracture rate, which is used to assess the continuity and stability of the model during dynamic data updates. Ultimately, the static background vacancy rate and dynamic trajectory breakage rate can not only quantitatively reflect the level of model integrity, but also specifically identify the types and locations of defects in model construction, effectively improving the operational reliability of the digital twin system. The ViBe (Visual Background Extractor) algorithm is a pixel-level background modeling and foreground detection algorithm. The historical multi-source scene data mentioned above only includes historical hydrographic data and historical ship AIS data. Because these data are used to test the constructed digital twin model, and because the acquired historical multi-source scene data is complete and free of data gaps, testing the digital twin model with historical multi-source scene data only reveals problems with the model itself, not defects in the acquired data. When a dynamic trajectory break occurs, it indicates a discontinuity in the temporal continuity of the ship or hydrographic data. This is usually due to inconsistent data update frequencies among the subunits in the model, resulting in some areas failing to receive or process data in a timely manner, causing trajectory incoherence.

[0079] According to an embodiment of the present invention, the defect compensation of the ship management digital twin model is performed according to the evaluation result to obtain an optimized digital twin model, specifically:

[0080] According to the evaluation results, subunits with gaps in three-dimensional scene construction in the ship management digital twin model are calibrated as scene-missing subunits, and subunits with inconsistent data update frequencies are calibrated as subunits with update frequencies to be optimized;

[0081] For the vacant sub-units of the scene, historical multi-source scene data of the target waters and environmental characteristic data of adjacent sub-units are obtained, and the geographical environmental data of the vacant sub-units are supplemented based on a spatiotemporal interpolation algorithm to generate supplementary data. The vacant sub-units of the scene are filled according to the supplementary data, which is the first defect compensation method;

[0082] For the subunit whose update frequency is to be optimized, extract the average of the data update frequencies of adjacent subunits, set the update frequency reference value of the subunit to be optimized according to the average, and analyze the data update delay of the subunit within a preset time window using a sliding window algorithm;

[0083] When the update delay exceeds the allowable fluctuation range of the reference value, the timing characteristics of the ship position data and hydrological data in the sub-unit are aligned and compensated based on the dynamic time warping algorithm, and a data update synchronization compensation channel with the adjacent sub-unit is established, which is the second defect compensation method;

[0084] Defect compensation is performed on the ship management digital twin model according to the first defect compensation method and the second defect compensation method to obtain an optimized digital twin model.

[0085] It should be noted that by compensating for defects in the ship management digital twin model, the model's integrity and timeliness can be effectively improved. The first defect compensation method uses a spatiotemporal interpolation algorithm to fill in the geographic environmental data of missing sub-units in the scene, achieving accurate restoration and repair of gaps in the three-dimensional scene construction, enhancing the model's spatial continuity and expression integrity. The second defect compensation method uses a dynamic time warping algorithm to align and synchronize the time series data of sub-units whose update frequencies are to be optimized, achieving consistency in the data update frequency between sub-units and improving the model's real-time dynamic response capabilities, thereby constructing a more accurate and stable optimized digital twin model. The adjacent sub-units are those within the preset range of the defective sub-unit. The environmental feature data of the adjacent sub-units includes features such as topography, seafloor topography, coastline, island distribution, and water depth. When constructing the data update synchronization compensation channel, the dynamic time warping (DTW) algorithm is first used to align the data time series of the target sub-unit with the adjacent sub-units. The minimum path mapping of the two data sequences is calculated, and the delayed data is nonlinearly stretched or compressed to eliminate the time phase difference. Subsequently, a distributed data buffer is deployed between adjacent subunits, encoding the compensated data with a unified timestamp and dynamically adjusting the injection rate based on the time weight of the sliding window. During channel operation, a master-slave clock synchronization protocol is employed, with the average update frequency of adjacent subunits serving as the reference period. An adaptive threshold trigger is set at the buffer entry. When data delay exceeds the window tolerance, the trigger activates the LSTM network to predict missing data, generating a compensation frame to fill the gap. A frequency correction signal is simultaneously sent to the target subunit, and the sensor sampling interval is adjusted via a PID controller to achieve frequency lock and content consistency for data updates across subunits. Its core principle is to eliminate phase differences through timing alignment, combining prediction compensation with feedback control to achieve adaptive frequency matching, forming a closed-loop synchronization channel and ensuring coordinated spatiotemporal updates of multi-source data.

