CN115294802A - AIS data-based ship navigation state intelligent identification method and system - Google Patents
AIS data-based ship navigation state intelligent identification method and system Download PDFInfo
- Publication number
- CN115294802A CN115294802A CN202210876444.6A CN202210876444A CN115294802A CN 115294802 A CN115294802 A CN 115294802A CN 202210876444 A CN202210876444 A CN 202210876444A CN 115294802 A CN115294802 A CN 115294802A
- Authority
- CN
- China
- Prior art keywords
- navigation
- speed
- ship
- port
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000012545 processing Methods 0.000 claims abstract description 68
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000010801 machine learning Methods 0.000 claims abstract description 16
- 238000012544 monitoring process Methods 0.000 claims abstract description 16
- 238000005265 energy consumption Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims description 29
- 230000011218 segmentation Effects 0.000 claims description 29
- 238000011049 filling Methods 0.000 claims description 17
- 238000004140 cleaning Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 4
- 238000004873 anchoring Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G3/00—Traffic control systems for marine craft
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Ocean & Marine Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Navigation (AREA)
Abstract
The invention provides an AIS data-based intelligent identification method and system for ship navigation states, wherein the method utilizes AIS data, adopts a machine learning algorithm to perform sectional processing according to the change of the speed, combines business logic to identify the navigation states, arranges relevant information of a navigation line according to a port where the navigation line passes, further performs ship navigation monitoring, calculates the congestion state and navigation energy consumption of the port where the navigation line passes, is convenient to adjust the operation navigation line in time, and is worthy of popularization and application.
Description
Technical Field
The invention relates to the technical field of shipping informatization, in particular to an AIS data-based intelligent ship navigation state identification method and system.
Background
Since 2020, the spread of new crown epidemic worldwide has affected most industries to varying degrees. The prominent effect on the marine market is port congestion.
The port is used as an important node of a global supply chain, and the congestion directly influences the operation efficiency of the ship. Before the end of 2021, the port was long-term congested due to the lack of labor resulting from massive infection of new crown viruses by handling workers in los angeles and long beach ports, causing a break in the container ship transport chain. According to the statistics of the los angeles port, 90% of ships directly drive to the anchoring, and are added into a waiting loading and unloading queue, and the average anchoring time reaches about 180 hours. At present, the industry has less research on invalid sailing (or referred to as a drifting state) of ships arriving at a port, the drifting time can only be checked manually through a ship view treasure, and the displacement state of the ship time and space cannot be solved by utilizing an algorithm.
Disclosure of Invention
The invention provides an intelligent ship navigation state identification method based on AIS data, aiming at solving the problems that the prior navigation state identification needs to manually search the drift time and can not carry out effective intelligent calculation, and the like. The invention also relates to an intelligent ship navigation state identification system based on the AIS data.
The technical scheme of the invention is as follows:
an AIS data-based intelligent ship navigation state identification method is characterized by comprising the following steps:
a data acquisition and processing step, namely acquiring AIS data in the ship navigation process, and cleaning, filling and down-sampling the AIS data;
a navigation speed segmentation processing step, which is to perform segmentation processing on the navigation speed in the collected and processed AIS data through a machine learning algorithm and calculate the average speed and mileage of each segment;
a navigation characteristic processing step, namely analyzing the segmented processing result according to the average speed and the mileage of each segment and by combining service logic, and identifying each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state;
a course information extraction step, namely identifying the starting time and the position of each segmental navigation state in the navigation process of the ship based on the obtained navigation state, carrying out time information matching on the navigation process of the ship, and calculating effective navigation time and drift time;
and a navigation monitoring and analyzing step, namely monitoring the navigation states of any two ports where the airlines pass through, carrying out navigation speed statistics and navigation speed change analysis, and further calculating the congestion state and navigation energy consumption of the ports where the airlines pass through.
Preferably, in the data collection processing step, the collected AIS data includes AIS static and dynamic information of the container ship, container shift port condition, and port information; the cleaning, filling and downsampling processing of the AIS data comprises removing data of which the AIS data loss amount exceeds a threshold value, repairing ships, having a missing starting port, having a missing ending port or being unknown at a port, removing data of which the starting port and the ending port are the same port, respectively matching the starting port and the ending port to corresponding countries and regions, complementally filling the data of which the AIS data loss amount does not exceed the threshold value by using a proximity value, and downsampling the instantaneous speed between any two ports.
Preferably, in the navigational speed segmentation processing step, a variable point is searched by a fusion lasso method or a variable point analysis method in a machine learning algorithm, and further the navigational speed in the collected and processed AIS data is segmented.
Preferably, in the step of processing the navigation characteristics, the obtained segmented navigation speed is interpreted by combining with business logic, a navigation section within a threshold range of a specific navigation speed represents normal navigation, a navigation section with the navigation speed close to 0 section and the duration exceeding a specific time threshold represents drift navigation, and the normal navigation includes navigation states of a low-speed staring section, a high-speed stable section and a deceleration abrupt change section.
Preferably, in the airline information extracting step, the airline-related information is further classified, including: matching the airlines to corresponding countries and regions according to port information of the airlines passing through; finding out the starting time and the ending time of the flight segment, and determining the time information of the relevant year of the flight line operation; finding the operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to destination ports, including recent congested ports and normal ports.
An intelligent identification system for ship navigation state based on AIS data is characterized by comprising a data acquisition processing module, a navigation speed segmentation processing module, a navigation characteristic processing module, a route information extraction module and a navigation monitoring analysis module which are connected in sequence,
the data acquisition and processing module is used for acquiring AIS data in the ship sailing process and cleaning, filling and down-sampling the AIS data;
the navigation speed segmentation processing module is used for carrying out segmentation processing on the navigation speed in the collected and processed AIS data through a machine learning algorithm and calculating the average speed and the mileage of each section;
the navigation characteristic processing module analyzes the segmented processing result according to the average speed and the mileage of each segment and by combining with service logic, and identifies each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state;
the route information extraction module is used for identifying the starting time and the position of each segmental navigation state in the navigation process of the ship based on the obtained navigation state, carrying out time information matching on the navigation process of the ship and calculating the effective navigation time and the drifting time;
the navigation monitoring and analyzing module monitors the navigation states of any two ports where the airline passes through, carries out navigation speed statistics and navigation speed change analysis, and further calculates the congestion state and navigation energy consumption of the port where the airline passes through.
Preferably, the AIS data collected by the data collection and processing module includes AIS static and dynamic information of the container ship, container shift port condition, and port information; the cleaning, filling and downsampling processing of the AIS data comprises removing data of which the AIS data loss amount exceeds a threshold value, repairing ships, having a missing starting port, having a missing ending port or being unknown at a port, removing data of which the starting port and the ending port are the same port, respectively matching the starting port and the ending port to corresponding countries and regions, complementally filling the data of which the AIS data loss amount does not exceed the threshold value by using a proximity value, and downsampling the instantaneous speed between any two ports.
Preferably, the navigational speed segmentation processing module searches for a variable point through a fusion lasso method or a variable point analysis method in a machine learning algorithm, and then performs segmentation processing on the navigational speed in the collected and processed AIS data.
Preferably, the navigation feature processing module interprets the obtained segmented navigation speed by combining with business logic, and represents a navigation section within a specific navigation speed threshold value range as normal navigation, and represents drift navigation for a navigation section with the navigation speed close to 0 section and the duration exceeding a specific time threshold value, wherein the normal navigation includes the navigation states of a low-speed staring section, a high-speed stable section and a deceleration abrupt change section.
Preferably, the route information extraction module further classifies route related information, including: matching the airline to corresponding countries and regions according to port information where the airline passes through; finding out the starting time and the ending time of the flight segment, and determining the time information of the relevant year of the flight line operation; finding the operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to destination ports, including recent congested ports and normal ports.
The invention has the beneficial effects that:
the invention provides an intelligent ship navigation state identification method based on AIS data, which can use a PostgreSQL database to store all necessary AIS data and identified navigation speed related information, adopts a machine learning algorithm to perform segmentation processing according to the change of the navigation speed, combines business logic to identify the navigation state, and arranges the related information of the navigation line according to a port through which the navigation line passes so as to monitor the navigation of a ship. The invention discloses an intelligent Identification method based on a ship navigation state in an Automatic Identification System (AIS), which utilizes AIS data, adopts a machine learning algorithm to perform segmentation processing according to the change of the navigation speed, calculates the average speed and mileage after segmentation to find and mark the time and the position of the start of an invalid voyage, judges the navigation state of a ship, calculates the effective navigation time and the drift time of the ship at a port to further evaluate the congestion state of the port, and monitors the navigation of the ship according to all ports where a navigation line passes. The data sources of the invention are all real navigation data-AIS data in the container shift history, thereby ensuring the authenticity and accuracy of the identified navigation state; a unified standard is defined for all navigation states, support is provided for navigation monitoring and port congestion analysis, ship type information of each air route is kept, and follow-up dynamic query of the navigation states, the air routes and the ship types between any two ports is facilitated. At present, the AIS system provides for the ship navigation state: 0 represents "sailing", 1 represents "mooring", and 5 represents "mooring". However, the navigation of 0 navigation cannot be recognized, so that the field is only seen to generate misjudgment, and effective navigation, invalid navigation and inefficient navigation cannot be distinguished. The invention can calculate the normal navigation time and the invalid navigation (i.e. drift) time, provides each time point of the segments of the AIS, and does not need to manually search the drift time. The intelligent ship navigation state identification method solves the problem that the conventional navigation state identification needs manual searching of the drift time and cannot carry out effective large-scale calculation, changes of the speed and the mileage of related ships are inspected by setting a destination port, the time and the position of the start of an invalid route are found and identified by calculating the average speed and the mileage in a segmented mode, and then the real ship waiting time is calculated to evaluate the port congestion state and judge the energy consumption level of ship navigation.
The invention also relates to an intelligent ship navigation state recognition system based on AIS data, which corresponds to the intelligent ship navigation state recognition method based on AIS data and can be understood as a system for realizing the intelligent ship navigation state recognition method based on AIS data.
Drawings
FIG. 1 is a flow chart of the intelligent identification method of the vessel navigation state based on AIS data.
FIG. 2 is a preferred flow chart of the intelligent identification method for the vessel navigation state based on AIS data.
FIG. 3 is a schematic diagram of the segmentation result of the present invention using the fused lasso method for the navigational speed segmentation.
FIG. 4 is a schematic diagram of a segmentation result of the present invention using a variable point analysis method for the cruise segmentation process.
Detailed Description
The present invention will be described with reference to the accompanying drawings.
The invention relates to an AIS data-based intelligent ship navigation state identification method, a flow chart of which is shown in figure 1, and the method comprises the following steps:
1. and a data acquisition and processing step, namely acquiring AIS data in the ship navigation process, and cleaning, filling and downsampling the AIS data.
The AIS static and dynamic information, container shift port hanging situation, port information and the like of the container ship are needed in the process of identifying the airline, and specifically comprise the MMSI (mean square root of ship), the ship type, the starting and ending time and the starting and ending port of each navigation section, the country and the region of the port, the latitude and longitude of the port and the like. After obtaining the AIS data, the AIS data is preprocessed to remove the AIS problems (e.g., the loss exceeds a threshold value, i.e., the loss is excessive), ship repair, data that the starting port and the ending port are missing or the port is unknown, and data that the starting port and the ending port are the same port, and the starting port and the ending port are respectively matched to the corresponding country and region. Due to the uneven collection of AIS data, it is necessary to downsample the instantaneous speed between any two ports, and one hour may be used as a standard here. The total mileage and the voyage time within the hour integral point time are calculated, and then the average speed is used to represent the voyage speed of the hour. Through down-sampling, the data volume is changed from tens of thousands of original data volumes, the time intervals are the same, meanwhile, the calculation cost is reduced, and the operation efficiency is improved.
For example, as shown in fig. 2, a preferred process may use a psycopg2 library in Python to connect with a PostgreSQL database, and retrieve AIS historical dynamic data of a container ship with an MMSI of 373233000 and relevant information of a port, including a country and a region to which the port belongs, and port latitude and longitude. Preprocessing AIS data, and performing data cleaning, filling and down-sampling: remove data that AIS is problematic (e.g., too many misses), ship repairs, missing or unknown ports at the port of origin and at the port of destination, and data that the port of origin and the port of destination are the same port. For example, the downsampling process is performed by taking one hour as a standard, and in addition, the missing data of which the AIS data missing amount does not exceed the threshold value can be complemented by using the interpolation of the adjacent values. The cleaned data is shown in table 1 (only the top 20 results are shown here), where the fields hour is the whole hour, total _ time is the actual voyage time in the whole hour, average _ speed is the average speed in the whole hour, and total _ dist is the actual voyage distance in the whole hour.
TABLE 1
hour | total_time | average_speed | total_dist | |
0 | 2022-01-27 13:00:00+08:00 | 0.683055556 | 10.32126881 | 7.05 |
1 | 2022-01-27 14:00:00+08:00 | 1.020555556 | 11.18018508 | 11.41 |
2 | 2022-01-27 15:00:00+08:00 | 0.9975 | 13.40350877 | 13.37 |
3 | 2022-01-27 16:00:00+08:00 | 0.993055556 | 13.63468531 | 13.54 |
4 | 2022-01-27 17:00:00+08:00 | 0.814166667 | 5.207778915 | 4.24 |
5 | 2022-01-27 18:00:00+08:00 | 1.039722222 | 0.086561582 | 0.09 |
6 | 2022-01-27 19:00:00+08:00 | 1 | 0.02 | 0.02 |
7 | 2022-01-27 20:00:00+08:00 | 1.050277778 | 0.019042581 | 0.02 |
8 | 2022-01-27 21:00:00+08:00 | 1.000833333 | 0.019983347 | 0.02 |
9 | 2022-01-27 22:00:00+08:00 | 1.048888889 | 0.028601695 | 0.03 |
10 | 2022-01-27 23:00:00+08:00 | 1.043055556 | 2.914513981 | 3.04 |
11 | 2022-01-28 00:00:00+08:00 | 1.008333333 | 10.99834711 | 11.09 |
12 | 2022-01-28 01:00:00+08:00 | 0.991666667 | 11.07226891 | 10.98 |
13 | 2022-01-28 02:00:00+08:00 | 1 | 10.85 | 10.85 |
14 | 2022-01-28 03:00:00+08:00 | 0.9975 | 10.68671679 | 10.66 |
15 | 2022-01-28 04:00:00+08:00 | 0.999444444 | 11.01612007 | 11.01 |
16 | 2022-01-28 05:00:00+08:00 | 1.000277778 | 11.61677312 | 11.62 |
17 | 2022-01-28 06:00:00+08:00 | 1.0025 | 11.13216958 | 11.16 |
18 | 2022-01-28 07:00:00+08:00 | 0.991944444 | 10.92803136 | 10.84 |
19 | 2022-01-28 08:00:00+08:00 | 0.948611111 | 10.83689605 | 10.28 |
2. And a navigation speed segmentation processing step, namely performing segmentation processing on the navigation speed in the collected and processed AIS data through a machine learning algorithm, and calculating the average speed and mileage of each section.
Preferably, based on the acquired and processed AIS data, the invention may excavate the navigation state by a fused lasso method (fused lasso algorithm) or a change point analysis (change point analysis) method in a machine learning algorithm, and search for a change point to further perform segmentation processing on the navigation speed.
Fused lasso is a sparse learning algorithm that can be used for signal processing and denoising. In the present invention, the speed is taken as an input data vector x with the length n, and an optimization problem of a coefficient vector theta with the length n is solved through the following model:
the model parameters rho and gamma respectively control the sparsity and smoothness of the coefficient vector theta, the coefficient vector theta is sparser when the model parameter rho is larger, and the coefficient vector theta is smoother when the model parameter gamma is larger, wherein the optimal model parameter can be determined by methods such as cross check.
The method of point-of-change analysis is a classical statistical method, and is commonly used for mutation detection of time-series data. In this application, the speed of flight is taken as the input signal x with the length n, and a maximum likelihood estimation problem needs to be solved. The method can automatically identify the dominant ones of the signals and determine the node locations of the signals. By traversing all possible time nodes, time nodes with large abrupt changes in mean or variance are mined out as change points. Generally, the sequence data change points that the method can find include: mean change points, variance change points, and mean and variance change points.
When the cruise analysis is realized, the packages of the R language used by the invention can be genlasso and changepoint.
And connecting a data storage database by using a pyycopg 2 database of Python for extracting historical navigation data of the container ship. And selecting any section of flight, taking the down-sampled flight speed as input data, and then respectively using a fused lasso algorithm or a variable point analysis method to obtain output data. The output data is a segmented constant sequence, where larger constants represent higher speeds and smaller constants represent lower speeds. The node of the output data is a speed mutation time point, and the interval between every two nodes represents the duration of the navigational speed. By analyzing the speed, every behavior during the course of the ship can be identified.
In the step, a historical navigational map of the ship is made according to data after processing such as cleaning. And invoking fused lasso or variable point analysis method algorithm to search for the variable point, wherein the two results of searching for the variable point are respectively shown in fig. 3 and fig. 4, which shows that the difference is not large, and the following only takes the result of the variable point analysis method as an example.
3. And a navigation characteristic processing step, namely analyzing the segmented processing result according to the average speed and the mileage of each segment and by combining service logic, and identifying each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state.
Because the fused lasso algorithm and the variable point analysis method only process the navigational speed from the statistical perspective, the obtained segmented navigational speed needs to be interpreted by combining business logic, the segment within the threshold range of the specific navigational speed represents normal navigation, the segment with the navigational speed close to 0 section and the duration exceeding the specific time threshold represents ineffective navigation (namely drifting), for example, the segment close to 20 sections (every hour in the sea) represents high-speed normal navigation, and the segment close to 0 section represents ineffective navigation (drifting). Finding out the starting time and the ending time of each segment, and removing the data with overlapping time to ensure that the found segments are not repeated in time. The results are added to the results of the previous step in chronological order, against historical data of the container ship, in order to ensure the integrity of the results and to facilitate subsequent monitoring of the voyage status.
The navigation feature processing step, in combination with the business logic, can identify approximately four navigation states, which are: the low-speed section of starting a voyage, high-speed stationary section, the sudden change section of slowing down floats the section of navigating, and this is the most complicated and that people are most interested in is to float the section of navigating, and the low-speed section of starting a voyage, high-speed stationary section and the sudden change section of slowing down all belong to the navigation state of normal navigation. According to historical experience, the voyage speed is less than 2 knots, the duration is more than 12 hours, and the voyage section with the mileage occupying port interval of more than 70% can be defined as an invalid voyage section. Through the analysis of the segmented results, as shown in table 2, the real sailing time of the ship with the MMSI of 373233000 is 496 hours.
TABLE 2
Wherein start _ time is the start time of the flight, end _ time is the end time of the flight, part _ dist is the flight distance of the flight, part _ time is the duration of the flight, part _ speed is the average speed of the flight, est _ speed is the estimated speed of the flight, and per _ dist is the accumulated ratio of the flight distance to the port distance.
As can be seen from table 2, the container ship starts to sail at a speed of less than 2 knots in 24 days 2 months for a duration of about 52 hours, and the sailing distance at this time exceeds the upper limit of the harbor spacing, so that it can be determined that the invalid sailing, i.e., the drifting, is started.
4. And a course information extraction step, namely extracting course information based on the obtained navigation state, identifying the starting time and the position of each segmental navigation state in the navigation process of the ship, carrying out time information matching on the navigation process of the ship, and calculating the effective navigation time and the drift time.
In order to avoid the problem that the found flight sections are not matched due to the lack of the flight speed in the speed segmentation process, the missing flight speed data needs to be supplemented, and interpolation data of the flight speed is added. Further, this step also classifies the relevant information of the route, including: matching the airline to corresponding countries and regions according to port information where the airline passes through; finding out the starting time and the ending time of the flight segment, and determining the time information of the relevant year of the flight line operation; finding the operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to the destination port, wherein the sailing states comprise a recent congested port and a normal port.
The container, the segment speed and the departure/arrival port information from 2022-01-23 to 2022-03-31 and the destination port is the beach port or the los angeles port are extracted, and as a result, as shown in table 3 and table 4, table 3 shows the navigation information (part) of the containers 2022-01-23 to 2022-03-31 for the containers whose destination port is the beach port or the los angeles port, and table 4 shows the navigation information (part) of the containers 2022-01-23 to 2022-03-31 for the containers whose destination port is the beach port or the los angeles port.
TABLE 3
TABLE 4
As can be seen from tables 3 and 4, most container ships have an invalid sailing (drifting) state.
5. And a navigation monitoring and analyzing step, namely monitoring the navigation states of any two ports where the airline passes through, carrying out navigation speed statistics and navigation speed change analysis, and further calculating the congestion state and navigation energy consumption of the port where the airline passes through.
Knowing any two ports, the navigation states passing through the two ports can be found out, and the navigation speed statistics and the navigation speed change analysis are carried out according to the defined air routes. That is, the variation analysis of the navigation state of the container liner routes passing through any two ports can be performed according to the identified navigation state, and then the port congestion analysis and the navigation energy consumption calculation can be performed.
The invention also relates to a ship navigation state intelligent recognition system based on AIS data, which corresponds to the ship navigation state intelligent recognition method based on AIS data and can be understood as a system for realizing the ship navigation state intelligent recognition method based on AIS data, and the system comprises a data acquisition processing module, a navigation speed segmentation processing module, a navigation characteristic processing module, a route information extraction module and a navigation monitoring analysis module which are connected in sequence, wherein the data acquisition processing module is used for acquiring AIS data in the ship navigation process and cleaning, filling and downsampling the AIS data; the navigation speed segmentation processing module is used for carrying out segmentation processing on the navigation speed in the collected and processed AIS data through a machine learning algorithm and calculating the average speed and the mileage of each section; the navigation characteristic processing module analyzes the segmented processing result according to the average speed and the mileage of each segment and by combining with service logic, and identifies each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state; the route information extraction module is used for identifying the starting time and the position of each segmental navigation state in the navigation process of the ship based on the obtained navigation state, carrying out time information matching on the navigation process of the ship and calculating the effective navigation time and the drifting time; the navigation monitoring and analyzing module monitors the navigation states of any two ports where the airline passes through, carries out navigation speed statistics and navigation speed change analysis, and further calculates the congestion state and navigation energy consumption of the port where the airline passes through.
Furthermore, the AIS data collected by the data collecting and processing module comprises AIS static and dynamic information of the container ship, container shift port condition and port information; the cleaning, filling and downsampling processing of the AIS data comprises removing data of which the AIS data loss amount exceeds a threshold value, repairing ships, having a missing starting port, having a missing ending port or being unknown at a port, removing data of which the starting port and the ending port are the same port, respectively matching the starting port and the ending port to corresponding countries and regions, complementally filling the data of which the AIS data loss amount does not exceed the threshold value by using a proximity value, and downsampling the instantaneous speed between any two ports.
Furthermore, the navigational speed segmentation processing module searches for a variable point through a fusion lasso method or a variable point analysis method in a machine learning algorithm so as to perform segmentation processing on the navigational speed in the collected and processed AIS data.
Furthermore, the sailing feature processing module is used for explaining the obtained segmental sailing speed by combining with business logic, and the sailing section within the threshold range of the specific sailing speed represents normal sailing, the sailing section with the sailing speed close to 0 section and the duration exceeding the threshold of the specific time represents drifting, and the normal sailing comprises sailing states of a low-speed sailing section, a high-speed stable section and a deceleration sudden-change section.
Further, the airline information extraction module classifies airline relevant information, including: matching the airlines to corresponding countries and regions according to port information of the airlines passing through; finding out the starting time and the ending time of the flight segment, and determining the time information of relevant years of the flight line operation; finding out an operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to destination ports, including recent congested ports and normal ports.
The AIS data is utilized, the fused lasso algorithm or the variable point analysis method for signal identification in statistical machine learning is adopted to perform sectional processing according to the change of the navigational speed, the navigational state is identified by combining the AIS data characteristics and the actual navigational service characteristics of the ship, the effective navigational time is further analyzed, support is provided for navigation planning based on big data, relevant information is collated according to ports through which the airline passes, the navigation monitoring of the ship is performed, the navigational state of the ship is automatically and rapidly identified, and particularly, the accurate invalid airline time can be directly obtained, so that the congestion condition of the airline is judged, the operation airline is conveniently adjusted in time, and the method and the system are worthy of popularization and application.
It should be noted that the above-described embodiments may enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the drawings and examples, it will be understood by those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. An AIS data-based intelligent ship navigation state identification method is characterized by comprising the following steps:
a data acquisition and processing step, namely acquiring AIS data in the ship navigation process, and cleaning, filling and down-sampling the AIS data;
a navigation speed subsection processing step, namely, the navigation speed in the collected and processed AIS data is subsection processed through a machine learning algorithm, and the average speed and the mileage of each section are calculated;
a navigation characteristic processing step, namely analyzing the segmented processing result according to the average speed and the mileage of each segment and by combining service logic, and identifying each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state;
a course information extraction step, namely identifying the starting time and the position of each segmental navigation state in the navigation process of the ship based on the obtained navigation state, carrying out time information matching on the navigation process of the ship, and calculating effective navigation time and drift time;
and a navigation monitoring and analyzing step, namely monitoring the navigation states of any two ports where the airline passes through, carrying out navigation speed statistics and navigation speed change analysis, and further calculating the congestion state and navigation energy consumption of the port where the airline passes through.
2. The intelligent identification method for the ship navigation state according to claim 1, wherein in the data acquisition and processing step, the acquired AIS data comprises AIS static and dynamic information of the container ship, container shift port condition and port information; the cleaning, filling and downsampling processing of the AIS data comprises the steps of removing data with the AIS data loss amount exceeding a threshold value, repairing ships, having a loss at a starting port, having a loss at an ending port or being unknown at a port, removing data with the starting port and the ending port being the same port, respectively matching the starting port and the ending port to corresponding countries and regions, completely filling the data with the AIS data loss amount not exceeding the threshold value by using a proximity value, and downsampling the instantaneous speed between any two ports.
3. The intelligent identification method for the ship navigation state according to claim 1, wherein in the navigation speed segmentation processing step, a variable point is searched by a fusion lasso method or a variable point analysis method in a machine learning algorithm, and then the navigation speed in the collected and processed AIS data is segmented.
4. The intelligent identification method for the ship navigation state according to one of claims 1 to 3, wherein in the navigation feature processing step, the obtained segmental navigation speeds are interpreted in combination with business logic, a navigation section within a specific navigation speed threshold value range represents normal navigation, a navigation section with the navigation speed close to 0 section and the duration exceeding a specific time threshold value represents drift navigation, and the normal navigation includes the navigation states of a low-speed starting section, a high-speed steady section and a deceleration abrupt section.
5. The intelligent identification method for the ship navigation state according to one of claims 1 to 3, wherein in the route information extraction step, the classification of the route-related information is further performed, and the classification comprises: matching the airline to corresponding countries and regions according to port information where the airline passes through; finding out the starting time and the ending time of the flight segment, and determining the time information of the relevant year of the flight line operation; finding out an operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to destination ports, including recent congested ports and normal ports.
6. An AIS data-based intelligent ship navigation state identification system is characterized by comprising a data acquisition processing module, a navigation speed segmentation processing module, a navigation characteristic processing module, a route information extraction module and a navigation monitoring analysis module which are connected in sequence,
the data acquisition and processing module is used for acquiring AIS data in the ship sailing process and cleaning, filling and down-sampling the AIS data;
the navigation speed segmentation processing module is used for carrying out segmentation processing on the navigation speed in the collected and processed AIS data through a machine learning algorithm and calculating the average speed and the mileage of each section;
the navigation characteristic processing module analyzes the segmented processing result according to the average speed and the mileage of each segment and by combining with service logic, and identifies each segmented navigation state in the navigation process of the ship to obtain the normal navigation state or the drifting navigation state;
the route information extraction module is used for identifying the starting time and the position of each segmental navigation state in the navigation process of the ship based on the obtained navigation state, carrying out time information matching on the navigation process of the ship and calculating the effective navigation time and the drifting time;
the navigation monitoring and analyzing module monitors the navigation states of any two ports where the airline passes through, carries out navigation speed statistics and navigation speed change analysis, and further calculates the congestion state and navigation energy consumption of the port where the airline passes through.
7. The intelligent ship navigation state recognition system of claim 6, wherein the AIS data collected by the data collection and processing module includes AIS static and dynamic information of the container ship, container liner port condition, and port information; the cleaning, filling and downsampling processing of the AIS data comprises removing data of which the AIS data loss amount exceeds a threshold value, repairing ships, having a missing starting port, having a missing ending port or being unknown at a port, removing data of which the starting port and the ending port are the same port, respectively matching the starting port and the ending port to corresponding countries and regions, complementally filling the data of which the AIS data loss amount does not exceed the threshold value by using a proximity value, and downsampling the instantaneous speed between any two ports.
8. The intelligent ship navigation state recognition system of claim 6, wherein the navigational speed segmentation processing module is configured to search for a variable point by a fusion lasso method or a variable point analysis method in a machine learning algorithm to segment the navigational speed in the collected and processed AIS data.
9. The intelligent identification system for ship voyage state according to one of claims 6 to 8, wherein the voyage feature processing module is used for interpreting the obtained segmental voyage speed in combination with business logic, and represents a normal voyage for a voyage in a range of a specific voyage speed threshold value, and represents a drift for a voyage in a voyage period with the voyage speed close to 0 knots and the duration exceeding a specific time threshold value, and the normal voyage comprises the voyage states of a low-speed starting period, a high-speed steady period and a deceleration abrupt change period.
10. The intelligent recognition system for ship navigation state according to one of claims 6 to 8, wherein the route information extraction module further classifies route related information, including: matching the airline to corresponding countries and regions according to port information where the airline passes through; finding out the starting time and the ending time of the flight segment, and determining the time information of the relevant year of the flight line operation; finding the operation subject of the ship and the air route according to the MMSI of the ship; and classifying the sailing states according to destination ports, including recent congested ports and normal ports.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210876444.6A CN115294802A (en) | 2022-07-25 | 2022-07-25 | AIS data-based ship navigation state intelligent identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210876444.6A CN115294802A (en) | 2022-07-25 | 2022-07-25 | AIS data-based ship navigation state intelligent identification method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115294802A true CN115294802A (en) | 2022-11-04 |
Family
ID=83824758
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210876444.6A Pending CN115294802A (en) | 2022-07-25 | 2022-07-25 | AIS data-based ship navigation state intelligent identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115294802A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SG11201601680VA (en) * | 2013-09-06 | 2016-04-28 | Nippon Yusen Kk | Device, program, recording medium and method for facilitating management of schedule of voyage |
KR20180111196A (en) * | 2017-03-31 | 2018-10-11 | 삼성중공업 주식회사 | Apparatus and method for determining route and speed of vessel, and recording medium |
CN109840358A (en) * | 2019-01-09 | 2019-06-04 | 上海海洋大学 | A kind of track segmentation method based on track time-domain difference |
US20210043091A1 (en) * | 2016-01-15 | 2021-02-11 | David Belu SOLOMON | Novel vessel systems and methods relating thereto |
CN113032502A (en) * | 2021-02-09 | 2021-06-25 | 北京工业大学 | Ship anomaly detection method based on improved track segment DBSCAN clustering |
CN113450596A (en) * | 2021-05-31 | 2021-09-28 | 武汉理工大学 | Space-time DP method based on ship track feature point extraction |
CN113780952A (en) * | 2021-09-16 | 2021-12-10 | 上海船舶运输科学研究所 | Method and system for analyzing transport capacity of container liner ship route |
CN113822630A (en) * | 2021-09-14 | 2021-12-21 | 中远海运科技股份有限公司 | AIS-based LNG ship transport tracking method and system |
-
2022
- 2022-07-25 CN CN202210876444.6A patent/CN115294802A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SG11201601680VA (en) * | 2013-09-06 | 2016-04-28 | Nippon Yusen Kk | Device, program, recording medium and method for facilitating management of schedule of voyage |
US20210043091A1 (en) * | 2016-01-15 | 2021-02-11 | David Belu SOLOMON | Novel vessel systems and methods relating thereto |
KR20180111196A (en) * | 2017-03-31 | 2018-10-11 | 삼성중공업 주식회사 | Apparatus and method for determining route and speed of vessel, and recording medium |
CN109840358A (en) * | 2019-01-09 | 2019-06-04 | 上海海洋大学 | A kind of track segmentation method based on track time-domain difference |
CN113032502A (en) * | 2021-02-09 | 2021-06-25 | 北京工业大学 | Ship anomaly detection method based on improved track segment DBSCAN clustering |
CN113450596A (en) * | 2021-05-31 | 2021-09-28 | 武汉理工大学 | Space-time DP method based on ship track feature point extraction |
CN113822630A (en) * | 2021-09-14 | 2021-12-21 | 中远海运科技股份有限公司 | AIS-based LNG ship transport tracking method and system |
CN113780952A (en) * | 2021-09-16 | 2021-12-10 | 上海船舶运输科学研究所 | Method and system for analyzing transport capacity of container liner ship route |
Non-Patent Citations (7)
Title |
---|
张丽勇;骆参驹;: "海上船舶航行数据特征的数据挖掘分析", 舰船科学技术, no. 16 * |
熊英格;徐卓立;刘好德;: "基于浮动车数据的交叉口范围动态划分方法", 交通信息与安全, no. 05 * |
王坚;黄?;刘艳英;陈森阳;: "厦门港船舶大气污染物排放量影响因素研究――船舶航行工况", 环境科学与管理, no. 02 * |
王建;史景聪;黄冬梅;郑小罗;何盛琪;张北辰;: "基于TDD的科考船航迹分段方法研究", 极地研究, no. 01, pages 112 - 120 * |
王征;张卫;彭传圣;刘瑛;秦翠红;: "中国近周边海域船舶排放清单及排放特征研究", 交通节能与环保, no. 01 * |
程立龙: "基于AIS数据的船舶轨迹停留特征时空分析研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》, no. 3, pages 9 - 11 * |
魏照坤;谢新连;潘伟;赵瑞嘉;: "基于朴素贝叶斯算法的船舶异常行为监测", 交通运输系统工程与信息, no. 06 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11719543B2 (en) | Course guidance method for efficient sailing of ship | |
CN113032502A (en) | Ship anomaly detection method based on improved track segment DBSCAN clustering | |
CN112862156B (en) | Ship path planning method based on ship track and ant colony algorithm | |
CN114066354A (en) | Intelligent air route recommendation method and system based on global ship historical track | |
CN110798805B (en) | Data processing method and device based on GPS track and storage medium | |
WO2023039894A1 (en) | Method and system for analyzing transport capacity of container liner route | |
CN113537386B (en) | Ship typical motion trail self-adaptive mining method based on improved K-Medoids clustering | |
CN112164247A (en) | Ship route prediction method based on ship track clustering | |
CN111640304B (en) | Automatic quantitative extraction method for traffic jam propagation characteristics of continuous flow traffic facility | |
CN110196962A (en) | A kind of air speed abnormality recognition method based on Density Estimator | |
CN114647705A (en) | Ship route extraction and trajectory classification method based on AIS data | |
CN112445847A (en) | Ship track prediction method and device based on Beidou navigation time-frequency data | |
CN113963210A (en) | Deep learning-based detection method and sorting system for waste data storage equipment | |
CN115050214B (en) | AIS data-based ship collision risk prediction method | |
CN114972918B (en) | Remote sensing image ship target identification method based on integrated learning and AIS data | |
CN113988213B (en) | Method and system for identifying repairing of ship in dock | |
CN115311229A (en) | Laser radar-based pavement disease detection and classification method and system and storage medium | |
CN108981728B (en) | Intelligent vehicle navigation map building method | |
CN114550498A (en) | Intelligent early warning system for ship deviation channel driven by massive AIS data | |
CN115294802A (en) | AIS data-based ship navigation state intelligent identification method and system | |
CN115456258A (en) | Method for predicting transportation capacity of competitor ship and computer readable medium | |
El Abidi et al. | Detection of Moroccan coastal upwelling using the alpha blending fusion technique of sea surface chlorophyll images and sea surface temperature images | |
Kwun et al. | Prediction of vessel arrival time using auto identification system data | |
Weiqiang et al. | A Cloud-based Approach for Ship Stay Behavior Classification using Massive Trajectory Data | |
CN114331042A (en) | Ship collision risk analysis method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |