CN115310533B - AIS-based offshore wind farm identification method and system - Google Patents
AIS-based offshore wind farm identification method and system Download PDFInfo
- Publication number
- CN115310533B CN115310533B CN202210936736.4A CN202210936736A CN115310533B CN 115310533 B CN115310533 B CN 115310533B CN 202210936736 A CN202210936736 A CN 202210936736A CN 115310533 B CN115310533 B CN 115310533B
- Authority
- CN
- China
- Prior art keywords
- ship
- wind power
- ais
- clusters
- suspected
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000009434 installation Methods 0.000 claims abstract description 72
- 238000011156 evaluation Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 7
- 238000005065 mining Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 5
- 238000011161 development Methods 0.000 abstract description 7
- 238000010248 power generation Methods 0.000 description 4
- 238000004873 anchoring Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 235000014435 Mentha Nutrition 0.000 description 1
- 241001072983 Mentha Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 235000014569 mints Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Wind Motors (AREA)
Abstract
The invention provides an offshore wind farm identification method and system based on AIS (automatic identification system), which are based on ship AIS dynamic data and ship data, and are combined with business logic to determine a suspected wind power installation ship and AIS characteristics of the wind power installation ship when the wind power installation ship is provided with fans, then a DBSCAN (distributed base station) clustering algorithm is adopted to cluster longitude and latitude position information of the suspected wind power installation ship meeting the AIS characteristics to obtain a plurality of first clusters, a Kmeans clustering algorithm is adopted to cluster each obtained first cluster again to obtain a plurality of second clusters, a business logic is combined to screen out clusters of suspected fans and clusters of the suspected offshore wind farms, the distance between each fan and other fans in each suspected offshore wind farm and the slope of the fan at the nearest to the fan are calculated, an offshore wind farm area is identified according to the number of fans in each suspected offshore wind farm, the distance between fans, the slope and ship destination port, and the ship port is warned to prohibit entering relevant areas and the development conditions of offshore wind farms are evaluated.
Description
Technical Field
The invention relates to the technical field of large data mining processing of offshore wind farms, in particular to an AIS-based offshore wind farm identification method and system.
Background
Wind power generation is one of the most mature and large-scale development potential discovery modes in renewable energy power generation technology, and land wind energy utilization is limited while land wind power plant construction rapidly develops, so that wind power generation is converted to offshore due to the problems of large occupied area, noise pollution and the like.
The offshore wind power generation system has the advantages of rich wind energy resources, stable unit operation, large single unit capacity, high annual utilization hour number, large energy output, small land occupation, small negative environmental influence and the like. Offshore wind farms are an important component in the development of clean low carbon energy and optimizing energy structures.
The newly increased production scale of the offshore wind power in China reaches 1690 kilowatts in 2021, the same ratio is increased by 340 months, and the accumulated installation scale reaches 2638 kilowatts. However, because the data about the offshore wind farm is less at present, accurate information about the position of the offshore wind farm cannot be obtained, and therefore, some ships can be mistakenly put into the offshore wind farm.
Disclosure of Invention
In order to solve the problems that the existing method for acquiring the position information of the offshore wind farm is not accurate enough, the invention provides an AIS-based offshore wind farm identification method, which is based on AIS data and ship data, adopts a DBSCAN clustering algorithm and a Kmeans clustering algorithm to perform multistage clustering, and combines business logic to discover and identify the offshore wind farm, so as to determine the position of the wind farm. The offshore wind farm range can be calibrated, the marine wind farm can be prevented from being wrongly put in, and the development condition of the offshore wind farm can be estimated. The invention further relates to an offshore wind farm identification system based on the AIS.
The technical scheme of the invention is as follows:
an AIS-based offshore wind farm identification method is characterized by comprising the following steps:
and a data acquisition step: collecting ship data, and performing wind power installation ship feature mining from the ship data according to business logic to determine suspected wind power installation ships;
a feature determination step: acquiring AIS dynamic data of a suspected wind power installation ship, and determining AIS characteristics of the wind power installation ship when a fan is installed according to the AIS dynamic data; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
clustering and calculating: clustering longitude and latitude position information of suspected wind power installation vessels meeting AIS characteristics by adopting a DBSCAN clustering algorithm to obtain a plurality of first clusters, clustering each obtained first cluster again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining the longitude and latitude coordinates of a central point of each second cluster, calculating the radius of the second cluster according to the longitude and latitude coordinates of the central point, and calculating the variance of the ship bow direction in the second cluster, the mean value and the variance of ship draft according to AIS dynamic data;
and a fan identification step: intelligently screening clusters with the radius, the variance of the ship heading, the mean value of ship draft and the variance meeting respective preset conditions from a plurality of second clusters, and storing the clusters as suspected fans in a database;
wind farm identification: clustering longitude and latitude coordinates of central points of all clusters which are screened out to serve as suspected fans in a database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, taking the third clusters as suspected offshore wind power plants, calculating the distance between each fan and other fans in each suspected offshore wind power plant and the slope of each fan closest to the fan, and automatically identifying the offshore wind power plant according to the number of fans in each suspected offshore wind power plant, the distance between the fans, the slope and a ship destination port.
