Fan blade grounding resistance testing method and equipment based on unmanned aerial vehicle-mounted mechanical arm

Abstract

The invention provides a fan blade grounding resistance testing method and equipment based on an unmanned aerial vehicle-mounted mechanical arm. The method comprises the following steps: extracting a fan blade image, enhancing fan blade image features and reducing the complexity of subsequent processing; segmenting the fan blade image sample to obtain a comprehensive image of a blade main body image and a lightning arrester image; performing feature extraction on the comprehensive image to obtain various features of the comprehensive image; performing identification and classification on the comprehensive image after the extraction of various features, positioning a target area of the lightning arrester, and calculating a parameter index of the area of the lightning arrester; and obtaining a lightning arrester identification result according to the parameter indexes. According to the invention, the working efficiency of the grounding resistance test of the fan is improved, the third-party test cost is saved, and the downtime of the fan is greatly reduced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of detection, and in particular, to a method and device for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm. Background Art

[0002] Wind power generation is a power generation method that uses wind energy to generate electric energy. The wind turbine blades capture wind energy, drive the main shaft to rotate, and finally drive the generator to generate electricity. Currently, the problems of damage and loss of wind turbine blade equipment caused by lightning disasters have been widely concerned by wind power generation enterprises and the industry. Among them, the lightning arrester equipment of the wind turbine blade will be damaged by the erosion of complex and harsh environments just like the wind turbine blade in a complex environment.

[0003] To avoid the risk of lightning strikes on wind turbine blades during thunderstorm weather, it is necessary to regularly test the grounding resistance value of the lightning arrester of the wind turbine blade. Obtaining the effective resistance value in advance can help the operation and maintenance personnel take preventive measures in advance to avoid a series of problems such as the damage of the wind turbine caused by lightning strikes on the blades, power outage, secondary forest fires, and property and personal safety losses caused by falling objects. In the daily maintenance of wind turbines, the grounding resistance test of the lightning arrester device is also the core part of the operation and maintenance content. Therefore, developing a method and device for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm to effectively overcome the defects in the above-mentioned related technologies has become an urgent technical problem in the industry. Summary of the Invention

[0004] In view of the above problems existing in the prior art, the embodiments of the present invention provide a method and device for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm.

[0005] In a first aspect, the embodiments of the present invention provide a method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm, including: extracting an image of the wind turbine blade, enhancing the features of the wind turbine blade image and reducing the complexity of subsequent processing; segmenting the wind turbine blade image sample to obtain a composite image of the blade main body image and the lightning arrester image; extracting features from the composite image to obtain various features of the composite image; performing recognition and classification on the composite image after various feature extractions, locating the target area of the lightning arrester and calculating the parameter index of the lightning arrester area; and obtaining the lightning arrester recognition result according to the parameter index.

[0006] Based on the above method embodiment content, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided by the embodiments of the present invention, the enhancing the features of the wind turbine blade image and reducing the complexity of subsequent processing includes: performing denoising, contrast enhancement, and normalization processing on the wind turbine blade image.

[0007] Based on the content of the above method embodiments, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the segmentation of the wind turbine blade image sample includes: using an image segmentation method based on deep learning to separate the comprehensive image including the blade main body image and the lightning arrester image from the blade background image to obtain the comprehensive image.

[0008] Based on the content of the above method embodiments, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the feature extraction of the comprehensive image to obtain various features of the comprehensive image includes: selecting features based on GLCM, features based on Tamura, statxture features, fractal features, the main axis direction feature of the grayscale image, and the spatial feature of the image as the features of the comprehensive image.

[0009] Based on the content of the above method embodiments, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the recognition and classification of the comprehensive image after various feature extractions, the positioning of the lightning arrester target area, and the calculation of the parameter index of the lightning arrester area include: using a classifier based on the convolutional neural network (CNN) for model training, and using the obtained classification model to recognize and classify the comprehensive image, locate the lightning arrester target area, and calculate the area of the lightning arrester.

[0010] Based on the content of the above method embodiments, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the denoising of the wind turbine blade image includes: analyzing the characteristics of the wind turbine blade image in the time domain and frequency domain, and using the difference between the noise and the image signal to perform targeted denoising processing, extracting effective information from the noisy wind turbine blade image, and improving the overall quality of the wind turbine blade image.

