A fan blade defect recognition system and method

Abstract

The application relates to the technical field of wind power generation, and discloses a wind turbine blade defect identification system and method, which comprises a UAV platform end and a remote management terminal, and the UAV platform end and the remote management terminal are wirelessly connected; the remote management terminal is used for acquiring UAV trajectory planning data and transmitting the UAV trajectory planning data to the UAV platform end; the UAV platform end is used for controlling the UAV to fly to a target scanning area based on the UAV trajectory planning data, scanning the wind turbine blade of the target scanning area, obtaining an infrared splicing image, and performing defect identification on the infrared splicing image to obtain a wind turbine blade defect identification result. The application realizes intelligent inspection of the wind turbine blade, effectively improves the inspection operation efficiency of the wind turbine blade, and saves operation and maintenance costs.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, specifically to a wind turbine blade defect identification system and method. Background Technology

[0002] Blades are key components of wind turbines, facing harsh environments such as heavy loads, high torque, and salt spray corrosion, which can easily lead to defects such as cracks, corrosion, and breakage. If these defects are not detected and repaired in time, the expansion of cracks and corrosion holes will cause the blades to break, resulting in huge economic losses and safety problems for the wind farm.

[0003] The inspection of wind turbine blades is usually carried out by inspection personnel working at heights, which poses certain personal safety risks and is also inefficient. Summary of the Invention

[0004] In view of this, the present invention provides a wind turbine blade defect identification system and method to solve the problems of certain personal safety hazards and low inspection efficiency in wind turbine blade inspection.

[0005] In a first aspect, the present invention provides a wind turbine blade defect identification system, the system comprising: a drone platform and a remote management terminal, wherein the drone platform and the remote management terminal are wirelessly connected.

[0006] The remote management terminal is used to acquire drone trajectory planning data and transmit the drone trajectory planning data to the drone platform.

[0007] On the UAV platform side, the system controls the UAV to fly to the target scanning area based on UAV trajectory planning data, scans the wind turbine blades in the target scanning area to obtain an infrared stitched image, and performs defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result.

[0008] This embodiment provides a wind turbine blade defect identification system. By scanning the wind turbine blades in the target scanning area to obtain an infrared stitched image, and performing defect identification on the infrared stitched image, the system obtains the wind turbine blade defect identification result. Through information interaction between the UAV platform and the remote management terminal, the system avoids the personal safety hazards and low inspection efficiency caused by inspection personnel working at heights during the inspection process. It realizes intelligent inspection of wind turbine blades, effectively improves the inspection and operation efficiency of wind turbine blades, and saves operation and maintenance costs.

[0009] In one optional implementation, the UAV platform includes: a first wireless data transmission module, a control module, and an optoelectronic device, wherein the first wireless data transmission module is wirelessly connected to a remote management terminal.

[0010] The first wireless data transmission module is used to receive UAV trajectory planning data sent by the remote management terminal and transmit the UAV trajectory planning data to the control module.

[0011] The control module is used to control the UAV to fly to the target scanning area based on UAV trajectory planning data; wherein the target scanning area is an area containing at least two wind turbine blades;

[0012] Optoelectronic equipment is used to scan the wind turbine blades in the target scanning area to obtain an infrared stitched image, and to identify defects in the infrared stitched image to obtain the wind turbine blade defect identification result.

[0013] This embodiment provides a wind turbine blade defect identification system. A control module controls a drone to fly to a target scanning area, which is set to include at least two wind turbine blades, maximizing the inspection area and significantly improving inspection efficiency. Furthermore, rapid scanning using photoelectric equipment inspects the wind turbine blades, avoiding the personal safety risks associated with traditional wind turbine blade inspection methods. The system also utilizes a hovering drone to take photos, further improving inspection efficiency and significantly reducing the maintenance costs of the wind turbine blades.

[0014] In one alternative embodiment, the optoelectronic device includes:

[0015] The stitching unit is used to scan the wind turbine blades in the target scanning area to obtain multiple blade scanning images, and then stitch the multiple blade scanning images together to obtain an infrared stitched image;

[0016] The identification unit is used to identify and classify blade defects in infrared stitched images to obtain the wind turbine blade defects and the location of defect pixels.

