A blade dust accumulation and damage monitoring system and method based on image recognition and laser radar

Abstract

The present application relates to the field of land wind power generation technology, and particularly to a blade dust accumulation and damage monitoring system and method based on image recognition and laser radar. The system comprises a shooting module, an image selection module, an image recognition module, a laser radar module, a flight module, a judgment module and a control module. The flight module is equipped with the shooting module and the laser radar module. The control module is used to send a first flight instruction and a shooting instruction to the flight module. The flight module flies to the front side of the wind turbine. After the flight module flies to the front side of the wind turbine, the shooting module receives the shooting instruction, shoots the wind turbine, obtains a plurality of first images, and sends the first images to the image selection module. Since the flight module flies along the blade at a first distance, the second distance obtained by the flight module is different from that obtained under normal circumstances when the blade has a problem. Therefore, whether the blade has dust accumulation or damage can be determined by the difference in the second distance.

Description

Technical Field

[0001] This invention relates to the field of onshore wind power generation technology, and in particular to a blade dust accumulation and damage monitoring system and method based on image recognition and lidar. Background Technology

[0002] With the rapid development of renewable energy, onshore wind power, as an important form of clean energy, has been widely applied across the country. For example, in the Turpan region, which boasts a vast wind belt with abundant wind resources, wind power has become a primary means of generating electricity. However, wind turbine malfunctions severely restrict their continuous and stable operation and power generation efficiency, resulting not only in power generation losses but also increased maintenance costs. Against this backdrop, developing an efficient and accurate fault diagnosis technology is crucial for improving the operational reliability and economy of wind turbines.

[0003] For example, in regions with unique geographical and climatic conditions like Turpan, the significant diurnal temperature range and unique basin topography result in different daytime and nighttime temperatures. Due to lower nighttime temperatures, dew easily condenses on the blades, and the environment is often windy with high winds. After prolonged operation, a large amount of dust easily accumulates on the blades. Dust accumulation alters the blade's surface morphology, disrupting its smoothness and preventing airflow from adhering closely to the blade surface. Ultimately, less wind energy is captured, the rotor speed decreases, and power generation is reduced. Furthermore, wind and sand erosion can damage some blades, also leading to a decrease in rotor speed and power generation. Therefore, effectively identifying blade dust accumulation and monitoring damage have become urgent problems to be solved in onshore wind power generation. Summary of the Invention

[0004] To address the aforementioned shortcomings, the present invention aims to propose a blade dust accumulation and damage monitoring system and method based on image recognition and lidar. This system effectively distinguishes whether dust accumulation or damage has occurred on the blades.

[0005] To achieve this objective, the present invention adopts the following technical solution: a blade dust accumulation and damage monitoring system based on image recognition and lidar, comprising an image capture module, an image selection module, an image recognition module, a lidar module, a flight module, a judgment module, and a control module;

[0006] The flight module is equipped with the camera module and the lidar module;

[0007] The control module is used to send the first flight command and the shooting command to the flight module;

[0008] After receiving the first flight command, the flight module flies to the side of the wind turbine. After the flight module flies to the side of the wind turbine, the shooting module receives the shooting command, takes pictures of the wind turbine, obtains several first images, and sends the first images to the image selection module.

[0009] The image selection module is used to receive a first image and select a number of first images in which the leaves are in the longest state from a number of first images as second images, and send the second images to the image recognition module.

[0010] The image recognition module is used to receive the second image and determine whether there is dust accumulation or damage based on the angular relationship between the blade and the tower in the second image. If there is dust accumulation or damage, a second flight command is sent to the flight module.

[0011] After receiving the second flight command, the flight module waits for the wind turbine to stop, then moves along the blade length direction at a fixed first distance and sends a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance, obtaining the distance between the lidar and the wheel hub as the second distance, and sending the first distance and the second distance to the judgment module.

[0012] The judgment module determines whether the blade has accumulated dust or is damaged based on the first distance and the second distance.

