Column equipment spraying control method and device based on unmanned aerial vehicle

Abstract

The invention discloses a column equipment spraying control method and device based on an unmanned aerial vehicle. The method comprises the steps that the unmanned aerial vehicle is controlled to collect an image of column equipment and recognize a to-be-sprayed point; controlling the unmanned aerial vehicle to collect flight test data and environment test parameters of the to-be-sprayed point; according to the flight test data and the environment test parameters of the to-be-sprayed point, an initial attitude deflection angle of the unmanned aerial vehicle during spraying is obtained through calculation; the unmanned aerial vehicle is controlled to fly to the hovering area of the to-be-sprayed point to hover at the initial attitude deflection angle, and spraying equipment is started to spray the to-be-sprayed point; real-time flight parameters and real-time environment data are collected in the spraying process, and the spraying direction of spraying equipment and the length of a spray head telescopic rod are adjusted according to the real-time environment data; and calculating a real-time attitude deflection angle of the unmanned aerial vehicle according to the real-time flight parameters and the real-time environment data, and adjusting the attitude of the unmanned aerial vehicle according to the real-time attitude deflection angle. According to the method, the spraying accuracy of the column equipment can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a method and device for controlling spraying of pole-mounted equipment based on unmanned aerial vehicles. Background Art

[0002] Traditional 10kV pole-mounted equipment maintenance usually relies on manual inspections, which has problems such as low recognition accuracy, low efficiency, and high safety risks. In addition, pole-mounted equipment is usually located at high altitudes, requiring personnel to perform high-altitude operations, which poses serious safety risks such as falling and electric shock. The safety and reliability of power equipment are crucial to the operation of the power system. As an important part of the power system, the stability and maintenance efficiency of 10kV pole-mounted equipment directly affect the operation of the power grid. Improving maintenance efficiency, reducing the risk of personal injury, and reducing maintenance costs have long been an urgent need of the power industry. In recent years, drone technology has been increasingly used in various fields. Its high degree of autonomy, flexibility, and operability make it an ideal choice for solving the challenges of high-altitude equipment maintenance. With the development of drone technology hardware and the reduction of costs, as well as the advancement of related software algorithms, it has become more feasible to achieve precise control and operation of drones in complex environments.

[0003] For example, patent text CN112520039B discloses a high-altitude spraying coaxial twin-propeller UAV combined with a mechanical arm and a control method, which includes a frame body with chamfers formed at the four corners of the frame body, and connecting arms extending outward are provided at the chamfer positions. A rotor assembly is installed at the outer end of the connecting arm. By operating the combination of the four-rotor coaxial twin-propeller UAV and the mechanical arm, the UAV can replace the operating workers to complete the high-altitude spraying operation, thereby reducing the risks caused by the complex on-site environment.

[0004] When UAVs are operating in the air, they need to maintain stable hovering and positioning in complex environments. However, they are easily displaced and shaken by the influence of wind, recoil and the magnetic field strength on the 10kv column. Although the existing UAV platforms can perform basic flight and operation, they still have deficiencies in fine control and anti-interference. The spraying equipment will cause unstable repair quality under the influence of injection pressure and recoil, and may miss some small but important defects, affecting the long-term operation safety of the equipment. The spraying process will be affected by wind speed and wind direction, which will lead to incomplete spraying, resulting in missing coating, or uneven spraying, resulting in waste of spraying materials. Summary of the invention

[0005] The present invention provides a method and device for controlling spraying of pole-mounted equipment based on an unmanned aerial vehicle, which can solve the problem of low spraying quality of pole-mounted equipment caused by the influence of wind force, recoil force and magnetic field strength.

[0006] A method for controlling spraying of column-mounted equipment based on an unmanned aerial vehicle, comprising:

[0007] Controlling the drone to collect images of the equipment on the pole, identifying points to be sprayed according to the images of the equipment on the pole, and storing the positions of the points to be sprayed;

[0008] Control the drone to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed;

[0009] According to the flight test data and the environmental test parameters of the point to be sprayed, the initial attitude deflection angle of the UAV during spraying is calculated;

[0010] Control the drone to fly to the hovering area of ​​the point to be sprayed, hover at the initial posture deflection angle, and start the spraying equipment to spray the point to be sprayed;

[0011] During the spraying process, real-time flight parameters and real-time environmental data are collected, and the spraying direction and the length of the telescopic rod of the spraying device are adjusted according to the real-time environmental data;

[0012] The real-time attitude deflection angle of the UAV is calculated according to the real-time flight parameters and the real-time environmental data, and the attitude of the UAV is adjusted according to the real-time attitude deflection angle.

[0013] Further, identifying the point to be sprayed according to the pole-mounted equipment image includes:

[0014] Preprocessing the pole-mounted device image to obtain a preprocessed image;

[0015] The preprocessed image is input into a pre-trained neural network model for feature extraction and defect classification to obtain insulation defect points, which are used as points to be sprayed.

[0016] Furthermore, the flight test data includes multiple spraying recoil forces, hovering jitter frequencies and test deflection angles; the environmental test data includes the heights of multiple points to be sprayed, wind speeds at the locations of the points to be sprayed and magnetic field strengths at the locations of the points to be sprayed;

[0017] According to the flight test data and the environmental test parameters of the spraying point, the initial attitude deflection angle of the UAV during spraying is calculated, including:

[0018] The spraying recoil, the hovering vibration frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are squared to obtain a quadratic term;

[0019] The spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are mutually multiplied in pairs to obtain an interaction term;

[0020] Using the least square method to perform regression analysis on the multiple quadratic terms, interaction terms and multiple test deflection angles, obtain regression coefficients, and fit error terms;

[0021] According to the quadratic term, the interaction term, the regression coefficient and the error term, a relationship between the deflection angle of the drone and the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed is obtained;

[0022] The initial posture deflection angle is obtained by calculation based on the relationship according to the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed.

[0023] Furthermore, the relationship is:

[0024] The deflection angle of the UAV is equal to the product of the initial deflection angle of the UAV, each regression coefficient and the corresponding quadratic term and interaction term, and the sum of the error term.

[0025] Further, the real-time environmental data includes real-time wind direction; adjusting the spraying direction and the length of the telescopic rod of the spraying equipment according to the real-time environmental data includes:

[0026] Determine whether the current spraying direction is windward or leeward according to the real-time wind direction;

[0027] When the current spraying direction is the windward direction, detect whether the current spraying direction is parallel to the real-time wind direction. If the current spraying direction is parallel to the real-time wind direction, determine that there is no need to adjust the length of the telescopic rod of the nozzle. If the current spraying direction is not parallel to the real-time wind direction, perform a secondary judgment on whether the length of the telescopic rod of the nozzle needs to be adjusted. If the length of the telescopic rod of the nozzle needs to be adjusted, adjust the length of the telescopic rod of the nozzle according to the real-time environmental data.