[0011] Based on the content of the above method embodiments, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the extraction of the wind turbine blade image includes: using an FPV auxiliary system, which consists of a network camera, a UAV-borne video transmitter, and a ground station system. The network camera is used for the robotic arm to accurately dock with the lightning arrester at a close distance. The data transmission method is an Ethernet port, and the obtained image is sent to the ground station system by the UAV-borne video transmitter to assist the operator in achieving accurate docking.

[0012] In a second aspect, an embodiment of the present invention provides a grounding resistance testing device for a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm, including: a first main module for extracting an image of the wind turbine blade, enhancing the features of the wind turbine blade image, and reducing the complexity of subsequent processing; a second main module for segmenting the wind turbine blade image sample to obtain a composite image of the blade main body image and the lightning arrester image; a third main module for extracting features from the composite image to obtain various features of the composite image; a fourth main module for identifying and classifying the composite image after various feature extractions, positioning the lightning arrester target area, and calculating the parameter index of the lightning arrester area; and a fifth main module for obtaining the lightning arrester recognition result according to the parameter index.

[0013] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0014] at least one processor, at least one memory, and a communication interface; wherein,

[0015] the processor, the memory, and the communication interface communicate with each other;

[0016] the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the grounding resistance testing method for the wind turbine blade based on the UAV-borne robotic arm provided in any one of the various implementation manners of the first aspect.

[0017] In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, and the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the grounding resistance testing method for the wind turbine blade based on the UAV-borne robotic arm provided in any one of the various implementation manners of the first aspect.

[0018] The grounding resistance testing method and device for the wind turbine blade based on the UAV-borne robotic arm provided by the embodiments of the present invention use the UAV to carry a grounding resistance testing robotic arm and combine intelligent algorithms to intelligently detect the grounding resistance of the wind turbine. While ensuring the safety of safety operation and maintenance personnel, the working efficiency of the grounding resistance testing of the wind turbine is improved, the third-party testing cost is saved, and the downtime of the wind turbine is also greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 Schematic flow chart of the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided by an embodiment of the present invention;

[0021] Figure 2 Schematic structural diagram of the device for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided by an embodiment of the present invention;

[0022] Figure 3 Schematic physical structure diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. In addition, the technical features in each embodiment or a single embodiment provided by the present invention can be combined with each other arbitrarily to form a feasible technical solution. Such combination is not restricted by the order of steps and / or the mode of structural composition, but must be based on the fact that those of ordinary skill in the art can implement it. When the combination of technical solutions conflicts with each other or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the protection scope required by the present invention. If there are step numbers in the following embodiments, they are only set for the convenience of explanation and illustration, and no limitation is imposed on the order between steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0024] An embodiment of the present invention provides a method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm. Referring to Figure 1 , the method includes: extracting an image of the wind turbine blade, enhancing the features of the wind turbine blade image and reducing the complexity of subsequent processing; segmenting the wind turbine blade image sample to segment a composite image of the blade main body image and the lightning arrester image; extracting features from the composite image to obtain various features of the composite image; performing recognition and classification on the composite image after various feature extractions, positioning the lightning arrester target area and calculating a parameter index of the lightning arrester area; and obtaining a lightning arrester recognition result according to the parameter index.

[0025] Based on the content of the above method embodiment, as an optional embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided by an embodiment of the present invention, the enhancing the features of the wind turbine blade image and reducing the complexity of subsequent processing includes: performing denoising, contrast enhancement, and normalization processing on the wind turbine blade image.

[0026] Specifically, the core of image preprocessing technology lies in applying a series of algorithms to reduce the noise level in the image and enhance its contrast, so as to restore the clarity and details of the image. In this process, denoising is a key step, mainly using filtering techniques based on the spatial domain. These techniques analyze the characteristics of the image in different domains (such as the time domain, frequency domain), and utilize the differences between noise and image signals to perform targeted denoising processing. Its ultimate goal is to extract useful information from the noisy image, effectively suppress noise, and thus improve the overall quality of the image.

[0027] Based on the content of the above method embodiments, as an alternative embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the segmentation of the wind turbine blade image sample includes: using an image segmentation method based on deep learning to separate the composite image including the blade main body image and the lightning arrester image from the blade background image to obtain the composite image.

