Universal self-adaptive offshore wind power equipment corrosion unmanned monitoring system and detection method

Abstract

The invention discloses a global self-adaptive offshore wind power equipment corrosion unmanned monitoring system and detection method, the system comprises an unmanned aerial vehicle main body, detection equipment and a telescopic mechanical arm for mounting and adjusting the detection equipment, the mechanical arm is mounted below the belly of the unmanned aerial vehicle main body, and the detection equipment is fixed at the movable end of the top of the mechanical arm. The detection method comprises the following steps: selecting a to-be-detected fan; a fan is searched, the coordinates of the fan are verified, cloud analysis and data arrangement are carried out through a central control room, and a visual model is generated to mark defect parts, mark positions and names; and recording data, analyzing and backing up. According to the method, the unmanned platform is used for detection, detection of the wind power facilities is completed in a short time, route planning is performed in advance through the unmanned platform, multiple facilities and even the whole wind power plant can be detected at a time, therefore, the purpose of expanding the coverage range is achieved, and the unmanned platform and the sensor are used for detection, so that the detection cost is reduced.

Description

Technical Field

[0001] The present invention relates to an equipment monitoring system and method, and in particular to a global adaptive unmanned monitoring system and detection method for corrosion of offshore wind power equipment. Background Art

[0002] As a clean energy generation method, offshore wind power has received extensive attention and investment. The number and capacity of offshore wind power facilities are growing rapidly worldwide. Data as of 2023 show that the capacity of offshore wind power facilities has reached 31.89 million kilowatts. However, the problem of corrosion is a huge challenge currently faced by offshore wind power facilities, and it has caused many accidents. At present, the corrosion detection of offshore wind power facilities still uses manual climbing, and ultrasonic technology, infrared technology, etc. are used to measure the coating thickness and rust condition for detection. For example, the Chinese invention patent with authorization number ZL200610124756.2 discloses a "local corrosion analysis scheme based on electrochemical noise nuclear device", which uses electrochemical technology to monitor the aging degree and cathodic protection efficiency of steel pipe pile coating in real time; at the same time, membrane resistance probe technology is used to monitor the atmospheric corrosion rate of electronic and electrical systems, as well as the amount of salt spray deposition in the wind power cabin, combined with remote network and cloud service technology, to achieve long-term online monitoring of electrical equipment corrosion rate, sacrificial anode protection efficiency and coating aging status. However, this detection method requires manual operation, which is more dangerous, and the cost and operation time are relatively long. For example, the Notus nondestructive testing system provided by Hongke uses terahertz technology, but it still cannot avoid the risks of manual operation of wind power equipment. Summary of the invention

[0003] Purpose of the invention: In view of the problems of existing manual detection technology, such as high risk, long time consumption, small coverage, high cost, etc., the present invention provides a global adaptive unmanned monitoring system and detection method for offshore wind power equipment corrosion. The system can accurately identify the surface corrosion and coating integrity of offshore wind turbines under marine conditions, and conduct a series of analysis and planning based on the detection results.

[0004] Technical solution: The global adaptive unmanned monitoring system for offshore wind power equipment corrosion includes a drone body, detection equipment, and a retractable mechanical arm for mounting and adjusting the detection equipment. The mechanical arm is mounted under the belly of the drone body, and the detection equipment is fixed at the top movable end of the mechanical arm.

[0005] Furthermore, the drone body includes drone wings, a motor, a carbon fiber frame, a drone control room, a fixed landing gear device and a fixed landing gear. The drone wings and the motor are assembled and fixed on the carbon fiber frame. The carbon fiber frame is installed around the drone control room. Landing gear is arranged on both sides of the drone control room. The landing gear includes a fixed landing gear device and a fixed landing gear.

[0006] Furthermore, a camera is provided in the control room of the drone.

[0007] Furthermore, the mechanical arm includes an upper rack and a lower rack corresponding to the upper rack, the upper rack and the lower rack are driven by gears, and the movable end of the upper rack is fixed to the detection equipment.

[0008] Furthermore, the gear is connected to the motor, the motor is connected to the single-chip microcomputer, the single-chip microcomputer is connected to the battery, and the single-chip microcomputer is placed on the single-chip microcomputer bracket.

[0009] Furthermore, shells are arranged on both sides of the upper rack and the lower rack, and the shells are connected by a connecting frame. A motor is embedded in one side of the shell, and a battery is hung on the other side.