[0028] When the current spraying direction is leeward, based on the pre-stored positions of the points to be sprayed, determine whether there is a point to be sprayed upwind of the current spraying direction. If there is a point to be sprayed, adjust the current spraying direction to the upwind direction to spray the corresponding point to be sprayed. If there is no point to be sprayed, control the drone to fly to the hovering area of ​​the next point to be sprayed.

[0029] Furthermore, the real-time environmental data also includes the diameter of the point to be sprayed and the real-time wind speed;

[0030] A second determination is made as to whether the length of the telescopic rod of the sprinkler head needs to be adjusted, and if the length of the telescopic rod of the sprinkler head needs to be adjusted, the length of the telescopic rod of the sprinkler head is adjusted according to the real-time environmental data, including:

[0031] Obtain the initial position coordinates of the nozzle of the spraying equipment and the position coordinates of the point to be sprayed;

[0032] Calculating the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment;

[0033] Calculate the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed;

[0034] Calculating an offset spraying range according to the spraying offset coordinates;

[0035] Determine whether the offset spraying range covers the point to be sprayed according to the offset spraying range, if it covers, determine that there is no need to adjust the length of the telescopic rod of the nozzle, if it does not cover, calculate the adjustment amount of the telescopic rod length according to the offset spraying range and the wind direction angle;

[0036] The length of the telescopic rod of the nozzle is adjusted according to the adjustment amount of the telescopic rod length.

[0037] Further, the spraying offset coordinates are calculated according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed, including:

[0038] Calculating the distance between the nozzle and the point to be sprayed according to the nozzle initial position coordinates and the position coordinates of the point to be sprayed;

[0039] Calculate the time it takes for the spray material to reach the spray point based on the distance between the nozzle and the spray point and the initial speed of the spray material;

[0040] The spraying offset coordinates are calculated based on the real-time wind speed, the time when the spraying material reaches the point to be sprayed, the wind direction angle, and the initial position coordinates of the spray head.

[0041] Furthermore, the adjustment amount of the telescopic rod length is the ratio of the difference between the radius of the point to be sprayed and the offset spraying range to the tangent value of the wind direction angle.

[0042] Furthermore, the real-time flight parameters include real-time spraying recoil and real-time shaking frequency, and the real-time environmental data also includes real-time wind speed and real-time magnetic field strength at the location of the spraying point;

[0043] Calculating a real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjusting the attitude of the UAV according to the real-time attitude deflection angle, including:

[0044] Based on the relationship, the real-time attitude deflection angle of the drone is calculated according to the real-time spraying recoil, the real-time shaking frequency, the real-time wind speed, the real-time magnetic field strength at the location of the spraying point, and the height of the spraying point;

[0045] Compare the real-time attitude deflection angle with the initial attitude deflection angle, and if the real-time attitude deflection angle is equal to the initial attitude deflection angle, control the UAV to maintain the initial attitude deflection angle;

[0046] If the real-time attitude deflection angle is less than the initial attitude deflection angle, calculating a first difference between the real-time attitude deflection angle and the initial attitude deflection angle, and controlling the UAV to increase the first difference from the real-time attitude deflection angle to the initial attitude deflection angle;

[0047] If the real-time attitude deflection angle is not equal to the initial attitude deflection angle, the current deflection angle of the UAV is adjusted to the real-time attitude deflection angle.

[0048] A column-mounted equipment spraying control device based on an unmanned aerial vehicle, comprising:

[0049] The acquisition module is used to control the drone to acquire images of the equipment on the pole, identify the points to be sprayed according to the images of the equipment on the pole, and store the positions of the points to be sprayed;

[0050] The test module is used to control the UAV to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed;

[0051] A first calculation module is used to calculate the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the spraying point;

[0052] A spraying control module is used to control the UAV to fly to the hovering area of ​​the point to be sprayed, hover at the initial attitude deflection angle, and start the spraying equipment to spray the point to be sprayed;

[0053] An adjustment module, used for collecting real-time flight parameters and real-time environmental data during the spraying process, and adjusting the spraying direction of the spraying equipment and the length of the spray head telescopic rod according to the real-time environmental data;

[0054] The second calculation module is used to calculate the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjust the attitude of the UAV according to the real-time attitude deflection angle.

[0055] Furthermore, the acquisition module identifies the point to be sprayed according to the image of the pole-mounted equipment, including:

[0056] Preprocessing the pole-mounted device image to obtain a preprocessed image;

[0057] The preprocessed image is input into a pre-trained neural network model for feature extraction and defect classification to obtain insulation defect points, which are used as points to be sprayed.

[0058] Further, the flight test data includes multiple spraying recoil forces, hovering jitter frequencies and multiple test deflection angles; the environmental test data includes the heights of multiple points to be sprayed, the wind speeds at the locations of the points to be sprayed and the magnetic field strengths at the locations of the points to be sprayed;

[0059] The first calculation module calculates the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the spraying point, including:

[0060] The spraying recoil, the hovering vibration frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are squared to obtain a quadratic term;

[0061] The spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are mutually multiplied in pairs to obtain an interaction term;

[0062] Using the least square method to perform regression analysis on the multiple quadratic terms, interaction terms and multiple test deflection angles, obtain regression coefficients, and fit error terms;

[0063] According to the quadratic term, the interaction term, the regression coefficient and the error term, a relationship between the deflection angle of the drone and the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed is obtained;

[0064] The initial posture deflection angle is obtained by calculation based on the relationship according to the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed.

[0065] Furthermore, the relationship is:

[0066] The deflection angle of the UAV is equal to the product of the initial deflection angle of the UAV, each regression coefficient and the corresponding quadratic term and interaction term, and the sum of the error term.

[0067] Furthermore, the real-time environmental data includes real-time wind direction; the adjustment module adjusts the spraying direction and the length of the spray head telescopic rod of the spraying equipment according to the real-time environmental data, including:

[0068] Determine whether the current spraying direction is windward or leeward according to the real-time wind direction;

[0069] When the current spraying direction is the windward direction, detect whether the current spraying direction is parallel to the real-time wind direction. If the current spraying direction is parallel to the real-time wind direction, determine that there is no need to adjust the length of the telescopic rod of the nozzle. If the current spraying direction is not parallel to the real-time wind direction, perform a secondary judgment on whether the length of the telescopic rod of the nozzle needs to be adjusted. If the length of the telescopic rod of the nozzle needs to be adjusted, adjust the length of the telescopic rod of the nozzle according to the real-time environmental data.

[0070] When the current spraying direction is leeward, based on the pre-stored positions of the points to be sprayed, determine whether there is a point to be sprayed upwind of the current spraying direction. If there is a point to be sprayed, adjust the current spraying direction to the upwind direction to spray the corresponding point to be sprayed. If there is no point to be sprayed, control the drone to fly to the hovering area of ​​the next point to be sprayed.