[0017] The determining unit is used to acquire the infrared field of view and the azimuth position of the wind turbine blades, and to determine the position of the wind turbine blades based on the azimuth position and the infrared field of view.

[0018] The calculation unit is used to calculate the location of the wind turbine blade defect based on the position of the wind turbine blade and the location of the defect pixel, and to determine the wind turbine blade defect identification result based on the wind turbine blade defect and the location of the wind turbine blade defect.

[0019] This embodiment provides a wind turbine blade defect recognition system that stitches together multiple scanned images of the blades to obtain an infrared stitched image. The system then identifies and classifies blade defects within the infrared stitched image, enabling rapid extraction and labeling of these defects. Furthermore, by calculating the location of the wind turbine blade defects based on the blade position and the defect pixel position, the system achieves rapid localization of wind turbine blade defects, thus improving the accuracy of defect recognition.

[0020] In one alternative implementation, the stitching unit is specifically used to arrange multiple blade scan images based on the azimuth position of the wind turbine blades to obtain an infrared stitched image.

[0021] This embodiment provides a wind turbine blade defect identification system that arranges multiple blade scan images based on the azimuth angle position of the wind turbine blades to obtain an infrared stitched image. The system can stitch together wind turbine blades within the target scanning area in a short time to form an infrared stitched image, thus improving image stitching efficiency and accuracy.

[0022] In one optional implementation, the identification unit is specifically used to identify and classify blade defects in the infrared stitched image using an image detection model, thereby obtaining the wind turbine blade defects and the location of defect pixels.

[0023] This embodiment provides a wind turbine blade defect identification system that uses an image detection model to extract and label wind turbine blade defects from infrared stitched images, thereby improving the accuracy of wind turbine blade defect identification and increasing inspection efficiency.

[0024] In one alternative embodiment, the optoelectronic device further includes:

[0025] The overlay unit is used to overlay the wind turbine blade defects and their locations onto the infrared stitched image, and then transmit the overlaid infrared stitched image to the remote management terminal via the first wireless data transmission module.

[0026] This embodiment provides a wind turbine blade defect identification system. By superimposing the wind turbine blade defects and their locations onto an infrared stitched image, users can quickly and accurately identify the wind turbine blade defects and their locations, thereby enabling maintenance of the wind turbine blades and ensuring their normal operation.

[0027] In one optional implementation, the remote management terminal includes: a data management terminal and a second wireless data transmission module, wherein the data management terminal is connected to the second wireless data transmission module and the second wireless data transmission module is connected to the first wireless data transmission module.

[0028] The data management terminal is used to acquire UAV trajectory planning data and transmit the UAV trajectory planning data to the first wireless data transmission module.

[0029] The second wireless data transmission module is used to transmit the UAV trajectory planning data to the UAV platform, and to receive an infrared stitched image with the wind turbine blade defects and their locations superimposed on it. The infrared stitched image with the wind turbine blade defects and their locations superimposed on it is then sent to the data management terminal for display.

[0030] This embodiment provides a wind turbine blade defect identification system. Through information interaction between a data management terminal and a second wireless data transmission module, it enables the transmission of UAV trajectory planning data and the display of wind turbine blade defects and their locations in infrared stitched images, laying the foundation for rapid maintenance of wind turbine blades.

[0031] In one alternative implementation, the drone platform also includes:

[0032] The inertial navigation module is used to collect the flight attitude information of the UAV; the flight attitude information of the UAV is used to control the flight of the UAV.

[0033] This embodiment provides a wind turbine blade defect identification system that achieves precise control of the UAV's flight attitude by using the UAV's flight attitude information, laying the foundation for the UAV to accurately scan the wind turbine blades in the target scanning area.

[0034] In one alternative implementation, the drone platform also includes:

[0035] The BeiDou module is used to collect the real-time location of the UAV and transmit it to the remote management terminal.