[0013] Preferably, the image selection module includes a first model detection submodule, a judgment value acquisition submodule, and a separation submodule:

[0014] The first model detection submodule is used to perform model detection on the first image and obtain the leaf frame in the first image;

[0015] The judgment value acquisition submodule is used to obtain the length of the bottommost blade frame as the first length, sort the first length by size, and obtain the top few first lengths as the second length;

[0016] Calculate the average value of the second length and use it as the judgment value;

[0017] The separation submodule is used to separate the first image whose length is greater than the judgment value and mark it as the second image.

[0018] Preferably, the image recognition module includes: a second model detection submodule, an extension line acquisition submodule, and a judgment submodule;

[0019] The second model detection submodule is used to obtain the two largest detection boxes in the second image as the first box, and send the first box to the extension line acquisition submodule;

[0020] The extension line acquisition submodule is used to acquire the first extension line of the center line of the width in the first frame respectively, take the included angle formed by the two first extension lines as the first included angle, and send the first included angle to the judgment submodule.

[0021] The judgment submodule is used to determine whether the first included angle is within a preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

[0022] Preferably, the judgment module includes a movement distance acquisition submodule, a calculation submodule, and a confirmation submodule;

[0023] The movement distance acquisition submodule is used to acquire the movement distance of the flight module as the third distance, and send the third distance to the calculation submodule;

[0024] The calculation submodule is used to receive the third distance, determine the vertical position of the flight module and the blade based on the third distance, obtain the fourth distance based on the blade parameters and the vertical position, and add the fourth distance to the first distance to obtain the fifth distance;

[0025] The sixth distance is calculated using the triangular relationship between the third and fifth distances, and then sent to the confirmation submodule.

[0026] The confirmation submodule is used to receive the fourth distance and determine whether the blade has dust accumulation or damage by the relationship between the sixth distance and the second distance.

[0027] Preferably, the blade is damaged when the sixth distance is greater than the second distance;

[0028] When the sixth distance is less than the second distance, dust accumulates on the blade.

[0029] A method for monitoring dust accumulation and damage on blades based on image recognition and lidar, applied to a blade dust accumulation and damage monitoring system based on image recognition and lidar, includes the following steps:

[0030] Step S1: Drive the drone to fly to the side of the wind turbine, take pictures of the wind turbine, and obtain several first images;

[0031] Step S2: Select several images from the first images in which the leaves are in their longest state, and use them as the second images;

[0032] Step S3: Based on the angular relationship between the blade and the tower in the second image, determine whether there is dust accumulation or damage. If there is dust accumulation or damage, wait for the wind turbine to stop, then drive the drone to move along the blade length direction at a fixed first distance and send a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance and obtaining the distance between the lidar and the hub as the second distance.

[0033] Step S4: Determine whether the blade has accumulated dust or is damaged based on the first distance and the second distance.

[0034] Preferably, step S2 includes the following steps:

[0035] Perform model detection on the first image to obtain the leaf frame in the first image;

[0036] Get the length of the bottommost leaf frame as the first length, sort the first lengths by size, get the top few first lengths as the second length, and calculate the average of the second lengths as the judgment value.

[0037] The first image, whose length is greater than the judgment value, is separated and marked as the second image.

[0038] Preferably, the step S3, which determines whether there is dust accumulation or damage based on the angular relationship between the blades and the tower in the second image, is as follows:

[0039] The two largest detection boxes in the second image are selected as the first box body.

[0040] Obtain the first extension line of the center line of the width of the first frame respectively, and take the included angle formed by the two first extension lines as the first included angle;

[0041] Determine whether the first included angle is within the preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

[0042] Preferably, step S4 is as follows:

[0043] Obtain the distance traveled by the flight module as the third distance;

[0044] The vertical position of the flight module and the blade is determined based on the third distance. The fourth distance is obtained based on the blade parameters and the vertical position. The fifth distance is obtained by adding the fourth distance to the first distance.

[0045] The sixth distance is calculated using the triangular relationship between the third and fifth distances;

[0046] When the sixth distance is greater than the second distance, it is determined that the blade is damaged;

[0047] When the sixth distance is less than the second distance, it is determined that dust has accumulated on the blade.

