A method and device for identifying waterway congestion based on AIS big data

Abstract

The present application provides a method and device for identifying waterway congestion based on AIS big data. The method provided by the present application includes: obtaining the ship AIS data of the target waterway; determining the time to be identified of the target waterway, dividing the time to be identified based on the distribution characteristics of the ship AIS data, and determining the data analysis time unit; extracting the corresponding data fragments of the ship AIS data based on the data analysis time unit; determining the multiple types of waterway congestion evaluation index values ​​corresponding to each time period according to the calculated values ​​of the multiple types of waterway congestion evaluation index values, and summarizing the waterway traffic characteristics within the time to be identified; clustering the waterway traffic characteristics, and determining the congestion degree of the target waterway within the time to be identified according to the clustering results. The method and device provided by the present application are used to calculate more accurate waterway congestion evaluation indicators, reflect the congestion situation during the passage of waterway in more detail, and improve navigation efficiency.

Description

Technical Field

[0001] The present application relates to the field of waterway monitoring technology, and in particular to a method and device for identifying waterway congestion based on AIS big data. Background Art

[0002] Against the backdrop of today's rapidly developing global shipping industry, identifying congested waterways is crucial. With the increasing demand for ocean transportation, waterway usage is increasing, leading to increasingly severe congestion. This not only impacts vessel efficiency but can also pose safety risks. Therefore, identifying and analyzing congestion in waterways provides a scientific basis for waterway management and scheduling optimization, thereby improving shipping safety and efficiency and ensuring unimpeded maritime traffic.

[0003] However, there are relatively few studies on waterway congestion, and many studies still draw on analytical methods from road and air traffic. Although some scholars have attempted to propose a definition of cross-channel congestion and construct complex network models, or introduced indicators such as the traffic speed index, traffic rate index, and stay time index, there are still certain shortcomings. First, existing research mostly focuses on congestion at the macro level and lacks in-depth analysis of specific waterway environments and ship characteristics. Second, indicators such as the traffic speed index and traffic rate index provide real-time traffic conditions at the micro level and can help identify congestion locations and times, but they cannot fully reflect the overall congestion level of waterways, especially in complex situations where multiple factors interact.

[0004] Therefore, there is an urgent need for a method that combines AIS data, ship motion characteristics, channel environment and other multi-dimensional factors to calculate more accurate waterway congestion evaluation indicators, reflect the congestion situation during the passage of waterways in more detail, and improve navigation efficiency. Summary of the Invention

[0005] In view of this, the present application provides a method and device for identifying waterway congestion based on AIS big data, which combines AIS data, ship motion characteristics, channel environment and other multi-dimensional factors to calculate a more accurate waterway congestion evaluation index, reflect the congestion situation during the passage of waterways in more detail, and improve navigation efficiency.

[0006] Specifically, this application is implemented through the following technical solutions:

[0007] In a first aspect, the present application provides a method for identifying waterway congestion based on AIS big data, the method comprising:

[0008] Acquire ship AIS data of a target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships;

[0009] Determining a time period for the target channel to be identified, dividing the time period to be identified based on distribution characteristics of the ship AIS data, and determining a data analysis time unit; the time period to be identified is composed of multiple time periods;

[0010] Extracting a data segment of the ship AIS data corresponding to each time unit based on the data analysis time unit;

[0011] Calculating, based on the data segments, multiple types of waterway congestion evaluation index values ​​within a data analysis time unit corresponding to the target waterway;

[0012] Determining the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods based on the multiple types of waterway congestion evaluation index values ​​corresponding to the different data segments in each of the time periods;

[0013] Summarizing the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain waterway traffic characteristics within the time period to be identified;

[0014] Clustering is performed on the waterway traffic characteristics, and the congestion degree of the target waterway within the time period to be identified is determined according to the clustering result.

[0015] The second aspect of the present application provides a waterway congestion identification device based on AIS big data, the device comprising an acquisition module, a determination module, an extraction module and a calculation module; wherein,

[0016] The acquisition module is used to acquire ship AIS data of a target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships;

[0017] The determination module is configured to determine a time period for the target channel to be identified, segment the time period to be identified based on the distribution characteristics of the ship's AIS data, and determine a data analysis time unit; the time period to be identified is composed of multiple time periods;

[0018] The extraction module is configured to extract a data segment of the ship AIS data corresponding to each time unit based on the data analysis time unit;

[0019] The calculation module is used to calculate the channel congestion evaluation index values ​​of multiple types within the data analysis time unit corresponding to the target channel based on the data segment;

[0020] The determining module is further configured to determine the multiple types of waterway congestion evaluation index values ​​corresponding to each time period based on the values ​​of the multiple types of waterway congestion evaluation index values ​​corresponding to the different data segments in each time period;

[0021] The acquisition module is further configured to aggregate the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain waterway traffic characteristics within the time period to be identified;

[0022] The determination module is further configured to cluster the waterway traffic characteristics and determine the congestion level of the target waterway within the time period to be identified based on the clustering results.

[0023] The present application provides a method and device for identifying waterway congestion based on AIS big data. First, by dynamically dividing the time period to be identified based on the distribution characteristics of ship AIS data, it can ensure that the data analysis time unit is closely related to the actual situation of the waterway. This dynamic time unit division can avoid the data loss or redundancy problems caused by fixed time periods, thereby improving the accuracy and flexibility of waterway congestion identification and effectively handling complex and frequently changing waterway environments. On this basis, through the comprehensive calculation of multiple types of waterway congestion evaluation index values, it can comprehensively capture the multi-dimensional characteristics of waterway congestion. This method not only focuses on a single flow rate or density, but also reflects the complex traffic behavior of the waterway, such as ship detention and converging effects, providing deeper insights for subsequent decision-making. By extracting multiple types of waterway congestion evaluation index values ​​corresponding to each time period within the identification time period, it can focus on the most representative information, reduce redundant data, and enhance data processing efficiency. Secondly, further cluster analysis can be used to identify time periods with different congestion levels, which can help form clear and operational classification standards. This process combines expert scoring and data learning to ensure a high degree of consistency between the analysis results and actual waterway management needs. The clustering results can not only reflect the current congestion status of the waterway, but also reveal potential trend changes and support the prediction of possible congestion in the future. It is not only suitable for current congestion assessment, but also provides a predictive basis for long-term traffic management and resource scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a flowchart of Example 1 of the waterway congestion identification method based on AIS big data provided by this application;