[0028] Specifically, the image segmentation algorithm based on deep learning mainly uses a deep neural network to automatically extract image features and perform pixel-level classification. The core of this technology lies in using a convolutional neural network (CNN) to identify different objects and regions in the image. After the image is input into the network, the CNN extracts the features of the image through multiple convolutional layers and pooling layers, and then performs classification and regression through fully connected layers. Finally, the classification label of each pixel is output to form a segmentation map. This method can learn multi-level features from simple to complex, so as to achieve precise segmentation of objects in the image.

[0029] Based on the content of the above method embodiments, as an alternative embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the extraction of features from the composite image to obtain multiple features of the composite image includes: selecting features based on GLCM, features based on Tamura, statxture features, fractal features, the main axis direction feature of the grayscale image, and the spatial feature of the image as the features of the composite image.

[0030] Specifically, the six features are as follows:

[0031] Features based on GLCM: Calculate texture features such as energy, contrast, uniformity, and correlation using the gray-level co-occurrence matrix (GLCM). GLCM is a generally recognized and effective method for image feature extraction. By calculating the gray values of pixel pairs in the image, co-occurrence matrices at multiple angles and distances can be obtained, and then features reflecting the image texture can be extracted.

[0032] Tamura-based features: According to Tamura texture parameters, roughness, contrast, and directionality features are extracted. Tamura feature extraction is an important method in remote sensing image analysis. It helps us understand and describe the texture features of images by quantifying six basic visual attributes of texture.

[0033] Statxture features: Combining structural and texture information, Statxture features are extracted. Statxture features reflect the statistical characteristics of texture, including mean, variance, entropy, contrast, etc., which can reflect the brightness distribution and changes of texture.

[0034] Fractal features: The complexity of the lightning arrester is described by parameters such as the fractal dimension. The fractal theory extracts the box dimension and information dimension from the signal as classification features. This kind of feature contains the variation rules of signal amplitude, frequency, and phase, and concentrates the difference information between various modulation methods.

[0035] Principal axis direction features of grayscale images: Calculate the principal axis direction of pixels. On the principal axis and the secondary axis, local grayscale change features in a total of 4 directions, namely the positive and negative half-axes of each, are extracted respectively.

[0036] Spatial features of images: Spatial feature extraction is usually carried out after the fully connected layer or the pooling layer. Global information and structural information of the image are obtained through some complex spatial transformations.

[0037] Based on the content of the above method embodiments, as an alternative embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, for the comprehensive image after various feature extractions, performing recognition and classification, positioning the target area of the lightning arrester and calculating the parameter index of the area of the lightning arrester, including: training a classifier based on the convolutional neural network (CNN) to obtain a classification model, using the classification model to perform recognition and classification on the comprehensive image, positioning the target area of the lightning arrester, and calculating the area of the lightning arrester.

[0038] Based on the content of the above method embodiments, as an alternative embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, denoising the wind turbine blade image, including: analyzing the features of the wind turbine blade image in the time domain and the frequency domain, and using the difference between the noise and the image signal to perform targeted denoising processing, extracting effective information from the wind turbine blade image containing noise, and improving the overall quality of the wind turbine blade image.

[0039] Based on the content of the above method embodiments, as an alternative embodiment, in the method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention, the extraction of the wind turbine blade image includes: using an FPV auxiliary system, which is composed of a network camera, a UAV-borne video transmitter, and a ground station system. The network camera is used for the precise docking of the robotic arm with the lightning arrester at a close distance. The data transmission method is through an Ethernet port. The acquired image is sent by the UAV-borne video transmitter to the ground station system to assist the operator in achieving precise docking.

[0040] The method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV)-borne robotic arm provided in the embodiments of the present invention uses a UAV to carry a grounding resistance test robotic arm and combines an intelligent algorithm to intelligently detect the grounding resistance of the wind turbine. While ensuring the safety of the maintenance personnel, it improves the working efficiency of the grounding resistance test of the wind turbine, saves the third-party test cost, and greatly reduces the downtime of the wind turbine.