[0071] Furthermore, the real-time environmental data also includes the diameter of the point to be sprayed and the real-time wind speed;

[0072] The adjustment module performs a secondary judgment on whether the length of the telescopic rod of the sprinkler head needs to be adjusted, and when the length of the telescopic rod of the sprinkler head needs to be adjusted, the length of the telescopic rod of the sprinkler head is adjusted according to the real-time environmental data, including:

[0073] Obtain the initial position coordinates of the nozzle of the spraying equipment and the position coordinates of the point to be sprayed;

[0074] Calculating the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment;

[0075] Calculate the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed;

[0076] Calculating an offset spraying range according to the spraying offset coordinates;

[0077] Determine whether the offset spraying range covers the point to be sprayed according to the offset spraying range, if it covers, determine that there is no need to adjust the length of the telescopic rod of the nozzle, if it does not cover, calculate the adjustment amount of the telescopic rod length according to the offset spraying range and the wind direction angle;

[0078] The length of the telescopic rod of the nozzle is adjusted according to the adjustment amount of the telescopic rod length.

[0079] Furthermore, the adjustment module calculates the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed, including:

[0080] Calculating the distance between the nozzle and the point to be sprayed according to the nozzle initial position coordinates and the position coordinates of the point to be sprayed;

[0081] Calculate the time it takes for the spray material to reach the spray point based on the distance between the nozzle and the spray point and the initial speed of the spray material;

[0082] The spraying offset coordinates are calculated based on the real-time wind speed, the time when the spraying material reaches the point to be sprayed, the wind direction angle, and the initial position coordinates of the spray head.

[0083] Furthermore, the adjustment amount of the telescopic rod length is the ratio of the difference between the radius of the point to be sprayed and the offset spraying range to the tangent value of the wind direction angle.

[0084] Furthermore, the real-time flight parameters include real-time spraying recoil and real-time shaking frequency, and the real-time environmental data also includes real-time wind speed and real-time magnetic field strength at the location of the spraying point;

[0085] The second calculation module calculates the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjusts the attitude of the UAV according to the real-time attitude deflection angle, including:

[0086] Based on the relationship, the real-time attitude deflection angle of the drone is calculated according to the real-time spraying recoil, the real-time shaking frequency, the real-time wind speed, the real-time magnetic field strength at the location of the spraying point, and the height of the spraying point;

[0087] Compare the real-time attitude deflection angle with the initial attitude deflection angle, and if the real-time attitude deflection angle is equal to the initial attitude deflection angle, control the UAV to maintain the initial attitude deflection angle;

[0088] If the real-time attitude deflection angle is not equal to the initial attitude deflection angle, the current deflection angle of the UAV is adjusted to the real-time attitude deflection angle.

[0089] The column-mounted equipment spraying control method and device based on drone provided by the present invention have at least the following beneficial effects:

[0090] (1) The relationship between the spraying recoil of the drone, the drone's vibration frequency, wind speed, magnetic field strength, the height of the spraying point and the deflection angle of the drone is analyzed. The attitude deflection angle is obtained from the relationship between the spraying recoil of the drone, the hovering vibration frequency of the drone, the height of the spraying point, the wind speed at the spraying point, and the magnetic field strength at the spraying point. The attitude deflection angle is dynamically adjusted so that the drone can maintain the best spraying angle during the operation, thereby improving the accuracy and uniformity of the spraying. The real-time adjustment of the attitude deflection angle can effectively compensate for the attitude deviation caused by environmental changes, enhance the stability of the drone during the operation, and reduce the problem of uneven spraying caused by vibration or wind speed changes. The automatic adjustment mechanism of the system reduces the need for manual intervention, making the operation more efficient and accurate, while also reducing the risk of human operation errors.

[0091] (2) The introduction of polynomial regression and its quadratic and interactive terms to predict the deflection angle can effectively compensate for the deviations that may occur in the actual operation of the UAV, ensure that the spraying material accurately covers the spraying point, reduce errors, improve the accuracy of attitude control, reduce the waste of spraying materials, and ensure that each spraying point can evenly obtain enough spraying material, thereby improving the overall repair quality; by predicting and setting the attitude deflection angle, the flight control system of the UAV can better cope with the dynamic changes in actual operations and improve the flight stability and control accuracy;

[0092] (3) The automatic adjustment of the telescopic rod length of the spray equipment nozzle can be calculated according to the real-time wind direction angle and the diameter of the insulation defect point, and the corresponding adjustment can be made, which can ensure that the spraying point is fully sprayed to avoid missed spraying or excessive spraying, thereby improving the maintenance effect and efficiency. Without manual intervention, the automated design can reduce the workload of operators and improve the consistency and accuracy of operations. It can reduce the spraying error caused by wind changes, ensure that the drone operation is carried out within a safe range, and improve the safety, accuracy and efficiency of operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 A flow chart of an embodiment of a method for controlling spraying of a pole-mounted device based on an unmanned aerial vehicle provided by the present invention.

[0094] Figure 2 A schematic structural diagram of an embodiment of a UAV-based column-mounted equipment spray control device provided by the present invention. DETAILED DESCRIPTION

[0095] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0096] refer to Figure 1In some embodiments, a method for controlling spraying of a column-mounted device based on a drone is provided, comprising:

[0097] S1. Control the drone to collect images of the equipment on the column, identify the points to be sprayed according to the images of the equipment on the column, and store the positions of the points to be sprayed;

[0098] S2, control the UAV to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed;

[0099] S3, calculating the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the point to be sprayed;

[0100] S4, controlling the drone to fly to the hovering area of ​​the point to be sprayed, hovering at the initial posture deflection angle, and starting the spraying equipment to spray the point to be sprayed;

[0101] S5, collecting real-time flight parameters and real-time environmental data during the spraying process, and adjusting the spraying direction of the spraying equipment and the length of the spray head telescopic rod according to the real-time environmental data;

[0102] S6. Calculate the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjust the attitude of the UAV according to the real-time attitude deflection angle.

[0103] Further, in step S1, identifying the point to be sprayed according to the pole-mounted equipment image includes:

[0104] S11, preprocessing the pole-mounted device image to obtain a preprocessed image;

[0105] S12, inputting the pre-processed image into a pre-trained neural network model for feature extraction and defect classification, obtaining insulation defect points, and using the insulation defect points as points to be sprayed.

[0106] Specifically, in step S11, the pole-mounted device image is preprocessed, including but not limited to using a filtering algorithm to remove noise in the image, applying contrast enhancement, histogram equalization, etc.

[0107] In step S12, the neural network model is trained, which may specifically include: labeling the insulation defect points in the training data set, providing supervision information for the deep learning model, and training the deep learning model using the labeled data. The model learns how to extract features from images and identify defects, evaluates the accuracy of the model on the validation set and the test set, and adjusts the model parameters to improve performance.

[0108] In addition, edge detection algorithms can be used to identify possible defect edges in the preprocessed image, and compared with templates of known defects to identify whether similar defects exist in the image.

[0109] Further, in step S2, the flight test data includes multiple spraying recoil forces, hovering jitter frequencies and test deflection angles; the environmental test data includes the heights of multiple points to be sprayed, the wind speeds at the points to be sprayed and the magnetic field strengths at the points to be sprayed.