[0048] One of the above technical solutions has the following advantages or beneficial effects: when dust accumulates on the blade, the blade will bulge, and the blade thickness will increase; conversely, when the blade is damaged, the blade will dent, and the blade thickness will decrease. Since the flight module flies while maintaining a first distance from the blade, the second distance obtained by the flight module when a problem occurs differs from the distance obtained under normal circumstances. Therefore, the difference in the second distance can be used to determine whether the blade is experiencing dust accumulation or damage. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the structure of one embodiment of the system of the present invention.

[0050] Figure 2 This is a schematic diagram illustrating the calculation of the first included angle in one embodiment of the present invention.

[0051] Figure 3 This is a schematic diagram illustrating the relationship between the flight module and the blades in one embodiment of the present invention.

[0052] Figure 4 This is a flowchart of one embodiment of the method of the present invention. Detailed Implementation

[0053] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0054] In the description of embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0055] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0056] like Figures 1-4 As shown, a blade dust accumulation and damage monitoring system based on image recognition and lidar includes an image capture module, an image selection module, an image recognition module, a lidar module, a flight module, a judgment module, and a control module.

[0057] The flight module is equipped with the camera module and the lidar module;

[0058] The control module is used to send the first flight command and the shooting command to the flight module;

[0059] After receiving the first flight command, the flight module flies to the side of the wind turbine. After the flight module flies to the side of the wind turbine, the shooting module receives the shooting command, takes pictures of the wind turbine, obtains several first images, and sends the first images to the image selection module.

[0060] The image selection module is used to receive a first image and select a number of first images in which the leaves are in the longest state from a number of first images as second images, and send the second images to the image recognition module.

[0061] The image recognition module is used to receive the second image and determine whether there is dust accumulation or damage based on the angular relationship between the blade and the tower in the second image. If there is dust accumulation or damage, a second flight command is sent to the flight module.

[0062] After receiving the second flight command, the flight module waits for the wind turbine to stop, then moves along the blade length direction at a fixed first distance and sends a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance, obtaining the distance between the lidar and the wheel hub as the second distance, and sending the first distance and the second distance to the judgment module.

[0063] The judgment module determines whether the blade has accumulated dust or is damaged based on the first distance and the second distance.

[0064] To better identify the status of the wind turbine, in this invention, the flight module is a drone equipped with a camera (image module) and a lidar (lidar module). First, the flight module flies to the side of the wind turbine to capture multiple first images. Since these first images include different positions of the turbine blades, making differentiation difficult, an image selection module is needed to select the first images. For example, it selects the second image where the blades are in their longest possible position. Because the blade position is confirmed at this point, the image is stable, and the angle between the blades and the tower can be used to determine if there is a problem with the blades. Therefore, when dust accumulates or the blades are damaged, the stress on the blades is affected. Under prolonged operation and wind conditions, the blades will bend to some extent, causing a change in the angle between the blades and the tower when viewed from the side. Therefore, the angle between the blades and the tower can be used to determine if there is dust accumulation or damage on the blades.

[0065] Although the image recognition module can identify blade problems, it cannot confirm whether the blades are damaged or have accumulated dust. This invention also sends a second flight command to the flight module, which is sent only after the wind turbine has stopped operating. Upon receiving the second flight command, the flight module flies to the blade and uses a lidar module to obtain the distance between the flight module and the blade (first distance), allowing the flight module to maintain this first distance while flying along the blade. During flight, the lidar module periodically rotates, obtaining the distance between itself and the blade root (second distance). Because dust accumulation on the blade causes it to bulge and increase its thickness, while damage causes it to dent and decrease its thickness, the second distance obtained by the flight module when the blade is damaged differs from the normal distance. Therefore, this difference in the second distance allows for the determination of whether the blade is damaged or has accumulated dust.

[0066] Preferably, the image selection module includes a first model detection submodule, a judgment value acquisition submodule, and a separation submodule:

[0067] The first model detection submodule is used to perform model detection on the first image and obtain the leaf frame in the first image;

[0068] The judgment value acquisition submodule is used to obtain the length of the bottommost blade frame as the first length, sort the first length by size, and obtain the top few first lengths as the second length;

[0069] Calculate the average value of the second length and use it as the judgment value;

[0070] The separation submodule is used to separate the first image whose length is greater than the judgment value and mark it as the second image.