[0025] Figure 2 A schematic diagram of catching up and convergence degree shown in an exemplary embodiment of the present application;

[0026] Figure 3 A schematic diagram of cross-convergence according to an exemplary embodiment of the present application;

[0027] Figure 4 This is a structural diagram of Example 1 of the waterway congestion identification device based on AIS big data provided in this application. DETAILED DESCRIPTION

[0028] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with this application.

[0029] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a," "the," and "the" used in this application are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0030] It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0031] Specific embodiments are given below to introduce the technical solutions of the present application in detail.

[0032] Figure 1 This is a flowchart of the first embodiment of the waterway congestion identification method based on AIS big data provided by this application. Figure 1 The method provided in this embodiment may include:

[0033] S101. Acquire ship AIS data of a target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships.

[0034] Specifically, the target waterway is a waterway where congestion is to be identified. The ship AIS data records the passage of different ships, including the ship's location information, speed, course, type and destination.

[0035] In specific implementation, the target waterway is determined based on the waterway with congestion to be identified. Through the data interface provided by the shore-based AIS receiving station, satellite AIS service or relevant maritime departments, the ship AIS data of the target waterway is extracted from the data interface at the set time interval and frequency, and the extracted ship AIS data is cleaned to remove missing values ​​and outliers.

[0036] S102: Determine a time period for the target waterway to be identified, divide the time period to be identified based on distribution characteristics of the ship AIS data, and determine a data analysis time unit; the time period to be identified is composed of multiple time periods.

[0037] Specifically, the target channel's pending identification time is set based on actual needs and is not limited in this embodiment. For example, in one embodiment, the target channel's pending identification time is 24 hours a day. The pending identification time is composed of multiple time periods, for example, the pending identification time is composed of 24 time periods.

[0038] It should be noted that because the AIS data of ships in the target channel changes dynamically during the identification time, it is necessary to segment the identification time and perform congestion identification based on the ship AIS data corresponding to the minimum time unit (data analysis time unit). The data analysis time unit is the basic time unit of the target channel, obtained by segmenting the target channel identification time, and the time length of each data analysis time unit is not exactly the same.

[0039] In a specific implementation, the method of dividing the time to be identified based on the distribution characteristics of the ship AIS data and determining the data analysis time units includes: determining a distribution curve of the number of ships passing through the target waterway within the time to be identified based on the ship AIS data; determining a change trend demarcation point on the ship number distribution curve based on a change trend of the ship number distribution curve; dividing the time to be identified based on the change trend demarcation point and determining the data analysis time units, wherein the time length of each data analysis time unit is not exactly the same.

[0040] Specifically, in conjunction with the above description, ship AIS data records the passage of different ships, including the number of different ships passing through the target channel at various time points during the time period to be identified. By acquiring the ship AIS data for the time period to be identified within the target channel, the number of different ships passing through the target channel at various time points is extracted to obtain time series data. Furthermore, the number of ships is counted according to a time window (e.g., per minute or per hour) to obtain a ship population distribution curve. The rate of change of the ship population distribution curve is calculated by numerical difference, and the local extreme value of the rate of change (e.g., the point where the rate of change changes from positive to negative or from negative to positive) is determined as the trend demarcation point. If there are too many trend demarcation points, a minimum time interval or minimum change amplitude can be set to further filter out important demarcation points. Furthermore, the time period to be identified is divided into several time periods based on the trend demarcation points, with each time period representing a data analysis time unit.

[0041] S103: Extracting a data segment of the ship AIS data corresponding to each time unit based on the data analysis time unit.

[0042] Specifically, as described above, ship AIS data records the passage of different ships at different time points, that is, it includes the passage of ships in different data analysis time units. In this step, the ship AIS data within each data analysis time unit is filtered based on the timestamp corresponding to that time unit, forming corresponding data segments, and obtaining the ship AIS data segments corresponding to each time unit.

[0043] S104. Calculate, based on the data segments, multiple types of waterway congestion evaluation index values ​​within the data analysis time unit corresponding to the target waterway.

[0044] Specifically, the waterway congestion evaluation index value represents the congestion level of the target waterway within a preset time period. Multiple types of waterway congestion evaluation index values ​​include detention rate, convergence rate, equivalent traffic volume, and saturation rate.

[0045] In a specific implementation, the calculation of multiple types of channel congestion evaluation index values ​​within the data analysis time unit corresponding to the target channel based on the data fragments includes: determining the data analysis time unit to be calculated, extracting the ship AIS data fragments of each ship in the target channel within the data analysis time unit to be calculated, and obtaining the data fragments; obtaining the real-time positions and trajectories of different ships in the target channel within the data analysis time unit to be calculated based on the data fragments, and calculating multiple types of channel congestion evaluation index values ​​according to the real-time positions and trajectories, wherein the multiple types include at least detention rate, convergence rate, equivalent traffic volume and saturation.

[0046] Specifically, the AIS data of each ship within the target waterway is filtered and extracted from the time unit of the data analysis to be calculated. For each data segment, the data segment is traversed to determine the real-time position of each ship within the target waterway within the corresponding time unit of the data analysis to be calculated. Furthermore, the trajectory of each ship within the target waterway is determined based on the change in the real-time position of each ship within the target waterway over time within each data segment. The congestion evaluation index value for each type of waterway is calculated based on the method for calculating the congestion evaluation index value for each type of waterway.