[0110] Specifically, during the spraying process, the material sprayed by the spraying equipment will exert a reaction force on the drone, namely, recoil. This force will cause the drone's attitude to change, affecting the accuracy and uniformity of the spraying. Understanding the spraying recoil will help adjust the drone's attitude and spraying parameters to ensure stability and spraying quality during the spraying process. The spraying recoil is obtained by measuring the recoil changes of the drone using sensors (such as force sensors or inertial measurement units IMU) during the spraying process. The drone's hovering jitter frequency reflects the tiny vibrations or swings of the drone during stable flight. These jitters affect the uniformity of the spraying. Analyzing the hovering jitter frequency can help adjust the drone's attitude control system and reduce jitter, thereby improving the accuracy of the spraying. The hovering jitter frequency is obtained by obtaining the drone's vibration data through the inertial measurement unit IMU and the acceleration sensor, and analyzing its frequency and amplitude to obtain it. The height of the point to be sprayed affects the spraying distance and spray distribution. The spraying effects at different heights may be different, affecting the uniformity of insulation repair. The height of the point to be sprayed is measured by the altimeter or The laser rangefinder obtains the height data for acquisition; the wind speed at the location of the spraying point will affect the direction and uniformity of the spray, and strong wind will cause the spray to deviate or be uneven. Adjusting the spraying direction and speed according to the wind speed can ensure that the spraying material reaches the spraying point accurately. The wind speed at the location of the spraying point is obtained by measuring the wind speed on the drone or on site using an anemometer or meteorological sensor; the magnetic field strength at the location of the spraying point affects the magnetic sensor of the drone, and then affects its attitude control system. Understanding the magnetic field strength can adjust the control algorithm of the drone to reduce the interference of the magnetic field on flight control. The magnetic field strength at the location of the spraying point is obtained by obtaining the magnetic field strength data of the environment through the magnetic field sensor; the deflection angle of the drone affects the accuracy of spraying, specifically affecting the spraying direction of the nozzle and the distribution of the spray. The spraying angle and the direction of the nozzle are adjusted according to the attitude change to ensure that the spraying effect at each insulation defect point is consistent. In this process, the drone obtains the test deflection angle of the drone during flight through the attitude sensor (such as gyroscope and accelerometer) on the drone.

[0111] Further, in step S3, the initial attitude deflection angle of the UAV during spraying is calculated based on the flight test data and the environmental test parameters of the spraying point, including:

[0112] S31, taking square values ​​of the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed, respectively, to obtain a quadratic term;

[0113] S32, performing mutual multiplication of the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed, to obtain an interaction term;

[0114] S33, using the least square method to perform regression analysis on the multiple quadratic terms, interaction terms and multiple test deflection angles, obtain regression coefficients, and fit error terms;

[0115] S34, according to the quadratic term, the interaction term, the regression coefficient and the error term, a relationship between the deflection angle of the drone and the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the location of the point to be sprayed and the magnetic field strength at the location of the point to be sprayed is obtained;

[0116] S35, obtaining the initial posture deflection angle by calculation based on the relationship according to the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed.

[0117] Specifically, in step S31, the spraying recoil force A1 is squared to obtain the spraying recoil force quadratic term A1 2 , take the square value of the hovering jitter frequency A2 to obtain the quadratic term of the hovering jitter frequency A2 2 , take the square value of the height A3 of the spray point to be sprayed, and obtain the quadratic term A3 of the height of the spray point to be sprayed 2 , take the square value of the wind speed A4 at the spraying point to obtain the quadratic term A4 of the wind speed at the spraying point 2 , take the square value of the magnetic field intensity A5 at the location of the spraying point to obtain the quadratic term A5 of the magnetic field intensity at the location of the spraying point 2 .

[0118] Further, in step S32, the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are multiplied in pairs, specifically:

[0119] Multiply the spraying recoil A1 and the hovering jitter frequency A2 to obtain the first interaction term (A1*A2);

[0120] Multiply the spraying recoil force A1 and the height A3 of the point to be sprayed to obtain the second interaction term (A1*A3);

[0121] The third interaction term (A1*A4) is obtained by multiplying the spraying recoil force A1 and the wind speed A4 at the location of the spraying point;

[0122] The fourth interaction term (A1*A5) is obtained by multiplying the spraying recoil force A1 and the magnetic field strength A5 at the location of the spraying point;

[0123] Multiply the hovering vibration frequency A2 and the height A3 of the point to be sprayed to obtain the fifth interaction term (A2*A3);

[0124] Multiply the hovering vibration frequency A2 and the wind speed A4 at the location of the spraying point to obtain the sixth interaction term (A2*A4);

[0125] The seventh interaction term (A2*A7) is obtained by multiplying the hovering vibration frequency A2 and the magnetic field strength A5 at the location of the spraying point;

[0126] The eighth interaction term (A3*A4) is obtained by multiplying the height A3 of the point to be sprayed and the wind speed A4 at the position of the point to be sprayed.

[0127] The ninth interaction term (A3*A5) is obtained by multiplying the height A3 of the point to be sprayed and the magnetic field strength A5 at the position of the point to be sprayed.

[0128] The wind speed A4 at the location of the spray point and the magnetic field strength A5 at the location of the spray point are multiplied to obtain the tenth interaction term (A4*A5);

[0129] In step S33, the regression coefficient is obtained by regression analysis of spraying recoil, hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, the magnetic field strength at the point to be sprayed, the corresponding quadratic term, and the interaction term with the test deflection angle of the drone by the least squares method. The regression coefficient can also be obtained by performing polynomial regression using the statsmodels or scikit-learn library in Python. The error term represents the difference or residual between the model prediction value and the actual observation value. The actual data usually contains noise (such as measurement error, random fluctuation), which will introduce error terms, resulting in differences between the actual observation value and the predicted value of the relational expression, which is specifically obtained by fitting the existing data.

[0130] Further, in step S34, the relationship is:

[0131] The deflection angle of the UAV is equal to the deflection angle of the UAV in the initial state, the product of each regression coefficient and the corresponding quadratic term and interaction term, and the sum of the error term.

[0132] Specifically, the relationship is expressed by the following formula:

[0133]

[0134] Among them, A1 is the spray recoil, A1 2 is the quadratic term of spray recoil, A2 is the hovering vibration frequency, A2 2 is the quadratic term of the hovering vibration frequency, A3 is the height of the point to be sprayed, A3 2 is the quadratic term of the height of the point to be sprayed, A4 is the wind speed at the point to be sprayed, 2 is the quadratic term of wind speed at the location of the spraying point, A5 is the magnetic field strength at the location of the spraying point, A5 2 is the quadratic term of the magnetic field intensity at the location of the spray point, (A1*A2) is the first interaction term; (A1*A3) is the second interaction term, (A1*A4) is the third interaction term, (A1*A5) is the fourth interaction term, (A2*A3) is the fifth interaction term; (A2*A4) is the sixth interaction term; (A2*A7) is the seventh interaction term, (A3*A4) is the eighth interaction term, (A3*A5) is the ninth interaction term; (A4*A5) is the tenth interaction term, β1, β2, β3…β 20 represents the regression coefficient, represents the error term, β0 represents the starting deflection angle of the UAV, and θ represents the deflection angle to be calculated.

