CN112734970B - Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology - Google Patents
Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology Download PDFInfo
- Publication number
- CN112734970B CN112734970B CN202011525960.1A CN202011525960A CN112734970B CN 112734970 B CN112734970 B CN 112734970B CN 202011525960 A CN202011525960 A CN 202011525960A CN 112734970 B CN112734970 B CN 112734970B
- Authority
- CN
- China
- Prior art keywords
- fault
- image
- unmanned aerial
- aerial vehicle
- fan
- 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
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/18502—Airborne stations
- H04B7/18506—Communications with or from aircraft, i.e. aeronautical mobile service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- 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
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Theoretical Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a system and a method for automatically inspecting an unmanned aerial vehicle in a wind power plant based on a LoRaWAN positioning technology, and belongs to the technical field of inspection of unmanned aerial vehicles in wind power plants. The unmanned aerial vehicle macroscopically patrols the wind turbine generator of the wind power plant and uploads a macroscopically patrolled image of the wind turbine generator through an LoRa terminal; preprocessing the uploaded image data; diagnosing a macroscopic inspection image of the wind turbine generator to find out a fault image; determining a shooting position for the fault image, and determining a fault fan according to the shooting position of the fault image and the position information of the fan generator set; controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection and shooting a high-definition image; and diagnosing the high-definition image shot by the microscopic inspection to find out a fault image and determine the fault type, and storing the high-definition fault image into an expert database for the next image diagnosis. The automation of patrolling and examining of wind-powered electricity generation field avoids manual operation unmanned aerial vehicle to patrol and examine the electric quantity on the route and consumes, improves unmanned aerial vehicle duration, realizes the low-power consumption location.
Description
Technical Field
The invention relates to a system and a method for inspecting an unmanned aerial vehicle in a wind power plant, in particular to a system and a method for automatically inspecting an unmanned aerial vehicle in a wind power plant based on LoRaWAN positioning technology.
Background
With the continuous breakthrough of wind power technology in China, the attention of wind power generation is paid more and more, and the scale of a wind power plant is gradually enlarged. If the unit is stopped due to faults, the generated energy is reduced, high maintenance cost is generated, and therefore the wind turbine needs to be regularly inspected. Traditionally, the inspection of fan blades mostly adopts the mode of manual inspection, if the inspection is observed by using a telescope and detected by workers at high altitude, the inspection method has the following defects: most wind power plants are widely distributed and have complex terrain, so that the traditional manual inspection is difficult, (2) the wind power plant inspection task is heavy, the traditional manual detection efficiency is low, the time cost is high, the shutdown loss is large, (3) the inspection personnel work high above the ground, and the safety risk is large.
Along with the continuous development of unmanned aerial vehicle technique, unmanned aerial vehicle patrols and examines and to have revealed very big use value, can effectively improve wind field and patrol and examine work level. Among the prior art, the personnel of patrolling and examining patrols and examines at field operation unmanned aerial vehicle much, and this kind of manual operation unmanned aerial vehicle patrols and examines and has higher requirement to the operation personnel, and in addition, the manual operation unmanned aerial vehicle can appear leaking clapping, condition such as wrong clap and under the bad condition of weather, the staff probably can't reach the scene, patrols and examines unmanned aerial vehicle and can't use. Furthermore, unmanned aerial vehicle duration is generally in 30 minutes, and when manual operation unmanned aerial vehicle patrolled and examined, once charged and supported the whole fan condition of unmanned aerial vehicle tour wind field.
In addition, the existing unmanned aerial vehicle positioning technology mostly adopts a single-point GPS positioning technology, the positioning accuracy is 5-10 meters, and although the requirement of people on the positioning accuracy can be met, the power consumption of the unmanned aerial vehicle positioning technology on equipment is large, and the cost is high. There are also remote locations of wind farms, and the satellites are often mobile and the radio signals are lost. Moreover, the traditional positioning devices such as Bluetooth, wi-Fi and ZigBee are large in power consumption and not suitable for positioning a wind field area with a large area.
Disclosure of Invention
In view of the deficiencies of the prior art, the object of the application is to provide an automatic inspection system and method for an unmanned aerial vehicle in a wind farm based on LoRaWAN positioning technology, aiming at solving the problems that the unmanned aerial vehicle is automatically inspected to replace a manual control unmanned aerial vehicle to inspect, the GPS positioning power consumption of the existing unmanned aerial vehicle is large, and the endurance time of the unmanned aerial vehicle is short.