[0047] Optionally, when the multiple types of waterway congestion evaluation index values ​​are equivalent traffic volumes, the calculating of the multiple types of waterway congestion evaluation index values ​​based on the real-time position and trajectory includes:

[0048] (1) Traverse the data segments and determine the size and quantity of each ship.

[0049] Specifically, equivalent traffic volume represents the volume of ships that a target waterway can handle during a specific time period. As described above, a data segment records the ship AIS data corresponding to each data analysis time unit, including the traffic conditions of different ship types. In this step, each data segment is traversed to determine the size and number of each ship contained in each data segment.

[0050] (2) Determine the type of each ship according to its size, and determine the equivalent coefficient based on the ship type; the equivalent coefficient is positively correlated with the size of the ship.

[0051] Specifically, the type of vessel is related to its size. A vessel's equivalence coefficient represents the relative influence of different types of vessels on waterway usage. It is positively correlated with vessel size; that is, the larger the vessel, the higher its equivalence coefficient. In conjunction with the above description, since in this embodiment, ships are categorized by size, different types of ships have different equivalence coefficients.

[0052] In practice, the system uses clustering methods (such as K-means and hierarchical clustering) to analyze the size information of different ships based on the collected size information. During the clustering process, a preset size difference threshold (a value set based on the large number of similar size information during the clustering process) is set. Ships with size differences below this threshold are classified as belonging to the same type. After all ships are classified, the average size (e.g., average length) of each type is calculated. By comparing the average sizes of ships of different types, ships with larger average sizes are assigned a higher equivalence factor.

[0053] For example, in one embodiment, the different types of ships determined by the clustering method include small ships, medium ships, relatively large ships, large ships, and very large ships. The average size of small ships is less than 25 meters, the average size of medium ships is 25-75 meters, the average size of relatively large ships is 75-125 meters, the average size of large ships is 125-175 meters, and the average size of very large ships is greater than 175 meters. The equivalent coefficient of small ships is 0.1, the equivalent coefficient of medium ships is 0.15, the equivalent coefficient of relatively large ships is 0.2, the equivalent coefficient of large ships is 0.25, and the equivalent coefficient of very large ships is 0.3.

[0054] (3) The product of the equivalent coefficient and the number of each ship is determined as the equivalent traffic volume of each ship, and the sum of the equivalent traffic volumes of each ship is determined as the equivalent traffic volume within the data analysis time unit corresponding to the target channel.

[0055] In specific implementation, for each type of ship, the product of the collected number of ships of that type and the corresponding equivalent coefficient is used to determine the equivalent traffic volume of that type of ship. Furthermore, the sum of the equivalent traffic volumes of all types of ships is determined as the equivalent traffic volume in the data analysis time unit corresponding to the target waterway. The equivalent traffic volume in the data analysis time unit corresponding to the target waterway can be calculated according to the following formula:

[0056] ;

[0057] Among them, the The equivalent traffic volume in the data analysis time unit corresponding to the target waterway; 、 、 、 、 is the equivalent coefficient corresponding to different ships; 、 、 、 、 The number of different ships.

[0058] For example, in one embodiment, combined with the above example, when the equivalent coefficient of a small ship is 0.1, the equivalent coefficient of a medium ship is 0.15, the equivalent coefficient of a larger ship is 0.2, the equivalent coefficient of a large ship is 0.25, and the equivalent coefficient of an ultra-large ship is 0.3, and the number of small ships is 30, the equivalent coefficient of a medium ship is 20, the equivalent coefficient of a larger ship is 10, the equivalent coefficient of a large ship is 10, and the equivalent coefficient of an ultra-large ship is 5, then the equivalent traffic volume in the data analysis time unit corresponding to the target channel is 0.1×30+0.15×20+0.2×10+0.25×10+0.3×5=12.

[0059] The method provided in this embodiment takes into account that the space occupied by ships of different sizes in the waterway and the impact on the waterway load are significantly different. The positive correlation between the size of the ship and the equivalent coefficient can more accurately reflect the different demands of various types of ships for waterway resources. Since small ships and large ships have different impacts on the congestion of the waterway, the calculated equivalent traffic volume can be closer to the actual traffic load through this differentiated equivalent coefficient. Moreover, by weighted summation of the equivalent coefficients of different ship types, complex ship types can be converted into a single equivalent traffic volume indicator. This quantitative assessment facilitates the establishment of a unified standard, making the traffic volume assessment comparable and operational. Since directly adding up the number of ships may ignore the actual impact of large ships, the equivalent coefficient can ensure that the congestion pressure of large ships on the waterway is fully considered, thereby improving the accuracy of the assessment.

[0060] Optionally, when the multiple types of waterway congestion evaluation index values ​​are saturation, the calculating the multiple types of waterway congestion evaluation index values ​​according to the real-time position and trajectory includes:

[0061] (1) Determine corresponding ship areas based on the real-time positions and trajectories of the different ships in the target waterway.

[0062] Specifically, saturation represents the ratio of the actual vessel flow in the target channel to its maximum capacity. A vessel's domain refers to the spatial area occupied by a vessel within the waterway. This domain is typically an elliptical or rectangular area, its specific shape determined by the vessel type, its real-time position, and its trajectory. This domain takes into account a safety buffer zone for vessels when navigating, turning, and avoiding other vessels, ensuring the safe passage of other vessels within the target channel.