[0086] According to an embodiment of the present invention, the process of obtaining ship density and weather condition data and historical ship density and weather condition data for each sub-area in the target waters, constructing a network transmission delay prediction model, and analyzing the ship density and weather condition data based on the network transmission delay prediction model to construct a network delay coefficient distribution map for the target waters is as follows:

[0087] Divide the target waters into N sub-areas, obtain historical ship density and historical weather data for each sub-area over a preset time period, the historical weather data including wind speed, rainfall intensity, and visibility parameters, and obtain historical network transmission delay data for each sub-area over a preset time period;

[0088] Aligning the historical network transmission delay data, historical ship density, and historical weather condition data in time series to construct a data analysis matrix;

[0089] performing a correlation analysis on the data analysis matrix based on the Pearson correlation coefficient to determine the impact of different ship densities and different weather conditions on network transmission delay, and obtaining impact data, the impact data including network delay coefficients corresponding to different ship densities and weather conditions;

[0090] constructing an impact feature vector based on the ship density and weather conditions according to the impact data, using the impact feature vector as an input feature of the prediction model, and constructing a training sample set with the network delay coefficient corresponding to the impact feature vector as an output target;

[0091] A decision tree algorithm is used to select feature splitting nodes for the training sample set. The weight of the input feature's impact on network delay is calculated based on the information gain rate. Feature splitting rules are constructed based on the weight ranking results. The branch structure of the decision tree model is generated through recursive splitting. A pre-pruning strategy is used to optimize the model complexity to form a network transmission delay prediction model.

[0092] Obtain ship density and weather condition data for each sub-area in the target waters, import the ship density and weather condition data into the network transmission delay prediction model to predict the network delay of each sub-area, and construct a network delay coefficient distribution map for the target waters.

[0093] It should be noted that in the target waters, the ship density and weather conditions will affect the transmission of multi-source scenario data of the target waters to the constructed ship management digital twin model; therefore, the Pearson correlation coefficient is used to analyze the correlation between historical ship density and historical weather conditions data and network transmission delay, and the impact of different ship densities and different weather conditions on network transmission delay is determined. Then, a network transmission delay prediction model is constructed through the decision tree algorithm. The model analyzes the correlation between historical ship density and weather data and network delay, establishes an accurate prediction mechanism, and can automatically identify the key factors affecting network delay under different environmental conditions, such as the dominant influencing factors in ship-dense areas or severe weather areas; secondly, feature selection based on information gain rate ensures that the model prioritizes the most predictive features for decision-making. The branch structure constructed by recursive segmentation can capture the complex nonlinear effects of ship density and various weather parameters on network delay; the pre-pruning strategy is used to effectively control the complexity of the model, avoid overfitting while maintaining the generalization ability of unknown data; the final generated network delay coefficient distribution map intuitively displays the differences in communication quality of each sub-area of ​​the target waters.

[0094] Figure 3 A flow chart of the present invention for correcting the ship position of the optimized digital twin model is shown.

[0095] According to an embodiment of the present invention, determining the real-time update hysteresis of each subunit of the optimized digital twin model according to the network delay coefficient distribution map, and correcting the ship position of the optimized digital twin model according to the update hysteresis, specifically:

[0096] S302: Map the network delay coefficient distribution map to each subunit in the optimized digital twin model, determine the network delay coefficient of each subunit, and obtain data transmission volume information of each subunit;

[0097] S304, calculating the real-time update hysteresis of each sub-unit during the data transmission process according to the network delay coefficient and data transmission volume information of each sub-unit;

[0098] S306: Acquire the real-time position data of the ship in each subunit of the optimized digital twin model, compare the timestamp of the real-time position data with the current system time, and obtain the actual delay duration of the ship position data. When the real-time update lag is greater than the actual delay duration, determine that the subunit has a ship position update lag;

[0099] S308, if a ship update lag occurs, constructing a ship motion prediction model based on historical ship navigation trajectory data, inputting the ship's real-time position data into the ship motion prediction model, generating a predicted navigation trajectory of the ship within the time span of the real-time update lag, and obtaining a predicted position of the ship based on the predicted navigation trajectory;

[0100] S310: Correct the ship position of the optimized digital twin model according to the predicted position, and monitor and manage the corrected ship position in real time.

