River-sea combined transportation network construction method and device based on AIS data, and storage medium

Abstract

The invention relates to an AIS data-based river-sea combined transportation network construction method and device and a storage medium, and the method comprises the steps: S1, obtaining ship AIS dynamic data and the like, and carrying out the data cleaning of the ship AIS dynamic data; s2, judging whether each data point is a data point in a parking state or not one by one; s3, segmenting the data points of each ship in a berthing state to obtain a plurality of candidate track segments; s4, combining all the candidate track segments to obtain a parking track segment; s5, obtaining a polygonal region of each port based on the port static data; s6, matching each data point in the parking state with the polygonal area of each port; s7, a river-sea combined transportation network is constructed based on the matched parking track segments, the nodes of the river-sea combined transportation network are ports, and the sides of the river-sea combined transportation network are direct routes connecting the two ports; and S8, weighting each edge of the river-sea combined transportation network. Compared with the prior art, the method has the advantages of high accuracy and the like.

Description

Technical Field

[0001] The present invention relates to the field of transportation, and in particular to a method, device and storage medium for constructing a river-sea transport network based on AIS data. Background Art

[0002] As an important part of my country's comprehensive transportation system, river-sea transport has not only promoted the development of regional economy, but also achieved cost reduction and efficiency improvement of logistics. The Yangtze River Economic Belt has become an ideal area for the development of river-sea transport due to its huge freight demand and large river ports and seaports. However, the Yangtze River Basin has complex topography and changeable navigation conditions, which will have a certain impact on the reliability of the river-sea transport network. At present, the research and development and application of river-sea transport have achieved certain achievements at home and abroad, but there are still the following shortcomings.

[0003] (1) Insufficient research on river-sea transport networks

[0004] Existing research mainly focuses on maritime networks, liner shipping networks and land multimodal transport networks, while relatively few studies are conducted on inland waterway shipping and river-sea transport networks. Most studies focus on analyzing port systems, competitiveness, synergies and transport solution optimization, but have limited comprehensive understanding of the entire river-sea transport network.

[0005] (2) Lack of research on new methods of network construction in the big data environment

[0006] The existing methods of constructing shipping networks and multimodal transport networks are difficult to directly apply to river-sea transport networks. Route information is usually not sufficient to distinguish whether it is a specific route for river-sea transport, and the information update of river-sea transport routes is usually slow. River-sea transport involves the connection between inland rivers and sea transport. This special mode of transportation requires the identification of the precise docking points of inland river and ocean sections. Summary of the invention

[0007] The purpose of the present invention is to provide a method, device and storage medium for constructing a river-sea transport network based on AIS data.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A method for constructing a river-sea transport network based on AIS data, comprising:

[0010] Step S1: Obtain ship AIS dynamic data, ship static data and port static data, and perform data cleaning on the ship AIS dynamic data;

[0011] Step S2: for each ship, determine whether each data point is a data point in a docked state;

[0012] Step S3: segmenting the data points of each ship in the docked state to obtain multiple candidate trajectory segments;

[0013] Step S4: merging all candidate trajectory segments to obtain one or more stop trajectory segments;

[0014] Step S5: obtaining polygonal areas of each port based on the static data of the port;

[0015] Step S6: Match each data point in the docking state with the polygonal area of ​​each port. If the data point is within the polygonal area of ​​any port, match the data point to the port;

[0016] Step S7: constructing a river-sea transport network based on the matched docking trajectory segments, wherein the nodes of the river-sea transport network are ports and the edges are direct routes connecting two ports;

[0017] Step S8: Assign weights to each edge of the river-sea transport network.

[0018] The data cleaning process in step S1 specifically includes:

[0019] Step S1-1: Aggregate all ship AIS dynamic data according to MMSI numbers to obtain data points of each ship;

[0020] Step S1-2: sorting the data points of each ship according to the timestamp to obtain the trajectory of each ship;

[0021] Step S1-3: After deleting the data points with abnormal speed in the trajectory of each ship, delete the data points whose longitude and latitude are outside the target range.