[0063] In specific implementations, the real-time positions and trajectories of different ships in the target channel are acquired and analyzed using appropriate fitting algorithms (such as polynomial regression or curve fitting). The paths and dynamic changes of different ships during navigation are obtained through fitting, and the nonlinear motion trajectories of each ship are determined. The fitted nonlinear motion trajectories are then analyzed in detail to observe the trajectory changes at different headings and speeds during navigation, particularly during high-speed navigation or turns. Based on the changing patterns of the motion trajectories, the ship's behavior under different navigation conditions can be identified. Based on the analysis results, the ship's domain is determined to be an elliptical area (because the ellipse more accurately reflects the spatial occupancy of a ship during navigation, especially its dynamic characteristics during turns and changes in course). Furthermore, the major and minor axes of the ship's domain are determined based on the ship's range of motion under different navigation conditions. When sailing at high speeds or turning, the major axis of the ship's domain can be determined based on the ship's maximum speed and acceleration, typically the maximum range of motion of the ship when sailing in a straight line. The minor axis of the ship's domain is determined based on the ship's minimum turning radius during turning. The turning radius is generally related to the ship's size and speed and is obtained by analyzing the specific trajectory data of each ship during turning.

[0064] (2) Calculating the ship territory areas of different ships based on the lengths of different ships and the corresponding ship territories in the data segments.

[0065] In a specific implementation, the product of the length of different ships and a first preset coefficient is determined as the major axis of the different ships, and the product of the length of different ships and a second preset coefficient is determined as the minor axis of the different ships. The first preset coefficient is greater than the second preset coefficient, and the product of the major axis of different ships, the minor axis of different ships, and pi is determined as the ship area of ​​different ships.

[0066] Specifically, the first preset coefficient and the second preset coefficient are set according to actual needs and are not limited in this embodiment. It should be noted that the first preset coefficient is greater than the second preset coefficient. For example, in one embodiment, the first preset coefficient is 8 and the second preset coefficient is 3.2. In specific implementation, the ship area of ​​different ships can be calculated according to the following formula:

[0067] ;

[0068] Among them, the For ships The area of ​​the ship's field; For ships Captain.

[0069] (3) The quotient of the sum of the ship area of ​​different ships and the actual navigable water area of ​​the target channel is determined as the saturation of the target channel within the data analysis time unit.

[0070] In specific implementation, the ship area areas of different ships are added together, and the quotient of the sum of the ship area areas of different ships and the actual navigable water area of ​​the target channel is determined as the saturation of the target channel in the data analysis time unit. That is, the saturation is calculated according to the following formula:

[0071] ;

[0072] ;

[0073] Among them, the The saturation of the target channel corresponding to the data analysis time unit; is the sum of the ship area of ​​all ships; is the actual navigable water area of ​​the target channel; For ships The area of ​​the ship's field; For ships Captain.

[0074] The method provided in this embodiment, on the one hand, takes into account that the movement characteristics of ships directly affect their demand and impact on the surrounding waters, and defines the occupancy range through the ship domain, so that the saturation assessment can dynamically reflect the actual congestion level of the waterway under the real-time navigation of the ship. Since ships require more space when they are at high speed or turning, the ship domain is defined based on these movement characteristics to ensure that the saturation calculation takes into account not only the quantity, but also the changes in space occupancy caused by the navigation status, thereby improving the accuracy of the calculation. In addition, by calculating the saturation by summing the domain area of ​​each ship and comparing it with the total water area of ​​the waterway, the actual demand for space of the ship in the waterway can be quantified, making the evaluation value more representative, especially in the case of mixed navigation of ships of different sizes. On the second hand, compared with previous methods that simply determine saturation by the number or size of ships, the area of ​​the ship domain includes the space demand of each ship in motion, which is more in line with the actual situation, making the saturation assessment more reasonable and comprehensive. When calculating saturation, not only the number of ships but also their territorial area is taken into account. This can avoid misjudgment of congestion caused by the increase in the number of ships, especially when there are a large number of ships but a small territorial area gathering, and can accurately identify them. Since certain specific types of ships (such as large transport ships or ships with faster speeds) have more obvious impacts in different environments, calculating saturation based on the ship's territorial area can better cope with the impact of these special ships in the waterway and effectively prevent congestion.

[0075] Furthermore, in conjunction with the above description, both equivalent traffic volume and saturation serve as channel congestion evaluation indicators for evaluating the congestion level of the target channel. Optionally, the channel congestion evaluation indicator also includes detention rate, which is the quotient of the difference between the number of ships entering and leaving the target channel during the corresponding data analysis time unit and the number of ships leaving the target channel. The detention rate reflects the flow of the target channel during the corresponding data analysis time unit; the higher the detention rate, the more congested the target channel.

[0076] In specific implementation, the detention rate of the target channel can be calculated according to the following formula:

[0077] ;

[0078] Among them, the is the retention rate of the target channel in the corresponding data analysis time unit; is the number of ships entering the target channel; is the number of ships leaving the target channel.

[0079] Optionally, the channel congestion evaluation index value also includes convergence degree, which represents the convergence trend between different ships; the convergence degree includes pursuit convergence degree and cross convergence degree, the pursuit convergence degree represents the convergence of different ships when sailing in the same direction, and the cross convergence degree represents the convergence of different ships when they are in different directions or crossing directions.

[0080] Specifically, Figure 2 This is a schematic diagram showing the convergence degree of an exemplary embodiment of the present application. Figure 2 The pursuit convergence degree includes two convergence characteristics: static and dynamic. The pursuit convergence degree can be calculated based on the following formula:

[0081] ;

[0082] Among them, the is the degree of convergence between different ships; It is the static part of the degree of convergence between different ships; It is the dynamic part of the pursuit and convergence between different ships; 、 、 is the adjustment parameter; For ships , ships the distance between For ships , ships The minimum distance between For ships , ships The maximum distance between 、 Ships , ships speed; is the pursuit convergence function.

