Wind power plant overhead line defect intelligent identification system based on unmanned aerial vehicle autonomous inspection

Abstract

The invention relates to the technical field of defect identification systems, in particular to a wind power plant overhead line defect intelligent identification system based on unmanned aerial vehicle autonomous inspection, which comprises an unmanned aerial vehicle control module, a data processing module, an image acquisition module, an image processing module, a defect identification module, a data storage module and a user interaction module, the data processing module obtains information about whether the unmanned aerial vehicle flies safely or not and transmits the information about whether the unmanned aerial vehicle flies safely or not to the user interaction module; the user interaction module provides a visual interface, and a user inquires whether the flight of the unmanned aerial vehicle is safe or not and then sends a signal for starting and stopping the image acquisition module; the image acquisition module carries out shooting according to the start-stop signal; the image processing module preprocesses the acquired image; and the defect identification module outputs the type and position of the defect. The safety of the inspection process of the unmanned aerial vehicle is ensured, and the acquisition precision of the image acquisition module is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of defect recognition systems, and in particular to an intelligent defect recognition system for overhead lines in wind farms based on autonomous inspection by unmanned aerial vehicles. Background Art

[0002] The defect intelligent identification system is a high-tech system that integrates drone inspection, image recognition, data analysis and intelligent decision-making. It is specially used for defect detection and maintenance of overhead lines (such as transmission lines, insulators, hardware, etc.) in wind farms. This system is designed to replace traditional manual inspections, improve inspection efficiency, reduce labor costs, and ensure the accuracy and timeliness of detection in complex environments.

[0003] The system is generally composed of the following parts: UAV platform, intelligent recognition module, data processing and analysis module, cloud platform and control platform. The UAV platform is equipped with a variety of sensors such as high-definition cameras, infrared imaging equipment, laser radar, etc., which are used to collect images, temperature data and three-dimensional structure information. It has autonomous flight control capabilities and can automatically inspect the overhead lines in the wind farm area according to the preset routes. It is also equipped with GPS and inertial navigation systems to achieve precise positioning and route control. The intelligent recognition module can analyze the images collected by the UAV based on machine learning and deep learning algorithms (such as convolutional neural network CNN, target detection algorithm, etc.) to identify the overhead lines on the overhead lines. Defects, such as aging of insulators, rusting of hardware, wear and tear of lines, looseness, entanglement of foreign objects, etc.; the data processing and analysis module is responsible for processing the data collected by the drone, including image enhancement, denoising, data format conversion, etc., and generating inspection reports through data analysis and statistics, listing the location, type, severity and other information of defects, to help maintenance personnel formulate maintenance plans; the cloud platform is used for centralized processing and storage of data. The cloud uses big data and artificial intelligence technology to analyze historical data and current inspection data, and predict possible defects and faults in the future; the control platform provides a graphical interface, allowing operation and maintenance personnel to view the inspection progress, analysis results and defect information in real time.

[0004] The application document with publication number CN117291872A discloses a system and method for identifying defect content during unmanned aerial vehicle line inspection, including a central control module, and a data acquisition module and a defect warning module connected thereto; the defect warning module warns of the identified equipment defects through an intelligent terminal; it is characterized in that: the central control module: identifies a number of power transmission equipment on the transmission line to be inspected, and matches the equipment defects corresponding to the number of power transmission equipment; evaluates and sorts the equipment defects based on historical data, and obtains a defect catalog of the number of power transmission equipment; extracts the equipment defects in the defect catalog in turn, and sets an acquisition feature sequence for the equipment defects; wherein the historical data includes historical inspection data and historical environmental data; and, selects a number of target defects corresponding to each power transmission equipment from the defect catalog, and integrates the number of target defects and the corresponding acquisition feature sequences into a target feature sequence; controls the data acquisition module to acquire image data based on the target feature sequence, and determines whether there is an equipment defect in the power transmission equipment through the image data; wherein the acquisition feature sequence includes the defect position, the acquisition position and the image type.