[0022] The process of deleting data points with abnormal speed in the trajectory of each ship in step S1-3 includes:

[0023] Step S1-3-1: Calculate the average speed of two data points adjacent to the target data point as a first average speed;

[0024] Step S1-3-2: Calculate the absolute value of the difference between the first average speed and the instantaneous speed of the corresponding target data point as the first difference;

[0025] Step S1-3-3: Calculate the ratio of the first difference and the instantaneous speed of the target data point as the first ratio, and determine whether the first ratio is greater than 50%. If so, the target data point is a data point with abnormal speed.

[0026] The determination process of a single data point in step S2 includes:

[0027] Step S2-1: determine whether the instantaneous speed of the data point is less than the speed threshold, if yes, execute step S2-2;

[0028] Step S2-2: Calculate the time difference between the data point and the previous data point, and determine whether the time difference between the data point and the previous data point is less than a first time interval threshold. If yes, execute step S2-3;

[0029] Step S2-3: Calculate the distance between the data point and the previous data point, and determine whether the distance between the data point and the previous data point is less than a preconfigured first distance interval threshold. If yes, determine that the data point is a data point in a docked state.

[0030] The step S3 comprises:

[0031] Step S3-1: creating a first candidate trajectory segment and selecting the candidate trajectory segment, and adding the first data point in the docked state to the selected candidate trajectory segment;

[0032] Step S3-2: Select a data point whose next state is a stopped state, and determine whether the time interval between the data point and the first data point in the selected candidate trajectory segment is less than a first time interval threshold, and whether the distance is less than a first distance interval threshold. If yes, add the data point to the selected candidate trajectory segment, otherwise, execute step S3-3;

[0033] Step S3-3: Create the next candidate trajectory segment and select the candidate trajectory segment, add the data point to the selected candidate trajectory segment, and determine whether there is a next data point in the docked state. If yes, proceed to step S3-2. Otherwise, multiple candidate trajectory segments are obtained.

[0034] The step S4 comprises:

[0035] Step S4-1: creating and selecting the first stop trajectory segment, and adding the first candidate trajectory segment to the selected stop trajectory segment;

[0036] Step S4-2: Select the next candidate trajectory segment, and determine whether the distance between the central data point of the selected candidate trajectory segment and the central data point of the previous candidate trajectory segment is less than the second distance threshold, and whether the time difference is less than the second time interval threshold. If yes, execute step S4-3, otherwise, execute step S4-4;

[0037] Step S4-3: Check the number of track points. If the check passes, add the selected candidate track segment to the selected stop track segment. Otherwise, execute step S4-4.

[0038] Step S4-4: Create the next docking trajectory segment, select the newly created docking trajectory segment, and add the selected candidate trajectory segment to the selected docking trajectory segment, and determine whether there are untraversed candidate trajectory segments. If yes, return to step S4-2, otherwise, all docking trajectory segments are obtained.

[0039] The step S4-3 is specifically as follows:

[0040] Step S4-3-1: Calculate the data points in the trajectory after the current selected stop trajectory segment and the current selected candidate trajectory segment are merged, as well as the time difference between the first data point and the last data point. If the data points in the merged trajectory are greater than the trajectory point threshold, and the time difference between the first data point and the last data point is greater than the third stop time threshold, the verification passes, and the selected candidate trajectory segment is added to the selected stop trajectory segment. Otherwise, execute step S4-4.

[0041] The step S5 comprises:

[0042] Step S5-1: Calculate the longitude and latitude of the center of gravity of the port based on the longitude and latitude of all anchorages in the port;

[0043] Step S5-2: According to the longitude and latitude of each port's anchorage, combined with the longitude and latitude of the port's centroid, obtain the angle between each anchorage and the port's centroid and any latitude ray, and sort them according to the angle;

[0044] Step S5-3: construct a polygonal area using the sequence of warp dimension values ​​of the sorted anchorages.

[0045] The step S7 comprises:

[0046] Step S7-1: taking the ports with matching trajectory points as nodes of the river-sea transport network;

[0047] Step S7-2: Generate an edge connecting the two ports according to the ports corresponding to the two docking trajectory segments.

[0048] The weighting process is implemented by using the number of ship visits and / or the length of stopover time between two ports.