[0071] The first detection submodule is equipped with a pre-trained YOLOv8s-OBB model, which can identify the blades and represent them as frames in the first image. Since the wind turbine is constantly moving and the blades are constantly rotating during the shooting process, it's impossible to determine the second image where the blades are at their longest. Therefore, this invention includes a judgment value acquisition submodule. This submodule only acquires the blades at the bottom, obtaining multiple first lengths and sorting them from largest to smallest. If only the largest first length is used to determine the second image, the conditions are relatively stringent, resulting in a relatively small number of second images, which is insufficient for subsequent judgments. Therefore, the average of the top 5% of the sorted first lengths is obtained to get the judgment value. This results in a relatively larger number of second images obtained through the judgment value, which is better suited for subsequent judgments.

[0072] Preferably, the image recognition module includes: a second model detection submodule, an extension line acquisition submodule, and a judgment submodule;

[0073] The second model detection submodule is used to obtain the two largest detection boxes in the second image as the first box, and send the first box to the extension line acquisition submodule;

[0074] The extension line acquisition submodule is used to acquire the first extension line of the center line of the width in the first frame respectively, take the included angle formed by the two first extension lines as the first included angle, and send the first included angle to the judgment submodule.

[0075] The judgment submodule is used to determine whether the first included angle is within a preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

[0076] The second model detection submodule is also equipped with a pre-trained YOLOV8s-OBB model, which can identify the frame of the blade and the tower. Since the second image obtained earlier is obtained by averaging, it cannot be guaranteed that the blade is at the bottom in the second image. Therefore, when judging with the tower, an angle range is set. If the first included angle is greater than or less than the angle range, it indicates that there is a problem with the blade, and it can be determined that the blade has dust accumulation or damage.

[0077] The following is an example illustration, such as... Figure 2As shown, the second model detection submodule detects the two largest frames: the frame of the bottommost blade and the frame of the tower. The centerlines of the two first frames are then connected, extending into two first extension lines with an angle α between them. It is then necessary to determine whether the angle α falls within a preset range. If not, it indicates a problem with the blade, suggesting dust accumulation or damage.

[0078] Preferably, the judgment module includes a movement distance acquisition submodule, a calculation submodule, and a confirmation submodule;

[0079] The movement distance acquisition submodule is used to acquire the movement distance of the flight module as the third distance, and send the third distance to the calculation submodule;

[0080] The calculation submodule is used to receive the third distance, determine the vertical position of the flight module and the blade based on the third distance, obtain the fourth distance based on the blade parameters and the vertical position, and add the fourth distance to the first distance to obtain the fifth distance;

[0081] The sixth distance is calculated using the triangular relationship between the third and fifth distances, and then sent to the confirmation submodule.

[0082] The confirmation submodule is used to receive the fourth distance and determine whether the blade has dust accumulation or damage by the relationship between the sixth distance and the second distance.

[0083] When the flight module executes the second flight command, it flies along the length of the blade, starting from the hub. The third distance acquired by the distance acquisition submodule is calculated from the blade root. Therefore, the flight module's position on the blade can be determined through this third distance calculation, and then the fourth distance can be calculated using the blade parameters. Figure 3 As shown, the first distance is H2, the second distance is H4, and the third distance is H3. Knowing the third distance H3, we can determine the vertical position of the flight module on the blade. From this position, we can find the blade thickness as D. At this point, the fourth distance H1 = D / 2. The fifth distance is then equal to H2 + H1. We can consider the fifth and third distances as the two legs of a right triangle. The sixth distance, the hypotenuse of the right triangle, can then be calculated from the fifth and third distances.

[0084] If damage occurs, the distance from the center to the edge of the blade is less than the fourth distance. Therefore, under the same first distance, the second distance obtained by the flight module is less than the calculated sixth distance.

[0085] When dust accumulates, the distance from the center to the edge of the blade is greater than the fourth distance. Therefore, under the same first distance, the second distance obtained by the flight module is greater than the calculated sixth distance. Thus, the relationship between the second distance and the sixth distance can be used to determine whether there is damage or dust accumulation.