[0005] The existing technology lacks safety monitoring of the UAV's flight status, which affects the accuracy of image acquisition and further affects the accuracy of defect identification. Summary of the invention

[0006] The purpose of the present invention is to monitor the flight status of a UAV so as to improve the accuracy of image acquisition. In view of the above-mentioned shortcomings, an intelligent recognition system for defects of overhead lines in wind farms based on autonomous inspection by UAVs is proposed.

[0007] The present invention adopts the following technical solution:

[0008] An intelligent recognition system for defects of overhead lines in wind farms based on autonomous inspection by unmanned aerial vehicles, the system comprising an unmanned aerial vehicle control module, a data processing module, an image acquisition module, an image processing module, a defect recognition module, a data storage module and a user interaction module; the unmanned aerial vehicle control module is used to set the inspection path, take-off and target area of ​​the unmanned aerial vehicle; the data processing module is used to collect flight data and environmental data of the unmanned aerial vehicle during the inspection process, obtain the safe flight factor of the unmanned aerial vehicle according to the flight data and the environmental data, obtain information on whether the flight of the unmanned aerial vehicle is safe according to the safe flight factor of the unmanned aerial vehicle, and transmit the information on whether the flight of the unmanned aerial vehicle is safe to the user interaction module; the user interaction module provides a visual interface, and after the user inquires about the information on whether the flight of the unmanned aerial vehicle is safe, a signal to start and stop the image acquisition module is issued; the image acquisition module performs shooting according to the start and stop signal; the image processing module preprocesses the collected images; the defect recognition module inputs the preprocessed images into the machine learning model for defect classification and recognition, and outputs the type and location of the defects; the data storage module is used to store the type and location of the defects, and transmit the type and location of the defects to the user interaction module.

[0009] Optionally, the data processing module includes an information storage submodule, an environment detection submodule, a flight detection submodule, a control submodule and a safety determination submodule; the information storage submodule is used to store the corresponding air pressure of the drone under an ideal working state, the corresponding speed of the drone under an ideal working state, the sea level air pressure, the molar mass of the air, the gravitational acceleration, the gas constant, the maximum flight altitude of the drone under an ideal working state, the minimum flight altitude of the drone under an ideal working state, the air density, the drag coefficient of the drone and the windward area of ​​the drone, and transmit them to the control submodule; the environment detection submodule is used to detect and obtain the absolute temperature of the environment, the real-time wind speed and the wind direction angle of the drone detected each time, and transmit them to the control submodule; the flight detection submodule is used to detect and obtain the airspeed of the drone detected by the airspeed meter, the real-time flight altitude of the drone, the heading angle of the drone and the speed of the drone detected by the GPS, and transmit them to the control submodule; the control submodule obtains the angle between the drone and the wind direction according to the wind direction angle of the drone and the heading angle of the drone, and obtains the angle between the drone and the wind direction according to the drag coefficient of the drone, the air density, the windward area of ​​the drone and the speed of the drone detected by the GPS The air resistance index is obtained according to the speed of the drone, the real-time wind speed, the drone airspeed detected by the airspeed meter, the angle between the drone and the wind direction and the air resistance index are used to obtain the real-time speed index of the drone, the total number of environmental absolute temperature detections is obtained according to the real-time flight altitude of the drone, the maximum flight altitude of the drone under an ideal working state and the minimum flight altitude of the drone under an ideal working state, the average value of the environmental absolute temperature is obtained according to the total number of environmental absolute temperature detections and the environmental absolute temperature detected each time, the air pressure index of the drone at the real-time altitude is obtained according to the sea level pressure, the molar mass of the air, the gravitational acceleration, the real-time flight altitude of the drone, the gas constant and the average value of the environmental absolute temperature, the safe flight factor of the drone is obtained according to the air pressure index of the drone at the real-time altitude, the corresponding air pressure of the drone under an ideal working state, the real-time speed index of the drone and the corresponding speed of the drone under an ideal working state, and the safe flight factor of the drone is transmitted to the safety determination submodule; the safety determination submodule obtains information on whether the drone flight is safe according to the safe flight factor of the drone, and transmits the information on whether the drone flight is safe to the user interaction module.