[0041] It should be noted that the robotic arm of the grounding resistance tester undertakes the important task of docking with the lightning arrester of the wind turbine blade. The rationality and flexibility of its design are the key points for completing the grounding resistance detection task. This structure uses a high-strength circular carbon fiber tubular structure as the main design body. For the contact area to increase the structural strength, 7075 aviation aluminum alloy material is used. The overall structure is lightweight, strong, and durable, and the wiring inside the tube is more concise and beautiful. To improve the versatility of the robotic arm, for the integration of the overall grounding resistance detection system, an independent and simplified method is adopted, that is: five main systems are designed at the front end of the robotic arm: a coating system, a grinding system, a probe system, a ranging system, and an FPV docking image assistance system. Among the five systems, the grinding system, the ranging system, and the FPV docking image assistance system require the electrical energy of the UAV. A multi-channel circuit module is built into the UAV system to provide sufficient electrical energy for each system. Among them, the FPV docking image assistance system is located directly above the robotic arm and is fixed; the ranging system is located at the bottom of the robotic arm and is fixed, and the power used is provided by the UAV power supply; the coating system, the grinding system, and the probe system form a fan-shaped structure with a spacing angle of 55°. A large-tension servo motor drives this wheel to drive the conversion of the three system structures, and the left and right conversion angles are each 55°.

[0042] Since the fan blades are in the external environment for a long time, exposed to wind, sun, cold and heat, it will not only cause certain damage to the fan blades themselves. The lightning arrester itself will also oxidize to a certain extent in this environment, and a metal oxide layer will be generated on the surface, mostly copper green. This phenomenon may lead to inaccurate resistance values measured by the grounding resistance tester, thus increasing the measurement difficulty and affecting the measurement efficiency. Therefore, the solution we came up with is that we first apply a certain amount of rust remover to the oxide layer on the surface of the lightning arrester through our coating system, and then use the grinding system to grind off the oxide layer on the surface of the lightning arrester. The mutual cooperation of the two systems can better remove the oxide layer on the lightning arrester of the fan blade. Therefore, the cooperation between the grinding system and the coating system enables the probe system to measure better, and can also improve the operation efficiency and accuracy.

[0043] The internal filling space of the coating system is 80*80*80mm, and it is made of a soft 3D printing material by means of light curing technology; the size of the special sponge is 100*100*100mm, and it has the characteristics of high density, strong elasticity, strong water absorption, and no corrosive effect on the rust remover. At the same time, the overall structure adopts a grid design, so that it can be more firmly inlaid with the sponge. The function conversion of this structure is controlled by a large-tension servo of the robotic arm through a driving gear, and the left-right conversion is 55°.

[0044] The main function of the grinding system is to grind the oxide layer on the surface of the lightning arrester. It mainly includes a high-power brushless motor, a transmission connection device, and a hard brush. The motor has the characteristics of high power, high ultimate load, and low speed. The power motor and the hard brush used for grinding are driven by a gear transmission device, and its rotation speed is 200 revolutions per minute, and the ultimate load is 9 kgf.cm. The power supply used for this drive is provided by the drone battery.

[0045] The probe system is an important component for connecting the ground resistance tester wire. It is located at the front end of the robotic arm and mainly consists of a probe, a buffer device, and a ground resistance tester wire. The probe of this component is made of copper. The probes are 25mm apart and arranged in a staggered manner, so that multiple probes can be contacted on a small area of ​​the lightning receptor surface. At present, at least 3 probes can be in contact with the lightning receptor, which effectively avoids the problem of the probe not being able to contact the wind turbine blade lightning receptor and improves the accuracy of the ground resistance tester. The buffer device has a built-in soft spring to alleviate the direct impact between the probe and the lightning receptor, reduce friction damage, and also help the stability of the aircraft. The other end of the probe system is connected to the ground resistance tester through a customized silver-plated, low-resistance, ultra-thin and soft insulated wire. That is, after the probe system contacts the lightning rod, the probe system and its wire, as well as the lightning rod and its ground wire and the ground resistance tester form a closed loop, thereby measuring the resistance value of the entire loop. The measured resistance value is subtracted from the inherent resistance value of the entire loop to calculate the resistance value of the entire lightning rod and wire, which is used to judge the status of the lightning rod. According to national standards, a ground resistance value greater than 4Ω indicates an abnormality and needs to be checked.