[0101] It should be noted that digital twin systems for ship management often face data update asynchrony due to network transmission delays. This is particularly true in densely populated areas or under adverse weather conditions. Unstable communication quality can significantly impact the real-time and accuracy of ship position data, leading to deviations between the ship's position in the digital twin model and its actual position, severely impacting the reliability of ship monitoring and scheduling decisions. A network delay coefficient distribution map reflects the differences in network transmission performance experienced by different sub-regions within the target waters during multi-source data transmission. This allows the system to accurately identify update delays in each sub-unit due to network transmission obstructions. Furthermore, by calculating the relationship between each sub-unit's data transmission volume and the network delay coefficient, the system dynamically determines the actual data lag in each sub-unit, enabling targeted prediction and correction for ships experiencing position update lags. Leveraging the ship motion prediction model, when the latest position information cannot be obtained in a timely manner, the ship's current position can be deduced from historical navigation trajectories, effectively avoiding position deviations and system misjudgments caused by data lags. Ultimately, this approach implements an intelligent compensation mechanism for the digital twin model in complex network environments, improving the real-time and accuracy of overall ship position updates and enhancing the model's credibility and dynamic management capabilities. The real-time update lag is the product of the network delay coefficient and the amount of data transmitted. When the real-time update lag is greater than the actual delay duration, it is determined that the subunit is experiencing ship position update lag. This is because the real-time update lag reflects the theoretical value of data transmission delay caused by the current network environment, while the actual delay duration represents the time lag of the ship position data actually received by the system. If the theoretically calculated update lag exceeds the actual observed delay duration, it indicates that the current network transmission delay has seriously affected the timeliness of data updates, resulting in the system being unable to obtain the latest ship position information on time.

[0102] According to an embodiment of the present invention, the further embodiment includes:

[0103] Real-time acquisition of navigation parameters for each vessel in the target area, including speed, steering angle, draft, and distance to the nearest obstacle. Priority coefficients for each vessel are calculated based on these parameters, with vessels whose steering angle change rate exceeds a threshold automatically receiving a higher priority level.

[0104] Based on the priority coefficient, a time slot allocation matrix is ​​constructed to divide the ship AIS signal transmission time slot into an emergency channel and a regular channel. An exclusive emergency channel is allocated to ships with a priority coefficient higher than a set value, and the remaining ships use a competitive regular channel and dynamically adjust the transmission interval according to the priority.

[0105] The channel allocation results are synchronized to the ship terminal and the digital twin model. The AIS message length of low-priority ships is compressed according to the actual channel occupancy rate, the latitude, longitude and heading core fields are retained, and a compressed ship status dataset is generated for model update.

[0106] After generating the compressed ship status data set for model updating, the method further includes:

[0107] identifying a group of ships with position drift according to the ship status dataset, and extracting original trajectory segments and compressed trajectory coordinates of the group of ships during the channel congestion period;

[0108] Establish a spatiotemporal correlation matrix for drifting ships, mark ships with blocked transmission in the same channel time slot as a correlation group, and construct trajectory continuity constraints based on the historical trajectory characteristics of the correlation group;

[0109] A sliding time window is used to perform Kalman filter preprocessing on the compressed trajectory of the associated group. A virtual anchor point is generated based on the geometric constraint that the relative positions of the ships remain unchanged. The group drift compensation vector is calculated based on the difference between the virtual anchor point and the real coordinate.

[0110] The group drift compensation vector is input into the spatiotemporal correction module of the digital twin model to perform collaborative offset correction on the displayed positions of all ships in the associated group.