[0010] Optionally, the environmental detection submodule includes a temperature detection unit and an anemometer; the temperature detection unit is used to detect and obtain the absolute temperature of the environment detected each time, and transmit it to the control submodule; the anemometer is used to detect and obtain the real-time wind speed and wind direction angle of the drone, and transmit it to the control submodule.

[0011] Optionally, the flight detection submodule includes an airspeed meter and a positioning unit; the airspeed meter is used to detect and obtain the airspeed of the UAV detected by the airspeed meter, and transmit it to the control submodule; the positioning unit is used to detect and obtain the real-time altitude of the UAV flight, the heading angle of the UAV and the speed of the UAV detected by GPS, and transmit them to the control submodule.

[0012] Optionally, when the control submodule calculates the safe flight factor of the drone, the following formula is satisfied:

[0013]

[0014] Among them, Hth is the safe flight factor of the UAV, AP is the air pressure index of the UAV at the real-time altitude, ap is the air pressure corresponding to the UAV under ideal working conditions, VEV is the real-time speed index of the UAV, and vev is the speed corresponding to the UAV under ideal working conditions; when the safe flight factor of the UAV is greater than or equal to the selection threshold of the safe flight factor of the UAV, it indicates that the UAV flight is unsafe; when the safe flight factor of the UAV is less than the selection threshold of the safe flight factor of the UAV, it indicates that the UAV flight is safe.

[0015] The beneficial effects achieved by the present invention are:

[0016] 1. Real-time monitoring of the UAV flight status, calculation of the safe flight factor based on flight data and environmental data, ensuring the safety of the UAV inspection process and improving the acquisition accuracy of the image acquisition module;

[0017] 2. The user interaction module provides users with a visual interface. Users can view the flight safety information and inspection defect information of the drone in real time and control the start or stop of the image acquisition module as needed, which improves the interactivity and feedback capabilities of the system;

[0018] 3. The data storage module saves the defect types and locations detected during the inspection process to facilitate subsequent data analysis and trend prediction.

[0019] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0021] Figure 2 It is a structural diagram of the data processing module in the present invention;

[0022] Figure 3It is a structural schematic diagram of the environment detection submodule in the present invention;

[0023] Figure 4 It is a structural schematic diagram of the flight detection submodule in the present invention;

[0024] Figure 5 is a relationship diagram in Embodiment 1 of the present invention;

[0025] Figure 6 This is a schematic diagram of the overall structure of Embodiment 2 of the present invention;

[0026] Figure 7 is a structural diagram of an image quality analysis module in Embodiment 2 of the present invention;

[0027] Figure 8 This is a relationship diagram in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0028] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual dimensions. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0029] Embodiment 1: This embodiment provides an intelligent identification system for wind farm overhead line defects based on autonomous inspection by drones. Figures 1 to 5 shown.

[0030] An intelligent recognition system for defects of overhead lines in wind farms based on autonomous inspection by unmanned aerial vehicles, the system comprising an unmanned aerial vehicle control module, a data processing module, an image acquisition module, an image processing module, a defect recognition module, a data storage module and a user interaction module; the unmanned aerial vehicle control module is used to set the inspection path, take-off and target area of ​​the unmanned aerial vehicle; the data processing module is used to collect flight data and environmental data of the unmanned aerial vehicle during the inspection process, obtain the safe flight factor of the unmanned aerial vehicle according to the flight data and the environmental data, obtain information on whether the flight of the unmanned aerial vehicle is safe according to the safe flight factor of the unmanned aerial vehicle, and transmit the information on whether the flight of the unmanned aerial vehicle is safe to the user interaction module; the user interaction module provides a visual interface, and after the user inquires about the information on whether the flight of the unmanned aerial vehicle is safe, a signal to start and stop the image acquisition module is issued; the image acquisition module performs shooting according to the start and stop signal; the image processing module preprocesses the collected images; the defect recognition module inputs the preprocessed images into the machine learning model for defect classification and recognition, and outputs the type and location of the defects; the data storage module is used to store the type and location of the defects, and transmit the type and location of the defects to the user interaction module.