[0036] This embodiment provides a wind turbine blade defect identification system that uses a Beidou module to achieve precise real-time positioning of a drone. The remote management terminal can control the inspection direction and hovering position based on the drone's real-time location, thereby improving inspection efficiency.

[0037] In a second aspect, the present invention provides a method for identifying wind turbine blade defects, applied to the wind turbine blade defect identification system of the first aspect or any corresponding embodiment thereof, the method comprising:

[0038] The remote management terminal acquires drone trajectory planning data and transmits the drone trajectory planning data to the drone platform.

[0039] The UAV platform controls the UAV to fly to the target scanning area based on the UAV trajectory planning data, scans the wind turbine blades in the target scanning area, obtains an infrared stitched image, and performs defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result. Attached Figure Description

[0040] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a structural block diagram of a wind turbine blade defect identification system according to an embodiment of the present invention;

[0042] Figure 2 This is a structural block diagram of the unmanned aerial vehicle platform according to an embodiment of the present invention;

[0043] Figure 3 This is a structural block diagram of an optoelectronic device according to an embodiment of the present invention;

[0044] Figure 4 This is a schematic diagram of an infrared stitched image of a wind turbine blade defect and its location, superimposed according to an embodiment of the present invention.

[0045] Figure 5 This is a structural block diagram of a remote management terminal according to an embodiment of the present invention;

[0046] Figure 6 This is a magnified schematic diagram of an infrared stitched image according to an embodiment of the present invention;

[0047] Figure 7 This is a flowchart illustrating a method for identifying defects in wind turbine blades according to an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] This invention provides a wind turbine blade defect identification system. It uses the infrared channel of an optoelectronic device to quickly scan wind turbine blades within a certain area, forming a clear spliced ​​strip image. AI (Artificial Intelligence) technology is then used to identify and extract defects such as cracks and gaps from the spliced ​​strip image. This effectively reduces the potential safety hazards of wind turbines caused by the expansion of defects in the blades, avoids personal safety accidents during high-altitude operations, saves on the cost of wind turbine blade inspection, and improves the economic benefits for enterprises.

[0050] This embodiment provides a wind turbine blade defect identification system. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0051] This embodiment provides a wind turbine blade defect identification system, such as Figure 1 As shown, it includes: a drone platform 101 and a remote management terminal 102, which are wirelessly connected.

[0052] The remote management terminal 102 is used to acquire UAV trajectory planning data and transmit the UAV trajectory planning data to the UAV platform terminal 101.

[0053] Specifically, the remote management terminal 102 sends the drone trajectory planning data to the drone platform terminal 101 according to the wind turbine blade inspection task; after completing the wind turbine blade inspection task, the remote management terminal 102 sends the recovery command to the drone platform terminal 101, and the drone platform terminal 101 controls the drone to return to the drone and complete the inspection.

[0054] The UAV platform 101 is used to control the UAV to fly to the target scanning area based on the UAV trajectory planning data, scan the wind turbine blades in the target scanning area, obtain an infrared stitched image, and perform defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result.

[0055] This embodiment provides a wind turbine blade defect identification system. By scanning the wind turbine blades in the target scanning area to obtain an infrared stitched image, and performing defect identification on the infrared stitched image, the system obtains the wind turbine blade defect identification result. Through information interaction between the UAV platform and the remote management terminal, the system avoids the personal safety hazards and low inspection efficiency caused by inspection personnel working at heights during the inspection process. It realizes intelligent inspection of wind turbine blades, effectively improves the inspection and operation efficiency of wind turbine blades, and saves operation and maintenance costs.

[0056] In some alternative implementations, such as Figure 2 As shown, the UAV platform 101 includes: a first wireless data transmission module 1011, a control module 1012, and an optoelectronic device 1013. The first wireless data transmission module 1011 is wirelessly connected to the remote management terminal 102.

[0057] The first wireless data transmission module 1011 is used to receive UAV trajectory planning data sent by the remote management terminal 102 and transmit the UAV trajectory planning data to the control module 1012.