[0046] The ranging system is mainly used to judge the distance between the mechanical arm and the lightning rod during the docking process. The rangefinder has a ranging range of 120 meters and a refresh rate of 300 Hz. After the data is collected, it is calculated by the flight control and transmitted by the onboard data transmission of the UAV to the ground control station of the UAV, providing assistance for accurate docking with the lightning rod.

[0047] The FPV auxiliary system uses a short-focus, wide-angle, 3-megapixel network camera, which is mainly used to assist the mechanical arm in precise docking during the close-range docking with the lightning receptor. The system consists of a network camera, an onboard image transmitter of a drone, and a ground station system. The data transmission method is the network port, and the acquired images are sent to the ground station system by the onboard image transmitter, thereby assisting the operator to achieve precise docking.

[0048] The technical solutions provided by various embodiments of the present invention use a drone equipped with a ground resistance test mechanical arm and combine it with an intelligent algorithm to perform intelligent detection of the ground resistance of the wind turbine. While ensuring the safety of safety operation and maintenance personnel, the working efficiency of the ground resistance test of the wind turbine is improved, the third-party testing costs are saved, and the downtime of the wind turbine is greatly reduced. The specific economic benefit indicators of the project are as follows:

[0049] The traditional operation and maintenance single-unit detection time is reduced from 5-6 hours to about 1 hour. The single-unit detection cost of traditional operation and maintenance using an external hanging basket is saved by about 2,500 yuan per unit.

[0050] Taking the average power of each wind turbine as 4MW, it can generate 4,000 kWh of electricity in one hour. Based on the calculation of 0.4 yuan per kWh of wind electricity, each inspection can save 5 hours. If a single wind turbine is inspected once a year, the power generation loss can be reduced by 8,000 yuan.

[0051] The practical application of this lightning protection grounding resistance testing drone can greatly improve the safety and supply efficiency of wind farms, and effectively solve problems such as insufficient intelligent upgrading and innovation capabilities, and limited safety production management methods.

[0052] Reduce the safety risks of staff. The project uses drone intelligent equipment to replace the traditional manual ground resistance test in the hanging basket mode. While completing the test work efficiently, it avoids the safety risks of on-site staff in high-altitude testing.

[0053] Improvement of renewable energy power generation. Compared with the traditional mode, the use of drones for ground resistance testing in this project reduces the downtime of wind turbines by 5 hours per unit, thus improving the power generation of renewable energy wind farms. Compared with traditional energy generation, renewable energy does not pollute the air environment, thus achieving a virtuous cycle of economic development and environmental protection.

[0054] The drone system is equipped with a visible light lens as standard, which can serve as an extension of the human eye to perform routine inspections of wind turbine blades and photovoltaic module appearance defects, as well as transmission lines and collector lines.

[0055] According to the actual mission scenario, such as infrared detection, point cloud scanning and other requirements, it is sufficient to replace the infrared lens or laser lens without the need to purchase a new UAV flight platform for adaptation;

[0056] In addition, the drone has a large load capacity and can assist in transporting equipment and cables to the tower, saving personnel time in going up and down the tower.

[0057] The implementation basis of each embodiment of the present invention is to implement programmed processing through a device with a processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention can be encapsulated into various modules. Based on this reality, on the basis of the above embodiments, an embodiment of the present invention provides a wind turbine blade grounding resistance test device based on an unmanned aerial vehicle mechanical arm, which is used to execute the wind turbine blade grounding resistance test method based on an unmanned aerial vehicle mechanical arm in the above method embodiment. Figure 2The device includes: a first main module, which is used to extract the wind turbine blade image, enhance the wind turbine blade image features and reduce the complexity of subsequent processing; a second main module, which is used to segment the wind turbine blade image samples and segment the comprehensive image of the blade main body image and the lightning receptor image; a third main module, which is used to extract features from the comprehensive image and obtain multiple features of the comprehensive image; a fourth main module, which is used to identify and classify the comprehensive image after multiple feature extractions, locate the lightning receptor target area and calculate the parameter index of the lightning receptor area; and a fifth main module, which is used to obtain the lightning receptor identification result according to the parameter index.