[0031] Optionally, the data processing module includes an information storage submodule, an environment detection submodule, a flight detection submodule, a control submodule and a safety determination submodule; the information storage submodule is used to store the corresponding air pressure of the drone under an ideal working state, the corresponding speed of the drone under an ideal working state, the sea level air pressure, the molar mass of the air, the gravitational acceleration, the gas constant, the maximum flight altitude of the drone under an ideal working state, the minimum flight altitude of the drone under an ideal working state, the air density, the drag coefficient of the drone and the windward area of ​​the drone, and transmit them to the control submodule; the environment detection submodule is used to detect and obtain the absolute temperature of the environment, the real-time wind speed and the wind direction angle of the drone detected each time, and transmit them to the control submodule; the flight detection submodule is used to detect and obtain the airspeed of the drone detected by the airspeed meter, the real-time flight altitude of the drone, the heading angle of the drone and the speed of the drone detected by the GPS, and transmit them to the control submodule; the control submodule obtains the angle between the drone and the wind direction according to the wind direction angle of the drone and the heading angle of the drone, and obtains the angle between the drone and the wind direction according to the drag coefficient of the drone, the air density, the windward area of ​​the drone and the speed of the drone detected by the GPS The air resistance index is obtained according to the speed of the drone, the real-time wind speed, the drone airspeed detected by the airspeed meter, the angle between the drone and the wind direction and the air resistance index are used to obtain the real-time speed index of the drone, the total number of environmental absolute temperature detections is obtained according to the real-time flight altitude of the drone, the maximum flight altitude of the drone under an ideal working state and the minimum flight altitude of the drone under an ideal working state, the average value of the environmental absolute temperature is obtained according to the total number of environmental absolute temperature detections and the environmental absolute temperature detected each time, the air pressure index of the drone at the real-time altitude is obtained according to the sea level pressure, the molar mass of the air, the gravitational acceleration, the real-time flight altitude of the drone, the gas constant and the average value of the environmental absolute temperature, the safe flight factor of the drone is obtained according to the air pressure index of the drone at the real-time altitude, the corresponding air pressure of the drone under an ideal working state, the real-time speed index of the drone and the corresponding speed of the drone under an ideal working state, and the safe flight factor of the drone is transmitted to the safety determination submodule; the safety determination submodule obtains information on whether the drone flight is safe according to the safe flight factor of the drone, and transmits the information on whether the drone flight is safe to the user interaction module.

[0032] Specifically, when the safety determination submodule makes a determination, the following principles are referred to: when the safety flight factor of the drone is greater than or equal to the selection threshold of the safety flight factor of the drone, it indicates that the drone flight is unsafe; when the safety flight factor of the drone is less than the selection threshold of the safety flight factor of the drone, it indicates that the drone flight is safe; the selection threshold of the safety flight factor of the drone is set by technical personnel in this field. Safe flight conditions ensure that the drone can effectively perform the inspection tasks of the overhead lines of wind farms, identify defects in a timely and accurate manner, and thus improve the safety and efficiency of maintenance. Therefore, it is crucial to closely integrate the calculation of the drone's safety flight factor with the inspection system to achieve efficient and intelligent drone inspections. According to the information that the drone flight is unsafe, the number of defect detections can be adjusted accordingly or the detection can be suspended, and the subsequent specific operation steps are set by technical personnel in this field.

[0033] Optionally, the environmental detection submodule includes a temperature detection unit and an anemometer; the temperature detection unit is used to detect and obtain the absolute temperature of the environment detected each time, and transmit it to the control submodule; the anemometer is used to detect and obtain the real-time wind speed and wind direction angle of the drone, and transmit it to the control submodule.

[0034] Optionally, the flight detection submodule includes an airspeed meter and a positioning unit; the airspeed meter is used to detect and obtain the airspeed of the UAV detected by the airspeed meter, and transmit it to the control submodule; the positioning unit is used to detect and obtain the real-time altitude of the UAV flight, the heading angle of the UAV and the speed of the UAV detected by GPS, and transmit them to the control submodule.

[0035] Optionally, when the control submodule calculates the safe flight factor of the drone, the following formula is satisfied:

[0036]

[0037] Among them, Hth is the safe flight factor of the UAV, AP is the air pressure index of the UAV at the real-time altitude, ap is the air pressure corresponding to the UAV under ideal working conditions, VEV is the real-time speed index of the UAV, and vev is the speed corresponding to the UAV under ideal working conditions.