[0086] Preferably, the blade is damaged when the sixth distance is greater than the second distance;

[0087] When the sixth distance is less than the second distance, dust accumulates on the blade.

[0088] A method for monitoring dust accumulation and damage on blades based on image recognition and lidar, applied to a blade dust accumulation and damage monitoring system based on image recognition and lidar, includes the following steps:

[0089] Step S1: Drive the drone to fly to the side of the wind turbine, take pictures of the wind turbine, and obtain several first images;

[0090] Step S2: Select several images from the first images in which the leaves are in their longest state, and use them as the second images;

[0091] Step S3: Based on the angular relationship between the blade and the tower in the second image, determine whether there is dust accumulation or damage. If there is dust accumulation or damage, wait for the wind turbine to stop, then drive the drone to move along the blade length direction at a fixed first distance and send a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance and obtaining the distance between the lidar and the hub as the second distance.

[0092] Step S4: Determine whether the blade has accumulated dust or is damaged based on the first distance and the second distance.

[0093] Preferably, step S2 includes the following steps:

[0094] Perform model detection on the first image to obtain the leaf frame in the first image;

[0095] Get the length of the bottommost leaf frame as the first length, sort the first lengths by size, get the top few first lengths as the second length, and calculate the average of the second lengths as the judgment value.

[0096] The first image, whose length is greater than the judgment value, is separated and marked as the second image.

[0097] Preferably, the step S3, which determines whether there is dust accumulation or damage based on the angular relationship between the blades and the tower in the second image, is as follows:

[0098] The two largest detection boxes in the second image are selected as the first box body.

[0099] Obtain the first extension line of the center line of the width of the first frame respectively, and take the included angle formed by the two first extension lines as the first included angle;

[0100] Determine whether the first included angle is within the preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

[0101] Preferably, step S4 is as follows:

[0102] Obtain the distance traveled by the flight module as the third distance;

[0103] The vertical position of the flight module and the blade is determined based on the third distance. The fourth distance is obtained based on the blade parameters and the vertical position. The fifth distance is obtained by adding the fourth distance to the first distance.

[0104] The sixth distance is calculated using the triangular relationship between the third and fifth distances;

[0105] When the sixth distance is greater than the second distance, it is determined that the blade is damaged;

[0106] When the sixth distance is less than the second distance, it is determined that dust has accumulated on the blade.

[0107] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0108] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A blade dust accumulation and damage monitoring system based on image recognition and lidar, characterized in that, It includes a shooting module, an image selection module, an image recognition module, a lidar module, a flight module, a judgment module, and a control module; The flight module is equipped with the camera module and the lidar module; The control module is used to send the first flight command and the shooting command to the flight module; After receiving the first flight command, the flight module flies to the side of the wind turbine. After the flight module flies to the side of the wind turbine, the shooting module receives the shooting command, takes pictures of the wind turbine, obtains several first images, and sends the first images to the image selection module. The image selection module is used to receive a first image and select a number of first images in which the leaves are in the longest state from a number of first images as second images, and send the second images to the image recognition module. The image recognition module is used to receive the second image and determine whether there is dust accumulation or damage based on the angular relationship between the blade and the tower in the second image. If there is dust accumulation or damage, a second flight command is sent to the flight module. After receiving the second flight command, the flight module waits for the wind turbine to stop, then moves along the blade length direction at a fixed first distance and sends a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance, obtaining the distance between the lidar and the wheel hub as the second distance, and sending the first distance and the second distance to the judgment module. The judgment module determines whether the blade has dust accumulation or damage based on the first distance and the second distance. The image selection module includes a first model detection submodule, a judgment value acquisition submodule, and a separation submodule: The first model detection submodule is used to perform model detection on the first image and obtain the leaf frame in the first image; The judgment value acquisition submodule is used to obtain the length of the bottommost blade frame as the first length, sort the first length by size, and obtain the top few first lengths as the second length; Calculate the average value of the second length and use it as the judgment value; The separation submodule is used to separate the first image whose length is greater than the judgment value and mark it as the second image.