[0058] The wind turbine blade grounding resistance test device based on an unmanned aerial vehicle mechanical arm provided by the embodiment of the present invention adopts Figure 2 Several modules in the system are intelligently detected by using a drone equipped with a ground resistance test robotic arm and combining it with an intelligent algorithm. While ensuring the safety of operation and maintenance personnel, the work efficiency of the ground resistance test of the wind turbine is improved, the third-party testing costs are saved, and the downtime of the wind turbine is greatly reduced.

[0059] It should be noted that the device in the device embodiment provided by the present invention can be used to implement the method in the above method embodiment as well as the method in other method embodiments provided by the present invention. The only difference is that the corresponding functional modules are set, and the principle is basically the same as the principle of the above device embodiment provided by the present invention. As long as the technical personnel in the field refer to the specific technical solutions in other method embodiments on the basis of the above device embodiment, obtain the corresponding technical means and the technical solutions composed of these technical means by combining technical features, the device in the above device embodiment can be improved on the premise of ensuring the practicality of the technical solution, thereby obtaining the corresponding device class embodiment, which is used to implement the methods in other method class embodiments. For example:

[0060] Based on the contents of the above-mentioned device embodiments, as an optional embodiment, the wind blade grounding resistance testing device based on an unmanned aerial vehicle-mounted robotic arm provided in the embodiments of the present invention also includes: a first sub-module, used to realize the enhanced wind blade image features and reduce the complexity of subsequent processing, including: denoising, contrast enhancement and normalization of the wind blade image.

[0061] Based on the contents of the above-mentioned device embodiment, as an optional embodiment, the wind turbine blade grounding resistance testing device based on an unmanned aerial vehicle-mounted robotic arm provided in the embodiment of the present invention further includes: a second sub-module, used to realize the segmentation of the wind turbine blade image samples, including: using an image segmentation method based on deep learning to segment the comprehensive image including the blade main body image and the lightning rod image from the blade background image to obtain the comprehensive image.

[0062] Based on the contents of the above-mentioned device embodiment, as an optional embodiment, the wind turbine blade grounding resistance testing device based on an unmanned aerial vehicle-mounted robotic arm provided in the embodiment of the present invention also includes: a third sub-module, used to realize the feature extraction of the comprehensive image to obtain multiple features of the comprehensive image, including: selecting GLCM-based features, Tamura-based features, statxture features, fractal features, main axis direction features of grayscale images and spatial features of images as features of the comprehensive image.

[0063] Based on the contents of the above-mentioned device embodiments, as an optional embodiment, the wind turbine blade grounding resistance testing device based on an unmanned aerial vehicle-mounted mechanical arm provided in the embodiments of the present invention further includes: a fourth submodule, used to realize the recognition and classification of the comprehensive image after multiple feature extractions, locate the target area of ​​the lightning receptor and calculate the parameter indicators of the lightning receptor area, including: using a classifier based on a convolutional neural network CNN for model training, and obtaining a classification model to recognize and classify the comprehensive image, locate the target area of ​​the lightning receptor, and calculate the area of ​​the lightning receptor.

[0064] Based on the contents of the above-mentioned device embodiments, as an optional embodiment, the wind blade grounding resistance testing device based on an unmanned aerial vehicle-mounted robotic arm provided in the embodiments of the present invention further includes: a fifth sub-module, used to realize the denoising of the wind blade image, including: analyzing the characteristics of the wind blade image in the time domain and the frequency domain, and using the difference between the noise and the image signal to perform targeted denoising processing, extracting effective information from the noisy wind blade image, and improving the overall quality of the wind blade image.

[0065] Based on the contents of the above-mentioned device embodiments, as an optional embodiment, the wind turbine blade grounding resistance testing device based on an unmanned aerial vehicle (UAV) robotic arm provided in the embodiment of the present invention further includes: a sixth sub-module, used to realize the extraction of wind turbine blade images, including: using an FPV auxiliary system, the FPV auxiliary system is composed of a network camera, an UAV-mounted image transmission and a ground station system, the network camera is used for the robotic arm to accurately dock with the lightning arrester at close range, the data transmission method is a network port, and the acquired image is sent to the ground station system by the airborne image transmission to assist the operating personnel to achieve accurate docking.