[0038] Optionally, the following formula is satisfied when the control submodule is calculated:

[0039]

[0040] VEV=V air -V wind ×cos(θ)-F;

[0041]

[0042] Among them, qy is the sea level pressure, m is the molar mass of air, g is the acceleration of gravity, h is the real-time altitude of the drone, r is the gas constant, and Δtemp is the average absolute temperature of the environment;

[0043] temp a is the ath detected ambient absolute temperature, and A is the total number of ambient absolute temperature detections;

[0044] h min h is the maximum altitude of the drone under ideal working conditions. max The minimum altitude at which the drone can fly under ideal working conditions;

[0045] V air is the airspeed of the UAV detected by the airspeed meter, V wind is the real-time wind speed, θ is the angle between the drone and the wind direction, and F is the air resistance index;

[0046] ρ is the air density, zl is the drag coefficient of the drone, s is the windward area of ​​the drone, v gps is the speed of the drone detected by GPS.

[0047] Refer to the relationship between the drone's air pressure index at real-time altitude / drone's real-time speed index and the drone's safe flight factor. The lighter the color (closer to yellow), the greater the value of the drone's safe flight factor, and the darker the color (closer to blue), the smaller the value of the drone's safe flight factor. It can be seen from the figure that the drone's air pressure index at real-time altitude / drone's real-time speed index and the drone's safe flight factor are not in a strictly proportional relationship, but show a certain fluctuation. This volatility may have the following reasons: Due to the formula It is not a direct linear relationship. The absolute value and fractional operations in the formula will introduce certain nonlinear effects. In theory, the air pressure index of the drone at the real-time altitude / the real-time speed index of the drone has a positive impact on the drone's safe flight factor. However, due to the influence of factors such as nonlinear formulas, ideal value settings, and product effects, the drone's safe flight factor in the figure fluctuates. This fluctuation does not completely violate the direct proportional trend, but is affected by the formula structure and parameter relationship.

[0048] When controlling the submodule calculation, refer to the following code:

[0049]

[0050]

[0051]

[0052] Specifically, the "ideal working state" mentioned in "the air pressure corresponding to the UAV under the ideal working state" and "the speed corresponding to the UAV under the ideal working state" can be understood as the operating state of the UAV under the designed or expected optimal flight conditions. In this state, the sensors built into the UAV are less disturbed and can perform their tasks efficiently and safely, such as inspection, shooting or other operations. The air pressure corresponding to the UAV under the ideal working state and the speed corresponding to the UAV under the ideal working state are obtained through query; the unit of the air pressure corresponding to the UAV under the ideal working state is hectopascals; the unit of the speed corresponding to the UAV under the ideal working state is meters per second.

[0053] The value of sea level pressure is 1013.25 hPa; the unit of molar mass of air is kilogram per mole; the value of gravity acceleration is 9.81 meters per square second; the unit of real-time altitude of the drone is meter; the value of gas constant is 287 J / (kg\cdotpK); the unit of average value of absolute ambient temperature is Kelvin.

[0054] The unit of the ambient absolute temperature detected each time is Kelvin.

[0055] The units of the maximum altitude and the minimum altitude of the drone under ideal working conditions are both meters. The maximum altitude and the minimum altitude of the drone under ideal working conditions are obtained through query.

[0056] The units of the drone airspeed and real-time wind speed detected by the airspeed meter are both meters per second; the angle between the drone and the wind direction can be understood as the angle between the drone's flight direction and the wind speed. The following example shows that when the wind direction is a headwind, that is, the wind direction is opposite to the drone's forward direction, the corresponding angle between the drone and the wind direction is 0°; when the wind direction is a tailwind, that is, the wind direction is the same as the drone's forward direction, the corresponding angle between the drone and the wind direction is 180°; the air resistance index can be understood as the impact of air resistance on the drone's flight.