[0058] Specifically, the first wireless data transmission module 1011 of the UAV platform 101 interacts wirelessly with the remote management terminal 102, mainly transmitting the real-time positioning information of the UAV and the infrared stitched image of the optoelectronic device 1013.

[0059] The control module 1012 is used to control the UAV to fly to the target scanning area based on the UAV trajectory planning data; wherein the target scanning area is an area containing at least two wind turbine blades.

[0060] Specifically, the control module 1012 mainly receives UAV trajectory planning data transmitted by the remote management terminal 102, and then controls the flight trajectory of the UAV.

[0061] Furthermore, the control module 1012 controls the UAV to fly to the planned first blade area (generally, there are no less than two wind turbine blades around the target scanning area) and hover. At this time, the remote management terminal 102 sends a start scanning command to the optoelectronic device 1013.

[0062] The optoelectronic device 1013 is used to scan the wind turbine blades in the target scanning area to obtain an infrared stitched image, and to identify defects in the infrared stitched image to obtain the wind turbine blade defect identification result.

[0063] Specifically, the infrared channel of the optoelectronic device 1013 is responsible for rapidly scanning the wind turbine blades in the target scanning area to form a clear infrared stitched image, and using AI algorithms to identify and mark the wind turbine blade defects in the infrared stitched image.

[0064] Furthermore, after receiving the start scanning command sent by the remote management terminal 102, the infrared channel of the optoelectronic device 1013 starts to drive the azimuth motor to perform a rapid scan of 0.5 seconds / revolution. The infrared channel of the optoelectronic device 1013 can trigger high-resolution infrared 4K imaging of multiple blades (such as 4) in the target scanning area within 0.5 seconds, thereby enabling the simultaneous completion of 360° panoramic stitching images of multiple blades.

[0065] This embodiment provides a wind turbine blade defect identification system. A control module controls a drone to fly to a target scanning area, which is set to include at least two wind turbine blades, maximizing the inspection area and significantly improving inspection efficiency. Furthermore, rapid scanning using photoelectric equipment inspects the wind turbine blades, avoiding the personal safety risks associated with traditional wind turbine blade inspection methods. The system also utilizes a hovering drone to take photos, further improving inspection efficiency and significantly reducing the maintenance costs of the wind turbine blades.

[0066] In some alternative implementations, such as Figure 3 As shown, the optoelectronic device 1013 includes:

[0067] The stitching unit 10131 is used to scan the wind turbine blades in the target scanning area to obtain multiple blade scanning images, and stitch the multiple blade scanning images together to obtain an infrared stitched image.

[0068] The identification unit 10132 is used to identify and classify blade defects in infrared stitched images to obtain the wind turbine blade defects and the location of defect pixels.

[0069] Specifically, while performing image stitching, AI is used to identify, extract, and label typical blade defects in the blade scan image to obtain the wind turbine blade defects and the location of defect pixels.

[0070] Furthermore, AI can be used to identify, extract, and label typical blade defects in the infrared stitched image output by the stitching unit 10131 to obtain the wind turbine blade defects and the location of defect pixels.

[0071] The determining unit 10133 is used to obtain the infrared field of view and the azimuth position of the wind turbine blade, and determine the position of the wind turbine blade based on the azimuth position and the infrared field of view.

[0072] Specifically, based on the azimuth angle of the wind turbine blades, the specific position (i.e., the position of the wind turbine blades) n of the wind turbine blades in the infrared mosaic image is calculated using the following formula:

[0073] n = azimuth position % (infrared field of view ÷ 1.2) + 1

[0074] The calculation unit 10134 is used to calculate the location of the wind turbine blade defect based on the position of the wind turbine blade and the location of the defect pixel, and to determine the wind turbine blade defect identification result based on the wind turbine blade defect and the location of the wind turbine blade defect.