[0049] Compared with the prior art, the present invention has the following beneficial effects: the candidate berthing trajectory segments of the ship are extracted and the berthing trajectory segments are merged by utilizing the ship's speed, the distance between adjacent trajectory points and the dwelling time, and by judging whether the ship's berthing point is within the polygon of the port anchorage and berth, the ship's berthing point is matched with the port node and a river-sea transport network is constructed, and the number of ship visits or the ship's dwelling time is used as the weight to construct the river-sea transport network, thereby realizing a model construction method that can truly reflect the actual situation of river-sea transport, and providing technical support for shipping companies, logistics companies and related departments in optimizing the river-sea transport network structure, improving risk management and formulating more effective transportation strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0051] Figure 2 The schematic diagram of the identification process of candidate docking trajectory segments of a ship;

[0052] Figure 3 Schematic diagram of the candidate trajectory segment merging process;

[0053] Figure 4 It is a flowchart of the convex hull method;

[0054] Figure 5 It is a topology model network model diagram;

[0055] Figure 6 This is the topological map of the river-sea transport network in the Yangtze River region obtained on the Space L topological model. DETAILED DESCRIPTION

[0056] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0057] A method for constructing a river-sea transport network based on AIS data, such as Figure 1 As shown, including:

[0058] Step S1: Obtain ship AIS dynamic data, ship static data and port static data, and perform data cleaning on the ship AIS dynamic data;

[0059] Before explaining the specific embodiments, the source of the AIS data of this application is explained first:

[0060] The data sources of this application mainly include ship data and port static data, among which the ship data include ship dynamic AIS data and ship static data. The ship AIS dynamic data, its data fields include longitude and latitude, timestamp, MMSI number, speed, turning angle, draft depth and other data; the ship static data includes the ship MMSI number, ship length, width, height, ship type and other information; the port static data includes port production data and port geographic information data, the port production data mainly includes port throughput and its designed throughput, port type; the port geographic information data includes the port geographical location, port matching radius and other information.

[0061] The description of each data field is shown in Table 1:

[0062] Table 1

[0063]

[0064] To facilitate subsequent operations, the data fields of the original AIS data need to be uniformly adjusted in data type and format. The adjusted AIS data example is shown in Table 2:

[0065] Table 2

[0066]

[0067]

[0068] After adjusting the AIS data format, it is necessary to clean the AIS data because it contains certain abnormal values ​​and duplicate values. The cleaning steps are as follows:

[0069] Step S1-1: Match the MMSI field of the AIS data with the MMSI field of the ship static data, remove the AIS data of the ship that is not in the ship static data, and then aggregate all the ship AIS dynamic data according to the MMSI number to obtain the data point of each ship;

[0070] Step S1-2: sorting the data points of each ship according to the timestamp to obtain the trajectory of each ship;

[0071] Step S1-3: After deleting the data points with abnormal speed in the trajectory of each ship, delete the data points whose longitude and latitude are outside the target range.

[0072] Now, we delete them based on the speed anomaly and then delete them based on whether the longitude and latitude are within the target range, so as to improve the screening accuracy of speed anomaly.

[0073] The process of deleting data points with abnormal speed in the trajectory of each ship in step S1-3 includes:

[0074] Step S1-3-1: Calculate the average speed of two data points adjacent to the target data point as a first average speed;

[0075] Step S1-3-2: Calculate the absolute value of the difference between the first average speed and the instantaneous speed of the corresponding target data point as the first difference;

[0076] Step S1-3-3: Calculate the ratio of the first difference and the instantaneous speed of the target data point as the first ratio, and determine whether the first ratio is greater than 50%. If so, the target data point is a data point with abnormal speed.

[0077] Step S2: for each ship, determine whether each data point is a data point in a docked state, wherein the determination process for a single data point includes:

[0078] Step S2-1: determine whether the instantaneous speed of the data point is less than the speed threshold, if yes, execute step S2-2;

[0079] Step S2-2: Calculate the time difference between the data point and the previous data point, and determine whether the time difference between the data point and the previous data point is less than a first time interval threshold. If yes, execute step S2-3;

[0080] Step S2-3: Calculate the distance between the data point and the previous data point, and determine whether the distance between the data point and the previous data point is less than a preconfigured first distance interval threshold. If yes, determine that the data point is a data point in a docked state.

[0081] In this embodiment, the speed threshold is 1 knot, the first time interval threshold is 5400 seconds, and the first distance interval threshold is 2000 meters.