[0057] The drag coefficient of a drone can be measured through a wind tunnel test. For example, a scaled-down model of the drone is made, a wind tunnel is selected, the wind tunnel is started to adjust the wind speed, multiple wind speed points are set for testing, and the corresponding values ​​are calculated by collecting data. The unit of the windward area of ​​the drone is square meters. Since drones are complex shapes, they can be simulated using computer-aided design (CAD) software and computational fluid dynamics (CFD) software. A three-dimensional model of the drone is constructed using CAD software (such as SolidWorks, AutoCAD, etc.). In the CFD software, the CAD model is imported and the boundary conditions of the fluid, such as flow velocity and direction, are set. The model is meshed for numerical calculations. The simulation is run in the CFD software to obtain the airflow distribution and pressure field. The software usually provides functions to calculate the windward area, or directly extracts data based on the projected area of ​​the model. The unit of the drone's speed detected by GPS is meters per second.

[0058] The above units are only examples, and those skilled in the art may set different units according to actual needs when implementing this solution.

[0059] This embodiment solves the problem of poor image acquisition accuracy of traditional recognition systems. Through the mutual cooperation of the UAV control module, data processing module, image acquisition module, image processing module, defect recognition module, data storage module and user interaction module, real-time monitoring of the UAV flight status is carried out, and the safe flight factor is calculated based on the flight data and environmental data to ensure the safety of the UAV inspection process and improve the acquisition accuracy of the image acquisition module.

[0060] Embodiment 2: This embodiment includes all the contents of Embodiment 1, and provides an intelligent identification system for wind farm overhead line defects based on autonomous inspection by drones. Figures 6 to 8 shown.

[0061] When the user inquires about the flight safety information of the drone, the user sends a signal to start the image acquisition module. At this time, the image acquisition module works, and the subsequent image processing module and defect recognition module also work synchronously to realize the defect recognition operation.

[0062] The wind farm overhead line defect intelligent identification system based on autonomous inspection by unmanned aerial vehicles also includes an image quality analysis module; the image quality analysis module is used to analyze the image and obtain information related to image brightness, color, effective feature points and edge strength, obtain image quality comprehensive evaluation indicators based on the information related to image brightness, color, effective feature points and edge strength, obtain information on whether the image quality is qualified based on the image quality comprehensive evaluation indicators, and transmit the information on whether the image quality is qualified to the user interaction module; the user interaction module queries the information on whether the image quality is qualified through a visual interface, and after the user queries the information on whether the image quality is qualified, a signal to start and stop the image processing module and the defect recognition module is issued. When the image quality is qualified, the image processing module and the defect recognition module continue to work, otherwise the image processing module and the defect recognition module are suspended.

[0063] Optionally, the image quality analysis module includes an ideal value setting submodule, an image parameter detection submodule, a data analysis submodule and an image quality determination submodule; the ideal value setting submodule is used to set an ideal average value of image brightness and transmit it to the data analysis submodule; the image parameter detection submodule is used to detect and obtain the edge strength of the image, the total number of valid feature points in the image, the standard deviation of the image brightness, the measured average value of the image brightness, the maximum color component in the image, the minimum color component in the image, the maximum value of the brightness in the image and the minimum value of the brightness in the image, and transmit them to the data analysis submodule; the data analysis submodule obtains the contrast of the image according to the maximum value of the brightness in the image and the minimum value of the brightness in the image, obtains the comprehensive image quality evaluation index according to the edge strength of the image, the total number of valid feature points in the image, the standard deviation of the image brightness, the measured average value of the image brightness, the ideal average value of the image brightness, the contrast of the image, the maximum color component in the image and the minimum color component in the image, and transmits the comprehensive image quality evaluation index to the image quality determination submodule; the image quality determination submodule obtains information on whether the image quality is qualified according to the comprehensive image quality evaluation index, and transmits the information on whether the image quality is qualified to the user interaction module.