[0111] It should be noted that in high-density ship encounter scenarios such as port anchorages, the simultaneous transmission of positioning data by large-scale AIS equipment can easily lead to wireless channel congestion, causing delays or packet loss in the transmission of ship status information, and thus causing problems such as asynchronous position updates of multiple ships in the digital twin model, trajectory breaks, or group coordinate offsets. Therefore, a dynamic priority assessment mechanism is used to accurately identify high-risk ships such as those turning and approaching obstacles and allocate exclusive channels. Message compression technology is used to alleviate channel load pressure. At the same time, a group drift compensation model is constructed based on the spatiotemporal correlation matrix and geometric constraints, which can effectively achieve technical improvements at three levels: First, the dynamic channel allocation strategy reduces the AIS data transmission delay of emergency avoidance ships, ensuring that key navigation status is synchronized to the digital twin platform in real time; second, through sliding window filtering and virtual anchor point generation algorithms, trajectory noise caused by message compression or packet loss is eliminated, the relative position relationship between ship groups is reconstructed, and the group drift error converges to a navigable accuracy range; finally, the closed-loop verification mechanism of the spatiotemporal correction module ensures the spatiotemporal consistency between the drift compensation vector and the actual navigation situation, avoiding overall model inaccuracy caused by local corrections, and significantly improving the reliability of dynamic supervision of ships in dense waters and the virtual-to-real mapping accuracy of the digital twin model.

[0112] Figure 4 A block diagram of a ship management system constructed based on a digital twin scenario of the present invention is shown.

[0113] The second aspect of the present invention further provides a ship management system 4 constructed based on a digital twin scenario, the system comprising: a memory 41 and a processor 42, wherein the memory includes a ship management method program constructed based on the digital twin scenario, and when the ship management method program constructed based on the digital twin scenario is executed by the processor, the following steps are implemented:

[0114] Acquire multi-source scene data of the target waters in real time, and construct a digital twin model of ship management in the target waters based on the multi-source scene data;

[0115] Performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result;

[0116] Compensating for defects in the ship management digital twin model according to the evaluation results to obtain an optimized digital twin model;

[0117] Obtaining ship density and weather condition data for each sub-area in the target waters, as well as historical ship density and weather condition data, to construct a network transmission delay prediction model; analyzing the ship density and weather condition data based on the network transmission delay prediction model to construct a network delay coefficient distribution map for the target waters;

[0118] The real-time update lag of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update lag.

[0119] The present invention discloses a ship management method and system based on digital twin scene construction. The method includes: acquiring multi-source scene data of the target waters in real time and constructing a corresponding digital twin model for ship management; conducting an integrity assessment on the constructed digital twin model and compensating for defects based on the assessment results to obtain an optimized digital twin model; further collecting ship density and weather condition data for each sub-area of ​​the target waters, constructing a network transmission delay prediction model, analyzing relevant data based on the model, and generating a network delay coefficient distribution map for the target waters; determining the real-time update lag of each sub-unit of the optimization model based on the delay distribution map, and correcting the ship position information in the digital twin model accordingly. This method can significantly improve the real-time performance and accuracy of the digital twin model and enhance the efficiency and safety of ship operation management in complex water environments.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0121] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0122] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0123] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0124] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0125] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A ship management method based on digital twin scenario construction, characterized in that: The following steps are involved: Acquire multi-source scene data of the target waters in real time, the multi-source scene data including geographic environment data, hydrological data, and ship AIS data of the target waters, and construct a digital twin model of ship management in the target waters based on the multi-source scene data; Performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result; Defect compensation is performed on the ship management digital twin model according to the evaluation results to obtain an optimized digital twin model, specifically: According to the evaluation results, subunits with gaps in three-dimensional scene construction in the ship management digital twin model are calibrated as scene-missing subunits, and subunits with inconsistent data update frequencies are calibrated as subunits with update frequencies to be optimized; For the vacant sub-units of the scene, historical multi-source scene data of the target waters and environmental characteristic data of adjacent sub-units are obtained, and the geographical environmental data of the vacant sub-units are supplemented based on a spatiotemporal interpolation algorithm to generate supplementary data. The vacant sub-units of the scene are filled according to the supplementary data, which is the first defect compensation method; For the subunit whose update frequency is to be optimized, extract the average of the data update frequencies of adjacent subunits, set the update frequency reference value of the subunit to be optimized according to the average, and analyze the data update delay of the subunit within a preset time window using a sliding window algorithm; When the update delay exceeds the allowable fluctuation range of the reference value, the timing characteristics of the ship position data and hydrological data in the sub-unit are aligned and compensated based on the dynamic time warping algorithm, and a data update synchronization compensation channel with the adjacent sub-unit is established, which is the second defect compensation method; Compensating for defects in the ship management digital twin model according to the first defect compensation method and the second defect compensation method to obtain an optimized digital twin model; Obtaining ship density and weather condition data, historical ship density and weather condition data, and historical network transmission delay data for each sub-area in the target waters, building a network transmission delay prediction model, and analyzing the ship density and weather condition data based on the network transmission delay prediction model to build a network delay coefficient distribution map for the target waters; The real-time update lag of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update lag.