[0075] Specifically, the defect pixels include azimuth pixels and pitch pixels. The location of the defect on the wind turbine blade is composed of the azimuth angle A and the pitch angle E. The formula for calculating the azimuth angle A is as follows:

[0076] A = Azimuth pixels / (1280 * 4) * 15° + 15° * (n - 1)

[0077] Furthermore, the formula for calculating the pitch angle E is as follows:

[0078] E = Pitch pixels / (1024 * 4) * 12°

[0079] This embodiment provides a wind turbine blade defect recognition system that stitches together multiple scanned images of the blades to obtain an infrared stitched image. The system then identifies and classifies blade defects within the infrared stitched image, enabling rapid extraction and labeling of these defects. Furthermore, by calculating the location of the wind turbine blade defects based on the blade position and the defect pixel position, the system achieves rapid localization of wind turbine blade defects, thus improving the accuracy of defect recognition.

[0080] In some optional implementations, the stitching unit 10131 is specifically used to arrange multiple blade scan images based on the azimuth position of the wind turbine blades to obtain an infrared stitched image.

[0081] Specifically, the number of infrared stitched images is calculated based on the field of view (i.e., infrared field of view angle) of the infrared channel. The formula for calculating the number of images N is as follows:

[0082] N = (360 / infrared field of view × 1.2)

[0083] The multiple of 1.2 is to prevent the image from being rendered with a 20% redundancy to account for error factors.

[0084] Furthermore, assuming a field of view of 18° and the number of images N = (360 / 18 × 1.2) = 24, the panoramic 360° stitching strips can be set to 4 rows, with each row consisting of 6 images scaled and stitched together proportionally. These images are then arranged from left to right according to the azimuth position of the wind turbine blades to obtain the infrared stitched image.

[0085] This embodiment provides a wind turbine blade defect identification system that arranges multiple blade scan images based on the azimuth angle position of the wind turbine blades to obtain an infrared stitched image. The system can stitch together wind turbine blades within the target scanning area in a short time to form an infrared stitched image, thus improving image stitching efficiency and accuracy.

[0086] In some optional implementations, the identification unit 10132 is specifically used to identify and classify blade defects in infrared stitched images using an image detection model, thereby obtaining the wind turbine blade defects and the location of defect pixels.

[0087] Specifically, since upgrading and improving the neural network framework can achieve target classification, target box coordinate determination, and target object determination, the neural network framework is used to identify and classify blade defects. The specific steps include: when identifying wind turbine blade defects, firstly, the blade defect dataset is used for training to enhance the generalization ability, and then the training samples in the blade defect dataset are input to iteratively train the pre-trained model to obtain the final trained model, i.e., the image detection model.

[0088] Furthermore, when implementing AI defect detection, the image triggered by scanning (multiple blade scan images or infrared stitched images) is input into the image detection model through the photoelectric device 1013, and the image detection model outputs the wind turbine blade defects and the location of defect pixels.

[0089] This embodiment provides a wind turbine blade defect identification system that uses an image detection model to extract and label wind turbine blade defects from infrared stitched images, thereby improving the accuracy of wind turbine blade defect identification and increasing inspection efficiency.

[0090] In some alternative embodiments, the optoelectronic device 1013 further includes:

[0091] The overlay unit 10135 is used to overlay the wind turbine blade defects and their locations onto the infrared stitched image, and then transmit the overlaid infrared stitched image to the remote management terminal 102 via the first wireless data transmission module 1011.

[0092] Specifically, the location of defects in the wind turbine blades (such as...) Figure 4 The area shown in the box is superimposed on the infrared stitched image. The infrared stitched image is transmitted to the remote management terminal 102 through the first wireless data transmission module 1011. After scanning the target scanning area once, the photoelectric device 1013 stops scanning.

[0093] This embodiment provides a wind turbine blade defect identification system. By superimposing the wind turbine blade defects and their locations onto an infrared stitched image, users can quickly and accurately identify the wind turbine blade defects and their locations, thereby enabling maintenance of the wind turbine blades and ensuring their normal operation.

[0094] In some alternative implementations, such as Figure 5 As shown, the remote management terminal 102 includes: a data management terminal 1021 and a second wireless data transmission module 1022. The data management terminal 1021 is connected to the second wireless data transmission module 1022, and the second wireless data transmission module 1022 is connected to the first wireless data transmission module 1011.