[0083] Further, Figure 3 This is a schematic diagram of the cross-convergence degree shown in an exemplary embodiment of the present application. Similar to the pursuit convergence degree, the calculation of the cross-convergence degree is also divided into static and dynamic. The static convergence situation is similar to the pursuit convergence degree. It should be noted that the dynamic convergence situation, when considering the dynamic convergence characteristics of the cross-convergence degree, simply using the relative distance divided by the relative time to indicate that the two ships have reached the minimum safe distance is not enough. It is necessary to establish a coordinate system, based on the coordinates of the two ships on the coordinate system at the corresponding time, calculate the relative distance, and solve the minimum time it takes for the two ships to reach the minimum safe distance. . Solution When, let , i.e. ship , ships exist The distance of the moment is equal to the ship , ships The minimum safe distance between the two. Calculate according to the following formula :

[0084] ;

[0085] Among them, the For ships , ships The minimum distance between 、 Ships , ships speed; is the current moment; For ships , ships The distance in the x direction at the initial moment; For ships , ships The distance in the y direction at the initial moment; For ships , ships The distance at the initial moment; For ships , ships The angle between them.

[0086] Furthermore, the cross-convergence degree can be calculated based on the following formula:

[0087] ;

[0088] Among them, the is the cross-convergence degree between different ships; is the static part of the cross-convergence between different ships; is the dynamic part of the cross-convergence between different ships; 、 、 is the adjustment parameter; For ships , ships the distance between The time it takes for the two ships to reach the minimum safe distance; For ships , ships The minimum distance between For ships , ships The maximum distance between them.

[0089] Furthermore, the cross-convergence degree is equal to the sum of the pursuit convergence degree and the cross-convergence degree, that is, , wherein the is the cross-convergence degree between different ships; is the degree of convergence between different ships; is the cross-convergence degree between different ships.

[0090] S105 , determining the multiple types of waterway congestion evaluation index values ​​corresponding to each time period based on the values ​​of the multiple types of waterway congestion evaluation index values ​​corresponding to different data segments in each time period.

[0091] Specifically, considering the description above, each time period contains multiple data segments. In this step, we compare the waterway congestion evaluation index values ​​corresponding to these data segments within each time period. Ultimately, we obtain multiple types of waterway congestion evaluation index values ​​(detention rate, convergence rate, equivalent traffic volume, and saturation) for each time period. In summary, we ultimately obtain multiple types of waterway congestion evaluation index values ​​for 24 time periods.

[0092] In specific implementation, the multiple types of channel congestion evaluation index values ​​corresponding to each time period are determined based on the numerical values ​​of the multiple types of channel congestion evaluation index values ​​corresponding to different data segments in each time period, including: for each time period within the time length to be identified, determining the multiple types of channel congestion evaluation index values ​​corresponding to each data segment in the time period; for the same type of channel congestion evaluation index value, comparing the channel congestion evaluation index values ​​corresponding to each data segment, and determining the maximum channel congestion evaluation index value as the channel congestion evaluation index value of the type in the time period.

[0093] Specifically, in combination with the above description, in this step, for each time period within the time length to be identified, multiple types of waterway congestion evaluation index values ​​corresponding to each data segment contained in each time period are determined (since they have already been calculated in the above steps, this step only needs to filter according to the time period). Furthermore, for each time period, the waterway congestion evaluation index values ​​of the same type corresponding to all data segments contained in the time period are compared (for example, the convergence corresponding to all data segments in the same time period are compared), and the waterway congestion evaluation index value with the largest value is determined as the waterway congestion evaluation index value of that type in the time period (for example, the convergence corresponding to all data segments in the same time period is determined as the convergence for the time period).

[0094] For example, in combination with the above description, the time period to be identified is 24 hours, and the time period corresponding to 12 noon is used as an example. There are 6 moments under 12 noon, namely moment 1, moment 2, moment 3, moment 4, moment 5, and moment 6. The retention rates corresponding to these 6 moments are compared in sequence, and the target retention rate with the largest retention rate is determined as the retention rate corresponding to 12 noon. Similarly, the convergence rates corresponding to these 6 moments are compared in sequence, and the target convergence rate with the largest convergence rate is determined as the convergence rate corresponding to 12 noon. The equivalent traffic volumes corresponding to these 6 moments are compared in sequence, and the target equivalent traffic volume with the largest equivalent traffic volume is determined as the equivalent traffic volume corresponding to 12 noon. The saturations corresponding to these 6 moments are compared in sequence, and the target saturation with the largest saturation is determined as the saturation corresponding to 12 noon.

[0095] S106 , summarizing the waterway congestion evaluation index values ​​of multiple types corresponding to each of the time periods to obtain waterway traffic characteristics within the time period to be identified.

[0096] Specifically, combined with the above description, each time period corresponds to multiple types of channel congestion evaluation index values, and the time to be identified is composed of multiple time periods. In this step, the multiple types of channel congestion evaluation index values ​​corresponding to all time periods need to be combined in chronological order to obtain the channel traffic characteristics of the target channel during the entire time to be identified.

[0097] In specific implementation, the channel congestion evaluation index values ​​of multiple types corresponding to each of the time periods are summarized to obtain the channel traffic characteristics within the time period to be identified, including: using the time unit corresponding to each time period as the time identifier of each group of channel congestion evaluation index values; each group of channel congestion evaluation index values ​​contains multiple types; arranging each group of channel congestion evaluation index values ​​according to the sequence of the time identifiers to obtain the channel traffic characteristics.

[0098] Specifically, in conjunction with the above description, each time period corresponds to multiple types of waterway congestion evaluation index values. In this step, the timestamp recorded in the time unit corresponding to each time period is used as the time identifier. Each set of waterway congestion evaluation index values ​​corresponding to each time period (multiple types of waterway congestion evaluation index values) is arranged in the order of the time identifiers, with the set of waterway congestion evaluation index values ​​corresponding to the time period with the earliest time identifier placed at the beginning and the set of waterway congestion evaluation index values ​​corresponding to the time period with the latest time identifier placed at the end. This results in the waterway traffic characteristics for the time period to be identified, including the waterway congestion evaluation index values ​​for all time periods.

[0099] S107 , clustering the waterway traffic characteristics, and determining the congestion level of the target waterway within the time period to be identified based on the clustering results.