[0135] Through polynomial regression and its quadratic and interaction terms, the effect and accuracy of drones in spraying operations can be significantly improved, ensuring that the spraying materials accurately reach the points to be sprayed and reducing uneven spraying. By predicting the deflection angle required by the drone during the spraying process and taking its inverse as the attitude deflection angle for actual operation, the deviations that may occur in the actual operation of the drone can be effectively compensated to ensure that the spraying materials accurately cover the points to be sprayed and reduce spraying errors. Accurate attitude control can reduce the waste of spraying materials and ensure that each point to be sprayed can evenly obtain enough spraying materials, thereby improving the overall spraying quality. By predicting and setting the attitude deflection angle, the flight control system of the drone can better cope with dynamic changes in actual operations, improve flight stability and control accuracy, and through automated prediction and adjustment mechanisms, reduce the need for manual intervention, reduce operational complexity, and improve operational efficiency and safety.

[0136] Further, in step S4, the drone is controlled to fly to the hovering area of ​​the point to be sprayed and hover at the initial attitude deflection angle, and the spraying equipment is turned on to spray the point to be sprayed. For example, assuming that when the drone is performing a spraying task, the spraying recoil is 3.5N, the hovering jitter frequency is 1.2Hz, the height of the point to be sprayed is 15m, the wind speed at the point to be sprayed is 6m / s, and the magnetic field strength is 50μT, the drone deflection angle is calculated by formula (1) to be 5°, that is, the initial attitude deflection angle of the drone should be set to 5° to compensate for the predicted deviation. When performing the spraying operation, the drone hovers according to the set initial attitude deflection angle, which can effectively compensate for the deviation caused by environmental changes or operation errors, thereby ensuring that the spraying material accurately covers the point to be sprayed.

[0137] Further, in step S5, the real-time environmental data includes real-time wind direction; adjusting the spraying direction and the length of the spray head telescopic rod of the spraying equipment according to the real-time environmental data includes:

[0138] S51, judging whether the current spraying direction is windward or leeward according to the real-time wind direction;

[0139] S52, when the current spraying direction is the windward direction, detecting whether the current spraying direction is parallel to the real-time wind direction, if the current spraying direction is parallel to the real-time wind direction, determining that there is no need to adjust the length of the telescopic rod of the nozzle, if the current spraying direction is not parallel to the real-time wind direction, performing a secondary judgment on whether the length of the telescopic rod of the nozzle needs to be adjusted, and if the length of the telescopic rod of the nozzle needs to be adjusted, adjusting the length of the telescopic rod of the nozzle according to the real-time environmental data;

[0140] S53, when the current spraying direction is leeward, determine whether there is a point to be sprayed in the upwind direction of the current spraying direction according to the pre-stored positions of each point to be sprayed; if there is a point to be sprayed, adjust the current spraying direction to the upwind direction to spray the corresponding point to be sprayed; if there is no point to be sprayed, control the drone to fly to the hovering area of ​​the next point to be sprayed.

[0141] Specifically, in step S51, the specific method for determining the windward direction or the leeward direction is: when the angle between the spraying direction of the drone and the real-time wind direction is within 180°, it is determined to be the windward direction; when the angle between the spraying direction of the drone and the real-time wind direction exceeds 180°, it is determined to be the leeward direction. By identifying whether the current spraying direction of the drone is the windward direction or the leeward direction according to the real-time wind direction, the spraying direction of the nozzle can be effectively adjusted to ensure that the spraying material can be accurately sprayed to the point to be sprayed, thereby improving the accuracy and efficiency of the spraying.

[0142] Furthermore, in step S52 and step S53, when the current spraying direction of the UAV is the upwind direction, if it is determined that the current direction is parallel to the real-time wind direction, the current spraying task can be continued without adjusting the length of the telescopic rod of the nozzle, thereby avoiding unnecessary adjustment of the telescopic rod and reducing waste of spraying materials and processing time; when the current spraying direction of the UAV is the leeward direction, it is identified whether there are points to be sprayed on the windward side, thereby avoiding spraying materials to areas that do not need to be sprayed, reducing unnecessary impact on the environment and waste of resources; real-time wind direction recognition and direction classification can automatically adjust the working order of the UAV without manual intervention, thereby improving the real-time and automation level of the operation.

[0143] Assume that there are multiple devices on a 10kv column that need to be sprayed. Initially, the drone is repairing the insulation of a certain device. The real-time monitoring shows that the wind direction is southwest-southwest, and the real-time wind direction is 30 degrees. The current spraying direction of the drone is northwest, which is 60 degrees different from the real-time wind direction, and belongs to the windward direction. The current direction is not parallel to the real-time wind direction angle, so a second judgment is required. When the current spraying direction of the drone is southeast, it belongs to the leeward direction. It recognizes that there is a point to be sprayed on the west side of the equipment corresponding to the current insulation repair task, and controls the drone to spray and repair the west side of the column to ensure that the spraying material can be effectively sprayed on the point to be sprayed.

[0144] The specific method of obtaining real-time wind direction is: obtaining it through the real-time wind direction information released by the weather station, or using the GPS device carried by the drone to obtain the wind direction of the current location.

[0145] Furthermore, the real-time environmental data also includes the diameter of the point to be sprayed and the real-time wind speed;

[0146] In step S52, a second determination is made as to whether the length of the telescopic rod of the sprinkler head needs to be adjusted. If the length of the telescopic rod of the sprinkler head needs to be adjusted, the length of the telescopic rod of the sprinkler head is adjusted according to the real-time environmental data, including:

[0147] S521, obtaining the initial position coordinates of the nozzle of the spraying equipment and the position coordinates of the point to be sprayed;

[0148] S522, calculating the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment;

[0149] S523, calculating the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed;

[0150] S524, calculating an offset spraying range according to the spraying offset coordinates;

[0151] S525, judging whether the offset spraying range covers the point to be sprayed according to the offset spraying range, and if so, determining that there is no need to adjust the length of the telescopic rod of the spray head, and if not, calculating the adjustment amount of the telescopic rod length according to the offset spraying range and the wind direction angle;

[0152] S526, adjusting the length of the telescopic rod of the nozzle according to the telescopic rod length adjustment amount.

[0153] Specifically, in step S521, the position coordinates of the point to be sprayed are (X1, Y1), and the initial position coordinates of the nozzle of the spraying equipment are (X0, Y0).

[0154] In step S522, the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment is α.

[0155] In step S523, the initial length of the telescopic rod of the nozzle is L0, the diameter of the point to be sprayed is D, the real-time wind speed is V, and the initial speed of the spraying material is V0.

[0156] According to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the sprayed material, the wind direction angle and the real-time wind speed, the spray offset coordinates are calculated, including:

[0157] According to the initial position coordinates of the nozzle (X0, Y0), the position coordinates of the point to be sprayed (X1, Y1) and the initial speed V0 of the spraying material, calculate the distance L1 between the nozzle and the point to be sprayed: Thus, the time t when the spraying material reaches the point to be sprayed is calculated:

[0158] According to the real-time wind speed V and the time t when the spray material arrives at the spraying point, the offset P is calculated first: P = V*t, and then the spray offset coordinates (X f , Y f ):X f =X0+P*cosα,Y f =Y0+P*sinα; that is, the coordinates of the drop point of the sprayed material after the offset are (X f , Y f ).