[0010] Furthermore, the battery is placed in a groove formed by a battery lower bracket and a battery side bracket, and the battery lower bracket and the battery side bracket are fixed to one side of the shell through studs.

[0011] Furthermore, the detection equipment is a thickness gauge for measuring the thickness of the coating.

[0012] Furthermore, the mechanical arm is mounted under the belly of the drone body through a connecting frame.

[0013] The detection method of the global adaptive unmanned offshore wind power equipment corrosion monitoring system comprises the following steps:

[0014] S1. Select the wind turbine to be inspected through the remote sensing map;

[0015] S2, find the fan and verify the coordinates of the fan. If it is the target fan, take photos and measure the thickness through the camera of the monitoring system to collect data and proceed to the next step; if not, find the target fan again;

[0016] S3, through the central control room (14), the cloud is used to analyze and organize the data, and a visual model is generated to mark the defective parts, and the locations and names are marked;

[0017] S4. Record data and analyze the backup.

[0018] The technical solution adopted by the present invention is to use an unmanned platform to fly around and collect data on a single offshore wind turbine and the entire offshore wind farm through a pre-set path planning algorithm, use a camera to visually detect corrosion on surface defects, and use an ATO coating thickness measuring device to detect the integrity of the surface coating. The detected data is uploaded to the cloud, and through a series of processing and visual analysis, a three-dimensional wind turbine model with defect identification is generated, which is convenient for intuitively seeing the location of the problem. And through subsequent analysis and processing procedures, long-term detection and maintenance planning are carried out for the wind farm.

[0019] The system mainly includes the main parts of the drone body, detection equipment and the robotic arm for mounting. It can be extended while keeping the center of gravity stable, so that the detection equipment is as close to the surface to be inspected as possible, while also leaving a certain space for the unmanned platform, maximizing the accuracy of detection while ensuring safety.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. By using unmanned platform detection, the detection of wind power facilities can be completed in a shorter time. It is expected that the detection of a single facility can be completed within 20 minutes, which greatly shortens the time required for detection compared to 4-6 hours per unit for manual detection. Thus, the purpose of shortening the time is achieved.

[0022] 2. Replace manual inspection methods with unmanned platform intelligent inspection methods, so that engineers no longer need to perform high-altitude operations such as climbing to inspect facilities, thereby reducing the danger of inspection.

[0023] 3. Pre-route planning is carried out through unmanned platforms. The current manual inspection method can only inspect a single facility at a time. Through autonomous planning flights over multiple wind power facilities in a large area, multiple facilities or even the entire wind farm can be inspected at one time, thereby achieving the purpose of expanding coverage.

[0024] 4. Use unmanned platforms and sensors to reduce the cost of testing. The cost of a single unit of manual testing can reach 12,500 yuan. The cost of testing by the method of the present invention can be reduced to about 40 yuan per unit. At the same time, the facility can be tested with very short downtime, thereby effectively reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of the detection scheme of the present invention;

[0026] Figure 2 Plan routes for unmanned systems to operate;

[0027] Figure 3 This is the hardware block diagram of the detection system;

[0028] Figure 4 This is an overall picture of the UAV equipment equipped with a robotic arm;

[0029] Figure 5 This is the front view of the unmanned detection system;

[0030] Figure 6 This is the overall picture of the robotic arm used to carry the testing equipment;

[0031] Figure 7This is a side elevation view of the robotic arm used to carry the testing equipment;

[0032] Figure 8 This is the AA section of the robotic arm used to carry the testing equipment;

[0033] Fig. 9 This is the cross-sectional CC diagram of the robotic arm used to carry the testing equipment;

[0034] Fig.10 This is the flow chart of the path planning principle;

[0035] Fig.11 It is a principle diagram of the detection process of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be further described below.