[0064] Specifically, when the image quality determination submodule makes a determination, the following principles are referred to: when the comprehensive image quality evaluation index is greater than or equal to the selection threshold of the comprehensive image quality evaluation index, it indicates that the image quality is qualified; when the comprehensive image quality evaluation index is less than the selection threshold of the comprehensive image quality evaluation index, it indicates that the image quality is unqualified; the selection threshold of the comprehensive image quality evaluation index is set by technical personnel in this field; judging whether the image quality is qualified or unqualified directly affects the subsequent defect recognition effect and recognition accuracy. The functions of image quality assessment include: ensuring data accuracy: high-quality images can more clearly display the defect characteristics (such as breakage, rust, cracks, etc.) on the line, thereby improving the accuracy of the recognition algorithm; avoiding misjudgment and missed judgment: low-quality images may cause the algorithm to misjudge or miss defects, reducing the reliability of the system; improving system efficiency: by evaluating image quality, the system can re-collect in time when unqualified images are collected, avoiding subsequent analysis and processing of invalid data, and improving overall efficiency.

[0065] Optionally, when the data analysis submodule calculates the comprehensive evaluation index of image quality, the following formula is satisfied:

[0066]

[0067] Τ=I max -I min .

[0068] Among them, Q is the comprehensive evaluation index of image quality, E is the edge strength of the image, N is the total number of valid feature points in the image, σ is the standard deviation of image brightness, μ s is the measured average value of image brightness, μ r is the ideal average value of image brightness, Τ is the image contrast, c max is the maximum color component in the image, c min is the minimum color component in the image;

[0069] I max is the maximum brightness in the image, I min is the minimum brightness value in the image.

[0070] The relationship between the total number of valid feature points in the reference image and the comprehensive evaluation index of image quality shows that as the total number of valid feature points in the image increases, the comprehensive evaluation index of image quality shows an increasing trend. This shows that increasing the number of valid feature points in the image will improve the comprehensive evaluation index of image quality.

[0071] When calculating the data analysis submodule, refer to the following code:

[0072]

[0073]

[0074] Specifically, the ideal average value of image brightness is set by technical personnel in this field; the calculation of "maximum color component in image" and "minimum brightness in image" can measure the intensity and purity of color, and color saturation can evaluate the quality of image; the total number of effective feature points in image can reflect the richness of information in image, and the larger the total number of effective feature points in image means that the image contains more details and recognizable elements, thereby improving the quality of image; the measured average value of image brightness can reflect the overall brightness level of image, when the measured average value of image brightness is too large, it will cause the image to be overexposed, when the measured average value of image brightness is too small, it will cause the image recognition to decrease; the standard deviation of image brightness can indicate the degree of brightness deviation from the mean, the larger the standard deviation of image brightness is, it usually means that there are more light-dark contrast in the image, and the brightness changes are obvious, such as obvious shadows, bright areas, etc. Such images often have high contrast, rich details, and can present more layering; with respect to image contrast, high contrast means that the brightness difference of image is significant, and objects and background can be clearly distinguished, making the image more vivid.

[0075] The above units are only examples, and those skilled in the art may set different units according to actual needs when implementing this solution.

[0076] This embodiment solves the problem that the traditional recognition system has a relatively single analysis. By analyzing the brightness, color, feature points and edge strength of the image, the system can filter out low-quality images to ensure that the images entering the defect recognition module are clear and complete in details.

[0077] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. Intelligent identification system for overhead line defects in wind farms based on autonomous inspection by drones, characterized by: The system includes a drone control module, a data processing module, an image acquisition module, an image processing module, a defect recognition module, a data storage module and a user interaction module; The drone control module is used to set the drone inspection path, take-off and target area; The data processing module is used to collect the flight data and environmental data of the drone during the inspection process, obtain the safe flight factor of the drone based on the flight data and environmental data, obtain information on whether the drone is safe to fly based on the safe flight factor of the drone, and transmit the information on whether the drone is safe to fly to the user interaction module; The user interaction module provides a visual interface, and after the user inquires whether the drone is safe to fly, a signal is sent to start and stop the image acquisition module; The image acquisition module performs shooting according to the start-stop signal; The image processing module pre-processes the collected image; The defect recognition module inputs the preprocessed image into the machine learning model for defect classification and recognition, and outputs the type and location of the defect; The data storage module is used to store the type and location of the defect, and transmit the type and location of the defect to the user interaction module.