In order to solve the technical problem, the automatic inspection system of the unmanned aerial vehicle in the wind farm based on the LoRaWAN positioning technology comprises the unmanned aerial vehicle, loRa modules, at least three LoRaWAN gateways distributed around the wind farm and an upper computer; the system comprises an unmanned aerial vehicle main control chip, a LoRa module, a LoRa wireless signal and a LoRa wireless signal, wherein the LoRa module is electrically connected with the unmanned aerial vehicle main control chip to form a LoRa terminal; the LoRa module is simultaneously in wireless communication connection with each LoRaWAN gateway node, and the upper computer comprises a network server, a control system and a fault recognition system; each LoRaWAN gateway node is in communication connection with a network server in the upper computer through the Ethernet;
the network server is used for receiving the position information of the wind turbine generator transmitted by the wind turbine generator; collecting data signals containing data sent by an LoRa module and timestamps added by gateway nodes from at least three LoRaWAN gateway nodes respectively; processing the same data frames received from all LoRaWAN gateways, including sequencing and grouping; according to the processed data frame, carrying out real-time unmanned aerial vehicle positioning calculation, and outputting and displaying the position information of the unmanned aerial vehicle; receiving a fault image obtained by macro inspection of the wind turbine generator system and sent by a fault identification system, reading the shooting time of the fault image from the received picture attribute information of the fault image, finding out a timestamp which is close to the shooting time of the fault image in a data frame timestamp, and matching the shooting position corresponding to the fault image according to the timestamp; determining a fault fan according to the shooting position of the fault picture and the position information of the wind generating set, and sending the position of the fault fan to a control system;
the control system is used for receiving and displaying meteorological condition information of the wind power plant from the meteorological station; receiving position information of the wind turbine generator transmitted by the wind turbine generator; planning an unmanned aerial vehicle routing inspection path, and sending the optimal routing inspection path of the fan to an LoRa terminal; receiving image data transmitted by an LoRa terminal in real time; carrying out image preprocessing on the macroscopic inspection image data of the wind turbine generator uploaded by the LoRa terminal, and sending the processed image data to a fault identification system; receiving and displaying the position of the fault fan transmitted by the network server; controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection, shooting a high-definition image, and transmitting the high-definition image of the fault fan to a fault identification system;
the fault recognition system is used for diagnosing the macroscopic inspection image data of the wind turbine generator and the high-definition image of the fault fan, which are received from the control system, and performing frame-by-frame comparison analysis on the macroscopic inspection image data and the high-definition image of the fault fan and the image data of the expert database to find out the fault image in the image: sending the diagnosed fault image in the macroscopic inspection image of the wind turbine generator to a network server; and judging the fault type of the high-definition fault image of the fault fan, and storing the high-definition fault image of the fault fan into an expert library for next detection, identification and comparison.
As a further improvement of the system, the unmanned aerial vehicle adopts a multi-rotor unmanned aerial vehicle.
In order to solve the technical problem, the automatic inspection method of the unmanned aerial vehicle in the wind power plant based on the LoRaWAN positioning technology comprises the following steps:
step 1: the unmanned aerial vehicle macroscopically inspects the wind turbine generator and shoots images, and the LoRa terminal uploads the macroscopically inspected images of the wind turbine generator;
step 2: preprocessing the macroscopic inspection image data of the wind turbine generator uploaded by the LoRa terminal;
and step 3: diagnosing the processed macroscopic inspection image of the wind turbine generator to find out a fault image;
and 4, step 4: determining a shooting position for the fault image, determining a fault fan according to the shooting position of the fault image and the position information of the wind generating set, and displaying the position of the fault fan;
and 5: controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection and shooting a high-definition image according to the position of the fault fan;
and 6: and diagnosing the high-definition image shot by the microscopic inspection to find out a fault image and determine the fault type, and storing the high-definition fault image into an expert library for the next image diagnosis.
As a further improvement of the method of the present invention, said step 1 comprises the steps of:
step 1.1: acquiring position information of a wind turbine generator;
step 1.2: planning a routing inspection path according to the position information of the wind turbine generator, and sending the optimal routing inspection path to an LoRa terminal;
step 1.3: the unmanned aerial vehicle macroscopically inspects the wind turbine generator according to the optimal inspection path, remotely photographs the wind turbine generator, and positions the unmanned aerial vehicle in real time by using a LoRaWAN technology;
step 1.3.1: the LoRa module sends a data frame;
step 1.3.2: at least 3 LoRaWAN gateways in the network range of the LoRa module receive the data frame, and each LoRaWAN gateway adds a timestamp to the data frame received by each LoRaWAN gateway;
step 1.3.3: each LoRaWAN gateway sends a data frame containing a timestamp to a network server;
step 1.3.4: the network server processes the same data frames received from all LoRaWAN gateways;
step 1.3.5: and performing unmanned aerial vehicle positioning calculation by using the processed data frame.
As a further improvement of the method of the present invention, in the step 1.2, a firefly algorithm is adopted to plan the routing inspection path:
encode a plurality of fans, and once can obtain a route with whole fans according to the code random arrangement, random arrangement can get many and patrol and examine the route many times, every route corresponds a glowworm, the number of initialization glowworm is N, it has N to solve to patrol and examine the route problem promptly that the fan is optimum, every glowworm corresponds an absolute brightness, the glowworm is removed according to the attraction rule that the glowworm that absolute brightness is little is attracted by the glowworm that absolute brightness is big, after removing, the glowworm that luminance is the biggest is exactly the optimal solution, the fan is optimum to patrol and examine the route promptly.
As a further improvement of the method of the present invention, the processing of the data frames in step 1.3.4 includes sorting and grouping, and the network server sorts the same data frames received from different LoRaWAN gateways according to the arrival time, and groups all metadata containing the data frames.
As a further improvement of the method of the present invention, in step 1.3.5, the method for performing positioning calculation of the unmanned aerial vehicle by using the processed data frame includes:
calculate the time difference that different gateways received to given data frame structure, convert the time difference of calculating into the distance difference, obtain the distance of loRa module to each LoRaWAN gateway to realize the location of loRa module, also realize unmanned aerial vehicle location.
As a further improvement of the method of the present invention, the method for diagnosing the images and finding out the fault images in the step 3 and the step 6 is: and finding out a fault image by a method of comparing the image data to be diagnosed with the image data stored in the expert database frame by frame.