[0159] Further, in step S524, according to the spraying offset coordinate (X f , Y f ), calculate the offset spray range R f : R f is the spray radius after offset.

[0160] Step S525: Determine whether the offset spraying range covers the point to be sprayed, and compare the offset spraying range R fThe radius of the spray point is D / 2. If R f ≥D / 2, the offset spray range can cover the area of ​​the spray point, and there is no need to adjust the length of the nozzle telescopic rod. If R f <D / 2, the length of the telescopic rod of the nozzle needs to be adjusted.

[0161] Specifically, the telescopic rod length adjustment ΔL is calculated as the ratio of the difference between the radius D / 2 of the point to be sprayed and the offset spray range Rf to the tangent value of the wind direction angle, specifically:

[0162]

[0163] △R=D / 2-R f .

[0164] The specific method for obtaining the diameter D of the point to be sprayed is: the drone collects the image of the point to be sprayed, and the size of the point to be sprayed is analyzed by image processing software. Professional power equipment detection instruments or infrared imaging technology can also be used to help accurately measure the size of the point to be sprayed.

[0165] In step S526, the length of the telescopic rod of the nozzle is adjusted according to the telescopic rod length adjustment amount. Specifically, when the offset spraying range is smaller than the radius of the point to be sprayed, the telescopic rod is extended by an amount of △L.

[0166] Assume that in an actual drone spraying operation, the initial distance for the spraying material to reach the point to be sprayed is 1 meter, the initial telescopic rod length is 0.5 meter, the wind direction angle is 45°, the nozzle spray angle is 30°, the wind speed is 2 meters per second, the initial velocity of the spraying material is 10 meters per second, the diameter of the point to be sprayed is 0.2 meters, and the radius is 0.1 meters. The initial spraying range is calculated to be 0.29 meters, and the offset of the wind speed to the spray landing point is 0.2 meters. Further calculation shows that the offset spraying range radius is 0.14 meters. Comparing the offset spraying range radius with the radius of the point to be sprayed, because the offset spraying range radius is larger than the radius of the point to be sprayed, the nozzle telescopic rod does not need to be adjusted.

[0167] Further, in step S6, the real-time flight parameters include real-time spraying recoil and real-time jitter frequency, and the real-time environmental data also includes real-time wind speed and real-time magnetic field strength at the location of the spraying point;

[0168] Calculating a real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjusting the attitude of the UAV according to the real-time attitude deflection angle, including:

[0169] S61, based on the relationship, calculate the real-time attitude deflection angle of the drone according to the real-time spraying recoil, the real-time shaking frequency, the real-time wind speed, the real-time magnetic field strength at the location of the spraying point, and the height of the spraying point;

[0170] S62, comparing the real-time attitude deflection angle with the initial attitude deflection angle, and if the real-time attitude deflection angle is equal to the initial attitude deflection angle, controlling the UAV to maintain the initial attitude deflection angle;

[0171] S63: If the real-time attitude deflection angle is not equal to the initial attitude deflection angle, the current deflection angle of the UAV is adjusted to the real-time attitude deflection angle.

[0172] Assume that the real-time data of the drone when performing the spraying task is as follows: the real-time spraying recoil is 3.4N, the real-time jitter frequency is 1.1Hz, the real-time wind speed is 7m / s, the real-time magnetic field strength is 52μT, and the height of the spraying point is 20m. The real-time attitude deflection angle of the drone is calculated to be 4°. Assuming that the initial attitude deflection angle set before is 5°, the real-time attitude deflection angle is smaller than the initial attitude deflection angle. According to the design, the current deflection angle of the drone is adjusted to 4° for spraying operation to ensure that the spraying operation is more precise and adapt to the attitude deviation caused by real-time environmental changes.

[0173] When the drone completes the spraying operation on all the points to be sprayed, a report can be generated. The report specifically includes: the detection position of the spraying point of the pole equipment, the number of detection points of the spraying point of the pole equipment, the area of ​​the spraying point, the spraying material, and the spraying repair results.

[0174] The report records the specific location of each spray point, which is helpful for subsequent maintenance and inspection. Each spray point can be accurately located to ensure the comprehensiveness of the maintenance work. The number of spray points detected and the area of ​​each defect are recorded to help evaluate the overall repair needs and workload. The spray materials used are recorded to help evaluate the effect and applicability of the materials, ensure that the selected materials meet the repair requirements, and facilitate future maintenance and quality control. The spray repair results can evaluate whether the repair has achieved the expected effect and promptly discover and solve any possible problems. The drone automatically generates reports, saving time for manual recording and report writing, and improving work efficiency.

[0175] refer to Figure 2 In some embodiments, a spray control device for a column-mounted device based on a drone is also provided, comprising:

[0176] The acquisition module 201 is used to control the drone to acquire the image of the equipment on the pole, identify the point to be sprayed according to the image of the equipment on the pole, and store the position of the point to be sprayed;

[0177] The test module 202 is used to control the UAV to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed;

[0178] The first calculation module 203 is used to calculate the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the spraying point;

[0179] The spraying control module 204 is used to control the UAV to fly to the hovering area of ​​the point to be sprayed, hover at the initial posture deflection angle, and start the spraying equipment to spray the point to be sprayed;

[0180] An adjustment module 205 is used to collect real-time flight parameters and real-time environmental data during the spraying process, and adjust the spraying direction and the length of the spray head telescopic rod of the spraying equipment according to the real-time environmental data;

[0181] The second calculation module 206 is used to calculate the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjust the attitude of the UAV according to the real-time attitude deflection angle.

[0182] Furthermore, the acquisition module 201 identifies the point to be sprayed according to the pole-mounted equipment image, including:

[0183] Preprocessing the pole-mounted device image to obtain a preprocessed image;

[0184] The preprocessed image is input into a pre-trained neural network model for feature extraction and defect classification to obtain insulation defect points, which are used as points to be sprayed.

[0185] Furthermore, the flight test data includes multiple spraying recoil forces, hovering jitter frequencies and test deflection angles; the environmental test data includes the heights of multiple points to be sprayed, wind speeds at the locations of the points to be sprayed and magnetic field strengths at the locations of the points to be sprayed;

[0186] The first calculation module 203 calculates the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the spraying point, including:

[0187] The spraying recoil, the hovering vibration frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are squared to obtain a quadratic term;

[0188] The spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are mutually multiplied in pairs to obtain an interaction term;

[0189] Using the least square method to perform regression analysis on the multiple quadratic terms, interaction terms and multiple test deflection angles, obtain regression coefficients, and fit error terms;

[0190] According to the quadratic term, the interaction term, the regression coefficient and the error term, a relationship between the deflection angle of the drone and the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed is obtained;

[0191] The initial posture deflection angle is obtained by calculation based on the relationship according to the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed.