[0082] Step S3: Figure 2 As shown, the data points of each ship in the docked state are segmented to obtain multiple candidate trajectory segments. Step S3 includes:

[0083] Step S3-1: creating a first candidate trajectory segment and selecting the candidate trajectory segment, and adding the first data point in the docked state to the selected candidate trajectory segment;

[0084] Step S3-2: Select a data point whose next state is a stopped state, and determine whether the time interval between the data point and the first data point in the selected candidate trajectory segment is less than a first time interval threshold, and whether the distance is less than a first distance interval threshold. If yes, add the data point to the selected candidate trajectory segment, otherwise, execute step S3-3;

[0085] Step S3-3: Create the next candidate trajectory segment and select the candidate trajectory segment, add the data point to the selected candidate trajectory segment, and determine whether there is a next data point in the docked state. If yes, proceed to step S3-2. Otherwise, multiple candidate trajectory segments are obtained.

[0086] The candidate trajectory segment of the ship is denoted as ST i =(s1,s2,…,s m ), (m≤n), where s m is the candidate trajectory segment ST i The mth stop of

[0087] The set of candidate ship docking trajectory segments is denoted as S ST ={ST1,ST2,…,ST j}, where ST j represents the jth candidate trajectory segment of the ship;

[0088] like Figure 2 As shown, traverse the trajectory points p1, p2, pn of a ship, initialize the candidate trajectory segment, candidate trajectory segment i = i + 1, and judge whether the speed, time and distance difference of the candidate trajectory segment meet formula 1. If so, add it to the candidate trajectory segment i, and finally form a complete set of docking trajectory segments.

[0089] Step S4: Merge all candidate trajectory segments to obtain one or more stop trajectory segments, such as Figure 4 As shown, including:

[0090] Step S4-1: creating and selecting the first stop trajectory segment, and adding the first candidate trajectory segment to the selected stop trajectory segment;

[0091] Step S4-2: Select the next candidate trajectory segment, and determine whether the distance between the central data point of the selected candidate trajectory segment and the central data point of the previous candidate trajectory segment is less than the second distance threshold, and whether the time difference is less than the second time interval threshold. If yes, execute step S4-3, otherwise, execute step S4-4;

[0092] Among them, the coordinates of the center point are the average of the longitude and latitude coordinates of each trajectory segment.

[0093] Step S4-3: Check the number of track points. If the check passes, add the selected candidate track segment to the selected stop track segment. Otherwise, execute step S4-4.

[0094] Step S4-4: Create the next docking trajectory segment, select the newly created docking trajectory segment, and add the selected candidate trajectory segment to the selected docking trajectory segment, and determine whether there are untraversed candidate trajectory segments. If yes, return to step S4-2, otherwise, all docking trajectory segments are obtained.

[0095] Among them, since the ship may dock during the journey, the rationality of the merged trajectory segment needs to be further determined. By detecting the total number of trajectory points and the docking time in the merged trajectory segment, the impact of these abnormal docking data can be minimized, and the docking trajectory segment of the ship docking in the port area can be screened out. Step S4-3 is specifically as follows:

[0096] Step S4-3-1: Calculate the data points in the trajectory after the current selected stop trajectory segment and the current selected candidate trajectory segment are merged, as well as the time difference between the first data point and the last data point. If the data points in the merged trajectory are greater than the trajectory point threshold, and the time difference between the first data point and the last data point is greater than the third stop time threshold, the verification passes, and the selected candidate trajectory segment is added to the selected stop trajectory segment. Otherwise, execute step S4-4.

[0097] In this embodiment, after merging, the coordinates of the center point of each docking track segment are used as the docking point coordinates, and the docking time is the sum of the docking time of the track points contained in the merged track segment. The following is an example of the merged data. Before processing, there are 300,000,000 AIS data, and finally 50387 ship docking point data are obtained. The distribution of ship docking point data is shown in Table 3.

[0098] Table 3

[0099]

[0100] Step S5: Based on the static data of ports, the polygonal area of ​​each port is obtained, such as Figure 5 As shown, including:

[0101] Step S5-1: Calculate the longitude and latitude of the center of gravity of the port based on the longitude and latitude of all anchorages in the port;

[0102] Step S5-2: According to the longitude and latitude of each port's anchorage, combined with the longitude and latitude of the port's centroid, obtain the angle between each anchorage and the port's centroid and any latitude ray, and sort them according to the angle;

[0103] Step S5-3: construct a polygonal area using the sequence of warp dimension values ​​of the sorted anchorages.