2. The wind farm overhead line defect intelligent identification system based on autonomous inspection by unmanned aerial vehicles according to claim 1 is characterized in that: The data processing module includes an information storage submodule, an environment detection submodule, a flight detection submodule, a control submodule and a safety determination submodule; The information storage submodule is used to store the corresponding air pressure of the drone under an ideal working state, the corresponding speed of the drone under an ideal working state, the sea level air pressure, the molar mass of the air, the gravitational acceleration, the gas constant, the maximum flight altitude of the drone under an ideal working state, the minimum flight altitude of the drone under an ideal working state, the air density, the drag coefficient of the drone and the windward area of ​​the drone, and transmit them to the control submodule; The environment detection submodule is used to detect and obtain the absolute temperature of the environment, the real-time wind speed and the wind direction angle of the drone each time it is detected, and transmit them to the control submodule; The flight detection submodule is used to detect and obtain the airspeed of the drone detected by the airspeed meter, the real-time altitude of the drone, the heading angle of the drone and the speed of the drone detected by the GPS, and transmit them to the control submodule; The control submodule obtains the angle between the drone and the wind direction according to the wind direction angle and the heading angle of the drone, obtains the air resistance index according to the drag coefficient of the drone, the air density, the windward area of ​​the drone and the speed of the drone detected by GPS, obtains the real-time speed index of the drone according to the real-time wind speed, the airspeed of the drone detected by the airspeed meter, the angle between the drone and the wind direction and the air resistance index, obtains the total number of environmental absolute temperature detections according to the real-time flight altitude of the drone, the maximum flight altitude of the drone under an ideal working state and the minimum flight altitude of the drone under an ideal working state, obtains the average value of the environmental absolute temperature according to the total number of environmental absolute temperature detections and the environmental absolute temperature detected each time, obtains the air pressure index of the drone at the real-time altitude according to the sea level pressure, the molar mass of the air, the gravitational acceleration, the real-time flight altitude of the drone, the gas constant and the average value of the environmental absolute temperature, obtains the safe flight factor of the drone according to the air pressure index of the drone at the real-time altitude, the corresponding air pressure of the drone under an ideal working state, the real-time speed index of the drone and the corresponding speed of the drone under an ideal working state, and transmits the safe flight factor of the drone to the safety determination submodule; The safety determination submodule obtains information on whether the drone is safe to fly based on the safe flight factor of the drone, and transmits the information on whether the drone is safe to fly to the user interaction module.

3. The wind farm overhead line defect intelligent identification system based on autonomous inspection by unmanned aerial vehicles according to claim 2, characterized in that: The environmental detection submodule includes a temperature detection unit and an anemometer; The temperature detection unit is used to detect and obtain the absolute temperature of the environment detected each time, and transmit it to the control submodule; The anemometer is used to detect and obtain the real-time wind speed and wind direction angle of the drone, and transmit them to the control submodule.

4. The wind farm overhead line defect intelligent identification system based on autonomous inspection by unmanned aerial vehicles according to claim 3 is characterized in that: The flight detection submodule includes an airspeed meter and a positioning unit; The airspeed meter is used to detect and obtain the airspeed of the drone detected by the airspeed meter, and transmit it to the control submodule; The positioning unit is used to detect and obtain the real-time altitude of the UAV, the heading angle of the UAV and the speed of the UAV detected by GPS, and transmit them to the control submodule.

5. The wind farm overhead line defect intelligent identification system based on autonomous inspection by unmanned aerial vehicles according to claim 4 is characterized in that: When the control submodule calculates the safe flight factor of the drone, the following formula is satisfied: Among them, Hth is the safe flight factor of the UAV, AP is the air pressure index of the UAV at the real-time altitude, ap is the air pressure corresponding to the UAV under ideal working conditions, VEV is the real-time speed index of the UAV, and vev is the speed corresponding to the UAV under ideal working conditions; when the safe flight factor of the UAV is greater than or equal to the selection threshold of the safe flight factor of the UAV, it indicates that the UAV flight is unsafe; when the safe flight factor of the UAV is less than the selection threshold of the safe flight factor of the UAV, it indicates that the UAV flight is safe.

Images