[0192] Furthermore, the relationship is:

[0193] The deflection angle of the UAV is equal to the product of the initial deflection angle of the UAV, each regression coefficient and the corresponding quadratic term and interaction term, and the sum of the error term.

[0194] Furthermore, the real-time environmental data includes real-time wind direction; the adjustment module 205 adjusts the spraying direction and the length of the spray head telescopic rod of the spraying equipment according to the real-time environmental data, including:

[0195] Determine whether the current spraying direction is windward or leeward according to the real-time wind direction;

[0196] When the current spraying direction is the windward direction, detect whether the current spraying direction is parallel to the real-time wind direction. If the current spraying direction is parallel to the real-time wind direction, determine that there is no need to adjust the length of the telescopic rod of the nozzle. If the current spraying direction is not parallel to the real-time wind direction, perform a secondary judgment on whether the length of the telescopic rod of the nozzle needs to be adjusted. If the length of the telescopic rod of the nozzle needs to be adjusted, adjust the length of the telescopic rod of the nozzle according to the real-time environmental data.

[0197] When the current spraying direction is leeward, based on the pre-stored positions of the points to be sprayed, determine whether there is a point to be sprayed upwind of the current spraying direction. If there is a point to be sprayed, adjust the current spraying direction to the upwind direction to spray the corresponding point to be sprayed. If there is no point to be sprayed, control the drone to fly to the hovering area of ​​the next point to be sprayed.

[0198] Furthermore, the real-time environmental data also includes the diameter of the point to be sprayed and the real-time wind speed;

[0199] The adjustment module 205 performs a secondary determination as to whether the length of the telescopic rod of the sprinkler head needs to be adjusted. If the length of the telescopic rod of the sprinkler head needs to be adjusted, the length of the telescopic rod of the sprinkler head is adjusted according to the real-time environmental data, including:

[0200] Obtain the initial position coordinates of the nozzle of the spraying equipment and the position coordinates of the point to be sprayed;

[0201] Calculating the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment;

[0202] Calculate the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed;

[0203] Calculating an offset spraying range according to the spraying offset coordinates;

[0204] Determine whether the offset spraying range covers the point to be sprayed according to the offset spraying range, if it covers, determine that there is no need to adjust the length of the telescopic rod of the nozzle, if it does not cover, calculate the adjustment amount of the telescopic rod length according to the offset spraying range and the wind direction angle;

[0205] The length of the telescopic rod of the nozzle is adjusted according to the adjustment amount of the telescopic rod length.

[0206] Furthermore, the adjustment module 205 calculates the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed, including:

[0207] Calculating the distance between the nozzle and the point to be sprayed according to the nozzle initial position coordinates and the position coordinates of the point to be sprayed;

[0208] Calculate the time it takes for the spray material to reach the spray point based on the distance between the nozzle and the spray point and the initial speed of the spray material;

[0209] The spraying offset coordinates are calculated based on the real-time wind speed, the time when the spraying material reaches the point to be sprayed, the wind direction angle, and the initial position coordinates of the spray head.

[0210] Furthermore, the adjustment amount of the telescopic rod length is the ratio of the difference between the radius of the point to be sprayed and the offset spraying range to the tangent value of the wind direction angle.

[0211] Furthermore, the real-time flight parameters include real-time spraying recoil and real-time shaking frequency, and the real-time environmental data also includes real-time wind speed and real-time magnetic field strength at the location of the spraying point;

[0212] The second calculation module 206 calculates the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjusts the attitude of the UAV according to the real-time attitude deflection angle, including:

[0213] Based on the relationship, the real-time attitude deflection angle of the drone is calculated according to the real-time spraying recoil, the real-time shaking frequency, the real-time wind speed, the real-time magnetic field strength at the location of the spraying point, and the height of the spraying point;

[0214] Compare the real-time attitude deflection angle with the initial attitude deflection angle, and if the real-time attitude deflection angle is equal to the initial attitude deflection angle, control the UAV to maintain the initial attitude deflection angle;

[0215] If the real-time attitude deflection angle is not equal to the initial attitude deflection angle, the current deflection angle of the UAV is adjusted to the real-time attitude deflection angle.

[0216] Please refer to the method embodiment for the specific working principle, which will not be repeated here.

[0217] The above-mentioned embodiments provide a method and device for controlling spraying of column-mounted equipment based on drones, which have at least the following beneficial effects:

[0218] (1) The relationship between the spraying recoil of the drone, the vibration frequency of the drone, the wind speed, the magnetic field strength, the height of the spraying point and the deflection angle of the drone is analyzed. The attitude deflection angle is obtained from the relationship between the spraying recoil of the drone, the hovering vibration frequency of the drone, the height of the spraying point, the wind speed at the spraying point, and the magnetic field strength at the spraying point. The attitude deflection angle is dynamically adjusted so that the drone can maintain the best spraying angle during the operation, thereby improving the accuracy and uniformity of the spraying. The real-time adjustment of the attitude deflection angle can effectively compensate for the attitude deviation caused by environmental changes, enhance the stability of the drone during the operation, and reduce the problem of uneven spraying caused by vibration or wind speed changes. The automatic adjustment mechanism of the system reduces the need for manual intervention, making the operation more efficient and accurate, while also reducing the risk of human operation errors.

[0219] (2) The introduction of polynomial regression and its quadratic and interactive terms to predict the deflection angle can effectively compensate for the deviations that may occur in the actual operation of the UAV, ensure that the spraying material accurately covers the spraying point, reduce errors, improve the accuracy of attitude control, reduce the waste of spraying materials, and ensure that each spraying point can evenly obtain enough spraying material, thereby improving the overall repair quality; by predicting and setting the attitude deflection angle, the flight control system of the UAV can better cope with the dynamic changes in actual operations and improve the flight stability and control accuracy;

[0220] (3) The automatic adjustment of the telescopic rod length of the spray equipment nozzle can be calculated according to the real-time wind direction angle and the diameter of the insulation defect point, and the corresponding adjustment can be made, which can ensure that the spraying point is fully sprayed to avoid missed spraying or excessive spraying, thereby improving the maintenance effect and efficiency. Without manual intervention, the automated design can reduce the workload of operators and improve the consistency and accuracy of operations. It can reduce the spraying error caused by wind changes, ensure that the drone operation is carried out within a safe range, and improve the safety, accuracy and efficiency of operations.

[0221] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for controlling spraying of column-mounted equipment based on drone, characterized in that: include: Controlling the drone to collect images of the equipment on the pole, identifying points to be sprayed according to the images of the equipment on the pole, and storing the positions of the points to be sprayed; Control the drone to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed; According to the flight test data and the environmental test parameters of the point to be sprayed, the initial attitude deflection angle of the UAV during spraying is calculated; Control the drone to fly to the hovering area of ​​the point to be sprayed, hover at the initial posture deflection angle, and start the spraying equipment to spray the point to be sprayed; During the spraying process, real-time flight parameters and real-time environmental data are collected, and the spraying direction and the length of the telescopic rod of the spraying device are adjusted according to the real-time environmental data; The real-time attitude deflection angle of the UAV is calculated according to the real-time flight parameters and the real-time environmental data, and the attitude of the UAV is adjusted according to the real-time attitude deflection angle.