[0104] In this embodiment, as shown in Table 4, the geographical location data examples of the longitude and latitude of the port anchorage are:

[0105] Table 4

[0106]

[0107] Step S6: Match each data point in the docking state with the polygonal area of ​​each port. If the data point is within the polygonal area of ​​any port, match the data point to the port;

[0108] Step S7: construct a river-sea transport network based on the matched docking trajectory segments. The nodes of the river-sea transport network are ports, and the edges are direct routes connecting two ports, including:

[0109] Step S7-1: taking the ports with matching trajectory points as nodes of the river-sea transport network;

[0110] Step S7-2: Generate an edge connecting the two ports according to the ports corresponding to the two docking trajectory segments.

[0111] Specifically, the process uses the Space L topological network model to construct a river-sea transport network. The nodes in the Space L network model are the ports, and the edges of the network are the routes of ships between the two ports.

[0112] like Figure 5 As shown in the figure, it is the Space L river-sea transport network model, which assumes that Shanghai Port is directly connected to Suzhou Port and Nantong Port; Suzhou Port is directly connected to Nantong Port and Yangzhou Port; Nantong Port is directly connected to Yangzhou Port; Wuhan Port, Yichang Port and Yueyang Port form another directly connected area; Shenzhen Port and Wuhu Port are separate nodes and are not directly connected to other nodes. Based on the above assumptions, the network adjacency matrix A can be obtained: dj , where if a ij =1, indicating that node i is connected to node j.

[0113]

[0114] like Figure 6 As shown, Figure 4 Based on the Space L topology model, the topology map of the river-sea transport network in the Yangtze River region was made, which matched a total of 581 river-sea transport ships and 21 port nodes.

[0115] The river-sea transport network constructed only with port nodes and routes is an unweighted topological network structure. However, in real situations, the number of ships docked at each port node in the network is different, and the ship docking time is different, which causes obvious differences in the attributes of each port node and route. Therefore, it is necessary to give weights to the constructed river-sea transport network in order to analyze the river-sea transport network more comprehensively and specifically.

[0116] Step S8: weighting each edge of the river-sea transport network, and the weighting process is implemented by the number of ship visits and / or the docking time between two ports.

[0117] Specifically, the weight is assigned by the number of visits of ships between ports, which can directly reflect the degree of shipping activities between ports. The more visits, the greater the transportation demand or route flow of ships between the two ports.

[0118] Alternatively, the length of time a ship stays at a port can be used as a weight, which mainly reflects the transport intensity of the port. The longer a ship stays at a port, the greater the demand for cargo loading and unloading operations at the port, and the greater the transport intensity of the port node.

[0119] The results obtained by the two weighting methods are basically the same. The basic parameters of the weighted network are as follows:

[0120] (1)V node ={v1,v2,…,v 21}: represents the network port node set, a total of 21 nodes;

[0121] (2)E edge ={e ij |v i ,v j ∈V node}: represents the edge set of the network, a total of 31 edges;

[0122] (3)W edge ={w ij |v i ,v j ∈V node}: represents the weight set of all edges.

[0123] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

Claims

1. A method for constructing a river-sea transport network based on AIS data, characterized in that: include: Step S1: Obtain ship AIS dynamic data, ship static data and port static data, and perform data cleaning on the ship AIS dynamic data; Step S2: for each ship, determine whether each data point is a data point in a docked state; Step S3: segmenting the data points of each ship in the docked state to obtain multiple candidate trajectory segments; Step S4: merging all candidate trajectory segments to obtain one or more stop trajectory segments; Step S5: obtaining polygonal areas of each port based on the static data of the port; Step S6: Match each data point in the docking state with the polygonal area of ​​each port. If the data point is within the polygonal area of ​​any port, match the data point to the port; Step S7: constructing a river-sea transport network based on the matched docking trajectory segments, wherein the nodes of the river-sea transport network are ports and the edges are direct routes connecting two ports; Step S8: Assign weights to each edge of the river-sea transport network.

2. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The data cleaning process in step S1 specifically includes: Step S1-1: Aggregate all ship AIS dynamic data according to MMSI numbers to obtain data points of each ship; Step S1-2: sorting the data points of each ship according to the timestamp to obtain the trajectory of each ship; Step S1-3: After deleting the data points with abnormal speed in the trajectory of each ship, delete the data points whose longitude and latitude are outside the target range.

3. The method for constructing a river-sea transport network based on AIS data according to claim 2, characterized in that: The process of deleting data points with abnormal speed in the trajectory of each ship in step S1-3 includes: Step S1-3-1: Calculate the average speed of two data points adjacent to the target data point as a first average speed; Step S1-3-2: Calculate the absolute value of the difference between the first average speed and the instantaneous speed of the corresponding target data point as the first difference; Step S1-3-3: Calculate the ratio of the first difference and the instantaneous speed of the target data point as the first ratio, and determine whether the first ratio is greater than 50%. If so, the target data point is a data point with abnormal speed.

4. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The determination process of a single data point in step S2 includes: Step S2-1: determine whether the instantaneous speed of the data point is less than the speed threshold, if yes, execute step S2-2; Step S2-2: Calculate the time difference between the data point and the previous data point, and determine whether the time difference between the data point and the previous data point is less than a first time interval threshold. If yes, execute step S2-3; Step S2-3: Calculate the distance between the data point and the previous data point, and determine whether the distance between the data point and the previous data point is less than a preconfigured first distance interval threshold. If yes, determine that the data point is a data point in a docked state.

5. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The step S3 comprises: Step S3-1: creating a first candidate trajectory segment and selecting the candidate trajectory segment, and adding the first data point in the docked state to the selected candidate trajectory segment; Step S3-2: Select a data point whose next state is a stopped state, and determine whether the time interval between the data point and the first data point in the selected candidate trajectory segment is less than a first time interval threshold, and whether the distance is less than a first distance interval threshold. If yes, add the data point to the selected candidate trajectory segment, otherwise, execute step S3-3; Step S3-3: Create the next candidate trajectory segment and select the candidate trajectory segment, add the data point to the selected candidate trajectory segment, and determine whether there is a next data point in the docked state. If yes, proceed to step S3-2. Otherwise, multiple candidate trajectory segments are obtained.

6. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The step S4 comprises: Step S4-1: creating and selecting the first stop trajectory segment, and adding the first candidate trajectory segment to the selected stop trajectory segment; Step S4-2: Select the next candidate trajectory segment, and determine whether the distance between the central data point of the selected candidate trajectory segment and the central data point of the previous candidate trajectory segment is less than the second distance threshold, and whether the time difference is less than the second time interval threshold. If yes, execute step S4-3, otherwise, execute step S4-4; Step S4-3: Check the number of track points. If the check passes, add the selected candidate track segment to the selected stop track segment. Otherwise, execute step S4-4. Step S4-4: Create the next docking trajectory segment, select the newly created docking trajectory segment, and add the selected candidate trajectory segment to the selected docking trajectory segment, and determine whether there are untraversed candidate trajectory segments. If yes, return to step S4-2, otherwise, all docking trajectory segments are obtained.

7. The method for constructing a river-sea transport network based on AIS data according to claim 6, characterized in that: The step S4-3 is specifically as follows: Step S4-3-1: Calculate the data points in the trajectory after the current selected stop trajectory segment and the current selected candidate trajectory segment are merged, as well as the time difference between the first data point and the last data point. If the data points in the merged trajectory are greater than the trajectory point threshold, and the time difference between the first data point and the last data point is greater than the third stop time threshold, the verification passes, and the selected candidate trajectory segment is added to the selected stop trajectory segment. Otherwise, execute step S4-4.

8. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The step S5 comprises: Step S5-1: Calculate the longitude and latitude of the center of gravity of the port based on the longitude and latitude of all anchorages in the port; Step S5-2: According to the longitude and latitude of each port's anchorage, combined with the longitude and latitude of the port's centroid, obtain the angle between each anchorage and the port's centroid and any latitude ray, and sort them according to the angle; Step S5-3: construct a polygonal area using the sequence of warp dimension values ​​of the sorted anchorages.

9. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The step S7 comprises: Step S7-1: taking the ports with matching trajectory points as nodes of the river-sea transport network; Step S7-2: Generate an edge connecting the two ports according to the ports corresponding to the two docking trajectory segments.

10. The method for constructing a river-sea transport network based on AIS data according to claim 1, characterized in that: The weighting process is implemented by using the number of ship visits and / or the length of stopover time between two ports.

Images