[0066] The method of the embodiment of the present invention is implemented by relying on electronic devices, so it is necessary to introduce the relevant electronic devices. Based on this purpose, the embodiment of the present invention provides an electronic device, such as Figure 3 As shown, the electronic device includes: at least one processor, a communication interface, at least one memory, and a communication bus, wherein at least one processor, the communication interface, and at least one memory communicate with each other through the communication bus. At least one processor can call the logic instructions in at least one memory to execute all or part of the steps of the method provided by the aforementioned various method embodiments.

[0067] In addition, the logic instructions in the at least one memory mentioned above can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each method embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0068] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0069] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0070] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. Based on this understanding, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and sometimes in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0071] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "includes..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. Any "predetermined threshold", "preset threshold" and other similar expressions that do not indicate specific values ​​can be determined by a person of ordinary skill in the art through simple experiments or corresponding debugging.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle (UAV) mechanical arm, characterized in that: include: Extract the fan blade image, enhance the fan blade image features and reduce the complexity of subsequent processing; Segment the wind turbine blade image sample to obtain a comprehensive image of the blade main body image and the lightning receptor image; extract features from the comprehensive image to obtain multiple features of the comprehensive image; The comprehensive image after multiple feature extraction is identified and classified, the target area of ​​the lightning receptor is located and the parameter index of the lightning receptor area is calculated; and the lightning receptor identification result is obtained according to the parameter index.

2. The method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 1, characterized in that: The method of enhancing the characteristics of the fan blade image and reducing the complexity of subsequent processing includes: performing denoising, contrast enhancement and normalization processing on the fan blade image.

3. The method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 2, characterized in that: The segmentation of the wind turbine blade image samples includes: using an image segmentation method based on deep learning to segment a comprehensive image including a blade main body image and a lightning receptor image from a blade background image to obtain the comprehensive image.

4. The method for testing the grounding resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 3 is characterized in that: The feature extraction of the comprehensive image obtains multiple features of the comprehensive image, including: selecting GLCM-based features, Tamura-based features, statxture features, fractal features, main axis direction features of grayscale images and image spatial features as the features of the comprehensive image.

5. The method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 4, characterized in that: The method of identifying and classifying the comprehensive image after multiple feature extraction, locating the target area of ​​the lightning receptor and calculating the parameter index of the area of ​​the lightning receptor comprises: using a classifier based on a convolutional neural network (CNN) to perform model training, identifying and classifying the comprehensive image with the obtained classification model, locating the target area of ​​the lightning receptor and calculating the area of ​​the lightning receptor.

6. The method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 5, characterized in that: The denoising of the fan blade image includes: analyzing the characteristics of the fan blade image in the time domain and the frequency domain, and using the difference between the noise and the image signal to perform targeted denoising processing, extracting effective information from the noisy fan blade image, and improving the overall quality of the fan blade image.

7. The method for testing the ground resistance of a wind turbine blade based on an unmanned aerial vehicle mechanical arm according to claim 6, characterized in that: The method of extracting wind turbine blade images includes: using an FPV auxiliary system, the FPV auxiliary system is composed of a network camera, an unmanned aerial vehicle (UAV) airborne image transmission and a ground station system, the network camera is used for the mechanical arm to accurately dock with the lightning arrester at a close distance, the data transmission method is a network port, and the acquired image is sent to the ground station system by the airborne image transmission to assist the operating personnel to achieve accurate docking.

8. A wind turbine blade grounding resistance test device based on an unmanned aerial vehicle mechanical arm, characterized in that: include: The first main module is used to extract the fan blade image, enhance the fan blade image features and reduce the complexity of subsequent processing; The second main module is used to segment the wind turbine blade image samples to obtain a comprehensive image of the blade main body image and the lightning receptor image; the third main module is used to extract features from the comprehensive image to obtain multiple features of the comprehensive image; the fourth main module is used to identify and classify the comprehensive image after multiple feature extraction, locate the lightning receptor target area and calculate the parameter indicators of the lightning receptor area; the fifth main module is used to obtain the lightning receptor identification result based on the parameter indicators.

9. An electronic device, characterized in that: include: At least one processor, at least one memory and a communication interface; wherein, The processor, memory and communication interface communicate with each other; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which cause a computer to execute the method of any one of claims 1 to 7.

Images