[0037] like Figure 1-6 As shown, the unmanned platform for global adaptive corrosion monitoring of offshore wind power equipment in this embodiment includes a drone body 1 carrying the platform, a detection device 3 and a retractable mechanical arm 2 for mounting and adjusting the detection device. The mechanical arm is mounted under the belly of the drone body, and the detection device 3 is fixed at the top movable end of the mechanical arm. The drone body 1 includes a drone wing 11, a motor 12, a carbon fiber frame 13, a drone control room 14, a fixed landing gear device 15 and a fixed landing gear 16. The drone wing 11 and the motor 12 are assembled and fixed on the carbon fiber frame 13. The carbon fiber frame 13 is installed around the drone control room 14. Landing devices are set on both sides of the drone control room 14, and the landing devices include a fixed landing gear device 15 and a fixed landing gear 16. The drone control room 14 is provided with a camera 17. The mechanical arm 2 includes an upper rack 21 and a lower rack 22 corresponding to the upper rack 21. The upper rack 21 and the lower rack 22 are connected and transmitted through a gear 23, and the detection device 3 is fixed at the movable end of the upper rack 21. The gear 23 is connected to the motor 24, the motor 24 is connected to the single chip microcomputer, and the single chip microcomputer is connected to the battery. A housing 25 is provided on both sides of the upper rack 21 and the lower rack 22, and the housing 25 is connected by a metal connecting frame 26. The motor 24 is embedded on one side of the housing 25, and the battery is hung on the other side. The battery is placed in a groove formed by a battery lower bracket 27 and a battery side bracket 28, and the battery lower bracket 27 and the battery side bracket 28 are fixed to one side of the housing 25 by studs. The detection device 3 is a thickness gauge for measuring the thickness of the coating.

[0038] This device mainly consists of three parts, namely, the UAV platform as a carrying platform, the mechanical arm 2 for adjusting the position of the detection equipment 3, and the detector for measuring the coating thickness. In terms of structure, the mechanical arm 2 is mounted under the belly of the UAV body 1 through the connecting frame 26, and is fixed with six M3 screws. The top movable end of the rack on the mechanical arm 2 is connected to the coating thickness gauge by clamping and locking. The mechanical arm 2 drives the gear 23 to rotate through the single-chip microcomputer controlled motor 24 loaded in the single-chip microcomputer bracket 29, and can achieve two mechanical forms, namely the storage form and the extension form. The storage form is used on the ground and during transportation. The gear 23 drives the upper rack 21 and the lower rack 22 of the mechanical arm 2 to retract to the center, and the main body and the clamped coating thickness gauge are received together under the UAV body, which reduces the wind resistance of the UAV during flight to a certain extent and can ensure that the equipment does not touch the ground. The extended form is used during flight inspection. The robotic arm 2 is extended forward as a whole, and the upper rack 21 equipped with a coating thickness gauge is extended forward and moved to the lower front position of the drone. At the same time, the lower rack 22 will also move backward accordingly to ensure the stability of the center of gravity, which greatly increases the safety and stability of the drone during inspection.

[0039] In terms of structure, the robot arm 2 is mounted under the belly of the drone and fixed with six screws. The movable end of the robot arm 2 is connected to the coating thickness gauge by clamping and locking. The robot arm 2 has two forms, namely the storage form and the extended form. The storage form is used on the ground and during transportation. The entire robot arm is retracted backwards, and the main body and the clamped thickness gauge are received together under the drone body 1, and it can ensure that the equipment does not touch the ground. The extended form is used during flight detection. The robot arm 2 is extended forward as a whole, and the detection equipment is moved to the lower front position of the drone to provide a close enough distance for approaching the surface of the facility for coating thickness measurement. In terms of circuit, the main body is the onboard computer carried by the drone and the single-chip microcomputer that controls the robot arm 2, which are stored in the central control shell. The drone flight control, camera, PID and other parts are connected to the onboard computer. The coating thickness gauge, rangefinder, and robot arm motor are connected to the single-chip microcomputer control, and then connected to the onboard computer through the single-chip microcomputer to realize data aggregation. The onboard computer is connected to the cloud and uploads data and images to the cloud for further analysis and processing.

[0040] The detection system uses MATLAB to perform path planning. After determining the geographic location, shape, size and other information of the detection target, the product will call the improved particle swarm optimization algorithm, input the information as parameters into the algorithm, and make the global optimal path planning.

[0041] The invention uses a variety of detection technologies, including machine vision, ATO photothermal method, etc., which can more accurately detect the entire wind power facility and can also perform more targeted detection in different environments. In terms of technical detection, the drone first searches for wind turbines in the environment according to the map, then matches the drone with the wind turbine model database, and then uses a variety of detectors and cameras on the drone to collect data according to the optimal path. During the detection process, Ultralytics is first used for target detection. During this process, a small target perception layer will be introduced to reduce the error caused by changes in target scale, and then an attention mechanism will be introduced to improve detection efficiency by fusing the two types of attention.