2. A ship management method based on digital twin scenario construction according to claim 1, characterized in that: The real-time acquisition of multi-source scene data of the target waters and the construction of a digital twin model of ship management in the target waters based on the multi-source scene data are specifically as follows: Acquire multi-source scene data of the target waters in real time, including geographical environment data, hydrological data, and ship AIS data of the target waters; Constructing a three-dimensional geographic model of the target water area based on the geographic environment data, and mapping the hydrological data to the three-dimensional geographic model to generate an initial water area model integrating geographic and hydrological features; A real-time motion model of the ship is constructed based on the ship AIS data, the real-time motion model of the ship is calibrated and matched with the spatial coordinate system of the initial water area model, and the calibrated real-time motion model of the ship is coupled with the initial water area model to construct a digital twin model of ship management in the target water area.

3. A ship management method based on digital twin scenario construction according to claim 1, characterized in that: The ship management digital twin model is subjected to a construction integrity assessment to obtain an assessment result, specifically: Discretize the surface of the ship management digital twin model according to a preset grid size, construct N model subunits, tile the subunits, and map them onto a two-dimensional plane to construct a two-dimensional display diagram of the ship management digital twin; Acquire historical multi-source scenario data of a preset time length in the target waters, build a test data connection channel for the ship management digital twin model, and import the historical multi-source scenario data into the ship management digital twin model for simulation operation according to the test data connection channel; Mapping the simulated operation of the ship management digital twin simulation to the two-dimensional display diagram, integrating the two-dimensional display diagrams during the simulation operation according to a preset frame rate, and constructing a two-dimensional dynamic display diagram of the simulation operation; The ViBe algorithm is introduced, and a static background model is established based on the initial frame image of the two-dimensional dynamic display image according to the ViBe algorithm. The subsequent frame sequence of the two-dimensional dynamic display image is input into the static background model frame by frame. The difference between the pixel value of the current frame and the pixel value of the initial frame image is calculated for each pixel point, and whether the pixel belongs to the dynamic foreground is determined according to a preset difference threshold, thereby forming a dynamic foreground pixel set; Decomposing the two-dimensional dynamic image frame by frame into a static background layer and a dynamic foreground layer according to the dynamic foreground pixel set; Performing image segmentation processing on the static background layer to extract the defect area outline, calculating the static background vacancy rate based on the mapping relationship between the defect area outline and the model sub-unit, performing multi-frame motion trajectory tracking on the dynamic foreground layer to generate a dynamic connectivity topology map, and calculating the dynamic trajectory fracture rate based on the dynamic connectivity topology map, wherein the motion trajectory includes the ship navigation trajectory and the hydrological flow trajectory; The construction integrity of the ship management digital twin model is evaluated based on the static background vacancy rate and the dynamic trajectory fracture rate to obtain an evaluation result. The evaluation result includes the construction completeness rate, the model construction vacancy position and the model defect items. The model defect items include the vacancy in the model three-dimensional scene construction and the inconsistent data update frequency of each sub-unit of the model.

4. A ship management method based on digital twin scenario construction according to claim 1, characterized in that: The method comprises obtaining ship density and weather condition data of each sub-area in the target waters, historical ship density and weather condition data, and historical network transmission delay data, constructing a network transmission delay prediction model, analyzing the ship density and weather condition data based on the network transmission delay prediction model, and constructing a network delay coefficient distribution map of the target waters, specifically: Divide the target waters into N sub-areas, obtain historical ship density and historical weather data for each sub-area over a preset time period, the historical weather data including wind speed, rainfall intensity, and visibility parameters, and obtain historical network transmission delay data for each sub-area over a preset time period; Aligning the historical network transmission delay data, historical ship density, and historical weather condition data in time series to construct a data analysis matrix; performing a correlation analysis on the data analysis matrix based on the Pearson correlation coefficient to determine the impact of different ship densities and different weather conditions on network transmission delay, and obtaining impact data, the impact data including network delay coefficients corresponding to different ship densities and weather conditions; constructing an impact feature vector based on the ship density and weather conditions according to the impact data, using the impact feature vector as an input feature of the prediction model, and constructing a training sample set with the network delay coefficient corresponding to the impact feature vector as an output target; A decision tree algorithm is used to select feature splitting nodes for the training sample set. The weight of the input feature's impact on network delay is calculated based on the information gain rate. Feature splitting rules are constructed based on the weight ranking results. The branch structure of the decision tree model is generated through recursive splitting. A pre-pruning strategy is used to optimize the model complexity to form a network transmission delay prediction model. Obtain ship density and weather condition data for each sub-area in the target waters, import the ship density and weather condition data into the network transmission delay prediction model to predict the network delay of each sub-area, and construct a network delay coefficient distribution map for the target waters.