2. The blade dust accumulation and damage monitoring system based on image recognition and lidar according to claim 1, characterized in that, The image recognition module includes: a second model detection submodule, an extension line acquisition submodule, and a judgment submodule; The second model detection submodule is used to obtain the two largest detection boxes in the second image as the first box, and send the first box to the extension line acquisition submodule; The extension line acquisition submodule is used to acquire the first extension line of the center line of the width in the first frame respectively, take the included angle formed by the two first extension lines as the first included angle, and send the first included angle to the judgment submodule. The judgment submodule is used to determine whether the first included angle is within a preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

3. The blade dust accumulation and damage monitoring system based on image recognition and lidar according to claim 1, characterized in that, The judgment module includes a movement distance acquisition submodule, a calculation submodule, and a confirmation submodule; The movement distance acquisition submodule is used to acquire the movement distance of the flight module as the third distance, and send the third distance to the calculation submodule; The calculation submodule is used to receive the third distance, determine the vertical position of the flight module and the blade based on the third distance, obtain the fourth distance based on the blade parameters and the vertical position, and add the fourth distance to the first distance to obtain the fifth distance; The sixth distance is calculated using the triangular relationship between the third and fifth distances, and then sent to the confirmation submodule. The confirmation submodule is used to receive the fourth distance and determine whether the blade has dust accumulation or damage by the relationship between the sixth distance and the second distance.

4. The blade dust accumulation and damage monitoring system based on image recognition and lidar according to claim 3, characterized in that, When the sixth distance is greater than the second distance, the blade is damaged; When the sixth distance is less than the second distance, dust accumulates on the blade.

5. A method for monitoring blade dust accumulation and damage based on image recognition and lidar, applied to the blade dust accumulation and damage monitoring system based on image recognition and lidar as described in any one of claims 1 to 4, characterized in that, Includes the following steps: Step S1: Drive the drone to fly to the side of the wind turbine, take pictures of the wind turbine, and obtain several first images; Step S2: Select several images from the first images in which the leaves are at their longest, and use them as the second images; Step S3: Based on the angular relationship between the blade and the tower in the second image, determine whether there is dust accumulation or damage. If there is dust accumulation or damage, wait for the wind turbine to stop, then drive the drone to move along the blade length direction at a fixed first distance and send a first control command to the lidar module. The first control command includes obtaining the distance between the lidar and the blade as the first distance and obtaining the distance between the lidar and the hub as the second distance. Step S4: Determine whether the blade has accumulated dust or is damaged based on the first distance and the second distance.

6. The method for monitoring blade dust accumulation and damage based on image recognition and lidar according to claim 5, characterized in that, Step S2 includes the following steps: Perform model detection on the first image to obtain the leaf frame in the first image; Get the length of the bottommost leaf frame as the first length, sort the first lengths by size, get the top few first lengths as the second length, and calculate the average of the second lengths as the judgment value. The first image, whose length is greater than the judgment value, is separated and marked as the second image.

7. The method for monitoring blade dust accumulation and damage based on image recognition and lidar according to claim 6, characterized in that, The step S3, which determines whether there is dust accumulation or damage based on the angular relationship between the blades and the tower in the second image, is as follows: The two largest detection boxes in the second image are selected as the first box body. Obtain the first extension line of the center line of the width of the first frame respectively, and take the included angle formed by the two first extension lines as the first included angle; Determine whether the first included angle is within the preset angle range. If the first included angle is within the preset angle range, it is determined that there is no problem with the blade. If the first included angle is not within the preset angle range, it is determined that the blade has dust accumulation or damage.

8. The method for monitoring blade dust accumulation and damage based on image recognition and lidar according to claim 5, characterized in that, The steps of step S4 are as follows: Obtain the distance traveled by the flight module as the third distance; The vertical position of the flight module and the blade is determined based on the third distance. The fourth distance is obtained based on the blade parameters and the vertical position. The fifth distance is obtained by adding the fourth distance to the first distance. The sixth distance is calculated using the triangular relationship between the third and fifth distances; When the sixth distance is greater than the second distance, it is determined that the blade is damaged; When the sixth distance is less than the second distance, it is determined that dust has accumulated on the blade.

Images