[0100] In a specific implementation, the waterway traffic characteristics are clustered, and the congestion degree of the target waterway within the time period to be identified is determined based on the clustering results, including: obtaining navigation information of the target waterway with different congestion degrees; scoring the navigation information using expert scoring; learning the scoring results to obtain the characteristics of navigation information under different waterway congestion degrees; calculating the standard waterway congestion degree evaluation index values ​​of each type under different waterway congestion degrees based on the navigation information characteristics; and clustering the waterway traffic characteristics using the standard waterway congestion degree evaluation index values ​​of each type under each waterway congestion degree as cluster centers.

[0101] Specifically, the navigation information of the target waterway at different congestion levels may be image information or the number of ships.

[0102] In specific implementation, navigation information for different numbers of ships in the target waterway is obtained, resulting in navigation information for different levels of congestion. Experts score the navigation information for each level of congestion in the target waterway and classify it into four categories: unobstructed, relatively unobstructed, slightly congested, and severely congested (each category contains several data segments). For each category, different types of waterway congestion evaluation indicators are extracted from the data segments corresponding to these categories. For each category of data segments, the average or median value of each type of waterway congestion evaluation indicator across all data segments is calculated as the standard waterway congestion evaluation indicator for that category of waterway congestion. Furthermore, cluster centers are initialized to the standard waterway congestion evaluation indicator for each type of waterway congestion evaluation indicator. For each data segment, its Euclidean distance to the cluster center is calculated. Based on this distance, the data segment is assigned to the closest cluster center, and the cluster centers are updated until convergence is achieved. Based on the clustering results, the waterway congestion level corresponding to the cluster containing the largest number of data segments is determined as the congestion level of the target waterway during the time period to be identified.

[0103] Optionally, the channel traffic characteristics are clustered, and the congestion level of the target channel within the time period to be identified is determined based on the clustering results, including: calculating the distance between the channel congestion level evaluation index value of each type in the channel traffic characteristics and multiple cluster centers, each cluster center representing a channel congestion level; determining the subclass to which each channel congestion level evaluation index value belongs based on the distance; determining a representative subclass based on the number of each channel congestion level evaluation index value in each subclass, and using the channel congestion level corresponding to the representative subclass as the congestion level of the target channel within the time period to be identified.

[0104] Specifically, the cluster center is the standard waterway congestion evaluation index value of each type under different waterway congestion conditions. In this step, the distance between the waterway congestion evaluation index value of each type in each time period in the waterway traffic characteristics and multiple cluster centers is calculated based on the Euclidean distance. For each type of waterway congestion evaluation index value, the cluster with the closest cluster center is determined as the subclass to which the waterway congestion evaluation index value belongs, and the waterway congestion evaluation index value is assigned to the subclass. After all the waterway congestion evaluation index values ​​are assigned, the subclass containing the largest number of waterway congestion evaluation index values ​​is determined as the representative subclass based on the number of waterway congestion evaluation index values ​​contained in each subclass, and the waterway congestion level corresponding to the representative subclass is determined as the congestion level of the target waterway during the time period to be identified.

[0105] For example, in one embodiment, subclass 1, subclass 2, subclass 3, and subclass 4 correspond to unimpeded, relatively unimpeded, slightly congested, and severely congested, respectively. Subclass 1 contains 10 waterway congestion evaluation index values, subclass 2 contains 20 waterway congestion evaluation index values, subclass 3 contains 15 waterway congestion evaluation index values, and subclass 4 contains 30 waterway congestion evaluation index values. Therefore, subclass 4 is a representative subclass, subclass 4 corresponds to severe congestion, and the congestion level of the target waterway during the time to be identified is severe congestion.

[0106] It should be noted that when there are multiple representative subcategories, when determining the waterway congestion level corresponding to the representative subcategories, it is necessary to combine the time units corresponding to the data segments in which the representative subcategories are located. If the data segments correspond to the same time unit, it is necessary to consider the combined impact of the congestion levels of these representative subcategories on the data segments and calculate the average of the scores corresponding to these representative subcategories (for example, if multiple representative subcategories correspond to mild congestion and severe congestion, the average of the scores corresponding to mild congestion and severe congestion will be determined as the congestion level). If the data segments correspond to different time units, for example, in the previous time unit, the representative subcategory corresponds to mild congestion, and in the next time unit, the representative subcategory corresponds to severe congestion, then the congestion level is determined to have changed from mild congestion to severe congestion.