[0095] The data management terminal 1021 is used to acquire UAV trajectory planning data and transmit the UAV trajectory planning data to the first wireless data transmission module 1011.

[0096] Specifically, when the wind turbine blade inspection task begins, the data management terminal 1021 sends the UAV trajectory planning data to the UAV platform terminal 101. The infrared channel of the optoelectronic device 1013 is powered on and cooled in advance. After the infrared channel imaging, the wind turbine blades in the target scanning area are scanned.

[0097] Furthermore, the data management terminal 1021, as a general-purpose server, runs management software and is responsible for issuing UAV trajectory planning plans and receiving stitched images and status information sent by the UAV platform terminal 101.

[0098] The second wireless data transmission module 1022 is used to transmit the UAV trajectory planning data to the UAV platform terminal 101, and to receive the infrared stitched image after superimposing the wind turbine blade defects and the defect positions of the wind turbine blades, and send the infrared stitched image after superimposing the wind turbine blade defects and the defect positions of the wind turbine blades to the data management terminal 1021 for display.

[0099] Specifically, such as Figure 6 As shown, after receiving the infrared stitched image, the data management terminal 1021 can display the infrared stitched image and zoom in as needed to view the defect details of the wind turbine blades in the infrared stitched image. At the same time, the infrared stitched image can be saved.

[0100] Furthermore, after the data management terminal 1021 completes the saving, it sends a command to the UAV platform terminal 101 to fly to the next blade area and hover for scanning. The same method is used to detect subsequent blades in sequence until all are completed.

[0101] This embodiment provides a wind turbine blade defect identification system. Through information interaction between a data management terminal and a second wireless data transmission module, it enables the transmission of UAV trajectory planning data and the display of wind turbine blade defects and their locations in infrared stitched images, laying the foundation for rapid maintenance of wind turbine blades.

[0102] In some alternative implementations, the drone platform 101 further includes:

[0103] The inertial navigation module 1014 is used to collect the flight attitude information of the UAV; the flight attitude information of the UAV is used to control the flight of the UAV.

[0104] Specifically, the inertial navigation module 1014 is responsible for collecting the attitude information of the UAV's roll, pitch, and heading, and then calculating the attitude information of the UAV's roll, pitch, and heading, and using the calculation results to control the flight of the UAV.

[0105] This embodiment provides a wind turbine blade defect identification system that achieves precise control of the UAV's flight attitude by using the UAV's flight attitude information, laying the foundation for the UAV to accurately scan the wind turbine blades in the target scanning area.

[0106] In some alternative implementations, the drone platform 101 further includes:

[0107] The Beidou module 1015 is used to collect the real-time location of the UAV and transmit the real-time location of the UAV to the remote management terminal 102.

[0108] Specifically, the Beidou module 1015 is responsible for providing high-precision real-time latitude, longitude, and altitude information of the UAV, which facilitates the remote management terminal 102 to adjust the position of the UAV.

[0109] This embodiment provides a wind turbine blade defect identification system that uses a Beidou module to achieve precise real-time positioning of a drone. The remote management terminal can control the inspection direction and hovering position based on the drone's real-time location, thereby improving inspection efficiency.

[0110] The following specific embodiment illustrates the workflow of a wind turbine blade defect identification system.

[0111] Example 1:

[0112] The workflow of a wind turbine blade defect identification system includes:

[0113] Step 1: When the wind turbine blade inspection task begins, the data management terminal sends the UAV trajectory planning technology to the UAV platform. The infrared channel of the photoelectric device is powered on and cooled in advance. After the infrared channel imaging, the UAV platform controls the UAV to fly to the first planned blade area (generally, there are no less than 2 wind turbine blades around the area) and hover.

[0114] Step 2: The remote management terminal sends a start scanning command to the photoelectric device. The infrared channel of the photoelectric device starts driving the azimuth motor to perform a rapid scan of 0.5 seconds per revolution. This method can trigger high-resolution infrared 4K imaging of multiple blades (e.g., 4) in the area within 0.5 seconds. At this time, a 360° panoramic stitched image can be completed for multiple blades at the same time. And according to the azimuth angle of the wind turbine blade, the specific position of the blade in the panoramic stitched image is calculated.