2. The method according to claim 1, characterized in that Identifying the point to be sprayed according to the image of the equipment on the pole, including: Preprocessing the pole-mounted device image to obtain a preprocessed image; The preprocessed image is input into a pre-trained neural network model for feature extraction and defect classification to obtain insulation defect points, which are used as points to be sprayed.

3. The method according to claim 1, characterized in that: The flight test data includes multiple spraying recoil forces, hovering jitter frequencies and test deflection angles; the environmental test parameters include the heights of multiple points to be sprayed, the wind speeds at the locations of the points to be sprayed and the magnetic field strengths at the locations of the points to be sprayed; According to the flight test data and the environmental test parameters of the spraying point, the initial attitude deflection angle of the UAV during spraying is calculated, including: The spraying recoil, the hovering vibration frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are squared to obtain a quadratic term; The spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed, and the magnetic field strength at the point to be sprayed are mutually multiplied in pairs to obtain an interaction term; Using the least square method to perform regression analysis on the multiple quadratic terms, interaction terms and multiple test deflection angles, obtain regression coefficients, and fit error terms; According to the quadratic term, the interaction term, the regression coefficient and the error term, a relationship between the deflection angle of the drone and the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed is obtained; The initial posture deflection angle is obtained by calculation based on the relationship according to the spraying recoil, the hovering jitter frequency, the height of the point to be sprayed, the wind speed at the point to be sprayed and the magnetic field strength at the point to be sprayed.

4. The method according to claim 3, characterized in that The relationship is: The deflection angle of the UAV is equal to the product of the initial deflection angle of the UAV, each regression coefficient and the corresponding quadratic term and interaction term, and the sum of the error term.

5. The method according to claim 1, characterized in that: The real-time environmental data includes real-time wind direction; adjusting the spraying direction and the length of the telescopic rod of the spraying equipment according to the real-time environmental data includes: Determine whether the current spraying direction is windward or leeward according to the real-time wind direction; When the current spraying direction is the windward direction, detect whether the current spraying direction is parallel to the real-time wind direction. If the current spraying direction is parallel to the real-time wind direction, determine that there is no need to adjust the length of the telescopic rod of the nozzle. If the current spraying direction is not parallel to the real-time wind direction, perform a secondary judgment on whether the length of the telescopic rod of the nozzle needs to be adjusted. If the length of the telescopic rod of the nozzle needs to be adjusted, adjust the length of the telescopic rod of the nozzle according to the real-time environmental data. When the current spraying direction is leeward, based on the pre-stored positions of the points to be sprayed, determine whether there is a point to be sprayed upwind of the current spraying direction. If there is a point to be sprayed, adjust the current spraying direction to the upwind direction to spray the corresponding point to be sprayed. If there is no point to be sprayed, control the drone to fly to the hovering area of ​​the next point to be sprayed.

6. The method according to claim 5, characterized in that The real-time environmental data also includes the diameter of the point to be sprayed and the real-time wind speed; A second determination is made as to whether the length of the telescopic rod of the sprinkler head needs to be adjusted, and if the length of the telescopic rod of the sprinkler head needs to be adjusted, the length of the telescopic rod of the sprinkler head is adjusted according to the real-time environmental data, including: Obtain the initial position coordinates of the nozzle of the spraying equipment and the position coordinates of the point to be sprayed; Calculating the angle between the real-time wind direction and the wind direction of the column supporting the column-mounted equipment; Calculate the spraying offset coordinates according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the spraying material, the wind direction angle and the real-time wind speed; Calculating an offset spraying range according to the spraying offset coordinates; Determine whether the offset spraying range covers the point to be sprayed according to the offset spraying range, if it covers, determine that there is no need to adjust the length of the telescopic rod of the nozzle, if it does not cover, calculate the adjustment amount of the telescopic rod length according to the offset spraying range and the wind direction angle; The length of the telescopic rod of the nozzle is adjusted according to the adjustment amount of the telescopic rod length.

7. The method according to claim 6, characterized in that The spray offset coordinates are calculated according to the initial position coordinates of the nozzle, the position coordinates of the point to be sprayed, the initial speed of the sprayed material, the wind direction angle and the real-time wind speed, including: Calculating the distance between the nozzle and the point to be sprayed according to the nozzle initial position coordinates and the position coordinates of the point to be sprayed; Calculate the time it takes for the spray material to reach the spray point based on the distance between the nozzle and the spray point and the initial speed of the spray material; The spraying offset coordinates are calculated based on the real-time wind speed, the time when the spraying material reaches the point to be sprayed, the wind direction angle, and the initial position coordinates of the spray head.

8. The method according to claim 6, characterized in that The adjustment amount of the telescopic rod length is the ratio of the difference between the radius of the point to be sprayed and the offset spraying range to the tangent value of the wind direction angle.

9. The method according to claim 3, characterized in that: The real-time flight parameters include real-time spraying recoil and real-time jitter frequency, and the real-time environmental data also includes real-time wind speed and real-time magnetic field strength at the location of the spraying point; Calculating a real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjusting the attitude of the UAV according to the real-time attitude deflection angle, including: Based on the relationship, the real-time attitude deflection angle of the drone is calculated according to the real-time spraying recoil, the real-time shaking frequency, the real-time wind speed, the real-time magnetic field strength at the location of the spraying point, and the height of the spraying point; Compare the real-time attitude deflection angle with the initial attitude deflection angle, and if the real-time attitude deflection angle is equal to the initial attitude deflection angle, control the UAV to maintain the initial attitude deflection angle; If the real-time attitude deflection angle is not equal to the initial attitude deflection angle, the current deflection angle of the UAV is adjusted to the real-time attitude deflection angle.

10. A spray control device for column-mounted equipment based on drone, characterized in that: include: The acquisition module is used to control the drone to acquire images of the equipment on the pole, identify the points to be sprayed according to the images of the equipment on the pole, and store the positions of the points to be sprayed; The test module is used to control the UAV to fly to the hovering area corresponding to each point to be sprayed for testing, and collect flight test data and environmental test parameters of the point to be sprayed; A first calculation module is used to calculate the initial attitude deflection angle of the UAV during spraying according to the flight test data and the environmental test parameters of the spraying point; A spraying control module is used to control the UAV to fly to the hovering area of ​​the point to be sprayed, hover at the initial attitude deflection angle, and start the spraying equipment to spray the point to be sprayed; An adjustment module, used for collecting real-time flight parameters and real-time environmental data during the spraying process, and adjusting the spraying direction of the spraying equipment and the length of the spray head telescopic rod according to the real-time environmental data; The second calculation module is used to calculate the real-time attitude deflection angle of the UAV according to the real-time flight parameters and the real-time environmental data, and adjust the attitude of the UAV according to the real-time attitude deflection angle.

Images