[0107] The method provided in this embodiment, firstly, dynamically divides the time period to be identified based on the distribution characteristics of vessel AIS data, ensuring that the data analysis time units are closely aligned with the actual waterway conditions. This dynamic time unit division avoids data loss or redundancy issues associated with fixed time periods, thereby improving the accuracy and flexibility of waterway congestion identification and effectively handling complex and frequently changing water transport environments. Furthermore, through the comprehensive calculation of multiple types of waterway congestion evaluation indicators, the multi-dimensional characteristics of waterway congestion can be fully captured. This method not only focuses on single flow or density but also reflects complex traffic behavior on the waterway, such as ship detention and converging effects, providing deeper insights for subsequent decision-making. By extracting multiple types of waterway congestion evaluation indicators corresponding to each time period within the time period to be identified, it is possible to focus on the most representative information, reduce redundant data, and enhance data processing efficiency. Secondly, further cluster analysis identifies time periods with different levels of congestion, which helps to form clear and actionable classification criteria. This process combines expert scoring with data learning to ensure high consistency between the analysis results and actual waterway management needs. The clustering results not only reflect the current congestion status of a waterway but also reveal potential trends and trends, supporting predictions of future congestion. This approach is not only applicable to current congestion assessments but also provides a predictive basis for long-term traffic management and resource scheduling. Thirdly, this embodiment takes into account the significant differences in the space occupied by ships of different sizes in a waterway and their impact on waterway load. By positively correlating ship size with the equivalence coefficient, the varying demands of each type of ship on waterway resources can be more accurately reflected. Since small and large ships have different impacts on waterway congestion, these differentiated equivalence coefficients ensure that the calculated equivalent traffic volume is closer to the actual traffic load. Furthermore, by weighting and summing the equivalence coefficients for different ship types, complex ship types can be converted into a single equivalent traffic volume indicator. This quantitative assessment facilitates the establishment of unified standards, making traffic volume assessments more comparable and operational. While simply adding up the number of ships may overlook the actual impact of large ships, the equivalence coefficient ensures that the congestion pressure of large ships on the waterway is fully accounted for, thereby improving assessment accuracy. Fourthly, considering that the movement characteristics of ships directly affect their demand for and impact on the surrounding waters, by defining the occupancy range of ship domains, saturation assessment can dynamically reflect the actual congestion level of the waterway under real-time ship navigation conditions. Since ships require more space when traveling at high speeds or turning, the definition of ship domains based on these movement characteristics ensures that saturation calculations take into account not only the number of ships, but also the changes in space occupancy caused by navigation status, thereby improving calculation accuracy. In addition, by calculating the saturation by summing the domain areas of each ship and comparing them with the total water area of ​​the waterway, the actual demand for space in the waterway can be quantified, making the assessment value more representative, especially when ships of different sizes are sailing together.Fifthly, compared to previous methods that simply determine saturation by the number or size of ships, the area of ​​the ship's territory includes the space requirements of each ship during movement, which is more in line with actual conditions and makes the saturation assessment more reasonable and comprehensive. When calculating saturation, not only the number of ships but also the area of ​​their territory is considered. This can avoid misjudgments of congestion caused by an increase in the number of ships, especially when there are many ships but a small area of ​​ships gathering. Since certain types of ships (such as large transport ships or ships with faster speeds) have a more obvious impact in different environments, calculating saturation based on the area of ​​the ship's territory can better deal with the impact of these special ships in the waterway and effectively prevent congestion.

[0108] Corresponding to the aforementioned embodiment of a method for identifying waterway congestion based on AIS big data, the present application also provides an embodiment of a device for identifying waterway congestion based on AIS big data.

[0109] Figure 4 This is a schematic diagram of the structure of the first embodiment of the waterway congestion identification device based on AIS big data provided by this application. Figure 4 The device provided in this embodiment includes an acquisition module 410, a determination module 420, an extraction module 430 and a calculation module 440; wherein,

[0110] The acquisition module 410 is used to acquire ship AIS data of a target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships;

[0111] The determination module 420 is configured to determine a target channel identification time, segment the identification time based on the distribution characteristics of the ship AIS data, and determine a data analysis time unit; the identification time is composed of multiple time periods;

[0112] The extraction module 430 is configured to extract a data segment of the ship AIS data corresponding to each time unit based on the data analysis time unit;

[0113] The calculation module 440 is used to calculate the channel congestion evaluation index values ​​of multiple types within the data analysis time unit corresponding to the target channel based on the data segment;

[0114] The determining module 420 is further configured to determine the multiple types of waterway congestion evaluation index values ​​corresponding to each time period based on the multiple types of waterway congestion evaluation index values ​​corresponding to the different data segments in each time period;

[0115] The acquisition module 410 is further configured to aggregate the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain waterway traffic characteristics within the time period to be identified;

[0116] The determination module 420 is further configured to cluster the waterway traffic characteristics and determine the congestion level of the target waterway within the time period to be identified based on the clustering results.

[0117] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0118] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0119] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0120] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for identifying waterway congestion based on AIS big data, characterized in that: The method comprises: Acquire ship AIS data of a target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships; Determine the time to be identified of the target waterway, divide the time to be identified based on the distribution characteristics of the ship AIS data, and determine the data analysis time unit; the time to be identified is composed of multiple time periods; the distribution characteristics represent the change trend demarcation point on the ship quantity distribution curve, and the ship quantity distribution curve represents the distribution of the number of ships passing through the target waterway within the time to be identified; Extracting, based on the data analysis time unit, a data segment of the ship AIS data corresponding to each data analysis time unit; Based on the data segments, calculating multiple types of waterway congestion evaluation index values ​​within the data analysis time unit corresponding to the target waterway; Determine the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods based on the values ​​of the multiple types of waterway congestion evaluation index values ​​corresponding to the different data segments in each of the time periods; Summarizing the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain the waterway traffic characteristics within the time period to be identified; combining the multiple types of waterway congestion evaluation index values ​​according to the chronological order of the waterway traffic characteristics; Clustering the waterway traffic characteristics, and determining the congestion degree of the target waterway within the time period to be identified according to the clustering result; Among them, the distance between each type of channel congestion evaluation index value in the channel traffic characteristics in each time period and multiple cluster centers is calculated, each cluster center represents a standard channel congestion index value under a channel congestion level; the subclass to which each channel congestion evaluation index value belongs is determined according to the distance; the subclass represents a certain type of channel congestion level; the representative subclass is determined according to the number of each channel congestion evaluation index value in each subclass, and the channel congestion level corresponding to the representative subclass is used as the congestion level of the target channel within the time period to be identified; the representative subclass represents the subclass containing the largest number of channel congestion evaluation index values.

2. The method according to claim 1, characterized in that The step of dividing the time to be identified based on the distribution characteristics of the ship AIS data and determining the data analysis time unit includes: Determine, based on the ship AIS data, a distribution curve of the number of ships passing through the target waterway within the time to be identified; Based on the change trend of the ship quantity distribution curve, determining a change trend demarcation point on the ship quantity distribution curve; The time length to be identified is divided based on the change trend dividing point to determine the data analysis time unit, and the time length of each data analysis time unit is not exactly the same.