5. A ship management method based on digital twin scenario construction according to claim 1, characterized in that: The real-time update hysteresis of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update hysteresis, specifically: Mapping the network delay coefficient distribution map to each subunit in the optimized digital twin model, determining the network delay coefficient of each subunit, and obtaining data transmission volume information of each subunit; Calculate the real-time update lag of each sub-unit during the data transmission process according to the network delay coefficient and data transmission volume information of each sub-unit; Acquire the real-time position data of the ship in each subunit of the optimized digital twin model, compare the timestamp of the real-time position data with the current system time, and obtain the actual delay duration of the ship position data; when the real-time update lag is greater than the actual delay duration, determine that the subunit has a ship position update lag; If a ship update lag occurs, a ship motion prediction model is constructed based on historical ship navigation trajectory data, the real-time position data of the ship is input into the ship motion prediction model, a predicted navigation trajectory of the ship within the time span of the real-time update lag is generated, and the predicted position of the ship is obtained based on the predicted navigation trajectory; The ship position of the optimized digital twin model is corrected according to the predicted position, and the corrected ship position is monitored and managed in real time.

6. A ship management system based on digital twin scenarios, characterized in that: The ship management system constructed based on the digital twin scenario includes a storage device and a processor. The storage device includes a ship management method program constructed based on the digital twin scenario. When the ship management method program constructed based on the digital twin scenario is executed by the processor, the following steps are implemented: Acquire multi-source scene data of the target waters in real time, the multi-source scene data including geographic environment data, hydrological data, and ship AIS data of the target waters, and construct a digital twin model of ship management in the target waters based on the multi-source scene data; Performing a construction integrity assessment on the ship management digital twin model to obtain an assessment result; Defect compensation is performed on the ship management digital twin model according to the evaluation results to obtain an optimized digital twin model, specifically: According to the evaluation results, subunits with gaps in three-dimensional scene construction in the ship management digital twin model are calibrated as scene-missing subunits, and subunits with inconsistent data update frequencies are calibrated as subunits with update frequencies to be optimized; For the vacant sub-units of the scene, historical multi-source scene data of the target waters and environmental characteristic data of adjacent sub-units are obtained, and the geographical environmental data of the vacant sub-units are supplemented based on a spatiotemporal interpolation algorithm to generate supplementary data. The vacant sub-units of the scene are filled according to the supplementary data, which is the first defect compensation method; For the subunit whose update frequency is to be optimized, extract the average of the data update frequencies of adjacent subunits, set the update frequency reference value of the subunit to be optimized according to the average, and analyze the data update delay of the subunit within a preset time window using a sliding window algorithm; When the update delay exceeds the allowable fluctuation range of the reference value, the timing characteristics of the ship position data and hydrological data in the sub-unit are aligned and compensated based on the dynamic time warping algorithm, and a data update synchronization compensation channel with the adjacent sub-unit is established, which is the second defect compensation method; Compensating for defects in the ship management digital twin model according to the first defect compensation method and the second defect compensation method to obtain an optimized digital twin model; Obtaining ship density and weather condition data, historical ship density and weather condition data, and historical network transmission delay data for each sub-area in the target waters, building a network transmission delay prediction model, and analyzing the ship density and weather condition data based on the network transmission delay prediction model to build a network delay coefficient distribution map for the target waters; The real-time update lag of each subunit of the optimized digital twin model is determined according to the network delay coefficient distribution map, and the ship position of the optimized digital twin model is corrected according to the update lag.

Images