[0042] The detection process is to use a drone to take photos of a wind turbine to be maintained at different positions and angles, collect a large number of wind turbine rust pictures, and finally obtain a wind turbine rust data set through screening and sorting. Then use YOLOv8 to train three models, namely the test set during training, the test set for evaluation model results, and the test set for evaluation model results. After the single data set is prepared, run train.py in YOLOv8 to start GPU training. By adjusting the number of training pictures (epochs), the number of convolutions for each training, and the number of iterations of training data, the weight file is finally obtained. After training, the model is tuned according to the recognition effect, and the parameters are fine-tuned, trained, and tested repeatedly. Finally, the training data with the most ideal PR curve and the model hyperparameters (HyperPara) that meet the requirements are obtained and applied in this product. After the model is trained, the optimal weight file is obtained and the content of the filming is detected in real time with the help of its content. Among them, the rusted part on the wind turbine will be detected and uploaded to the cloud for subsequent processing and evaluation, and a subsequent processing plan will be given.

[0043] The above is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any technician in the relevant technical field, without departing from the scope of the technical solution of the present invention, makes any form of equivalent replacement or modification to the technical solution and technical content disclosed in the present invention, which does not depart from the content of the technical solution of the present invention and still falls within the protection scope of the present invention.

Claims

1. A global adaptive unmanned monitoring system for corrosion of offshore wind power equipment, characterized by: The invention comprises an unmanned aerial vehicle body (1), a detection device (3) and a retractable mechanical arm (2) for mounting and adjusting the detection device, wherein the mechanical arm (2) is mounted below the belly of the unmanned aerial vehicle body (1), and the detection device (3) is fixed at the top movable end of the mechanical arm (2).

2. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 1 is characterized by: The drone body (1) comprises a drone wing (11), a motor (12), a carbon fiber frame (13), a drone control room (14), a fixed landing gear device (15) and a fixed landing gear (16); the drone wing (11) and the motor (12) are assembled and fixed on the carbon fiber frame (13); the carbon fiber frame (13) is installed around the drone control room (14); landing gears are arranged on both sides of the drone control room (14); the landing gears comprise a fixed landing gear device (15) and a fixed landing gear (16); and a camera (17) is arranged in the drone control room (14).

3. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 1 is characterized by: The mechanical arm (2) comprises an upper rack (21) and a lower rack (22) corresponding to the upper rack (21); the upper rack (21) and the lower rack (22) are driven by a gear (23); and the movable end of the upper rack (21) is fixed with a detection device (3).

4. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 3 is characterized by: The gear (23) is connected to the motor (24), the motor (24) is connected to the single-chip microcomputer, the single-chip microcomputer is connected to the battery, and the single-chip microcomputer is placed on a single-chip microcomputer bracket (29).

5. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 4 is characterized by: A housing (25) is arranged on both sides of the upper rack (21) and the lower rack (22); the housing (25) is connected via a connecting frame (26); a motor (24) is embedded on one side of the housing (25) and a battery is hung on the other side.

6. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 5 is characterized by: The battery is placed in a groove formed by a battery lower bracket (27) and a battery side bracket (28), and the battery lower bracket (27) and the battery side bracket (28) are fixed to one side of the housing (25) by means of studs.

7. The global adaptive unmanned offshore wind power equipment corrosion monitoring system according to claim 1 is characterized by: The mechanical arm (2) is mounted below the belly of the drone body (1) via a connecting frame (26).

8. The detection method of the global adaptive unmanned offshore wind power equipment corrosion monitoring system according to any one of claims 1 to 7, characterized in that: The steps include: S1. Select the wind turbine to be inspected through the remote sensing map; S2, find the fan and verify the coordinates of the fan. If it is the target fan, take photos and measure thickness through the camera of the monitoring system to collect data and proceed to the next step; if not, find the target fan again; S3, through the central control room (14), the cloud is used to analyze and organize the data, and a visual model is generated to mark the defective parts, and the locations and names are marked; S4. Record data and analyze the backup.

9. An intelligent algorithm for path planning of a global adaptive unmanned offshore wind power equipment corrosion monitoring system according to any one of claims 1 to 7, characterized in that: The steps include: (1) Initialize the system and set the initial values ​​of control variables and other parameters; (2) Randomly initialize the position and velocity of each particle in the particle swarm; (3) Fitness calculation to obtain individual optimal solutions and global optimal solutions; (4) Determine whether the maximum number of iterations has been reached. If yes, stop the iteration, save the global optimal value, and end; if no, iterate and update the current population to generate a new particle population; (5) Fitness calculation to obtain individual optimal solutions and global optimal solutions.

Images