3. The method according to claim 1, characterized in that The calculating, based on the data segments, multiple types of waterway congestion evaluation index values ​​within the data analysis time unit corresponding to the target waterway includes: Determine a time unit for data analysis to be calculated, extract ship AIS data segments of each ship in the target waterway within the time unit for data analysis to be calculated, and obtain the data segments; Based on the data segments, the real-time positions and trajectories of different ships in the target waterway within the data analysis time unit to be calculated are obtained, and multiple types of waterway congestion evaluation index values ​​are calculated according to the real-time positions and trajectories, and the multiple types include at least detention rate, convergence rate, equivalent traffic volume and saturation.

4. The method according to claim 1, characterized in that: The determining of the multiple types of waterway congestion evaluation index values ​​corresponding to each time period based on the values ​​of the multiple types of waterway congestion evaluation index values ​​corresponding to different data segments in each time period includes: For each time period within the time length to be identified, determining multiple types of waterway congestion evaluation index values ​​corresponding to each data segment in the time period; For the same type of waterway congestion evaluation index values, the waterway congestion evaluation index values ​​corresponding to each data segment are compared, and the maximum waterway congestion evaluation index value is determined as the waterway congestion evaluation index value of the type in the time period.

5. The method according to claim 1, characterized in that The summarizing of the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain the waterway traffic characteristics within the time period to be identified includes: The time corresponding to each time period is used as the time mark of each group of waterway congestion evaluation index values; each group of waterway congestion evaluation index values ​​includes multiple types; The waterway congestion evaluation index values ​​of each group are arranged according to the sequence of the time marks to obtain the waterway traffic characteristics.

6. The method according to claim 1, characterized in that The clustering of the waterway traffic characteristics and determining the congestion degree of the target waterway within the time period to be identified according to the clustering result includes: Acquiring navigation information of the target waterway at different congestion levels; the navigation information is image information or the number of ships; Scoring the navigation information using expert scoring; As a result of learning and scoring, the characteristics of navigation information under different degrees of waterway congestion are obtained; the characteristics of navigation information are evaluation value indicators of different types of waterway congestion degrees; Calculate the evaluation index value of the congestion degree of each type of standard waterway under different waterway congestion degrees according to the characteristics of the navigation information; Clustering the waterway traffic characteristics using the evaluation index values ​​of each type of standard waterway congestion degree at each waterway congestion degree as cluster centers; Among them, experts score the navigation information of the target waterway with different congestion levels, and mark them into different congestion level categories, each of which contains several data segments; for each category, different types of waterway congestion level evaluation value indicators in the data segments corresponding to the category are extracted respectively; for the data segments of each category, the average value or median value of each type of waterway congestion level evaluation value indicator in all data segments is calculated as the standard waterway congestion level evaluation index value under the category of waterway congestion level.

7. The method according to claim 3, characterized in that When the multiple types of waterway congestion evaluation index values ​​are equivalent traffic volumes, the multiple types of waterway congestion evaluation index values ​​are calculated according to the real-time position and trajectory, including: Traversing the data segments to determine the size and quantity of each ship; Determining the type of each ship according to the size of each ship, and determining an equivalent coefficient based on the ship type; the equivalent coefficient is positively correlated with the size of the ship; The product of the equivalent coefficient and the number of ships of each type is determined as the equivalent traffic volume of each type of ships, and the sum of the equivalent traffic volumes of all types of ships is determined as the equivalent traffic volume within the data analysis time unit corresponding to the target waterway.

8. The method according to claim 3, characterized in that When the multiple types of waterway congestion evaluation index values ​​are saturation, the multiple types of waterway congestion evaluation index values ​​are calculated according to the real-time position and trajectory, including: Determining corresponding ship areas based on the real-time positions and trajectories of the different ships in the target waterway; Calculating the ship territory areas of different ships based on the lengths of different ships in the data segments and the corresponding ship territories; The quotient of the sum of the ship domain areas of different ships and the actual navigable water area of ​​the target waterway is determined as the saturation of the target waterway within the data analysis time unit corresponding to the target waterway.

9. A waterway congestion identification device based on AIS big data, characterized in that: The device includes an acquisition module, a determination module, an extraction module and a calculation module; wherein, The acquisition module is used to acquire the ship AIS data of the target waterway; the target waterway is a waterway whose congestion is to be identified, and the ship AIS data records the passage conditions of different ships; The determination module is used to determine the time to be identified of the target waterway, divide the time to be identified based on the distribution characteristics of the ship AIS data, and determine the data analysis time unit; the time to be identified is composed of multiple time periods; the distribution characteristics represent the change trend demarcation point on the ship quantity distribution curve, and the ship quantity distribution curve represents the distribution of the number of ships passing through the target waterway within the time to be identified; The extraction module is used to extract the data segment of the ship AIS data corresponding to each data analysis time unit based on the data analysis time unit; The calculation module is used to calculate the evaluation index values ​​of the congestion degree of multiple types of waterways in the data analysis time unit corresponding to the target waterway based on the data segment; The determination module is further used to determine the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods based on the values ​​of the multiple types of waterway congestion evaluation index values ​​corresponding to the different data segments in each of the time periods; The acquisition module is further used to summarize the multiple types of waterway congestion evaluation index values ​​corresponding to each of the time periods to obtain the waterway traffic characteristics within the time period to be identified; the waterway traffic characteristics combine the multiple types of waterway congestion evaluation index values ​​in chronological order; The determination module is further used to cluster the waterway traffic characteristics and determine the congestion level of the target waterway within the time to be identified according to the clustering result; Among them, the distance between each type of channel congestion evaluation index value in the channel traffic characteristics in each time period and multiple cluster centers is calculated, each cluster center represents a standard channel congestion index value under a channel congestion level; the subclass to which each channel congestion evaluation index value belongs is determined according to the distance; the subclass represents a certain type of channel congestion level; the representative subclass is determined according to the number of each channel congestion evaluation index value in each subclass, and the channel congestion level corresponding to the representative subclass is used as the congestion level of the target channel within the time period to be identified; the representative subclass represents the subclass containing the largest number of channel congestion evaluation index values.

Images