[0115] Step 3: Since upgrading and improving the neural network framework can achieve target classification, target bounding box coordinate determination, and target object identification, a neural network framework is used to identify and classify blade defects. The specific steps include: When identifying wind turbine blade defects, firstly, the model is trained using a blade defect dataset to enhance generalization ability. Then, training samples from the blade defect dataset are input to iteratively train the pre-trained model, resulting in the final trained model, i.e., the image detection model. In implementing AI defect detection, images triggered by scanning (multiple blade scan images or infrared stitched images) are input into the image detection model via photoelectric equipment. The image detection model outputs the wind turbine blade defects and the location of the defect pixels.

[0116] Step 4: The photoelectric device combines the high-precision positioning information of the Beidou module with the position of the image pixels where the wind turbine blade defects are located in the model output, and superimposes the specific location information onto the infrared stitched image. The infrared stitched image and the original image (i.e., the blade scan image) are transmitted to the remote management terminal through the first wireless data transmission module. After scanning one circle, the photoelectric device stops scanning.

[0117] Step 5: After receiving the stitched image, the remote management terminal displays the infrared stitched image and allows users to zoom in on the image as needed to view the details of the wind turbine blade defects; at the same time, the infrared stitched image can be saved.

[0118] Step 6: After the remote management terminal completes the saving, it sends a command to the drone platform. The drone platform then controls the drone to fly to the next blade area and hover. The same method is used to detect the subsequent blades in sequence until all are completed.

[0119] Step 7: After the inspection is completed, the remote management terminal sends a recovery command to the drone platform. The drone platform then controls the drone to return, and the entire inspection process is complete.

[0120] According to an embodiment of the present invention, a method for identifying defects in wind turbine blades is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0121] This embodiment provides a method for identifying wind turbine blade defects, which can be used in the aforementioned wind turbine blade defect identification system. Figure 7 This is a flowchart of a wind turbine blade defect identification method according to an embodiment of the present invention, such as... Figure 7 As shown, the process includes the following steps:

[0122] Step S701: The remote management terminal acquires the drone trajectory planning data and transmits the drone trajectory planning data to the drone platform.

[0123] In step S702, the UAV platform controls the UAV to fly to the target scanning area based on the UAV trajectory planning data, scans the wind turbine blades in the target scanning area to obtain an infrared stitched image, and performs defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result.

[0124] This embodiment of a wind turbine blade defect identification method is applied to, for example, Figure 1 The illustrated embodiment is a wind turbine blade defect identification system; therefore, the specific implementation of steps S701 and S702 can be found in the preceding text. Figure 1 The corresponding descriptions of the illustrated embodiments are not repeated here.

[0125] It is understandable that the function and beneficial effects of the method in this embodiment are the same as those of the previous embodiment. Figure 1 The function and beneficial effects of the wind turbine blade defect identification system in the illustrated embodiment are corresponding and will not be repeated here.

[0126] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this application.

[0127] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0131] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of this application, essentially, or the parts that contribute to the prior art, or parts of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0132] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A wind turbine blade defect identification system, characterized in that, The system includes: a drone platform and a remote management terminal, which are wirelessly connected. The remote management terminal is used to acquire UAV trajectory planning data and transmit the UAV trajectory planning data to the UAV platform. The UAV platform is used to control the UAV to fly to the target scanning area based on the UAV trajectory planning data, scan the wind turbine blades in the target scanning area to obtain an infrared stitched image, and perform defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result. The UAV platform includes: a first wireless data transmission module, a control module, and an optoelectronic device, wherein the first wireless data transmission module is wirelessly connected to the remote management terminal; The first wireless data transmission module is used to receive UAV trajectory planning data sent by the remote management terminal and transmit the UAV trajectory planning data to the control module; The control module is used to control the UAV to fly to the target scanning area based on the UAV trajectory planning data; wherein, the target scanning area is an area containing at least two wind turbine blades; The optoelectronic device is used to scan the wind turbine blades in the target scanning area to obtain an infrared stitched image, and to perform defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result. The optoelectronic device includes: The stitching unit is used to scan the wind turbine blades in the target scanning area to obtain multiple blade scanning images, and stitch the multiple blade scanning images together to obtain the infrared stitched image; The identification unit is used to identify and classify blade defects in the infrared stitched image to obtain the wind turbine blade defects and the location of defect pixels. The determining unit is used to acquire the infrared field of view and the azimuth angle of the wind turbine blades, and to determine the position of the wind turbine blades based on the azimuth angle of the wind turbine blades and the infrared field of view; wherein, based on the azimuth angle of the wind turbine blades, the specific position n of the wind turbine blades in the infrared stitched image is calculated using the following formula: The calculation unit is used to calculate the location of the wind turbine blade defect based on the position of the wind turbine blade and the location of the defect pixel, and to determine the wind turbine blade defect identification result based on the wind turbine blade defect and the defect location of the wind turbine blade; the defect pixel includes azimuth pixels and pitch pixels, and the wind turbine blade defect location is composed of azimuth angle A and pitch angle E, wherein the formula for calculating azimuth angle A is as follows: The formula for calculating the pitch angle E is as follows: The stitching unit is specifically used to arrange the multiple blade scan images based on the azimuth angle position of the wind turbine blades to obtain the infrared stitched image; wherein, the number of infrared stitched images is calculated based on the field of view of the infrared channel, and the formula for calculating the number of images N is as follows: The multiple of 1.2 is to prevent errors during image imaging, which would require a 20% redundancy; assuming an infrared field of view of 18°, the number of images... The panoramic 360° stitching strips are set to 4 rows, with each row consisting of 6 images scaled and stitched together proportionally. These images are then arranged from left to right according to the azimuth position of the wind turbine blades to obtain the infrared stitched image.

2. The system according to claim 1, characterized in that, The identification unit is specifically used to identify and classify blade defects in the infrared stitched image using an image detection model, thereby obtaining the wind turbine blade defects and the location of the defect pixels.

3. The system according to claim 1, characterized in that, The optoelectronic device further includes: The overlay unit is used to overlay the wind turbine blade defect and the defect location of the wind turbine blade onto the infrared stitched image, and transmit the infrared stitched image after overlaying the wind turbine blade defect and the defect location of the wind turbine blade to the remote management terminal through the first wireless data transmission module.

4. The system according to claim 3, characterized in that, The remote management terminal includes: a data management terminal and a second wireless data transmission module, wherein the data management terminal is connected to the second wireless data transmission module, and the second wireless data transmission module is connected to the first wireless data transmission module; The data management terminal is used to acquire UAV trajectory planning data and transmit the UAV trajectory planning data to the first wireless data transmission module. The second wireless data transmission module is used to transmit the UAV trajectory planning data to the UAV platform, and to receive an infrared stitched image of the wind turbine blade defects and their locations superimposed on it, and to send the infrared stitched image of the wind turbine blade defects and their locations superimposed on it to the data management terminal for display.

5. The system according to claim 1, characterized in that, The drone platform also includes: An inertial navigation module is used to collect the flight attitude information of the UAV; the flight attitude information of the UAV is used to control the flight of the UAV.

6. The system according to claim 1, characterized in that, The drone platform also includes: The Beidou module is used to collect the real-time location of the UAV and transmit the real-time location of the UAV to the remote management terminal.

7. A method for identifying defects in wind turbine blades, characterized in that, The method, applied to the wind turbine blade defect identification system as described in any one of claims 1-6, comprises: The remote management terminal acquires the drone trajectory planning data and transmits the drone trajectory planning data to the drone platform. The UAV platform controls the UAV to fly to the target scanning area based on the UAV trajectory planning data, scans the wind turbine blades in the target scanning area to obtain an infrared stitched image, and performs defect identification on the infrared stitched image to obtain the wind turbine blade defect identification result.

Images