As a further improvement of the method of the present invention, said step 4 comprises the steps of:
step 4.1: reading the shooting time of the fault image from the picture attribute information of the fault image;
step 4.2: comparing the shooting time of the fault picture with the time stamp of the data frame, and acquiring the time stamp which is closest to the shooting time of the fault picture in the time stamp so as to match the shooting position corresponding to the fault image;
step 4.3: and matching the fan closest to the shooting position of the fault picture according to the shooting position of the fault picture and the position information of the wind generating set, namely obtaining the fan with the fault.
As a further improvement of the method, in the step 5, controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection and shooting a high-definition image includes the following steps:
when the unmanned aerial vehicle is controlled to fly to the hub of the designated fan, four paths of the front edge, the rear edge, the windward side and the leeward side of each blade are patrolled and examined, high-definition image shooting is carried out, and multi-angle microscopic image data collection of the failed fan is completed.
The invention has the beneficial effects that:
1. the automation of the inspection of the wind power plant is completely realized, the manpower resource is saved, and the inspection level is improved.
2. The mode of planning the whole wind field routing inspection path before routing inspection avoids the consumption of electric quantity on the path of manually operated unmanned aerial vehicle routing inspection, and can solve the problem of poor endurance of the unmanned aerial vehicle.
3. By adopting LoRaWAN positioning technology, only LoRa equipment is needed, and extra hardware cost is not needed.
4. The LoRa terminal has the advantages of lowest power consumption, lowest cost, lowest environmental influence and smallest size. For a wind power plant with wide distribution, higher positioning accuracy is not needed, and compared with the characteristics of high accuracy of a GPS (global positioning system) but the highest power consumption, the LoRaWAN positioning technology is a more suitable scheme.
5. The wind power plant is built at an open position, and the LoRaWAN positioning technology is applied to the wind power plant, so that the multipath transmission of signals can be effectively reduced, and the positioning precision is improved.
6. Wind-powered electricity generation field is built in remote positions such as mountain area more, adopts GPS location probably to appear the poor condition of signal, and adopts LoRaWAN positioning technology realizing the low power dissipation, under the long-distance condition of transmission, unmanned aerial vehicle high-speed removal can keep communication stable.
Drawings
Fig. 1 is a schematic structural diagram of an automatic inspection system of an unmanned aerial vehicle in a wind farm based on a LoRaWAN positioning technology in the embodiment;
fig. 2 is a schematic structural diagram of a module assembly in the upper computer in the present embodiment;
fig. 3 is a flow chart of an automatic inspection method for a wind farm unmanned aerial vehicle based on a LoRaWAN positioning technology in the embodiment;
FIG. 4 is a flow chart of a macro routing inspection process in the method of the present invention;
FIG. 5 is a flow chart of a process for planning a routing inspection path by using a firefly algorithm in the method of the present invention;
FIG. 6 is a flowchart of the process of determining a faulty blower at step 4 in the method of the present invention.
Detailed Description
To facilitate an understanding of the present application, the present application will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present application are given in the accompanying drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
For solving a great deal of defect that current manual control unmanned aerial vehicle patrols and examines and to patrol and examine and bring, current unmanned aerial vehicle GPS location consumption is big and the problem that unmanned aerial vehicle duration is short, the solution that this application provided is: automatic inspection system and method for unmanned aerial vehicle in wind power station based on LoRaWAN positioning technology.
As shown in FIG. 1, automatic system of patrolling and examining of wind-powered electricity generation field unmanned aerial vehicle based on LoRaWAN location technology, including unmanned aerial vehicle, the LoRa module of setting on the unmanned aerial vehicle fuselage, distribute at least three LoRaWAN gateway and the host computer around the wind-powered electricity generation field. In this embodiment, what the loRa module chooseed for use is the wireless loRa module of SX1278, and the loRa module constitutes the loRa terminal with unmanned aerial vehicle main control chip electric connection, and unmanned aerial vehicle main control chip modulates the loRa wireless signal transmission with data through the loRa module and goes out.
The upper computer comprises a network server, a control system and a fault identification system, as shown in fig. 2. The LoRa module is simultaneously connected with each LoRaWAN gateway node wireless connection, and each LoRaWAN gateway node is connected with the network server in the host computer through the ethernet respectively again simultaneously.
The network server is used for receiving the position information of the wind turbine generator transmitted by the wind turbine generator; collecting data signals containing data sent by an LoRa module and timestamps added by gateway nodes from at least three LoRaWAN gateway nodes respectively; processing the same data frames received from all LoRaWAN gateways, including sequencing and grouping; according to the processed data frame, carrying out real-time unmanned aerial vehicle positioning calculation, and outputting and displaying the position information of the unmanned aerial vehicle; receiving a fault picture sent by a fault identification system, reading the shooting time of the fault picture from the picture attribute information of the fault picture received by the fault identification system, finding out a time stamp which is close to the shooting time of the fault picture in the data frame time stamps, and matching the shooting position corresponding to the fault picture according to the time stamp; and determining the fault fan according to the shooting position of the fault picture and the position information of the wind generating set, and sending the position of the fault fan to the control system.
The control system is used for receiving and displaying the meteorological condition information of the wind power plant from the meteorological station; receiving wind turbine generator position information transmitted by a wind turbine generator; planning an unmanned aerial vehicle routing inspection path, and sending the optimal routing inspection path of the fan to an LoRa terminal; receiving image data transmitted by an LoRa terminal in real time; carrying out image preprocessing on the macroscopic inspection image data of the wind turbine generator uploaded by the LoRa terminal, and sending the processed image data to a fault identification system; receiving and displaying the position of the fault fan transmitted by the network server; and controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection and shooting a high-definition image, and transmitting the high-definition image of the fault fan to the fault identification system.