Preferably, in the data collecting step, the ship profile data includes a ship status field, a ship equipment field, and a ship type field, the ship status field is included in a service field, a dismantling field, and an order field, and the ship type field includes a providing offshore service field and a supporting platform field.
Preferably, in the step of determining the characteristics, the AIS characteristics include speed to ground and state of voyage.
Preferably, in the clustering and calculating step, when the DBSCAN clustering algorithm is adopted for clustering, a contour coefficient is also adopted as an evaluation index of the clustering effect.
Preferably, in the clustering and calculating step, after obtaining the plurality of second clusters, the minimum time and the maximum time in the clusters are reserved, and the time difference is calculated according to the minimum time and the maximum time and is used as an evaluation index of the installation efficiency of the wind power installation ship.
An AIS-based offshore wind farm identification system is characterized by comprising a data acquisition module, a characteristic determination module, a clustering and calculating module, a fan identification module and a wind farm identification module which are connected in sequence,
the data acquisition module acquires ship data, and performs wind power installation ship feature mining from the ship data according to the business logic to determine a suspected wind power installation ship;
the characteristic determining module is used for acquiring AIS dynamic data of the suspected wind power installation ship and determining AIS characteristics of the wind power installation ship when the wind power installation ship installs the fan according to the AIS dynamic data; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
the clustering and calculating module is used for clustering longitude and latitude position information of suspected wind power installation vessels meeting AIS characteristics by adopting a DBSCAN clustering algorithm to obtain a plurality of first clusters, clustering each obtained first cluster again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining longitude and latitude coordinates of a central point of each second cluster, calculating the radius of the second cluster according to the longitude and latitude coordinates of the central point, and calculating the variance of the ship heading, the mean value of ship draught and the variance in the second clusters according to AIS dynamic data;
the fan identification module intelligently screens clusters with the radius, the variance of the ship bow direction and the mean and variance of ship draft meeting respective preset conditions from a plurality of second clusters, and stores the clusters as suspected fans in a database;
the wind power plant identification module adopts a DBSCAN clustering algorithm to cluster longitude and latitude coordinates of central points of all clusters which are screened out as suspected wind power plants in a database to obtain a plurality of third clusters, the third clusters are used as suspected offshore wind power plants, the distance between each wind power plant and other wind power plants in each suspected offshore wind power plant and the slope between each wind power plant and the nearest wind power plant are calculated, and the offshore wind power plant is automatically identified according to the number of wind power plants in each suspected offshore wind power plant, the distance between the wind power plants, the slope and the ship destination port.
Preferably, the ship profile data includes a ship status field, a ship equipment field, and a ship type field, the ship status field being included in the operation field, the disassembly field, and the order field, and the ship type field including a provide offshore service field and a support platform field.
Preferably, the AIS features include speed to ground and navigational status.
Preferably, in the clustering and calculating module, when the DBSCAN clustering algorithm is adopted for clustering, a contour coefficient is also adopted as an evaluation index of the clustering effect.
Preferably, in the clustering and calculating module, after obtaining a plurality of second clusters, the minimum time and the maximum time in the clusters are reserved, and the time difference is calculated according to the minimum time and the maximum time and is used as an evaluation index of the installation efficiency of the wind power installation ship.
The beneficial effects of the invention are as follows:
according to the offshore wind farm identification method based on the AIS, based on ship AIS dynamic data and ship data, service logic is combined to determine the suspected wind power installation ship and AIS characteristics of the wind power installation ship when a fan is installed, a DBSCAN clustering algorithm is adopted to cluster longitude and latitude position information of the suspected wind power installation ship meeting the AIS characteristics, a plurality of first clusters (namely fan clusters) are obtained, the obtained fan clusters are subjected to reclustering by adopting a Kmeans clustering algorithm, a plurality of second clusters are obtained, the accurate position of each fan can be obtained, and the longitude and latitude coordinates of the center point of each second cluster are obtained and are used as the accurate position of each fan. Clustering the center points of each fan by adopting a DBSCAN clustering algorithm, clustering a plurality of fan points into a plurality of wind power stations, screening out clusters of suspected fans and clusters of suspected offshore wind power stations by combining relevant business logic, calculating the distance between each fan and other fans in each suspected offshore wind power station and the slope of each fan and the nearest fan, automatically identifying the offshore wind power station area according to the number of fans in each suspected offshore wind power station, the distance between fans, the slope and the ship destination port, warning ships to prohibit entering relevant areas, and further intelligently evaluating the installation efficiency of wind power installation vessels and the development condition of offshore wind power.
The invention also relates to an AIS-based offshore wind farm identification system, which corresponds to the AIS-based offshore wind farm identification method, and can be understood as a system for realizing the AIS-based offshore wind farm identification method, and the system comprises a data acquisition module, a characteristic determination module, a clustering and calculating module, a fan identification module and a wind farm identification module which are sequentially connected, wherein the modules work cooperatively with each other, perform multistage clustering by adopting a DBSCAN clustering algorithm and a Kmeans clustering algorithm based on AIS data and ship data, find and automatically identify an offshore wind farm by combining business logic, determine the position of the wind farm, automatically calibrate the offshore wind farm area, warn ships to prohibit entering related areas, and intelligently evaluate the development condition of offshore wind power.
Drawings
FIG. 1 is a flow chart of the AIS-based offshore wind farm identification method of the present invention.
Fig. 2 is a schematic illustration of the path of travel of the bridge vessel.
FIG. 3 is a schematic diagram of a wind farm at sea in Qingzhou Yangjiang Guangdong.
Detailed Description
The present invention will be described below with reference to the accompanying drawings.
The invention relates to an AIS-based offshore wind farm identification method, which comprises the following steps in sequence, as shown in a flow chart of FIG. 1:
and a data acquisition step: collecting ship data, and performing wind power installation ship feature mining from the ship data according to business logic to determine suspected wind power installation ships; specifically, first, the psycopg2 in the Python language (which is a PostgreSQL database interface in the Python language) is used to connect to a PostgreSQL database, ship data is queried from the PostgreSQL database, and after the above data is obtained, the ship data is specifically used: a ship status field, a ship equipment gearescriptionstatic field and a ship type shiptype field. The ship field mainly describes the current operational state of the ship, including: fields such as operation, disassembly, order and the like; the gearescriptifenarrative field mainly describes the equipment conditions of the ship; the shiptype field mainly describes the detailed type of the ship, including: offshore Support vessel Offshore Support Vessel, support Platform, etc.
By inquiring related data and referring to business logic (or business knowledge), it is determined that the wind power installation ship needs to have a lifting function, mainly works as offshore service, and possibly has a self-elevating platform. The inquiry data determines a number of wind power installation vessels and the time period for installing fans. By sorting the data of the known wind power installation vessels and the vessels with fans installed in parallel, the wind power installation vessel is determined to be a vessel with the following characteristics: the Geardscriptivenarrativ field contains 'SWL', meaning a crane; the shift field contains 'In Service/communication', meaning that the ship is currently In operation; the shiptype field contains: 'Offshore SupportVessel', 'jack up' and the like, 'Offshore Support Vessel' indicates that the vessel is primarily dedicated to providing offshore service, and 'jack up' indicates that the vessel is provided with a jack-up platform. And querying all ships with the characteristics in the database according to the conditions, and using the ships as suspected wind power installation ships.
A feature determination step: acquiring AIS dynamic data of a suspected wind power installation ship, and determining AIS characteristics of the wind power installation ship when a fan is installed according to the AIS dynamic data and by combining business knowledge; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
specifically, all AIS dynamic data of suspected wind power installation vessels are queried from a PostgreSQL database, after the data are obtained, the AIS dynamic data are preprocessed, and fields used by the AIS dynamic data comprise: status, lon, lat, sog, cog, hdg, draught and dest fields. Wherein status= {0,1,5}, status (status) field is: 1 is anchoring, 5 is berthing, and 0 is sailing; lon and lat are longitude and latitude positions; sog is the real-time ground speed of the ship; hdg is the heading, i.e., bow orientation; draugh is the real-time draft of the ship; dest is the port of interest of the ship.
When a ship is provided with a fan, the navigational speed is close to 0, the heading is kept unchanged within a certain period of time, off-shore operation draft is not too large, AIS state data of the wind power installation ship for installing the fan is determined to have the following AIS characteristics according to AIS state data mining of the known wind power installation ship and the period of time for installing the fan, the AIS state data is derived from the suspected wind power installation ship, the AIS characteristics comprise the ground navigational speed and the navigational state, wherein the ground navigational speed sog is 0, the navigational state is status=1, namely the current state is 1, namely the anchoring state.
Clustering and calculating: clustering longitude and latitude position information of suspected wind power installation vessels meeting AIS characteristics by adopting a DBSCAN clustering algorithm to obtain a plurality of first clusters serving as fan clusters, clustering each obtained fan cluster again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining longitude and latitude coordinates of a central point of each fan cluster, calculating the radius of the fan cluster according to the longitude and latitude coordinates of the central point, and calculating the variance of the ship heading in the fan cluster, the mean value and the variance of ship draught according to AIS dynamic data;
specifically, clustering longitude and latitude position information of a suspected wind power installation ship meeting ground speed sog as 0 and anchoring in a navigation state by adopting a DBSCAN clustering algorithm to obtain a plurality of fan clusters, clustering each fan cluster formed by clustering again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, acquiring longitude and latitude coordinates of a central point of each second cluster, calculating longitude and latitude distances from each point to the central point in the cluster according to the position coordinates of the central point, and taking the maximum value of the distances as the radius of the cluster. Meanwhile, the minimum time and the maximum time in the second clusters are reserved, the time difference is calculated according to the minimum time and the maximum time and is used as an evaluation index of the installation efficiency of the wind power installation ship, the variance of the ship heading hdg in the second clusters and the mean value and the variance of the ship draft are calculated according to AIS dynamic data, the variances are all used as parameters of each second cluster, and each second cluster is suspected to be a fan.
The DBSCAN clustering algorithm is an algorithm in machine learning non-supervision learning, and compared with other clustering algorithms, the DBSCAN clustering algorithm is fast in clustering speed, can effectively process noise points and discover spatial clusters with any shape, and does not need to specify the number of clusters. The parameters only need to pay attention to the neighborhood parameters (epsilon, minPts), wherein epsilon is the neighborhood radius, the neighborhood of the minPts number in the radius of one neighborhood is considered as one cluster, and the algorithm is more convenient and quick to find the optimal parameters. When the DBSCAN clustering algorithm is used for clustering, the Euclidean distance of the longitude and latitude values is directly adopted, and the longitude and latitude distance is not adopted, because: 1) The longitude and latitude distance is complex to calculate, and the algorithm is slow for a large amount of data; 2) The latitude and longitude distances and the euclidean distance of the latitude and longitude values are not greatly different in the invention. The Euclidean distance of decimal longitude and latitude values is adopted as the distance function of DBSCAN clusters for algorithm efficiency. And the contour coefficient in machine learning is used as an evaluation index of the clustering effect. The closer the contour coefficient is to 1, the better the clustering effect, the closer to-1, indicating that sample i should be more classified into other clusters.
The Kmeans clustering algorithm is an algorithm in machine learning unsupervised learning, and compared with other clustering algorithms, the algorithm only aims at clustering of regular-shape clusters and the number of classes needs to be specified in advance. The Kmeans clustering algorithm principle is as follows: for a given sample set x= { x 1 ,x 2 ,...,x n The sample set is divided into K clusters according to the distance between samples. The points in the clusters are connected as closely as possible, and the distance between the clusters is as large as possible. I.e. assume the goal is to partition k clusters { C 1 ,C 2 ,C 3 ,...,C k Then the goal is to minimize the square error E:
in the above, x isSample set, k is the number of clusters, μ i Is cluster C i Mean vector, mu i The expression is:
and a fan identification step: combining business knowledge, intelligently screening clusters with the radius, the variance of the ship heading, the mean value of ship draft and the variance meeting respective preset conditions from a plurality of second clusters, and storing the clusters as suspected fans into a database;
specifically, by inquiring related data and combining business knowledge, the radius of each offshore wind turbine is found to be about 45-60 meters, the ship heading of the wind power installation ship is not excessively adjusted when the wind power installation ship installs the wind turbines, and as the offshore wind turbines are all in offshore areas and the draft is not excessively large and approximately unchanged, the second cluster with the radius within 100 meters, the draft mean within 7 meters and the draft variance within 5 is selected and used as a suspected wind turbine to be stored in a database for subsequent use.
Wind farm identification: clustering longitude and latitude coordinates of central points of all clusters which are screened out to serve as suspected fans in a database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, taking the third clusters as suspected offshore wind power plants, calculating the distance between each fan and other fans in each suspected offshore wind power plant and the slope of each fan closest to the fan, and automatically identifying the offshore wind power plant according to the number of fans in each suspected offshore wind power plant, the distance between the fans, the slope and a ship destination port.
Specifically, clustering longitude and latitude coordinates of central points of all clusters which are screened out to be suspected fans in a database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, and taking the formed third clusters as suspected offshore wind farms. The fans in the offshore wind farm are regularly arranged in a straight line mode, the number of the fans in the wind farm is more than 10 fans, the intervals among the fans are equal, the distance between each fan and other fans in each suspected offshore wind farm and the slope of each fan closest to the fan are calculated, and if the number of the fans in a certain cluster (namely a certain suspected offshore wind farm) in a third cluster is more than 10, the distances among the fans are approximately equal, the slopes are approximately the same, and the ship destination ports are the same, the cluster is automatically judged to be an offshore wind farm.
Examples:
and a data acquisition step: and obtaining the ship names and the fan installation time of a plurality of wind power installation ships by inquiring the related news of the offshore wind farm in China. The databases are queried for the ship's materials, which are found to have the same commonality, all special ships, and the Geardscriptivelnarrativ fields all contain ' SWL ', indicating that the ship is equipped with a crane for lifting fan blades. The shiptype field contains: 'Offshore Support Vessel', 'jack up' and the like, 'Offshore Support Vessel' means that the vessel is primarily dedicated to providing offshore services, and 'jack up' means that the vessel is provided with a jack-up platform.
Taking the bridge Fu boat as an example, the bridge Fu boat completes the installation of a T31# fan in the three period of the sand raking of the Yangjiang of three gorges in 2021 for 3 months and about 3 months and 27 days, and the boat track is in a ring cluster, so that the track characteristics of the installed fans are met. A feature determination step: AIS data of bridge vessels 2021.03.01-2021.04.10 are queried in a database, and track data in the time period are mapped by using a PLotly visualization module in Python language. As shown in fig. 2, it is obvious that during this period, the track forms 3 annular clusters, which are tracks when the fan is installed, and by mining the ground speed sog, (status) field status, ship heading hdg and ship destination port dest data in these 3 clusters, the following features are found: sog is close to 0, status=1, i.e. the ship state is moored. The AIS characteristic for the wind power installation vessel is set to sog near 0, status=1 when installing the wind turbine.
Clustering and calculating: the query database contains 'SWL' in the Geardscriptivelnarrativ field, and the shiptype field contains: the ship of 'Offshore SupportVessel', 'jack up' etc. is queried for the ship 2021 all year round and meets sog =0, status=1, and the latitude and longitude position information is clustered by using a DBSCAN clustering algorithm to obtain a plurality of first clusters, which are fan clusters, near the offshore wind farm in Qingzhou, yangjiang, yue, namely, AIS dynamic data with longitude lon in the interval [111.45,111.68] and latitude lat in the interval [21.22,21.38 ]. And using the contour coefficient as an evaluation method, selecting the optimal parameter as epsilon=0.01 and mints=10. And for each fan cluster obtained by clustering, clustering each fan cluster formed by clustering again by using a Kmeans clustering algorithm to obtain a plurality of second clusters, setting the number k of the second clusters as 1, obtaining the center point position coordinates of each second cluster, calculating the longitude and latitude distance from each point in the second cluster to the center point according to the center point position coordinates, and taking the maximum value of the distance as the radius of the second cluster. Meanwhile, the minimum time and the maximum time in the second clusters are reserved, the time difference is calculated to serve as the duration, the variance of the ship heading hdg, the variance and the mean of the real-time draft of the ship are calculated, and the results are reserved as parameters of each second cluster. And a fan identification step: according to relevant business knowledge, when the wind power installation ship is provided with a fan, the ship heading is basically kept unchanged, the length of the fan blade is between 45 and 60 meters, the wind fields are all in an offshore area, and the draft is not too large. Thus, preferably, classes with a radius within 100 meters, a variance within hdg within 5, and a mean of draugh within 7 are selected, which are considered fans, and stored in the database for later use in the identification of offshore wind farms.
Wind farm identification: and clustering the longitude and latitude coordinates of the central points of all clusters which are screened out to be used as suspected fans in the database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, namely obtaining the wind farm cluster within the range of the wind farm in Qingzhou of Yangjiang in Yue in 2021.
As shown in fig. 3, it is obvious that the fans in the offshore wind farm are regularly arranged in a straight line, the number of fans in the wind farm is more than 10 fans, the intervals among the fans are equal, the distance between each fan is approximately 530 meters, dest is 'YANG JIANG', and then the cluster can be determined to be an offshore wind farm. The opentreet map also shows that there are many installed clusters in this area.
The invention also relates to an AIS-based offshore wind farm identification system which corresponds to the AIS-based offshore wind farm identification method and can be understood as a system for realizing the method, wherein the system comprises a data acquisition module, a characteristic determination module, a clustering and calculating module, a fan identification module and a wind farm identification module which are connected in sequence,
the data acquisition module is used for acquiring ship data, and carrying out wind power installation ship feature mining from the ship data according to the business logic so as to determine a suspected wind power installation ship;
the characteristic determining module is used for acquiring AIS dynamic data of the suspected wind power installation ship and determining AIS characteristics of the wind power installation ship when the wind power installation ship is provided with a fan according to the AIS dynamic data and combining business knowledge; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
clustering and calculating module, clustering longitude and latitude position information of suspected wind power installation ship meeting AIS characteristic by using DBSCAN clustering algorithm to obtain a plurality of first clusters, clustering each obtained first cluster again by using Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining longitude and latitude coordinates of a central point of each second cluster, calculating radius of the second cluster according to longitude and latitude coordinates of the central point, and calculating variance of ship heading, mean value of ship draught and variance in the second cluster according to AIS dynamic data;
the fan identification module is used for screening clusters with the radius, the ship heading variance, the ship draft mean value and the ship draft variance meeting respective preset conditions from a plurality of second clusters by combining business knowledge, and storing the clusters as suspected fans into a database;
the wind power plant identification module is used for clustering longitude and latitude coordinates of central points of all clusters which are screened out to serve as suspected fans in a database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, the third clusters are used as suspected offshore wind power plants, distances between each fan and other fans in each suspected offshore wind power plant and slopes between each fan and the nearest fan are calculated, and the offshore wind power plant is automatically identified according to the number of fans in each suspected offshore wind power plant, the distances among the fans, the slopes and the ship destination port.
Preferably, the ship profile data includes a ship status field including a service field, a dismantling field, and an order field, a ship equipment field, and a ship type field including a provide offshore service field and a support platform field.
Preferably, the AIS features include speed to ground and navigational status.
Preferably, in the clustering and calculating module, when the DBSCAN clustering algorithm is adopted for clustering, the contour coefficient is also adopted as an evaluation index of the clustering effect.
Preferably, in the clustering and calculating module, after obtaining a plurality of second clusters, the minimum time and the maximum time in the clusters are reserved, and the time difference is calculated according to the minimum time and the maximum time and is used as an evaluation index of the installation efficiency of the wind power installation ship.
The invention provides an objective and scientific offshore wind farm identification method and system based on AIS, which are based on AIS data and ship data, perform multistage clustering by adopting a DBSCAN clustering algorithm and a Kmeans clustering algorithm, discover and identify an offshore wind farm by combining business knowledge, determine the position of the wind farm, can calibrate the range of the offshore wind farm, prevent the ship from entering by mistake, and evaluate the development condition of the offshore wind farm.
It should be noted that the above-described embodiments will enable those skilled in the art to more fully understand the invention, but do not limit it 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 the present invention may be modified or equivalent, and in all cases, all technical solutions and modifications which do not depart from the spirit and scope of the present invention are intended to be included in the scope of the present invention.
Claims (10)
1. An AIS-based offshore wind farm identification method is characterized by comprising the following steps:
and a data acquisition step: collecting ship data, and performing wind power installation ship feature mining from the ship data according to business logic to determine suspected wind power installation ships;
a feature determination step: acquiring AIS dynamic data of a suspected wind power installation ship, and determining AIS characteristics of the wind power installation ship when a fan is installed according to the AIS dynamic data; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
clustering and calculating: clustering longitude and latitude position information of suspected wind power installation vessels meeting AIS characteristics by adopting a DBSCAN clustering algorithm to obtain a plurality of first clusters, clustering each obtained first cluster again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining the longitude and latitude coordinates of a central point of each second cluster, calculating the radius of the second cluster according to the longitude and latitude coordinates of the central point, and calculating the variance of the ship bow direction in the second cluster, the mean value and the variance of ship draft according to AIS dynamic data;
and a fan identification step: intelligently screening clusters with the radius, the variance of the ship heading, the mean value of ship draft and the variance meeting respective preset conditions from a plurality of second clusters, and storing the clusters as suspected fans in a database;
wind farm identification: clustering longitude and latitude coordinates of central points of all clusters which are screened out from a database by adopting a DBSCAN clustering algorithm to obtain a plurality of third clusters, taking the third clusters as suspected offshore wind power plants, calculating the distance between each fan and other fans in each suspected offshore wind power plant and the slope of each fan closest to the fan, automatically identifying the offshore wind power plant according to the number of fans in each suspected offshore wind power plant, the distance between the fans, the slope and a ship destination port, and automatically judging that the suspected offshore wind power plant is the offshore wind power plant if the number of fans in a certain suspected offshore wind power plant is more than 10, the distances between the fans are equal, the slopes are the same, and the ship destination ports are the same.
2. The AIS-based offshore wind farm identification method of claim 1, wherein in the data gathering step the vessel profile data comprises a vessel status field, a vessel equipment field, and a vessel type field, the vessel status field comprising an operational field, a dismantling field, and an order field, the vessel type field comprising a provide offshore service field and a support platform field.
3. The method of claim 1, wherein in the step of determining the characteristics, the AIS characteristics include speed to ground and navigational status.
4. An AIS based offshore wind farm identification method according to any of claims 1 to 3, wherein in the clustering and calculating step, when clustering is performed by using a DBSCAN clustering algorithm, a profile coefficient is also used as an evaluation index of the clustering effect.
5. The method for identifying an offshore wind farm based on AIS according to claim 4, wherein in the clustering and calculating steps, after obtaining the plurality of second clusters, a minimum time and a maximum time in the clusters are reserved, and a time difference is calculated according to the minimum time and the maximum time and is used as an evaluation index of the installation efficiency of the wind power installation vessel.
6. An AIS-based offshore wind farm identification system is characterized by comprising a data acquisition module, a characteristic determination module, a clustering and calculating module, a fan identification module and a wind farm identification module which are connected in sequence,
the data acquisition module acquires ship data, and performs wind power installation ship feature mining from the ship data according to the business logic to determine a suspected wind power installation ship;
the characteristic determining module is used for acquiring AIS dynamic data of the suspected wind power installation ship and determining AIS characteristics of the wind power installation ship when the wind power installation ship installs the fan according to the AIS dynamic data; the AIS dynamic data comprise ship longitude and latitude position information, ground speed, ship heading, ship draft and ship destination port;
the clustering and calculating module is used for clustering longitude and latitude position information of suspected wind power installation vessels meeting AIS characteristics by adopting a DBSCAN clustering algorithm to obtain a plurality of first clusters, clustering each obtained first cluster again by adopting a Kmeans clustering algorithm to obtain a plurality of second clusters, obtaining longitude and latitude coordinates of a central point of each second cluster, calculating the radius of the second cluster according to the longitude and latitude coordinates of the central point, and calculating the variance of the ship heading, the mean value of ship draught and the variance in the second clusters according to AIS dynamic data;
the fan identification module intelligently screens clusters with the radius, the variance of the ship bow direction and the mean and variance of ship draft meeting respective preset conditions from a plurality of second clusters, and stores the clusters as suspected fans in a database;
the wind power plant identification module adopts a DBSCAN clustering algorithm to cluster longitude and latitude coordinates of central points of all clusters which are screened out as suspected wind power plants in a database to obtain a plurality of third clusters, the third clusters are used as suspected offshore wind power plants, the distance between each wind power plant and other wind power plants in each suspected offshore wind power plant and the slope of each wind power plant and the nearest wind power plant are calculated, the offshore wind power plant is automatically identified according to the number of wind power plants in each suspected offshore wind power plant, the distance between the wind power plants, the slope and the ship destination port, and if the number of wind power plants in a certain suspected offshore wind power plant is greater than 10, the distances between the wind power plants are equal, the slopes are the same, and the ship destination port is the same, the suspected offshore wind power plant is automatically judged to be the offshore wind power plant.
7. The AIS-based offshore wind farm identification system of claim 6, wherein the vessel profile data comprises a vessel status field comprising an operational field, a dismantling field, and an order field, a vessel equipment field, and a vessel type field comprising a provide offshore service field and a support platform field.
8. The AIS-based offshore wind farm identification system of claim 6, wherein the AIS features comprise speed to ground and navigational status.
9. The offshore wind farm identification system based on the AIS according to any one of claims 6 to 8, wherein in the clustering and calculating module, when the DBSCAN clustering algorithm is adopted for clustering, a profile coefficient is also adopted as an evaluation index of the clustering effect.
10. The offshore wind farm identification system based on AIS according to claim 9, wherein the clustering and calculating module reserves the minimum time and the maximum time in the cluster after obtaining the plurality of second clusters, and calculates the time difference according to the minimum time and the maximum time as an evaluation index of the installation efficiency of the wind power installation vessel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210936736.4A CN115310533B (en) | 2022-08-05 | 2022-08-05 | AIS-based offshore wind farm identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210936736.4A CN115310533B (en) | 2022-08-05 | 2022-08-05 | AIS-based offshore wind farm identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115310533A CN115310533A (en) | 2022-11-08 |
| CN115310533B true CN115310533B (en) | 2024-01-19 |
Family
ID=83860581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210936736.4A Active CN115310533B (en) | 2022-08-05 | 2022-08-05 | AIS-based offshore wind farm identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115310533B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102031151B1 (en) * | 2019-03-20 | 2019-10-15 | 씨벤티지 주식회사 | Method for providing route of vessel |
| AU2020102354A4 (en) * | 2020-09-21 | 2020-10-29 | Tianjin Research Institute For Water Transport Engineering.M.O.T. | Morning and early warning method for coastal port ship operation conditions |
| CN112487659A (en) * | 2020-12-15 | 2021-03-12 | 国网江苏省电力有限公司经济技术研究院 | Optimal design method and system for offshore wind farm current collection system |
| CN112613455A (en) * | 2020-12-29 | 2021-04-06 | 武汉理工大学 | Ship anchoring gathering area identification method and device |
| CN112906830A (en) * | 2021-04-14 | 2021-06-04 | 武汉理工大学 | Automatic generation method of optimal ship route based on AIS big data |
| CN113538820A (en) * | 2021-07-14 | 2021-10-22 | 华能国际电力江苏能源开发有限公司 | Offshore wind farm submarine cable monitoring and protecting system and method |
| CN113962321A (en) * | 2021-10-30 | 2022-01-21 | 中远海运科技股份有限公司 | Ship-to-ship loading and unloading goods identification processing method and system |
| CN113988213A (en) * | 2021-11-12 | 2022-01-28 | 中远海运科技股份有限公司 | Ship in-dock repair identification method and system |
| CN114444315A (en) * | 2022-01-30 | 2022-05-06 | 中远海运科技股份有限公司 | Ship station avoidance route simulation method and system |
| CN114781244A (en) * | 2022-03-25 | 2022-07-22 | 国网河南省电力公司安阳供电公司 | Grouping and parameter optimization method in wind power plant |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11460003B2 (en) * | 2019-09-18 | 2022-10-04 | Inventus Holdings, Llc | Wind turbine damage detection system using machine learning |
-
2022
- 2022-08-05 CN CN202210936736.4A patent/CN115310533B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102031151B1 (en) * | 2019-03-20 | 2019-10-15 | 씨벤티지 주식회사 | Method for providing route of vessel |
| AU2020102354A4 (en) * | 2020-09-21 | 2020-10-29 | Tianjin Research Institute For Water Transport Engineering.M.O.T. | Morning and early warning method for coastal port ship operation conditions |
| CN112487659A (en) * | 2020-12-15 | 2021-03-12 | 国网江苏省电力有限公司经济技术研究院 | Optimal design method and system for offshore wind farm current collection system |
| CN112613455A (en) * | 2020-12-29 | 2021-04-06 | 武汉理工大学 | Ship anchoring gathering area identification method and device |
| CN112906830A (en) * | 2021-04-14 | 2021-06-04 | 武汉理工大学 | Automatic generation method of optimal ship route based on AIS big data |
| CN113538820A (en) * | 2021-07-14 | 2021-10-22 | 华能国际电力江苏能源开发有限公司 | Offshore wind farm submarine cable monitoring and protecting system and method |
| CN113962321A (en) * | 2021-10-30 | 2022-01-21 | 中远海运科技股份有限公司 | Ship-to-ship loading and unloading goods identification processing method and system |
| CN113988213A (en) * | 2021-11-12 | 2022-01-28 | 中远海运科技股份有限公司 | Ship in-dock repair identification method and system |
| CN114444315A (en) * | 2022-01-30 | 2022-05-06 | 中远海运科技股份有限公司 | Ship station avoidance route simulation method and system |
| CN114781244A (en) * | 2022-03-25 | 2022-07-22 | 国网河南省电力公司安阳供电公司 | Grouping and parameter optimization method in wind power plant |
Non-Patent Citations (6)
| Title |
|---|
| Maritime Traffic Route Detection Framework Based on Statistical Density Analysis From AIS Data Using a Clustering Algorithm;Jeong-Seok Lee 等;《IEEE Access》;第10卷;全文 * |
| SCADA Data-Based Working Condition Classification for Condition Assessment of Wind Turbine Main Transmission System;Huanguo Chen 等;《energies》;第14卷(第21期);全文 * |
| 基于CNN-集成学习的多风电机组故障诊断;叶祎旎 等;《工业工程》;第25卷(第1期);全文 * |
| 基于历史航迹的船舶任意点到港的航线规划算法;段俊利 等;《上海船舶运输科学研究所学报》;第43卷(第2期);全文 * |
| 海量航行特定数据远程检索系统设计;刘佳;《舰船科学技术》;第42卷(第16期);全文 * |
| 考虑通航环境因素的智能船舶航线规划;潘伟 等;《上海海事大学学报》;第42卷(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115310533A (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111427100B (en) | Typhoon center positioning method and device and typhoon path generation method | |
| JP5435418B2 (en) | Ocean current data assimilation method and system | |
| CN107193060B (en) | A kind of multipath Typhoon Storm Surge Over method for quick predicting and system | |
| CN104520861A (en) | Method for predicting wind power density | |
| CN112348050B (en) | Anemometer arrangement method based on wind characteristics along high-speed rail | |
| CN114910980B (en) | Wind ring forecasting method based on subjective path intensity forecasting and parameterized wind field model | |
| CN115310533B (en) | AIS-based offshore wind farm identification method and system | |
| CN115114873A (en) | Ocean environment recurrence period design standard calculation method and system | |
| CN115326075A (en) | Path planning method for realizing wind field global automatic inspection based on unmanned aerial vehicle | |
| CN111666725A (en) | Anemometer tower planning and site selection method and system suitable for wind power plant with non-complex terrain | |
| CN104392113A (en) | Method for estimating wind speed of cold air wind on offshore sea surface | |
| CN112749847B (en) | Ship route determining method and device and electronic equipment | |
| CN115693666B (en) | Method and system for determining generating capacity of offshore wind farm based on satellite inversion | |
| Solbrekke et al. | Norwegian offshore wind power—Spatial planning using multi‐criteria decision analysis | |
| CN112364301B (en) | Slope length-based near-ground wind speed statistics downscaling method | |
| CN114047508B (en) | Sea radar detection range statistical evaluation method and system | |
| CN116070735A (en) | Yellow sea green tide distribution area based on side length and azimuth difference rule and drift prediction initial field manufacturing method thereof | |
| CN115146742A (en) | Offshore wind farm unit grouping and flagship machine selection method suitable for farm group control | |
| CN112364300A (en) | Near-ground wind speed statistics downscaling correction method based on relative slope length | |
| Tranberg et al. | Determining offshore wind installation times using machine learning and open data | |
| CN113988349A (en) | Numerical model forecasting product objective release method based on space-time local model | |
| CN112308275A (en) | Optimal pitch angle identification method and equipment of wind generating set | |
| CN117152372B (en) | Three-dimensional digital geographic information service platform for wind power engineering | |
| CN117574329B (en) | Nitrogen dioxide refined space distribution method based on ensemble learning | |
| CN117910973A (en) | Marine biomass methanol full life cycle assessment and carbon emission reduction optimization 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |