Ship anchorage identification method based on AIS data

Abstract

The application provides a ship habitual anchorage identification method based on AIS data, and belongs to the technical field of ship habitual anchorage identification. A parallel processing framework is designed, AIS data of ships in one year is researched, and the overall navigation situation of ships in a water area is analyzed based on AIS historical data to determine the ship habitual anchorage. The steps are as follows: firstly, AIS data in the anchoring period of the ship is screened by parallel analysis of the AIS data; secondly, the plan of the research area is processed by gridding to identify the anchoring position of the ship; finally, the ship habitual anchorage is determined in combination with the actual situation. The application can efficiently and accurately identify the ship habitual anchorage according to the AIS data generated in the navigation process of the ship, and provide a decision basis for the traffic organization and management of the water traffic management department and the planning and design of the channel network.

Description

Technical Field

[0001] This invention belongs to the field of ship habitual anchorage identification technology, and involves ship AIS data processing, especially large-scale parallel data processing and ship navigation situation analysis methods. Background Technology

[0002] With the increasing development of shipping and trade, improving the standardization of waterway system operations and rationally planning and designing coastal waterway networks have become crucial foundations for further enhancing waterway safety and efficiency. Due to the complex natural conditions of coastal waters and the significant differences in navigation habits among vessels of different tonnages and cargo types, vessels often need to anchor outside of designated anchorages by the transportation authorities during actual navigation. Therefore, to further improve the scientific and rational organization and management of waterway traffic, it is urgent to analyze and identify the specific locations of vessels' habitual anchorages based on their navigation records.

[0003] Automatic Identification System (AIS) data consists of real-time information broadcast by ships during navigation, including their location and navigation status. Ships update their latitude and longitude coordinates approximately every 6 minutes, and this data is received by shore-based base stations and stored in a database. Some techniques exist for analyzing ship collision risks and predicting maritime traffic flow based on AIS data. However, existing technologies primarily focus on the logical analysis of the data itself. In practice, the water areas involved in these projects are generally large, with AIS data reaching up to 2TB per year. Commonly used decoding algorithms are time-consuming, and computers are prone to failure due to prolonged continuous high-speed operation. Therefore, there is still no mature technology for efficiently reading and analyzing large-scale AIS data in busy waterways and identifying ships' habitual anchorages based on the analysis results.

[0004] This invention designs a framework for parallel processing of AIS data and proposes a method for determining habitual anchorages in waterways based on historical AIS data analysis. Summary of the Invention

[0005] The purpose of this invention is to provide a method for quickly reading and parsing AIS data and analyzing the habitual anchorages of ships in waterways based on the data processing results.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for identifying habitual anchorages of ships based on AIS data is proposed. A parallel processing framework is designed to study ship AIS data from the past year, separating the data reading and processing steps and executing them in batches simultaneously. Based on historical AIS data analysis, the overall navigational behavior of ships in the waterway is determined to identify habitual anchorages.

[0008] The specific steps are as follows:

[0009] Step one: Parallel analysis of AIS data

[0010] AIS data processing is mainly divided into two steps of data reading and data analysis, and the sequence is clear. Data can be processed in segments, but it must be read first and then analyzed. There is no other coupling relationship between the two. Therefore, try to make full use of the parallel computing capability of multi-core computers, divide the data (for example, by a week), start thread 1 to read the ship AIS data from the database, and store it in the computer memory, and start thread 2 to analyze the data read. After the data analysis starts, thread 1 can continue to read new data, and the read data is queued for thread 2 processing.

[0011] Among them, at least the Maritime Mobile Service Identify (MMSI), longitude, latitude, and time fields should be retained during data reading.

[0012] The data analysis process is shown in steps two to four.

[0013] Step two: Screening AIS data of the ship anchoring period

[0014] According to the ship navigation speed to determine whether the ship is anchored, the AIS data record of the ship during anchoring is screened, and the specific steps are as follows:

[0015] (1) According to the ship MMSI code, extract the navigation track of each ship, and arrange the data records in ascending order according to the time field.

[0016] (2) For each ship's AIS record, starting from the second time point t, calculate the real-time speed of ship i as follows:

[0017] dx = (x i,t -x i,t-1 ) × 111 × cos

[0018] dy = (y i,t -y i,t-1 ) × 111

[0019] Where dx represents the moving distance of the ship in the longitude direction, in kilometers. x i,t represents the longitude of ship i at time t. θ represents the latitude at time t. dy represents the moving distance of the ship in the longitude direction, in kilometers. y i,t represents the latitude of ship i at time t.

[0020]

[0021] Where vi,t Let Δt represent the speed of ship i at time t, and let Δt represent the actual time difference between time t and time t-1.

[0022] (3) For each ship's AIS records, filter and retain the records corresponding to times when the speed is less than 2 knots, and delete the rest of the records.

[0023] Step 3: Rasterize the study area plan to identify ship anchorage locations.

[0024] The study waters were divided into grids, with grid sides ranging from 100m to 500m. The number of ships anchored within each grid was counted. The specific steps are as follows:

[0025] (1) Extract the date from the "Time" field in the AIS record, remove the hour, minute and second information to form the "Date" field, and denote the newly generated AIS data set as R.

[0026] (2) Based on the ship's AIS records, cycle through all grids within the study area. If the point determined by the "longitude" and "latitude" of a record falls exactly within grid i, then mark that record. Filter the marked records to form a set R. i ,have

[0027] (3) Based on the fields "MMSI" and "Date", R i A filtering process is performed to remove duplicate values, thus preventing signals emitted by ships anchored at the same location on the same day from being counted repeatedly. The filtering method is as follows:

[0028]

[0029] in This indicates the filtering results.

[0030] (4) Count the number of ships anchored in each grid to form a schematic diagram of anchorage distribution density. The method for calculating the number of ships anchored in grid i is as follows:

[0031]

[0032] Step 4: Determine the ship's customary anchorage based on actual conditions.

[0033] Based on the anchorage distribution diagram obtained from AIS data analysis, and taking into account anchorage planning and water area characteristics (such as whether it involves military or fishing use), the areas where more than 20 ships are anchored within two months are identified as anchorages.

[0034] The effects and benefits of this invention are:

[0035] The application can efficiently and accurately identify the ship anchorage habit according to the AIS data generated during the ship navigation process, so as to provide decision basis for the traffic organization management and the channel network planning and design of the water traffic management department. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is a schematic diagram of ship anchorage distribution, in which the color depth represents the relative size of the number of anchored ships in each grid. The darker the color, the more ships anchored in the grid. Among them, 1, 2 and 3 represent part of the identified habit anchorage. DETAILED DESCRIPTION

[0037] The application will be further described below in combination with specific embodiments.

[0038] Step one: parallel analysis of AIS data

[0039] The ship AIS database is used to obtain the ship AIS data near Hangzhou Bay within the time range of January 2022-February 2022, and the spatial range is a rectangular region between two points (31.00N, 121.59E) and (30.23N, 122.51E). The data size is about 2.55G, and there are 13332341 records.

[0040] The method proposed in the application is used to analyze the AIS data. Through statistics, the decoding speed reaches an average of 22262.89 rows of AIS messages per second.

[0041] Step two: screening AIS data of the ship anchoring time period

[0042] According to the method proposed in step two in the content of the application, the ship sailing speed is calculated and the AIS message records of the ship sailing speed less than 2 knots are retained, and other records are deleted. A total of 7046178 valid AIS messages are obtained, containing information of ** ships.

[0043] Step three: grid processing of the research area plan, identifying the ship anchoring position

[0044] According to the method proposed in step three in the content of the application, the research area is divided into 500m×500m grids. The number of anchored ships in each grid is calculated to form a schematic diagram of ship anchorage distribution. The color depth represents the relative size of the number of anchored ships in the grid, and the darker the color, the more ships anchored. The results are shown in the attached Figure 1 .

[0045] Step four: determine the ship habit anchorage in combination with the actual situation

[0046] According to the calculation results of the ship anchoring position, in combination with the actual situation, it can be known that the ship anchoring positions are relatively dense near (30.542N, 121.933E) and (30.5468N, 122.247E).

[0047] The above-described embodiments only express the implementation of the present application, but cannot be interpreted as a limitation on the scope of the patent of the present application. It should be pointed out that, for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which all belong to the protection scope of the present application.

Claims

1. An AIS data-based ship habitual anchorage identification method, characterized in that, A parallel processing framework is designed to study the AIS data of ships within one year, and the overall navigation situation of ships in the water area is analyzed based on the AIS historical data to determine the habitual anchorage of the ships. The steps are as follows: Step one: parallel analysis of AIS data AIS data processing includes data reading and data analysis, and the two steps are separated and executed in batches and synchronously. In the data reading process, at least the following four fields should be retained: ship mobile communication service identification code MMSI, longitude, latitude, and time; Step two: determine whether the ship is anchored according to the ship navigation speed, and screen the AIS data records of the ship during anchoring: (1) According to the ship MMSI code, the navigation track of each ship is extracted, and the data records are arranged in ascending order according to the time field; (2) For each ship's AIS record, from the second time point t, the real-time speed of ship i is calculated as follows: ； ； wherein, represents the moving distance of the ship in the longitude direction, in units of kilometers; represents the longitude of the ship i at time t; represents the latitude of the ship at time t; represents the moving distance of the ship in the longitude direction, in units of kilometers; represents the latitude of the ship i at time t; ； wherein, denotes the speed of the ship i at time t, denotes the actual time difference of the ship at time t and t-1. (3) For each ship's AIS record, the records corresponding to the time when the speed is less than 2 knots are screened and retained, and the remaining records are deleted; Step three: grid processing of the plan view of the study area to identify the anchoring position of the ship The study area is grid processed to count the number of anchored ships in each grid: (1) According to the "time" field in the AIS record, the date is extracted, the hour, minute and second information is removed, the "date" field is formed, and the newly generated AIS data set is recorded as ; (2) According to the ship AIS records, make a cycle for all grids in the study area; if the point determined by "longitude" and "latitude" in the record falls in the grid , mark the record; screen the marked records to form a set , ; (3) Based on the fields "MMSI" and "Date" To avoid duplicate counting of signals emitted by ships anchored at the same location on the same day, a filtering process is performed to remove duplicate values. The filtering method is as follows: ； wherein, represents the screening result; (4) The number of anchored ships in each grid is counted to form an anchor distribution density diagram. The number of anchored ships in each grid is calculated by the following formula: wherein, N is the number of anchored ships in each grid, A is the area of each grid, and V is the volume of each ship. ； Step four: according to the anchoring position distribution diagram obtained by analyzing the AIS data, considering the anchorage planning and the nature of the water area, and combining the actual situation to determine the habitual anchorage of the ship.

2. The AIS data-based ship habitual anchorage identification method according to claim 1, characterized in that, The grid length of step three is 100m-500m.

3. The AIS data-based ship habitual anchorage identification method according to claim 1, characterized in that, In step four, the area where the grid with more than 20 ships anchored within 2 months is determined as the anchorage.

Images