And the fault identification system is used for diagnosing the macroscopic inspection image data of the wind turbine generator and the high-definition image of the fault fan, which are received from the control system, and performing frame-by-frame comparison analysis on the macroscopic inspection image data and the high-definition image of the fault fan and the image data of the expert database to find out the fault image in the inspection image: sending the diagnosed fault image in the macroscopic inspection image of the wind turbine generator to a network server; and judging the fault type of the high-definition fault image of the fault fan, and storing the high-definition fault image of the fault fan into an expert database for next detection, identification and comparison.
When carrying out wind-powered electricity generation field unmanned aerial vehicle and patrolling and examining automatically, need do earlier and patrol and examine the preparation work before, include: selecting a proper unmanned aerial vehicle; arranging an unmanned aerial vehicle automatic airport in a wind power plant; setting a meteorological station in a wind power plant to detect the meteorological environment of the wind power plant; a technician checks the meteorological condition of the wind power plant and determines whether the meteorological condition of the wind power plant meets the patrol condition or not; the technical staff checks the health and the electric quantity status of the unmanned aerial vehicle equipment, etc. In this embodiment, the size is 1.5m for the unmanned aerial vehicle automatic airport in the spacious region in wind power plant, with the fixing bolt reinforcement. The automatic airport of unmanned aerial vehicle adopts the photovoltaic power supply, is provided with the line network. Because many rotor unmanned aerial vehicle have stably hover, advantages such as easy operation, the event is selected many rotor unmanned aerial vehicle and is patrolled and examined work.
After the preparation work is done, the automatic inspection method for the unmanned aerial vehicle in the wind farm based on the automatic inspection system for the unmanned aerial vehicle in the wind farm based on the LoRaWAN positioning technology comprises the following steps as shown in FIG. 3:
step 1: unmanned aerial vehicle patrols and examines and divide into the macroscopical inspection of wind turbine generator system and the microcosmic inspection of single fan. Firstly, macro inspection of the wind turbine is carried out, and a LoRa terminal uploads a macro inspection image of the wind turbine; the specific process is shown in fig. 4, and comprises the following steps:
step 1.1, inputting the position information of the wind turbine into a control system and a network server in an upper computer.
And 1.2, planning a routing inspection path in the control system by using an artificial intelligence algorithm according to the position information of the wind turbine generator, and sending the planned optimal routing inspection path of the fan to the LoRa terminal.
In the present embodiment: the artificial intelligence algorithm adopts a firefly algorithm. The firefly algorithm is a heuristic clustering intelligent algorithm, and the basic idea is as follows: randomly generating a group of random solutions, simulating the characteristic that the firefly tends to fly to a brighter place, iterating by establishing a multi-mode continuous function optimization problem, and continuously updating the solutions in the iteration process until the optimal solutions are achieved. In this embodiment, the fans are encoded to 1,2,3, \8230;, n, and a path, for example {1,2,3, \8230;, n } can be obtained by randomly arranging n fans once, and each path corresponds to a firefly, i.e., P = { P } 1 ，p 2 ，…，p n }. The number of initialization firefly is N, and the fan is optimum to patrol and examine the route problem and has N to solve promptly, and every firefly corresponds an absolute brightness, and the firefly is moved by the attraction rule that the firefly that absolute brightness is big attracts according to the firefly that absolute brightness is little, and its absolute brightness value is big more, and the potential solution that the firefly represented is more excellent. After the movement is finished, the firefly with the maximum brightness is the optimal solution, namely, the optimal routing inspection path of the fan, namely, the shortest path of routing inspection of the unmanned aerial vehicle.
The specific process is shown in fig. 5, and comprises the following steps:
step 1.2.1: establishing a proper coordinate system according to the inspection space, setting a coordinate origin, and setting coordinates (x) of each fan i ，y i ，z i )。
Step 1.2.2: and calculating a fan distance matrix according to the coordinates of each fan and defining an objective function.
The space distance between the two fans is calculated according to the formula (1):
wherein i represents the ith fan; j represents the jth fan; d ij The space distance of the fan is i, j; x is a radical of a fluorine atom i 、y i 、z i Respectively corresponding coordinates of an x axis, a y axis and a z axis of the fan i in a coordinate system; x is the number of j 、y j 、z j Respectively corresponding coordinates of an x axis, a y axis and a z axis of the fan j in a coordinate system;
the distance matrix is calculated according to equation (2):
wherein D is the space distance D from the fan ij A distance matrix is formed; n is the number of the fans;
the objective function is calculated according to equation (3):
wherein x is ij According to equation (4) to calculate
Step 1.2.3: setting the maximum iteration times and initial parameters of the firefly algorithm;
in this embodiment, the maximum number of iterations of the firefly algorithm is set to M. The initial parameters include the number of fireflies N, the maximum attraction force, and the light absorption coefficient. Wherein, the attraction formula is calculated according to the formula (5):
wherein, beta ij Representing the attractive force between i and j fireflies; beta is a 0 Which represents the attraction of fireflies at the origin, is usually taken to be 1. Gamma is a light absorption coefficient, has certain influence on the optimization effect of the algorithm, and is generally taken as 0.01,100]. Suppose the solution of firefly i is P (i) = { P i1 ，p i2 ，…，p in The solution of firefly j is P (j) = { P j1 ,p j2 ，…,p jn }。r ij Calculating the individual distance of the firefly according to the formula (6):
step 1.2.4: initializing N fireflies
(i =1,2, \8230;, N), representing the solution of polling N paths, initializes the luminance information. In order to obtain the optimal path, the brightest firefly needs to be found, that is, the superiority and inferiority of the solution are determined by the absolute brightness of the firefly, which can be calculated according to equation (7):
step 1.2.5: and judging whether the iteration times of the algorithm reach M, if so, executing the step 1.2.9, otherwise, executing the next step.
Step 1.2.6: calculating the distance r between fireflies ij And attractive force beta ij The firefly moves and updates the brightness according to the attraction rule.
Step 1.2.7: selecting firefly with the maximum brightness for self-adaptive local adjustment, and for certain firefly P (i) = { P = i1 ,p i2, …,p in I.e. the solution before adjustment, randomly chosen points are transposed by parity, e.g. p i1 And p i2 Exchange, p i3 And p i4 And exchanging to obtain a new sequence, namely the adjusted solution, so as to carry out local adjustment, comparing the absolute brightness of the solution before and after the local adjustment, and keeping the solution with large absolute brightness, namely the local optimal solution.
Step 1.2.8: the number of iterations M = M +1, step 1.2.5 is performed.
Step 1.2.9: and finding out a global optimal solution from all the fireflies, and optimizing the global optimal solution by using a 2-opt algorithm to obtain an optimal fan routing inspection path.
The found global optimal solution may have crossed edges, and the embodiment utilizes a 2-opt algorithm to exchange two edges each time, compares the edge lengths before and after the exchange, and eliminates the crossed edges to obtain the optimal solution with the shortest path, namely the optimal routing inspection path of the fan.
Step 1.2.10: and outputting the optimal path and length, and drawing a route map.
Step 1.3: the unmanned aerial vehicle carries out macroscopic inspection on the wind turbine generator according to the optimal inspection path of the fan, carries out long-distance image shooting on the wind turbine generator and carries out real-time positioning on the unmanned aerial vehicle by utilizing a LoRaWAN technology;
the variable-focus camera carried by the unmanned aerial vehicle is used for shooting remote images of the wind turbine generator, and macroscopic inspection image data of the wind turbine generator are uploaded to the control system through wireless transmission.
When the unmanned aerial vehicle patrols and examines, control system controls unmanned aerial vehicle route and flight attitude. The rotational speed of four screws of unmanned aerial vehicle through control system control changes unmanned aerial vehicle's airspeed and direction among this embodiment, through realizing the instruction of six hundred times per second, comes stable control unmanned aerial vehicle's route of patrolling and examining and flight gesture. When four screw rotational speeds are the same, a screw accelerates, and unmanned aerial vehicle can fly towards the screw direction with higher speed, and two arbitrary screw rotational speeds are greater than two in addition, and unmanned aerial vehicle alright realizes rotatoryly. Through the change of four screw speed, unmanned aerial vehicle can constantly change the airspeed and the direction of oneself.
Utilize LoRaWAN technique to fix a position unmanned aerial vehicle in real time in this embodiment. The LoRa module that unmanned aerial vehicle main control chip connects is connected with at least three LoRaWAN gateway, and LoRaWAN gateway is connected with network server, and LoRaWAN gateway sends data package to network server, and network server carries out backstage data processing, confirms unmanned aerial vehicle positional information. The LoRaWAN positioning technology realizes positioning through Time Difference Of Arrival (TDOA) Of data generated by three or more LoRaWAN gateways, and additional hardware support is not needed in the positioning process. When the uplink data of one LoRa module is received by three or more gateways, the positioning can be realized. The uplink data may be a general data frame structure and may not be specific location information. The gateways share the same Time base, and add a super-high resolution Time stamp to each received data packet, each gateway reports the Arrival Time, and the final node position is determined based on a position calculation algorithm Of TDOA (Time Difference Of Arrival).
The specific positioning steps are shown as follows:
step 1.3.1: the LoRa module sends a data frame;
step 1.3.2: at least 3 LoRaWAN gateways in the network range of the LoRa module receive the data frame, and each LoRaWAN gateway adds a timestamp to the data frame received by each LoRaWAN gateway;
step 1.3.3: each LoRaWAN gateway sends a data frame containing the timestamp to a network server;
the network server receives the upstream frame containing the LoRaWAN gateway accurate timestamp, and simultaneously, the signal level, arrival time, signal-to-noise ratio and frequency error are also forwarded to the network server as part of the upstream frame.
Step 1.3.4: the network server processes the same data frame received from all LoRaWAN gateways;
the processing of the data frames comprises sorting and grouping, the network server sorts the same data frames received by the network server from different LoRaWAN gateways according to the arrival time, and all metadata containing the data frames are grouped.
Step 1.3.5: and inputting the processed data frame into a network server, performing unmanned aerial vehicle positioning calculation by using a positioning calculation algorithm, and outputting and displaying the position information of the unmanned aerial vehicle.
By utilizing a positioning calculation algorithm, the network server calculates the time difference received by different gateways for a given frame structure, and converts the calculated time difference into a distance difference, so that the distance from the LoRa module to each LoRaWAN gateway can be obtained. The time difference between the two LoRaWAN gateways is known, and the location of the LoRa module can be placed in a hyperbola. The time differences of the LoRaWAN gateways are known, a plurality of hyperbolas are drawn, and LoRa module positions are located on the hyperbolas. The nodical positional information that is exactly the loRa module of many hyperbolas to realize fixing a position loRa module location, also realize unmanned aerial vehicle location.
Step 2: the control system preprocesses the macro inspection image data of the wind turbine generator uploaded by the LoRa terminal and sends the processed image data to the fault recognition system.
The technical personnel preprocess the received macroscopic inspection image data of the wind turbine generator through the control system, for example, delete the wrong or unusable pictures, enhance the unclear pictures by using an image enhancement technology, prepare for fault identification, and send the processed image data to the fault identification system.
And step 3: and the fault identification system diagnoses the received macroscopic inspection image of the wind turbine generator to find out a fault image and sends the fault image to the network server.
And the fault identification system compares the received macroscopic inspection image data of the wind turbine generator with the image data stored in the expert database frame by frame through a deep learning algorithm, finds out a fault image in the inspection image and sends the fault image to the network server.
And 4, step 4: the network server determines a shooting position for the received fault image, determines a fault fan according to the shooting position of the fault image and the position information of the wind generating set, and sends the position of the fault fan to the control system; specifically, as shown in fig. 6, the method includes the following steps:
step 4.1: the network server reads the shooting time of the fault image from the picture attribute information of the fault image received by the network server.
Step 4.2: in a network server, comparing the fault picture shooting time with a data frame time stamp received by the network server, and acquiring the time stamp closest to the fault picture shooting time in the time stamps so as to match the shooting position corresponding to the fault picture;
step 4.3: according to the shooting position of the fault picture and the position information of the wind generating set, the network server matches the fan closest to the shooting position of the fault picture, namely the fan is the fault fan, and the position of the fault fan is sent to the control system.
And 5: the control system displays the position of the fault fan, and technicians send instructions through the control system to control the unmanned aerial vehicle to fly to the specified fault fan for microscopic inspection and shoot high-definition images to obtain microscopic inspection images of the fault fan;
when the unmanned aerial vehicle flies to the appointed fan hub, four paths of the front edge, the rear edge, the windward side and the leeward side of a single blade are patrolled and examined, 12 paths are patrolled and examined by each fan, high-definition image shooting is carried out, and multi-angle microscopic patrolling and examining image data acquisition of a fault fan is completed.
Step 6: and uploading the microcosmic inspection high-definition image of the fault fan to a fault recognition system, comparing and analyzing the microcosmic inspection high-definition image with image data in an expert database, determining fault types such as blade cracks, oil stains, sand holes, gel coat falling and the like, and storing the high-definition fault image of the fault fan into the expert database for next image diagnosis, so that the recognition and diagnosis rate is continuously improved.
It should be understood that various modifications and changes can be made by those skilled in the art without departing from the spirit of the invention, and still fall within the scope of the invention.
Claims (4)
1. A method for automatically inspecting an unmanned aerial vehicle in a wind power plant based on a LoRaWAN positioning technology is characterized in that an unmanned aerial vehicle automatic inspection system in the wind power plant based on the LoRaWAN positioning technology is adopted, and the system comprises the unmanned aerial vehicle, loRa modules, at least three LoRaWAN gateways and an upper computer, wherein the LoRaWAN gateways are distributed around the wind power plant; the LoRa module and the unmanned aerial vehicle main control chip are electrically connected to form a LoRa terminal, and the unmanned aerial vehicle main control chip modulates a data packet into a LoRa wireless signal through the LoRa module and sends the LoRa wireless signal out; the LoRa module is simultaneously in wireless communication connection with each LoRaWAN gateway node, and the upper computer comprises a network server, a control system and a fault recognition system; each LoRaWAN gateway node is in communication connection with a network server in the upper computer through the Ethernet;
the network server is used for receiving the position information of the wind turbine generator transmitted by the wind turbine generator; collecting data signals containing data sent by an LoRa module and timestamps added by gateway nodes from at least three LoRaWAN gateway nodes respectively; processing the same data frames received from all LoRaWAN gateways, including sequencing and grouping; performing real-time unmanned aerial vehicle positioning calculation according to the processed data frame; receiving a fault image obtained by macro inspection of the wind turbine generator system and sent by a fault identification system, reading the shooting time of the fault image from the received picture attribute information of the fault image, finding out a timestamp which is close to the shooting time of the fault image in a data frame timestamp, and matching the shooting position corresponding to the fault image according to the timestamp; determining a fault fan according to the shooting position of the fault picture and the position information of the wind generating set, and sending the position of the fault fan to a control system;
the control system is used for receiving the position information of the wind turbine generator transmitted by the wind turbine generator; planning an unmanned aerial vehicle routing inspection path, and sending the optimal routing inspection path of the fan to an LoRa terminal; receiving image data transmitted by an LoRa terminal in real time; carrying out image preprocessing on the macroscopic inspection image data of the wind turbine generator uploaded by the LoRa terminal, and sending the processed image data to a fault identification system; receiving and displaying the position of the fault fan transmitted by the network server; controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection, shooting a high-definition image, and transmitting the high-definition image of the fault fan to a fault identification system;
the fault recognition system is used for diagnosing the macroscopic inspection image data of the wind turbine generator and the high-definition image of the fault fan, which are received from the control system, and performing frame-by-frame comparison analysis on the macroscopic inspection image data and the high-definition image of the fault fan and the image data of the expert database to find out the fault image in the image: sending the diagnosed fault image in the macroscopic inspection image of the wind turbine generator to a network server; judging the fault type of the high-definition fault image of the fault fan, and storing the high-definition fault image of the fault fan into an expert library for next detection, identification and comparison;
the method is characterized by comprising the following steps:
step 1: the method comprises the steps that an unmanned aerial vehicle is controlled to conduct macroscopic inspection on a wind turbine generator of a wind power plant and shoot images, and a LoRa terminal uploads the macroscopic inspection images of the wind turbine generator;
step 2: preprocessing the macroscopic inspection image data of the wind turbine generator uploaded by the LoRa terminal;
and step 3: diagnosing the processed macroscopic inspection image of the wind turbine generator to find out a fault image;
and 4, step 4: determining a shooting position for the fault image, determining a fault fan according to the shooting position of the fault image and the position information of the wind generating set, and displaying the position of the fault fan;
and 5: controlling the unmanned aerial vehicle to fly to a specified fault fan for microscopic inspection and shooting a high-definition image according to the position of the fault fan;
step 6: diagnosing the high-definition image shot by the microscopic inspection to find out a fault image and determine the fault type, and storing the high-definition fault image into an expert library for the next image diagnosis;
the step 1 comprises the following steps:
step 1.1: acquiring position information of a wind turbine generator;
step 1.2: planning a routing inspection path according to the position information of the wind turbine generator, and sending the optimal routing inspection path to an LoRa terminal;
step 1.3: the unmanned aerial vehicle carries out macroscopic inspection on the wind turbine generator according to the optimal inspection path, carries out long-distance image shooting on the wind turbine generator, and carries out real-time positioning on the unmanned aerial vehicle by utilizing a LoRaWAN technology;
step 1.3.1: the LoRa module sends a data frame;
step 1.3.2: at least 3 LoRaWAN gateways in the network range of the LoRa module receive the data frame, and each LoRaWAN gateway adds a timestamp to the data frame received by each LoRaWAN gateway;
step 1.3.3: each LoRaWAN gateway sends a data frame containing a timestamp to a network server;
step 1.3.4: the network server processes the same data frames received from all LoRaWAN gateways;
step 1.3.5: performing unmanned aerial vehicle positioning calculation by using the processed data frame;
in the step 1.2, a firefly algorithm is adopted to plan the routing inspection path:
coding a plurality of fans, randomly arranging all the fans once according to codes to obtain a path, randomly arranging for multiple times to obtain a plurality of routing inspection paths, wherein each path corresponds to one firefly, the number of the initialized fireflies is N, namely N solutions exist in the problem of the optimal routing inspection path of the fan, each firefly corresponds to an absolute brightness, the firefly moves according to the attraction rule that the firefly with small absolute brightness is attracted by the firefly with large absolute brightness, and after the movement is finished, the firefly with the largest brightness is the optimal solution, namely the optimal routing inspection path of the fan;
the processing of the data frames in the step 1.3.4 includes sorting and grouping, the network server sorts the same data frames received by the network server from different LoRaWAN gateways according to the arrival time, and groups all metadata including the data frames;
the method for performing positioning calculation of the unmanned aerial vehicle by using the processed data frame in the step 1.3.5 comprises the following steps:
calculate the time difference that different gateways received to given data frame structure, convert the time difference of calculating into the distance difference, obtain the distance of loRa module to each LoRaWAN gateway to realize the location of loRa module, also realize unmanned aerial vehicle location.
2. The automatic inspection method for the unmanned aerial vehicle in the wind farm based on the LoRaWAN positioning technology according to claim 1, wherein the method for diagnosing the images and finding out the fault images in the steps 3 and 6 is as follows: and finding out a fault image by a method of comparing the image data to be diagnosed with the image data stored in the expert database frame by frame.
3. The LoRaWAN positioning technology-based automatic inspection method for unmanned aerial vehicles in wind farms according to claim 1, wherein the step 4 comprises the following steps:
step 4.1: reading the shooting time of the fault image from the picture attribute information of the fault image;
step 4.2: comparing the fault picture shooting time with the data frame time stamp, and acquiring the time stamp which is closest to the fault picture shooting time in the time stamps so as to match the shooting position corresponding to the fault image;
step 4.3: and matching the fan closest to the shooting position of the fault picture according to the shooting position of the fault picture and the position information of the wind generating set, namely obtaining the fan with the fault.
4. The automatic inspection method for the unmanned aerial vehicle in the wind farm based on the LoRaWAN positioning technology according to claim 1, wherein in the step 5, the unmanned aerial vehicle is controlled to fly to a specified fault fan for microscopic inspection and high-definition image shooting, and the method comprises the following steps: when the unmanned aerial vehicle is controlled to fly to the hub of the designated fan, four paths of the front edge, the rear edge, the windward side and the leeward side of each blade are patrolled and examined, high-definition image shooting is carried out, and multi-angle microscopic image data collection of the failed fan is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011525960.1A CN112734970B (en) | 2020-12-22 | 2020-12-22 | Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011525960.1A CN112734970B (en) | 2020-12-22 | 2020-12-22 | Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734970A CN112734970A (en) | 2021-04-30 |
| CN112734970B true CN112734970B (en) | 2022-12-27 |
Family
ID=75605431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011525960.1A Active CN112734970B (en) | 2020-12-22 | 2020-12-22 | Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734970B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113365370B (en) * | 2021-05-24 | 2022-09-02 | 内蒙古工业大学 | Intelligent mobile system based on LoRa technique |
| CN113506345A (en) * | 2021-07-28 | 2021-10-15 | 上海纽盾科技股份有限公司 | Unmanned aerial vehicle security situation sensing method and system for equipment inspection |
| CN114582040B (en) * | 2022-05-05 | 2022-08-09 | 中国长江三峡集团有限公司 | Intelligent inspection system and method for wind power generation equipment |
| CN115661970B (en) * | 2022-12-26 | 2023-05-12 | 海外远景(北京)科技有限公司 | Wind power equipment inspection system based on image recognition technology |
| CN117115936A (en) * | 2023-09-07 | 2023-11-24 | 中国三峡新能源(集团)股份有限公司河南分公司 | Unmanned aerial vehicle-based wind power inspection method and device |
| CN117370838B (en) * | 2023-10-17 | 2024-04-05 | 国网电力空间技术有限公司 | Unmanned helicopter power inspection data processing method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105222760A (en) * | 2015-10-22 | 2016-01-06 | 一飞智控(天津)科技有限公司 | The autonomous obstacle detection system of a kind of unmanned plane based on binocular vision and method |
| CN109767141A (en) * | 2019-03-07 | 2019-05-17 | 辽宁北斗卫星导航平台有限公司 | A kind of monitoring method of patrolling railway, device, medium and equipment |
| CN111751002A (en) * | 2020-05-15 | 2020-10-09 | 国网浙江省电力有限公司嘉兴供电公司 | Intelligent charged equipment fault diagnosis method based on infrared thermal imaging |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103612756B (en) * | 2013-12-12 | 2015-10-28 | 中国科学院自动化研究所 | The multi-rotor aerocraft of electric power line inspection and based on its system |
| CN109376587A (en) * | 2018-09-05 | 2019-02-22 | 福州日兆信息科技有限公司 | Communication iron tower intelligent inspection system and method are surveyed in detection based on Internet of Things |
| CN109839954A (en) * | 2019-02-22 | 2019-06-04 | 国家电网有限公司 | A kind of multi-rotor unmanned aerial vehicle intelligent inspection system |
| CN110703800A (en) * | 2019-10-29 | 2020-01-17 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle-based intelligent identification method and system for electric power facilities |
-
2020
- 2020-12-22 CN CN202011525960.1A patent/CN112734970B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105222760A (en) * | 2015-10-22 | 2016-01-06 | 一飞智控(天津)科技有限公司 | The autonomous obstacle detection system of a kind of unmanned plane based on binocular vision and method |
| CN109767141A (en) * | 2019-03-07 | 2019-05-17 | 辽宁北斗卫星导航平台有限公司 | A kind of monitoring method of patrolling railway, device, medium and equipment |
| CN111751002A (en) * | 2020-05-15 | 2020-10-09 | 国网浙江省电力有限公司嘉兴供电公司 | Intelligent charged equipment fault diagnosis method based on infrared thermal imaging |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734970A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112734970B (en) | Automatic inspection system and method for unmanned aerial vehicle in wind farm based on LoRaWAN positioning technology | |
| CN112633535B (en) | Intelligent inspection method and system for photovoltaic power station based on unmanned aerial vehicle image | |
| He et al. | Research of multi-rotor UAVs detailed autonomous inspection technology of transmission lines based on route planning | |
| CN110033453A (en) | Based on the power transmission and transformation line insulator Aerial Images fault detection method for improving YOLOv3 | |
| CN107357313A (en) | Power transmission line fault maintenance system and method based on unmanned aerial vehicle inspection image | |
| CN113759960A (en) | Unmanned aerial vehicle-based fan blade and tower barrel inspection identification system and method | |
| CN114898232B (en) | Photovoltaic power station unmanned aerial vehicle inspection method and system based on photovoltaic group string data analysis | |
| CN111044044B (en) | Electric unmanned aerial vehicle routing inspection route planning method and device | |
| JP2018021491A (en) | System and flight route generating method | |
| CN112329977A (en) | Wind power prediction system for extreme scene | |
| CN108229587A (en) | A kind of autonomous scan method of transmission tower based on aircraft floating state | |
| CN110910440B (en) | Power transmission line length determination method and system based on power image data | |
| CN103984355A (en) | Routing inspection flying robot and overhead power line distance prediction and maintaining method | |
| CN112598637A (en) | Automatic flight method for routing inspection of blades of wind turbine generator in blade area | |
| CN207424682U (en) | A kind of centralized monitoring system for unmanned plane polling transmission line | |
| CN113296537A (en) | Electric power unmanned aerial vehicle inspection method and system based on electric power tower model matching | |
| CN113589837A (en) | Electric power real-time inspection method based on edge cloud | |
| CN115326075A (en) | Path planning method for realizing wind field global automatic inspection based on unmanned aerial vehicle | |
| CN111561943A (en) | Robot inspection method and system | |
| CN114548228A (en) | Unmanned aerial vehicle power grid inspection method and system based on SOC and MESH networking | |
| CN116866512A (en) | Photovoltaic power station inspection system and operation method thereof | |
| CN114610070A (en) | Unmanned aerial vehicle-cooperated wind power plant intelligent inspection method | |
| KR102377859B1 (en) | Photovoltaic panel management system. | |
| CN212515006U (en) | Accurate positioner of unusual target suitable for unmanned aerial vehicle photovoltaic patrols and examines scene | |
| CN116930112A (en) | Sky-ground integrated real-time monitoring system and method for carbon dioxide in park |
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 |