CN120803002B

CN120803002B - Unmanned aerial vehicle-based wind driven generator inspection method apparatus, device and medium

Info

Publication number: CN120803002B
Application number: CN202511290448.6A
Authority: CN
Inventors: 张海涛; 谢琦; 艾坤; 包京哲; 刘海峰
Original assignee: Hefei Zhongke Leinao Intelligent Technology Co ltd
Current assignee: Hefei Zhongke Leinao Intelligent Technology Co ltd
Priority date: 2025-09-10
Filing date: 2025-09-10
Publication date: 2025-11-14
Anticipated expiration: 2045-09-10
Also published as: CN120803002A

Abstract

This invention relates to the field of unmanned aerial vehicle (UAV) equipment technology, and discloses a method, apparatus, equipment, and medium for inspecting wind turbines based on UAVs. First, with the UAV reaching its initial cruise position and its battery level within a preset range, an image of the top of the wind turbine is acquired. Then, based on the top image of the wind turbine, the UAV is controlled to fly to the center point of the wind turbine blades and acquire an image of the swept surface. Next, the actual error distance of the swept surface is determined based on the swept surface image. If the actual error distance of the swept surface meets the preset conditions for the swept surface, the reference flight direction of the wind turbine components is determined. Then, based on the component safety distance, the component inspection image overlap rate, and the number of component inspection shots, a sequence set of single-step movement distances for each component and the total movement distance of the component are determined. Finally, images are acquired based on the component reference flight direction, the sequence set of single-step movement distances for each component, and the total movement distance of the component to obtain a set of component inspection images.

Description

Unmanned aerial vehicle-based wind driven generator inspection method apparatus, device and medium

Technical Field

The invention relates to the technical field of unmanned aerial vehicle equipment, in particular to a wind driven generator inspection method, device, equipment and medium based on unmanned aerial vehicle.

Background

With the development of unmanned aerial vehicle technology, it is normal to use unmanned aerial vehicle to patrol and examine wind power equipment.

In the related technology, although the problem of high-altitude safety is partially solved by manually operating the unmanned aerial vehicle to patrol and fly, the operation complexity is high, a professional flying hand is required to control the unmanned aerial vehicle, the flying path depends on manual planning, and the unmanned aerial vehicle is difficult to dynamically adjust under the complex climate condition, so that the image is easy to blur or miss. Therefore, a new unmanned aerial vehicle-based wind driven generator inspection method needs to be provided.

Disclosure of Invention

The embodiments of the present specification aim to solve at least one of the technical problems in the related art to some extent. Therefore, the embodiment of the specification provides a wind driven generator inspection method, device, equipment and medium based on an unmanned aerial vehicle.

The embodiment of the specification provides a wind driven generator inspection method based on an unmanned aerial vehicle, which comprises the following steps:

Under the condition that the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, acquiring a top image of a fan, wherein the cruising initial position comprises coordinates and a safety height of the wind driven generator;

Based on the fan top image, controlling the unmanned aerial vehicle to fly to the center point of a fan blade of the wind driven generator and collecting a fan sweep surface image;

Determining a swept surface actual error distance based on the fan swept surface image;

Determining a component reference flight direction of the wind driven generator under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface;

Determining a single-step moving distance sequence set of the component and a total moving distance of the component based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times;

and acquiring an image based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set.

The embodiment of the specification provides a wind-driven generator inspection device based on unmanned aerial vehicle, the device includes:

The system comprises a fan top image acquisition module, a power supply module and a power supply module, wherein the fan top image acquisition module is used for acquiring a fan top image under the condition that the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, and the cruising initial position comprises coordinates and a safety height of the wind driven generator;

The fan swept surface image acquisition module is used for controlling the unmanned aerial vehicle to fly to the center point of a fan blade of the wind driven generator and acquiring a fan swept surface image based on the fan top image;

The sweep face actual error distance determining module is used for determining the sweep face actual error distance based on the fan sweep face image;

the component reference flight direction determining module is used for determining the component reference flight direction of the wind driven generator under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface;

The component moving data determining module is used for determining a component single-step moving distance sequence set and a component total moving distance based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times;

And the component inspection image acquisition module is used for acquiring images based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set.

The present description embodiment provides a computer device, a memory storing a computer program, and a processor implementing the steps of the method according to any of the embodiments above when the computer program is executed by the processor.

The present description provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method according to any of the above embodiments.

The present description provides a computer program product comprising instructions which, when executed by a processor of a computer device, enable the computer device to perform the steps of the method of any one of the embodiments described above.

In the above-described embodiments, first, when the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, a fan top image is acquired, wherein the cruising initial position includes coordinates and a safety height of the wind power generator. And then, based on the image of the top of the fan, controlling the unmanned aerial vehicle to fly to the center point of the fan blade of the wind driven generator and collecting the image of the sweeping surface of the fan. Next, a swept surface actual error distance is determined based on the fan swept surface image. And under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface, determining the reference flying direction of the component of the wind driven generator. Then, a component single step movement distance series set and a component total movement distance are determined based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times. And finally, acquiring an image based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set. The implementation process dynamically optimizes the flight track and shooting parameters of the unmanned aerial vehicle by fusing the environment data in the fan image and the multi-mode sensor information in real time and combining a visual image recognition algorithm. The detection precision and efficiency under complex scenes are improved while manual intervention is reduced, so that the contradiction between cost and efficiency is effectively solved.

Drawings

Fig. 1 is a flowchart of an inspection method of a wind driven generator based on an unmanned plane according to an embodiment of the present disclosure;

Fig. 2 is a schematic flow chart of controlling the unmanned aerial vehicle to fly to the center point of the fan blade and collecting the image of the sweeping surface of the fan according to the embodiment of the present disclosure;

FIG. 3 is a schematic flow chart of determining an actual error distance of a swept surface according to the embodiments of the present disclosure;

FIG. 4a is a schematic flow chart of determining the coordinates of the foot drop of a blade according to the embodiment of the present disclosure;

FIG. 4b is a schematic diagram of a plurality of key points corresponding to a top image of a blower according to an embodiment of the present disclosure;

FIG. 5 is a schematic flow chart of determining the actual error distance of the top part according to the embodiment of the present disclosure;

FIG. 6a is a schematic flow chart for correcting the actual error distance at the top according to the embodiment of the present disclosure;

FIG. 6b is a top schematic view of the corrected top actual error distance provided by the embodiments of the present disclosure;

fig. 7 is a schematic flow chart of determining an axial flight direction of a wind wheel according to an embodiment of the present disclosure;

fig. 8 is a schematic flow chart of acquiring a windward image of a fan according to an embodiment of the present disclosure;

FIG. 9 is a schematic flow chart of determining the actual error distance of the windward side according to the embodiment of the present disclosure;

FIG. 10a is a schematic flow chart for correcting the actual error distance of the windward side according to the embodiment of the present disclosure;

FIG. 10b is a schematic view of the windward side after correcting the actual error distance of the windward side according to the embodiment of the present disclosure;

FIG. 11 is a schematic flow chart of determining a reference flight direction of a component of a wind turbine according to an embodiment of the present disclosure;

FIG. 12 is a schematic flow chart of determining a single step distance sequence set of blades and a total distance of movement of a component according to an embodiment of the present disclosure;

FIG. 13 is a schematic flow chart of determining a single step distance sequence set of tower and total distance of tower movement provided in an embodiment of the present disclosure;

fig. 14a is a schematic flow chart of implementing blade inspection according to the embodiment of the present disclosure;

FIG. 14b is a schematic view of a inspection wind turbine blade and tower provided in embodiments of the present disclosure;

fig. 15 is a schematic flow chart of implementing tower inspection according to an embodiment of the present disclosure;

FIG. 16 is a schematic flow chart of acquiring a lee side image of a fan according to an embodiment of the present disclosure;

FIG. 17 is a schematic flow chart of determining the actual error distance of the leeward side according to the embodiment of the present disclosure;

FIG. 18 is a schematic flow chart of correcting the actual error distance of the leeward side according to the embodiment of the present disclosure;

Fig. 19 is a schematic view of an inspection device for an unmanned aerial vehicle-based wind turbine according to an embodiment of the present disclosure;

Fig. 20 is a schematic diagram of an internal structure of a computer device according to an embodiment of the present disclosure.

Detailed Description

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.

In the related art, the manual hanging basket inspection mode relies on an operation and maintenance person to climb a tower (the height is over 100 meters) of the wind driven generator, and parts such as blades, cabins and the like are inspected in a short distance through visual or handheld equipment (such as a thermal imager). However, in the manual hanging basket inspection mode, the inspection time of each wind driven generator is as long as 2-3 hours, the annual average high-altitude accident rate is as high as 0.3 per mill, and the falling risk exists. Since manual inspection can only rely on limited observation, the omission factor is as high as 20%, especially for micro-cracks (such as cracks less than or equal to 2 mm).

In the related art, a ground telescope/high-power mirror detection mode realizes blade inspection by remotely capturing blade images of a wind driven generator through erection of high-resolution equipment. However, the ground telescope/high power mirror detection mode is limited by light and weather, and the single coverage rate is less than 30%. Furthermore, the omission rate for microcracks is as high as 42% and the annual effective detection window for offshore wind farms is typically less than 100 days. The ground telescope/high power mirror detection mode has strong subjectivity and lacks traceability.

In the related art, a vibration/temperature sensor is deployed at key positions such as a gear box and a bearing of a wind driven generator in an on-line monitoring mode of a fixed sensor, and data are transmitted in real time through an SCADA system. However, the fixed sensor on-line monitoring method only covers the internal components of the wind turbine, and cannot detect the external damage (such as lightning mark and corrosion) of the blade of the wind turbine. Furthermore, incompatibility with manual recording and unmanned data protocols results in failure prediction accuracy below 60%.

The prior art has the defects of efficiency bottleneck (such as depending on shutdown and low coverage), safety risk (such as high-altitude operation), data fracture (such as difficult integration of multi-source information), blade diagnosis blind areas (such as difficult detection of external damage) and the like, so that the annual average damage rate of the blades of the wind driven generator is up to 9.5%, and the operation and maintenance cost is high.

In the related art, the manual unmanned aerial vehicle patrols and flies, which partially solves the problem of high-altitude safety, but still has a significant bottleneck. On one hand, the operation complexity is high, a professional flying hand is required to control, the flying path depends on manual planning, and the flying path is difficult to dynamically adjust under complex climatic conditions (such as turbulence or sudden wind speed change), so that image blurring or missed detection is easy to cause. On the other hand, the cost is higher, the single machine of the professional unmanned aerial vehicle needs to be matched with laser radar, a thermal imager and other equipment, the single flight duration is less than or equal to 25 minutes, and the comprehensive operation and maintenance cost is increased by frequently replacing the battery.

Based on the analysis, the embodiment of the specification provides a wind driven generator inspection method based on an unmanned aerial vehicle. Firstly, under the condition that the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, acquiring a top image of a fan, wherein the cruising initial position comprises coordinates and a safety height of a wind driven generator. And then, based on the image of the top of the fan, controlling the unmanned aerial vehicle to fly to the center point of the fan blade of the wind driven generator and collecting the image of the sweeping surface of the fan. Next, a swept surface actual error distance is determined based on the fan swept surface image. And under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface, determining the reference flying direction of the component of the wind driven generator. Then, a component single step movement distance series set and a component total movement distance are determined based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times. And finally, acquiring an image based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set. The implementation process dynamically optimizes the flight track and shooting parameters of the unmanned aerial vehicle by fusing the environment data in the fan image and the multi-mode sensor information in real time and combining a visual image recognition algorithm. The detection precision and efficiency under complex scenes are improved while manual intervention is reduced, so that the contradiction between cost and efficiency is effectively solved.

Referring to fig. 1, the method for inspecting a wind driven generator based on an unmanned aerial vehicle may include the following steps:

s110, acquiring a top image of the fan under the condition that the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range.

The cruising initial position comprises coordinates and a safety height of the wind driven generator.

Specifically, before the inspection starts, the tower height of the wind driven generator is obtained through pre-measurement or other modesBlade lengthAnd coordinates of the wind power generator. Because the safety distance is needed between the unmanned aerial vehicle and the wind driven generator to ensure the safety of the unmanned aerial vehicle, the safety distance between the top unmanned aerial vehicle and the wind driven generator is arranged. Safety height. Controlling unmanned aerial vehicle to vertically take off and pull up to safe heightAnd then, the vertex flies to the coordinate of the wind driven generator, and the unmanned aerial vehicle head is adjusted to a preset direction (such as the north direction) so as to start inspection.

After the unmanned aerial vehicle flies to the cruising initial position, whether the electric quantity is in a preset range (such as 25-100%) needs to be judged, and whether the unmanned aerial vehicle has the electric quantity for returning or not is judged so as to meet the requirement of inspection. If the electric quantity is not in the preset range, the unmanned aerial vehicle is indicated to have insufficient electric quantity at the moment and is insufficient for carrying out subsequent inspection requirements, so that the task fails and the unmanned aerial vehicle needs to be controlled to return. If the electric quantity is within the preset range, whether the self-adaptive inspection parameters are loaded or not needs to be judged. And if the self-adaptive inspection parameters are not loaded, loading the self-adaptive inspection parameters. And if the self-adaptive inspection parameters are loaded, adjusting the angle of a top camera control cradle head of the unmanned aerial vehicle to be 0 degrees, and then acquiring images to obtain a top image of the fan.

S120, controlling the unmanned aerial vehicle to fly to the center point of a fan blade of the wind driven generator based on the image of the top of the fan and collecting the image of a sweeping surface of the fan.

S130, determining the actual error distance of the sweeping surface based on the fan sweeping surface image.

And S140, determining the reference flying direction of the component of the wind driven generator under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface.

S150, determining a single-step moving distance sequence set of the components and the total moving distance of the components based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times.

S160, acquiring images based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set.

Specifically, the wind driven generator has two fan swept surfaces, and under the condition that at least one fan sweep surface is not subjected to inspection, image recognition, calculation and analysis are performed on images of the top of the fan, so that whether the unmanned aerial vehicle is in a preset central range of the top of the wind driven generator is judged. If the unmanned aerial vehicle is in a preset central range at the top of the wind driven generator, the unmanned aerial vehicle is controlled to fly to the central point of a fan blade of the wind driven generator by using a control algorithm, and then the head direction of the unmanned aerial vehicle is adjusted so as to acquire images, so that a fan swept surface image is obtained.

Then, image recognition and calculation are carried out on the fan sweep surface image, and the actual error distance of the sweep surface is determined. And under the condition that the actual error distance of the sweep surface meets the preset condition of the sweep surface, the unmanned aerial vehicle is at the center point of the actual blade. And then, analyzing and calculating by using the fan sweep surface image, and determining the reference flying direction of the component of the wind driven generator. And calculating by utilizing the safety distance of the components, the overlapping rate of the inspection images of the components and the inspection shooting times of the components, and determining a single-step moving distance sequence set of the components and the total moving distance of the components so as to control the flying distance of the unmanned aerial vehicle. And then, controlling the unmanned aerial vehicle to fly along the component reference flying direction according to the component single-step moving distance sequence set by using a control algorithm, and acquiring images after finishing one single-step moving distance so as to finally obtain a component inspection image set. Finally, the unmanned aerial vehicle is controlled by a control algorithm to fly along the direction opposite to the reference flying direction of the component according to the total moving distance of the component, and the unmanned aerial vehicle returns to the center point of the actual blade so as to carry out subsequent inspection or return.

It should be noted that, because there are two fan swept surfaces of the wind driven generator, after the component inspection image set of one side fan swept surface is determined through the above process, the unmanned aerial vehicle is controlled to fly back to the cruising initial position again, then the electric quantity is judged again to be in the preset range, if the electric quantity is not in the preset range, the unmanned aerial vehicle is controlled to fly back. If the electric quantity is in the preset range, performing subsequent operation to obtain a component inspection image set of the swept surface of the fan at the other side, and then returning.

In the above embodiment, first, under the condition that the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, a fan top image is obtained, wherein the cruising initial position comprises coordinates and a safety height of the wind driven generator. And then, based on the image of the top of the fan, controlling the unmanned aerial vehicle to fly to the center point of the fan blade of the wind driven generator and collecting the image of the sweeping surface of the fan. Next, a swept surface actual error distance is determined based on the fan swept surface image. And under the condition that the actual error distance of the swept surface meets the preset condition of the swept surface, determining the reference flying direction of the component of the wind driven generator. Then, a component single step movement distance series set and a component total movement distance are determined based on the component safety distance, the component inspection image overlapping rate and the component inspection shooting times. And finally, acquiring an image based on the component reference flight direction, the component single-step moving distance sequence set and the component total moving distance to obtain a component inspection image set. The implementation process dynamically optimizes the flight track and shooting parameters of the unmanned aerial vehicle by fusing the environment data in the fan image and the multi-mode sensor information in real time and combining a visual image recognition algorithm. The detection precision and efficiency under complex scenes are improved while manual intervention is reduced, so that the contradiction between cost and efficiency is effectively solved.

In some embodiments, referring to fig. 2, the fan top image corresponds to a top image center point, and based on the fan top image, controlling the unmanned aerial vehicle to fly to the fan blade center point of the wind turbine and collecting the fan swept surface image may include the steps of:

s210, identifying based on the top image of the fan, and determining the outline of the first target area.

S220, calculating based on the first target area outline, and determining the vertical foot coordinates of the blade.

S230, determining the actual error distance of the top based on the first target area outline, the blade foot drop coordinates and the top image center point.

S240, determining the axial flight direction of the wind wheel based on the vertical foot coordinates of the blades under the condition that the actual error distance of the top meets the preset condition of the top.

S250, controlling the unmanned aerial vehicle to fly to the center point of the fan blade and collecting an image of the sweeping surface of the fan based on the axial flight direction of the wind wheel, the horizontal safety distance and the safety height of the blade.

Specifically, an image recognition algorithm is utilized to recognize an image of the top of the fan, and a first target area outline is determined. Next, based on the obtained first target area profile, a geometrical calculation method is used to determine the blade foot drop coordinates. After the vertical foot coordinates of the blade are determined, the first target area outline and the top image center point are combined to conduct proportional conversion between the image distance and the actual distance, and the actual error distance of the top is determined. When the actual error distance at the top meets the preset condition at the top, namely when the actual error is within the allowable range, the position of the unmanned aerial vehicle is shown to be within the preset center range at the top of the wind driven generator, and then the axial flight direction of the wind wheel is determined by analyzing the position relation of the vertical foot coordinates of the blades and applying a geometric principle and a course planning algorithm. And finally, based on the determined axial flight direction of the wind wheel, simultaneously combining two key parameters of the horizontal safety distance and the blade safety height, controlling the unmanned aerial vehicle to fly to the center point of the blade of the fan and collecting the image of the swept surface of the fan.

In the above embodiment, the first target area contour is determined based on the identification of the fan top image, the calculation is performed based on the first target area contour, the blade foot drop coordinate is determined, the top actual error distance is determined based on the first target area contour, the blade foot drop coordinate and the top image center point, the wind wheel axial flight direction is determined based on the blade foot drop coordinate under the condition that the top actual error distance meets the top preset condition, the unmanned aerial vehicle is controlled to fly to the fan blade center point and the fan sweep surface image is collected based on the wind wheel axial flight direction, the horizontal safety distance and the blade safety height, and a data basis is provided for follow-up inspection of the blades and the tower.

In some embodiments, referring to fig. 3, the fan sweep image corresponds to a sweep image center point, and determining the actual error distance of the sweep based on the fan sweep image may include the steps of:

S310, identifying based on the fan sweep surface image, and determining a second target area outline.

S320, determining the center of mass coordinates of the wind wheel center structure based on the outline of the second target area.

S330, determining the actual error distance of the swept surface based on the second target area outline, the center of mass coordinates of the wind wheel center structure and the center point of the image of the swept surface.

Specifically, an image recognition algorithm is utilized to recognize the fan sweep surface image, and a second target area contour is determined. And then, calculating the obtained profile of the second target area through a centroid calculation formula to determine the centroid coordinates of the wind wheel center structure. After the barycenter coordinates of the wind wheel center structure are determined, the barycenter coordinates are combined with the outline of the second target area and the center point of the image of the swept surface to conduct proportional conversion between the image distance and the actual distance, and the actual error distance of the swept surface is determined.

In the above embodiment, the second target area contour is determined based on the identification of the fan swept surface image, the center of mass coordinate of the wind wheel center structure is determined based on the second target area contour, the center of mass coordinate of the wind wheel center structure and the center point of the swept surface image, and the actual error distance of the swept surface is determined based on the second target area contour, the center of mass coordinate of the wind wheel center structure and the center point of the swept surface image, so that a data basis is provided for the subsequent correction of the unmanned plane swept surface.

In some embodiments, referring to fig. 4a, the first target area profile includes a first blade-split profile, a second blade-split profile, and a nacelle-split profile, and the calculating based on the first target area profile, determining the blade-drop coordinates may include the steps of:

s410, calculating based on the first blade segmentation contour, and determining the barycenter coordinates of the first blade.

And S420, calculating based on the second blade segmentation contour, and determining the barycenter coordinates of the second blade.

And S430, calculating based on the cabin segmentation contour, and determining cabin centroid coordinates.

S440, calculating based on the first blade centroid coordinates, the second blade centroid coordinates and the cabin centroid coordinates to determine the blade foot drop coordinates.

Specifically, the image recognition algorithm is utilized to recognize the top image of the fan, so that the recognized segmentation contour in the image can be determined. Under the condition that the fan top image is determined to comprise a first blade segmentation contour, a second blade segmentation contour and a cabin segmentation contour corresponding to the largest two blades, calculating the first blade segmentation contour through a centroid calculation formula, and determining a first blade centroid coordinate. And calculating the second blade segmentation contour through a centroid calculation formula, and determining the centroid coordinates of the second blade. And calculating the cabin segmentation contour through a centroid calculation formula, and determining cabin centroid coordinates. And finally, calculating the distance between the center of mass coordinate of the engine room and the connecting line of the center of mass coordinate of the first blade and the center of mass coordinate of the second blade by using a vertical projection method, and further determining the vertical foot coordinate of the blade. And if the image at the top of the fan does not comprise at least one of the first blade segmentation contour, the second blade segmentation contour and the cabin segmentation contour corresponding to the largest two blades, controlling the unmanned aerial vehicle to return. For example, referring to FIG. 4b, the green dot in FIG. 4b is the blade centroid, the red dot is the nacelle centroid, the deep blue and the point at the blade centroid is the blade foot.

Illustratively, the contour is segmented based on the first bladeCalculating to determine the barycenter coordinates of the first blade). Segmenting the contour based on the second bladeCalculating and determining the barycenter coordinates of the second blade). Cabin-based segmentation of contoursCalculating and determining the barycenter coordinates of the engine room. Calculating the cabin centroid coordinates by the following formulaTo the first blade centroid coordinatesWith the second blade centroid coordinatesBlade foot drop coordinates of the connection line between them:

Wherein, the Is a parameter.

In the above embodiment, the calculation is performed based on the first blade segmentation contour, the first blade centroid coordinates are determined, the calculation is performed based on the second blade segmentation contour, the second blade centroid coordinates are determined, the calculation is performed based on the cabin segmentation contour, the cabin centroid coordinates are determined, the calculation is performed based on the first blade centroid coordinates, the second blade centroid coordinates and the cabin centroid coordinates, the blade foot drop coordinates are determined, and a data basis is provided for subsequent fan top correction of the unmanned aerial vehicle.

In some embodiments, referring to fig. 5, determining the top actual error distance based on the first target area profile, the blade foot drop coordinates, and the top image center point may include the steps of:

and S510, calculating based on the blade foot drop coordinates and the top image center point, and determining the top image error distance.

S520, determining a top scaling factor based on the area of the cabin dividing contour and the actual area of the cabin top.

S530, determining the actual error distance of the top based on the error distance of the top image and the top proportion coefficient.

Specifically, center point calculation is performed by using the top image of the fan, and the center point of the top image is determined. Then, the distance between the coordinates of the foot of the blade and the center point of the top image is calculated based on the geometric projection relationship, the distance value between the foot and the center point of the image is obtained, and the distance is taken as the error distance of the top image. And calculating the cabin segmentation contour through a contour area calculation formula, and determining the area of the cabin segmentation contour. Next, the calculated area of the nacelle dividing contour is compared with the actual area of the nacelle roof, and the roof scaling factor is determined by scaling the areas of the two. And finally, calculating according to the top image error distance and the top proportion coefficient, and determining the top actual error distance. The actual area of the top of the engine room is an adaptive inspection parameter. For example, referring to fig. 4b, the sky blue point in fig. 4b is the top image center point.

Illustratively, the top image center point is=) The vertical foot coordinates of the blade areThe area of the cabin dividing contour isThe actual area of the top of the engine room is. Calculating the actual error distance between the center point of the top image and the top by the following formula:

Wherein, the Is the width of the image at the top of the fan,Is the height of the image at the top of the fan,For the top image error distance,As a top scale factor of the coefficient of proportionality,Is the top actual error distance.

In the above embodiment, the calculation is performed based on the blade vertical foot coordinates and the top image center point, the top image error distance is determined, the top proportion coefficient is determined based on the area of the cabin segmentation contour and the cabin top actual area, the top actual error distance is determined based on the top image error distance and the top proportion coefficient, and a data basis is provided for the subsequent correction of the fan top of the unmanned aerial vehicle.

In some embodiments, referring to fig. 6a, the method may further comprise the steps of:

and S610, under the condition that the actual error distance at the top does not meet the preset condition at the top, calculating based on the vertical foot coordinates of the blade and the center point of the top image, and determining to correct the flight direction.

S620, after the actual error distance of the flight top of the unmanned aerial vehicle is controlled based on the corrected flight direction, the corrected top image of the fan is acquired.

Wherein, the corrected fan top image corresponds to the corrected image center point.

Specifically, under the condition that the actual error distance of the top does not meet the preset condition of the top, the position of the unmanned aerial vehicle is not located in the preset center range of the top of the wind driven generator, and the deviation phenomenon exists, so that the position of the unmanned aerial vehicle needs to be corrected, the accuracy and the reliability of the whole inspection flow are improved, and the problem that inspection data are inaccurate or key parts are omitted due to position deviation is avoided. And calculating the angle between the current position of the unmanned aerial vehicle and the vertical foot coordinates of the blade according to the geometric relationship between the vertical foot coordinates of the blade and the center point of the top image by taking the preset direction of the unmanned aerial vehicle as a reference, and determining the corrected flight direction. The clockwise direction is indicated as positive. And then controlling the unmanned aerial vehicle to reach a required position after correcting the actual error distance of the flight top of the flight direction by using a control algorithm, and then acquiring an image to obtain a corrected fan top image. And calculating a central point by using the corrected top image of the fan, and determining the central point of the corrected image. Wherein the top preset condition may be less than the center point correction threshold。

Illustratively, based on leaf drop foot coordinatesAnd a top image center point=) Calculating, determining corrected flight direction:

And S630, identifying based on the corrected fan top image, and determining the outline of the correction target area.

And S640, calculating based on the outline of the correction target area, and determining the correction blade foot drop coordinates.

The corrected target area outline corresponds to corrected cabin centroid coordinates.

Specifically, the corrected fan top image is identified by using an image identification algorithm, so that the identified segmentation contour in the image can be determined. Under the condition that the corrected fan top image comprises the largest segmentation contour corresponding to the two blades and the corrected cabin segmentation contour, the segmentation contours corresponding to the two blades are respectively calculated through a centroid calculation formula, and the centroid coordinates corresponding to the two blades are determined. And calculating the corrected cabin segmentation contour through a centroid calculation formula, and determining corrected cabin centroid coordinates. And finally, calculating and correcting the distance between the center of mass coordinates of the engine room and the connecting line of the center of mass coordinates corresponding to the two blades by using a vertical projection method, and further determining and correcting the vertical foot coordinates of the blades. And if the corrected fan top image does not comprise at least one of the segmentation contours corresponding to the maximum two blades and the corrected cabin segmentation contours, controlling the unmanned aerial vehicle to return.

S650, determining the corrected top actual error distance based on the corrected target area outline, the corrected blade foot drop coordinates and the corrected image center point.

Specifically, the distance between the coordinates of the vertical foot of the correction blade and the center point of the correction image is calculated based on the geometric projection relationship, the distance value between the vertical foot and the center point of the image is obtained, and the distance is taken as the correction top image error distance. And calculating the corrected cabin segmentation contour through a contour area calculation formula, and determining the area of the corrected cabin segmentation contour. And comparing the calculated area of the corrected cabin dividing contour with the actual area of the cabin top, and determining a corrected top scaling factor by performing area scaling conversion on the calculated area of the corrected cabin dividing contour and the actual area of the cabin top. And finally, calculating according to the corrected top image error distance and the corrected top proportion coefficient, and determining the corrected top actual error distance.

S660, comparing the corrected top actual error distance with the top preset condition, and performing iteration execution on the condition that the corrected top actual error distance does not meet the top preset condition and the iteration number does not meet the iteration preset number until the corrected top actual error distance meets the top preset condition, taking the corrected cabin centroid coordinate as the cabin centroid coordinate, and taking the corrected blade vertical foot coordinate as the blade vertical foot coordinate.

Specifically, the corrected top actual error distance is compared with the top preset condition. And judging the number of position correction times which are executed by the current unmanned aerial vehicle and determining the iteration times under the condition that the corrected top actual error distance does not meet the top preset condition. If the iteration times do not meet the preset iteration times, the process can be re-executed until the corrected top actual error distance meets the top preset condition, at the moment, the corrected cabin centroid coordinates are taken as cabin centroid coordinates, and the corrected blade foot drop coordinates are taken as blade foot drop coordinates. And if the iteration times meet the preset iteration times, controlling the unmanned aerial vehicle to return. Wherein the iteration preset times can be the maximum correction times of a single inspection stage。

For example, referring to fig. 6b, fig. 6b is an image with corrected top actual error distance meeting top preset conditions.

In the embodiment, the unmanned aerial vehicle flight movement and the acquisition of the inspection image are more accurately realized by correcting the position of the unmanned aerial vehicle.

In some embodiments, referring to fig. 7, the wind wheel axial flight direction includes a windward flight direction and a leeward flight direction, and determining the wind wheel axial flight direction based on the blade foot-drop coordinates may include the following steps if the actual error distance at the top meets the preset condition at the top:

S710, determining the windward flight direction based on the cabin centroid coordinates and the blade foot drop coordinates.

S720, determining the leeward flight direction based on the blade foot drop coordinates and the cabin centroid coordinates.

Specifically, under the condition that the actual error distance of the top meets the preset condition of the top, the angle between the current position of the unmanned aerial vehicle and the nacelle is calculated according to the geometric relationship between the barycenter coordinate of the nacelle and the vertical foot coordinate of the blade by taking the preset direction of the unmanned aerial vehicle as a reference, and the windward flight direction is determined. The clockwise direction is indicated as positive.

And calculating the angle between the cabin and the current position of the unmanned aerial vehicle according to the geometric relationship between the coordinates of the vertical feet of the blades and the coordinates of the mass center of the cabin by taking the preset direction of the unmanned aerial vehicle as a reference, and determining the flight direction of the leeward surface. The clockwise direction is indicated as positive.

It should be noted that the inspection sequence may be preset. According to the actual situation, the inspection of the windward side can be selected, and then the inspection of the leeward side can be performed, or the inspection of the leeward side can be performed, and then the inspection of the windward side can be performed.

In the above embodiment, the unmanned aerial vehicle is controlled by determining the flight direction of the unmanned aerial vehicle, so that the inspection of the wind driven generator is completed later.

In some embodiments, referring to fig. 8, the fan sweep image includes a fan windward image, and based on the axial flight direction of the wind wheel, the horizontal safety distance, and the blade safety height, controlling the unmanned aerial vehicle to fly to the center point of the fan blade and collecting the fan sweep image may include the following steps:

s810, controlling the horizontal flight horizontal safety distance of the unmanned aerial vehicle based on the windward flight direction, and then flying the safe height of the blade downwards to enable the unmanned aerial vehicle to fly to the center point of the fan blade.

S820, adjusting the head orientation of the unmanned aerial vehicle based on the windward flight direction and collecting the windward image of the fan.

The windward image of the fan corresponds to a center point of the windward image.

Specifically, blade lengthSafety distance to wind power generatorThe sum of the distances between the two is the blade safety height. In order to ensure the safety of the unmanned aerial vehicle in the inspection process, the unmanned aerial vehicle is controlled to horizontally fly along the windward flight direction by using a control algorithm, so that the unmanned aerial vehicle can effectively avoid fan blades in the flight process and can be kept in a proper range for detection. The unmanned aerial vehicle can effectively avoid the blades of the wind driven generator in the flight process, and can be maintained in a proper range for detection. After the horizontal flight is completed and the horizontal safety distance is reached, the unmanned aerial vehicle is controlled to fly downwards to ensure the safety height of the blades, so that the unmanned aerial vehicle flies to the center point of the fan blades. After the wind turbine blade center point is reached, the unmanned aerial vehicle also needs to adjust the head orientation of the unmanned aerial vehicle based on the windward flight direction, so that the head is ensured to be aligned to the windward side of the wind turbine, and then image acquisition is carried out, so that a wind turbine windward side image is obtained. And calculating a center point by using the windward image of the fan, and determining the center point of the windward image. Wherein the horizontal safe distance is an adaptive inspection parameter.

In the above embodiment, the horizontal flight horizontal safety distance of the unmanned aerial vehicle is controlled based on the windward flight direction, and then the safety height of the blades is flown downwards, so that the unmanned aerial vehicle flies to the center point of the fan blade, the head orientation of the unmanned aerial vehicle is adjusted based on the windward flight direction, the windward image of the fan is collected, and a data basis is provided for follow-up inspection of the blades and the tower.

In some embodiments, referring to fig. 9, the second target area profile includes a hub split profile, and determining the rotor center structure centroid coordinates based on the second target area profile may include the steps of:

S910, calculating based on the hub segmentation profile, and determining the hub centroid coordinate.

Specifically, in the process of inspecting the windward side, an image recognition algorithm is utilized to recognize the image of the windward side of the fan, and the recognized segmentation contour in the image can be determined. Under the condition that the windward side image of the fan comprises the segmentation contours of the three blades, the segmentation contours of the tower and the hub segmentation contours, the hub segmentation contours are calculated through a centroid calculation formula, and the hub centroid coordinates are determined. And if the windward side image of the fan does not comprise at least one of the segmentation contour of the three blades, the segmentation contour of the tower barrel and the segmentation contour of the hub, controlling the unmanned aerial vehicle to return.

Determining the swept surface actual error distance based on the second target area profile, the rotor wheel center structure centroid coordinates, and the swept surface image center point may include the steps of:

S920, calculating based on the hub centroid coordinates and the center point of the windward image, and determining the windward image error distance.

S930, determining a windward side proportionality coefficient based on the area of the hub split contour and the actual area of the hub.

S940, determining the actual error distance of the windward side based on the error distance of the windward side image and the windward side proportionality coefficient.

Specifically, the distance between the center of mass coordinate of the hub and the center point of the windward image is calculated based on the geometric projection relationship, the distance value between the hub and the center point of the image is obtained, and the distance is used as the error distance of the windward image. And calculating the hub segmentation contour through a contour area calculation formula, and determining the area of the hub segmentation contour. Next, the calculated area of the hub split profile is compared with the actual area of the hub, and the windward scaling factor is determined by performing an area scaling of the two. And finally, calculating according to the windward image error distance and the windward proportional coefficient, and determining the windward actual error distance. The actual area of the hub is an adaptive inspection parameter.

Illustratively, the center point of the windward image is=The center of mass coordinate of the hub is) The area of the hub split contour isThe actual area of the hub is. Calculating the actual error distance between the center point of the windward image and the windward through the following formula:

Wherein, the The width of the image of the windward side of the fan,The image height of the windward side of the fan,For the windward image error distance,As the proportion coefficient of the windward side,Is the actual error distance of the windward side.

In the above embodiment, the calculation is performed based on the hub segmentation profile, the hub centroid coordinates are determined, the calculation is performed based on the hub centroid coordinates and the center point of the windward image, the windward image error distance is determined, the windward scaling factor is determined based on the area of the hub segmentation profile and the actual area of the hub, the windward actual error distance is determined based on the windward image error distance and the windward scaling factor, and a data basis is provided for the subsequent windward correction of the unmanned aerial vehicle.

In some embodiments, referring to fig. 10a, the method may further comprise the steps of:

s1010, under the condition that the actual error distance of the windward side does not meet the preset condition of the swept side, calculating based on the hub centroid coordinates and the center point of the windward side image, and determining the windward correction flight direction.

S1020, after the actual error distance of the windward side of the unmanned aerial vehicle is controlled to fly based on the windward correction flying direction, the corrected windward side image is acquired.

The corrected windward image corresponds to a center point of the windward correction image.

Specifically, when the actual error distance of the windward side does not meet the preset condition of the swept side, the position of the unmanned aerial vehicle is not located in the preset center range of the windward side of the wind driven generator, and the deviation phenomenon exists, so that the position of the unmanned aerial vehicle needs to be corrected, the accuracy and the reliability of the whole inspection flow are improved, and the problem that inspection data are inaccurate or key parts are omitted due to position deviation is avoided. And calculating an angle between the current position of the unmanned aerial vehicle and the hub centroid coordinates according to the geometric relationship between the hub centroid coordinates and the windward image center point by taking the preset direction of the unmanned aerial vehicle as a reference, and determining the windward correction flight direction. The clockwise direction is indicated as positive. And then controlling the unmanned aerial vehicle to fly in the windward correcting flight direction by using a control algorithm, reaching a required position after the actual error distance of the windward is corrected, and then acquiring an image to obtain a corrected windward image. And calculating a central point by using the corrected windward image, and determining the central point of the windward corrected image. Wherein the sweep surface preset condition may be less than the center point correction threshold。

Illustratively, based on hub centroid coordinates) And a windward image center point=Calculating to determine windward corrected flight direction:

S1030, identifying based on the corrected windward image, and determining the contour of the windward correction target area.

S1040, based on the contour of the windward correction target area, the center of mass coordinates of the corrected hub and the center point of the windward correction image, the corrected windward actual error distance is determined.

The profile of the windward correction target area corresponds to the barycenter coordinate of the correction hub.

Specifically, the corrected windward image is identified by using an image identification algorithm, and the contour of the identified windward correction target area in the image can be determined. Under the condition that the contour of the correcting target area facing the wind is determined to comprise the split contour of three blades and the corrected hub split contour, calculating the corrected hub split contour through a centroid calculation formula, and determining the corrected hub centroid coordinates.

And calculating and correcting the distance between the center of mass coordinate of the hub and the center point of the windward correction image based on the geometric projection relationship, obtaining a distance value between the hub and the center point of the image, and taking the distance as the windward correction image error distance. And calculating the corrected hub split profile through a profile area calculation formula, and determining the area of the corrected hub split profile. And comparing the calculated area of the corrected hub split profile with the actual area of the hub, and determining a windward correction proportionality coefficient by performing area proportionality conversion on the calculated area of the corrected hub split profile and the actual area of the hub. And finally, calculating according to the windward correction image error distance and the windward correction proportionality coefficient, and determining the corrected windward actual error distance.

S1050, comparing the corrected windward actual error distance with a preset condition of the swept surface, and iteratively executing the process until the corrected windward actual error distance meets the preset condition of the swept surface and the corrected windward image is used as a windward image of the fan under the condition that the corrected windward actual error distance does not meet the preset condition of the swept surface and the iteration times meet the preset iteration times of the windward.

Specifically, the corrected windward actual error distance is compared with the preset condition of the sweeping surface. And judging the number of position correction times which are executed by the current unmanned aerial vehicle and determining the iteration times under the condition that the corrected windward actual error distance does not accord with the preset condition of the swept surface. If the iteration times do not meet the preset number of windward iterations, the process can be re-executed until the corrected windward actual error distance meets the preset condition of the swept surface, and the corrected windward image is used as the windward image of the fan. And if the iteration times meet the preset number of the windward iteration times, controlling the unmanned aerial vehicle to return. Wherein the windward iteration preset times can be the maximum correction times of a single inspection stage。

For example, referring to fig. 10b, fig. 10b is an image of the corrected windward actual error distance meeting the preset condition of the swept surface.

In some embodiments, referring to fig. 11, the second target area further includes a third blade-dividing contour, a fourth blade-dividing contour, a fifth blade-dividing contour, and a tower-dividing contour, and determining the component reference flight direction of the wind turbine may include the following steps if the actual error distance of the swept surface meets the preset condition of the swept surface:

S1110, calculating based on the third blade segmentation contour, and determining the mass center coordinates of the third blade.

And S1120, calculating based on the fourth blade segmentation contour, and determining the mass center coordinates of the fourth blade.

And S1130, calculating based on the fifth blade segmentation contour, and determining the mass center coordinates of the fifth blade.

S1140, calculating based on the tower segmentation contour, and determining the coordinates of the mass center of the tower.

S1150, determining the third blade flight direction based on the third blade centroid coordinates and the wind wheel center structure centroid coordinates.

S1160, determining the flight direction of the fourth blade based on the mass center coordinates of the fourth blade and the mass center coordinates of the center structure of the wind wheel.

S1170, determining the flight direction of the fifth blade based on the mass center coordinates of the fifth blade and the mass center coordinates of the center structure of the wind wheel.

S1180, determining the flight direction of the tower barrel based on the barycenter coordinates of the tower barrel and the barycenter coordinates of the center structure of the wind wheel.

Specifically, the image recognition algorithm is utilized to recognize the windward side image of the fan, and the recognized segmentation contour in the image can be determined. Under the condition that the windward side image of the fan comprises a third blade segmentation contour, a fourth blade segmentation contour, a fifth blade segmentation contour, a tower barrel segmentation contour and a wind wheel center structure segmentation contour, calculating the third blade segmentation contour through a centroid calculation formula, and determining a third blade centroid coordinate. And calculating the fourth blade segmentation contour through a centroid calculation formula to determine the fourth blade centroid coordinates. And calculating the fifth blade segmentation contour through a centroid calculation formula, and determining the centroid coordinates of the fifth blade. And calculating the tower barrel segmentation profile through a mass center calculation formula, and determining the mass center coordinates of the tower barrel.

And calculating the angle of the unmanned aerial vehicle, which needs to fly towards the third blade, according to the geometric relationship between the centroid coordinates of the third blade and the centroid coordinates of the wind wheel center structure by taking the preset direction of the unmanned aerial vehicle as a reference, and determining the flight direction of the third blade. And calculating the angle of the unmanned aerial vehicle, which needs to fly towards the fourth blade, according to the geometric relationship between the mass center coordinates of the fourth blade and the mass center coordinates of the central structure of the wind wheel, and determining the flight direction of the fourth blade. And calculating the angle of the unmanned aerial vehicle, which needs to fly towards the fifth blade, according to the geometric relationship between the barycenter coordinates of the fifth blade and the barycenter coordinates of the central structure of the wind wheel, and determining the flight direction of the fifth blade. And calculating the angle of the unmanned aerial vehicle, which needs to fly towards the tower, according to the geometric relationship between the barycenter coordinates of the tower and the barycenter coordinates of the central structure of the wind wheel, and determining the flight direction of the tower. The clockwise direction is indicated as positive.

In the process of carrying out the windward inspection, the center of mass coordinate of the wind wheel center structure is the center of mass coordinate of the hub, and in the process of carrying out the leeward inspection, the center of mass coordinate of the wind wheel center structure is the center of mass coordinate of the engine room.

In the embodiment, the inspection direction of the blade and the tower barrel is determined, so that the subsequent unmanned aerial vehicle can acquire the inspection image more accurately.

In some embodiments, referring to fig. 12, where the part safety distance includes a blade length, the part inspection image overlap rate includes a blade inspection image overlap rate, the part inspection shot number includes a blade inspection shot number, and the part single step movement distance sequence set and the part total movement distance are determined based on the part safety distance, the part inspection image overlap rate, and the part inspection shot number, the method may include the steps of:

s1210, determining the single shooting coverage length of the blade based on the length of the blade, the overlapping rate of the blade inspection images and the number of blade inspection shooting times.

S1220, calculating based on the single shooting coverage length and the overlapping rate of the blade inspection images, and determining the first step distance of the blade and the subsequent step length of the blade.

S1230, determining a single-step moving distance sequence set of the blade based on the first step distance of the blade, the subsequent step length of the blade and the inspection shooting times of the blade.

S1240, determining the total moving distance of the blade based on the blade length and the blade single shooting coverage length.

Specifically, when the blade inspection is performed, in order to enable the inspection image to completely cover the blade, the overlapping rate (for example, 20%) of the blade inspection image is set. And calculating by using the length of the blade, the overlapping rate of the blade inspection images and the number of times of blade inspection shooting, and determining the single-shot coverage length of the blade to represent the length of the blade which can be effectively covered by the single Zhang Xun inspection images. And dividing the coverage length of the blade shot once by an integer 2 to obtain the first step distance of the blade. And calculating by utilizing the overlapping rate of the single shot coverage length of the blade and the blade inspection image, and determining the subsequent step length of the blade. The repeated specific times of the follow-up step length of the blade can be determined according to the blade inspection shooting times, so that the distribution of shooting points can meet the requirements of complete coverage of images and set overlapping rate in the whole blade inspection process. In this way, the blade first step distance and the blade subsequent step length repeated a certain number of times constitute a blade single step distance sequence set. The single-step moving distance sequence set is used for guiding the unmanned aerial vehicle to gradually move towards the blade tip area from the actual blade center point, and the distance of each single-step moving is set. And determining the total moving distance of the blade by a corresponding calculation method based on the length of the blade and the single shooting coverage length of the blade. The total moving distance of the blade is used for setting the distance from the blade tip to the center point of the actual blade after the single blade is inspected.

Illustratively, the blade length isThe overlapping rate of the blade inspection images isThe blade inspection shooting times are as follows. The calculation is performed by the following formula:

calculating blade single shot coverage length :

Calculating a set of single step distance sequences for a blade:

Blade first step distance (actual blade center point→first shooting point center)

Blade follow-up step length (distance between shooting points)

Blade single step distance sequence set:

calculating the total moving distance of the blade :

It should be noted that, since the length of each blade of the wind power generator is uniform, the single step distance sequence set and the total moving distance of each blade in the windward inspection process and the single step distance sequence set and the total moving distance of each blade in the leeward inspection process are uniform.

In the above embodiment, the dynamically planned patrol distance is achieved by determining the single step sequence set of moving distances of the blades and the total moving distance of the blades.

In some embodiments, referring to fig. 13, where the component safety distance includes a tower safety height, the component inspection image overlapping rate includes a tower inspection image overlapping rate, the component inspection shooting times include a tower inspection shooting times, and determining the component single step distance sequence set and the component total movement distance based on the component safety distance, the component inspection image overlapping rate, and the component inspection shooting times may include the steps of:

s1310, determining the single shooting coverage length of the tower barrel based on the safety height of the tower barrel, the overlapping rate of the inspection images of the tower barrel and the inspection shooting times of the tower barrel.

S1320, calculating based on the single shooting coverage length of the tower and the overlapping rate of the tower inspection images, and determining the first step distance of the tower and the subsequent step length of the tower.

S1330, determining a single-step moving distance sequence set of the tower barrel based on the first step distance of the tower barrel, the subsequent step length of the tower barrel and the inspection shooting times of the tower barrel.

S1340, determining the total moving distance of the tower based on the tower safety height and the tower single shooting coverage length.

Specifically, when the tower inspection is performed, in order to enable the inspection image to completely cover the tower, the overlapping rate (for example, 20%) of the tower inspection image is set. And calculating by utilizing the safe height of the tower, the overlapping rate of the inspection images of the tower and the inspection shooting times of the tower, and determining the single shooting coverage length of the tower, wherein the single coverage length represents the length of the tower which can be effectively covered by the inspection images of the single Zhang Xun. And dividing the coverage length of the tower barrel shot once by an integer 2 to obtain the first step distance of the tower barrel. And calculating the overlapping rate of the single shooting coverage length of the tower and the tower inspection image, and determining the subsequent step length of the tower. The repeated specific times of the follow-up step length of the tower can be determined according to the inspection shooting times of the tower, so that the distribution of shooting points can meet the requirements of complete coverage of images and set overlapping rate in the whole inspection process of the tower. Thus, the tower first step distance and the tower subsequent step repeated a certain number of times form a blade single step distance sequence set. The single-step moving distance sequence set is used for guiding the unmanned aerial vehicle to gradually move from the actual blade center point to the bottom area of the tower, and the distance of each single-step moving is set. And determining the total moving distance of the tower barrel by a corresponding calculation method based on the safety height of the tower barrel and the single shooting coverage length of the tower barrel. The total moving distance of the tower is used for setting the distance from the bottom of the tower to the center point of the actual blade after the tower is inspected.

Illustratively, the fan tower height isThe required inspection shooting times of the tower barrel are as followsThe lowest safe height of the tower isThe overlapping rate of the inspection images is as follows. Based on the height of the tower drum of the fanThe times of inspection and shooting of the tower drumAnd determining the safety height of the tower. The calculation is performed by the following formula:

Calculating single shooting coverage length of tower :

Calculating single-step moving distance sequence set of tower barrel:

First step distance of tower (actual blade center point→first shooting point center)

Tower follow-up step length (distance between shooting points)

Tower single step movement distance sequence set

Calculating the total moving distance of the tower:

It should be noted that, because the safe heights of the towers are consistent, the single-step moving distance sequence set of the towers and the total moving distance of the towers in the windward inspection process are consistent, and the single-step moving distance sequence set of the towers and the total moving distance of the towers in the leeward inspection process are consistent.

In the above embodiment, the dynamically planned patrol distance is achieved by determining the single-step distance sequence set of the tower and the total distance of movement of the tower.

In some embodiments, referring to fig. 14a, image acquisition is performed based on a component reference flight direction, a component single step distance sequence set, and a component total distance of movement, resulting in a component inspection image set, which may include the steps of:

s1410, controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the third blade flying direction and the single-step moving distance sequence set of the blade, and acquiring an image to obtain a third blade inspection image set.

S1420, controlling the unmanned aerial vehicle to fly to the center point of the actual blade based on the total moving distance of the blade.

Specifically, based on the flight control system of the unmanned aerial vehicle, the unmanned aerial vehicle is ensured to start from the center point of the actual blade through a control algorithm by combining the single-step movement distance sequence set of the flight direction of the third blade and the blade, and the inspection flight is carried out along the flight direction of the third blade. In the process that the unmanned aerial vehicle carries out inspection flight along the third blade flight direction, the unmanned aerial vehicle gradually advances according to the single-step moving distance sequence set of the blades, and after the single-step moving distance flight is completed each time, image data are acquired in real time, and a third blade inspection image at a corresponding position is acquired. After each flight and image acquisition, a single-step moving distance inspection task is completed, and finally a complete third blade inspection image set is formed. After the unmanned aerial vehicle completes the third blade inspection image acquisition, the unmanned aerial vehicle can adjust the flight direction and reverse the total moving distance of the flying blade based on the flight control system and the control algorithm, so that the unmanned aerial vehicle returns to fly to the center point of the actual blade, and the inspection image acquisition of the next blade is conveniently carried out.

Illustratively, the third blade has a flight direction ofThe single step moving distance sequence set of the blade isThe total moving distance of the blades is. Controlling unmanned aerial vehicle from actual blade central point towardsThe first flight distance is thatAnd (5) carrying out third blade inspection image acquisition after flying. Then continue to faceThe direction of flight is that each flight distance isRepeatingAnd (3) carrying out third blade inspection image acquisition after each flight. After the flight of the single-step moving distance sequence set of the blade is completed, a third blade inspection image set can be obtained. Finally, controlling the unmanned aerial vehicle to fly in the reverse directionReturning to the actual blade center point.

In the process of inspection on the windward side, the inspection of the third blade is performed once according to the process, and in the process of inspection on the leeward side, the inspection of the third blade is performed once according to the process.

S1430, controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the fourth blade flying direction and the single-step moving distance sequence set of the blade, and acquiring images to obtain a fourth blade inspection image set.

S1440, controlling the unmanned aerial vehicle to fly to the center point of the actual blade based on the total moving distance of the blade.

Specifically, based on the unmanned aerial vehicle flight control system, the unmanned aerial vehicle is ensured to start from the actual blade center point through a control algorithm by combining the fourth blade flight direction and the single-step movement distance sequence set of the blades, and the inspection flight is carried out along the fourth blade flight direction. In the process that the unmanned aerial vehicle carries out inspection flight along the fourth blade flight direction, the unmanned aerial vehicle gradually advances according to the blade single-step movement distance sequence set, and after each single-step movement distance flight is completed, image data are acquired in real time, and a fourth blade inspection image at a corresponding position is acquired. After each flight and image acquisition, a single-step moving distance inspection task is completed, and finally a complete fourth blade inspection image set is formed. After the unmanned aerial vehicle completes the fourth blade inspection image acquisition, the unmanned aerial vehicle can adjust the flight direction and reverse the total moving distance of the flying blade based on the flight control system and the control algorithm, so that the unmanned aerial vehicle returns to fly to the center point of the actual blade, and the inspection image acquisition of the next blade is conveniently carried out.

S1450, controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the flight direction of the fifth blade and the single-step moving distance sequence set of the blade, and acquiring an image to obtain a fifth blade inspection image set.

S1460, controlling the unmanned aerial vehicle to fly to the center point of the actual blade based on the total moving distance of the blade.

Specifically, based on the unmanned aerial vehicle flight control system, combining the fifth blade flight direction and the blade single-step movement distance sequence set, and ensuring that the unmanned aerial vehicle starts from the actual blade center point through a control algorithm to carry out inspection flight along the fifth blade flight direction. In the process that the unmanned aerial vehicle carries out inspection flight along the flight direction of the fifth blade, the unmanned aerial vehicle gradually advances according to the single-step moving distance sequence set of the blade, and after the single-step moving distance flight is completed each time, image data are acquired in real time, and a fifth blade inspection image at a corresponding position is acquired. After each flight and image acquisition, a single-step moving distance inspection task is completed, and finally a complete fifth blade inspection image set is formed. After the unmanned aerial vehicle completes the image acquisition of the fifth blade inspection, the unmanned aerial vehicle can adjust the flight direction and reverse the total moving distance of the flying blade based on a flight control system and a control algorithm, so that the unmanned aerial vehicle returns to fly to the center point of the actual blade, and the image acquisition of the next blade inspection is conveniently carried out.

For example, referring to FIG. 14b, FIG. 14b is an image of a patrol wind turbine blade and tower.

In the above embodiment, by inspecting each blade of the wind turbine, the acquisition of the inspection image of each blade is realized, so that the operation and maintenance detection of each blade is realized subsequently.

In some embodiments, referring to fig. 15, image acquisition is performed based on a component reference flight direction, a component single step travel distance sequence set, and a component total travel distance to obtain a component inspection image set, which may include the steps of:

S1510, controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the tower flight direction and the single-step moving distance sequence set of the tower, and acquiring images to obtain a tower inspection image set.

S1520, controlling the unmanned aerial vehicle to fly to the center point of the actual blade based on the total moving distance of the tower.

Specifically, based on the unmanned aerial vehicle's flight control system, combine tower section of thick bamboo flight direction and tower section of thick bamboo single step to move distance sequence set, ensure through control algorithm that unmanned aerial vehicle from actual blade central point, carry out the inspection flight along tower section of thick bamboo flight direction. In the process that the unmanned aerial vehicle carries out inspection flight along the tower barrel flight direction, the unmanned aerial vehicle gradually advances according to the tower barrel single-step movement distance sequence set, and after each single-step movement distance flight is completed, image data are acquired in real time, and a tower barrel inspection image at a corresponding position is acquired. After each flight and image acquisition, a single-step moving distance inspection task is completed, and a complete tower inspection image set is finally formed. After the unmanned aerial vehicle completes the tower inspection image acquisition, the unmanned aerial vehicle can adjust the flight direction and reverse the total movement distance of the flying tower based on the flight control system and the control algorithm, so that the unmanned aerial vehicle returns to fly to the center point of the actual blade.

For example, referring to FIG. 14b, FIG. 14b is an image of a patrol wind turbine blade and tower. In the process of inspection on the windward side, the inspection of the tower barrel is carried out once according to the process, and in the process of inspection on the leeward side, the inspection of the tower barrel is carried out once according to the process.

In the embodiment, the tower barrel of the wind driven generator is inspected, so that the acquisition of the inspection image of the tower barrel is realized, and the operation and maintenance detection of the tower barrel is realized later.

In some embodiments, referring to fig. 16, the fan sweep image includes a fan lee image, and based on the wind wheel axial flight direction, the horizontal safety distance, and the blade safety height, controlling the unmanned aerial vehicle to fly to the fan blade center point and collecting the fan sweep image may include the steps of:

S1610, controlling the horizontal flight horizontal safety distance of the unmanned aerial vehicle based on the leeward flight direction, and then flying the safe height of the blade downwards to enable the unmanned aerial vehicle to fly to the center point of the fan blade.

S1620, adjusting the head direction of the unmanned aerial vehicle based on the leeward flight direction and collecting the leeward image of the fan.

The fan leeward image corresponds to a leeward image center point.

Specifically, blade lengthSafety distance to wind power generatorThe sum of the distances between the two is the blade safety height. In order to ensure the safety of the unmanned aerial vehicle in the inspection process, the unmanned aerial vehicle is controlled to fly horizontally along the leeward flight direction by using a control algorithm, so that the unmanned aerial vehicle can effectively avoid fan blades in the flight process and can be maintained in a proper range for detection. The unmanned aerial vehicle can effectively avoid the blades of the wind driven generator in the flight process, and can be maintained in a proper range for detection. After the horizontal flight is completed and the horizontal safety distance is reached, the unmanned aerial vehicle is controlled to fly downwards to ensure the safety height of the blades, so that the unmanned aerial vehicle flies to the center point of the fan blades. After the wind turbine blade center point is reached, the unmanned aerial vehicle also needs to adjust the aircraft nose orientation of the unmanned aerial vehicle based on the leeward flight direction, ensures that the aircraft nose is aligned to the windward side of the wind turbine, and then performs image acquisition to obtain a wind turbine leeward image. And calculating a central point by using the leeward image of the fan, and determining the central point of the leeward image. Wherein the horizontal safe distance is an adaptive inspection parameter.

In the above-mentioned embodiment, based on leeward plane flight direction control unmanned aerial vehicle horizontal flight horizontal safety distance, then downward flight blade safety height makes unmanned aerial vehicle fly to fan blade central point, based on leeward plane flight direction adjustment unmanned aerial vehicle's aircraft nose orientation and gather fan leeward plane image, for follow-up carry out blade and tower section of thick bamboo and patrol and examine and provide data basis.

In some embodiments, referring to fig. 17, the second target area profile includes a nacelle-still-split profile, and determining rotor-center-structure centroid coordinates based on the second target area profile may include the steps of:

s1710, calculating based on the cabin segmentation contour, and determining cabin centroid coordinates.

Specifically, in the inspection process of the leeward side, an image recognition algorithm is utilized to recognize the image of the leeward side of the fan, and the recognized segmentation contour in the image can be determined. Under the condition that the fan leeward side image is determined to comprise the split contour of the three blades, the split contour of the tower barrel and the cabin split contour, calculating the cabin split contour through a centroid calculation formula, and determining cabin centroid coordinates. And if the fan lee surface image is determined not to comprise at least one of the segmentation contour of the three blades, the segmentation contour of the tower barrel and the cabin segmentation contour, controlling the unmanned aerial vehicle to return.

Determining the swept surface actual error distance based on the second target area profile, the rotor hub structure centroid coordinates, and the swept surface image center point may include the steps of:

S1720, calculating based on the cabin centroid coordinates and the leeward image center point, and determining the leeward image error distance.

S1730, determining the leeward scaling factor based on the area of the cabin dividing contour and the actual cabin area.

S1740, determining the actual error distance of the leeward based on the error distance of the leeward image and the scaling factor of the leeward.

Specifically, the distance between the center of mass coordinate of the engine room and the center point of the leeward image is calculated based on the geometric projection relationship, the distance value between the engine room and the center point of the image is obtained, and the distance is taken as the leeward image error distance. And calculating the cabin segmentation contour through a contour area calculation formula, and determining the area of the cabin segmentation contour. Next, the calculated area of the nacelle dividing contour is compared with the actual nacelle area, and the leeward scaling factor is determined by scaling the areas of the two. And finally, calculating according to the leeward image error distance and the leeward proportionality coefficient to determine the leeward actual error distance. Wherein the actual area of the nacelle is an adaptive inspection parameter.

Illustratively, the leeward image center point is=) The mass center coordinates of the engine room are as followsThe area of the cabin dividing contour isThe actual area of the engine room is. Calculating the actual error distance between the center point of the leeward image and the leeward by the following formula:

Wherein, the Is the width of the image of the lee surface of the fan,Is the image height of the lee surface of the fan,Is the error distance of the leeward image,Is the proportion coefficient of the lee surface,Is the actual error distance of the lee surface.

In the above embodiment, the calculation is performed based on the cabin division profile, the cabin centroid coordinates are determined, the calculation is performed based on the cabin centroid coordinates and the leeward image center point, the leeward image error distance is determined, the leeward scaling factor is determined based on the area of the cabin division profile and the cabin actual area, the leeward actual error distance is determined based on the leeward image error distance and the leeward scaling factor, and a data basis is provided for the subsequent windward correction of the unmanned aerial vehicle.

In some embodiments, referring to fig. 18, the method may further comprise the steps of:

s1810, under the condition that the actual error distance of the leeward side does not meet the preset condition of the swept side, calculating based on the cabin centroid coordinates and the center point of the leeward side image, and determining the leeward correction flight direction.

S1820, after the actual error distance of the leeward side of the unmanned aerial vehicle is controlled based on the leeward correction flight direction, the corrected leeward side image is acquired.

The corrected leeward image corresponds to a leeward correction image center point.

Specifically, when the actual error distance of the leeward side does not meet the preset condition of the swept side, the position of the unmanned aerial vehicle is not located in the preset center range of the leeward side of the wind driven generator, and the deviation phenomenon exists, so that the position of the unmanned aerial vehicle needs to be corrected, the accuracy and the reliability of the whole inspection flow are improved, and the problem that inspection data are inaccurate or key parts are omitted due to position deviation is avoided. And calculating an angle between the current position of the unmanned aerial vehicle and the cabin centroid coordinates according to the geometric relationship between the cabin centroid coordinates and the leeward image center points by taking the preset direction of the unmanned aerial vehicle as a reference, and determining the leeward correction flight direction. The clockwise direction is indicated as positive. And then, controlling the unmanned aerial vehicle to fly the actual error distance of the leeward surface along the leeward correcting flying direction by using a control algorithm, reaching a required position, and then acquiring an image to obtain a corrected leeward surface image. And calculating a central point by using the corrected leeward image, and determining the central point of the leeward corrected image. Wherein the sweep surface preset condition may be less than the center point correction threshold。

Illustratively, based on cabin centroid coordinatesAnd leeward image center point=) Calculating to determine the direction of the leeward corrected flight:

S1830, identifying based on the corrected leeward image, and determining the contour of the leeward correction target area.

S1840, correcting the outline of the target area, correcting the barycenter coordinates of the cabin and correcting the center point of the image based on the leeward, and determining the corrected leeward actual error distance.

The contour of the leeward correction target area corresponds to the corrected cabin centroid coordinates;

Specifically, the corrected windward image is identified by using an image identification algorithm, and the contour of the identified leeward correction target area in the image can be determined. Under the condition that the leeward correction target area contour is determined to comprise the split contour of three blades and the correction cabin split contour, calculating the correction cabin split contour through a centroid calculation formula, and determining the correction cabin centroid coordinates.

And calculating and correcting the distance between the cabin centroid coordinates and the leeward correction image center point based on the geometric projection relationship, obtaining a distance value between the cabin and the image center point, and taking the distance as the leeward correction image error distance. And calculating the corrected cabin segmentation contour through a contour area calculation formula, and determining the area of the corrected cabin segmentation contour. And comparing the calculated area of the corrected cabin dividing contour with the actual cabin area, and determining a leeward correction proportionality coefficient by performing area proportionality conversion on the calculated area and the actual cabin area. And finally, calculating according to the leeward correction image error distance and the leeward correction proportionality coefficient, and determining the corrected leeward actual error distance.

S1850, comparing the corrected leeward actual error distance with the preset condition of the sweep surface, and iteratively executing the process until the corrected leeward actual error distance meets the preset condition of the sweep surface and the corrected leeward image is used as the fan leeward image under the condition that the corrected leeward actual error distance does not meet the preset condition of the sweep surface and the iteration number meets the preset iteration number of the leeward surface.

Specifically, the corrected leeward actual error distance is compared with the preset condition of the sweeping surface. And judging the number of position correction times which are executed by the current unmanned aerial vehicle and determining the iteration times under the condition that the corrected leeward actual error distance does not accord with the preset condition of the swept surface. If the iteration times do not meet the preset leeward iteration times, the process can be re-executed until the corrected leeward actual error distance meets the preset sweep surface conditions, and the corrected leeward image is used as the fan leeward image. And if the iteration times meet the preset iteration times of the lee surface, controlling the unmanned aerial vehicle to return. Wherein the preset number of leeward iteration can be the maximum correction number of single inspection stage。

In some embodiments, adjustment of the focal length multiple of the camera of the unmanned aerial vehicle can be achieved according to the obtained image and the calculated parameters, so that more accurate image acquisition is achieved.

The embodiment of the specification provides an unmanned aerial vehicle-based wind driven generator inspection device 1900, referring to fig. 19, the unmanned aerial vehicle-based wind driven generator inspection device 1900 comprises a fan top image acquisition module 1910, a fan swept surface image acquisition module 1920, a swept surface actual error distance determination module 1930, a component reference flight direction determination module 1940, a component movement data determination module 1950, and a component inspection image acquisition module 1960.

A fan top image obtaining module 1910, configured to obtain a fan top image when the unmanned aerial vehicle flies to a cruising initial position and the electric quantity is in a preset range, where the cruising initial position includes coordinates and a safety height of the wind driven generator;

The fan swept surface image acquisition module 1920 is used for controlling the unmanned aerial vehicle to fly to the center point of the fan blade of the wind driven generator and acquiring a fan swept surface image based on the fan top image;

A swept surface actual error distance determination module 1930 for determining a swept surface actual error distance based on the fan swept surface image;

A component reference flight direction determining module 1940, configured to determine a component reference flight direction of the wind turbine when the actual error distance of the swept surface meets a preset condition of the swept surface;

a component movement data determining module 1950 for determining a component single step movement distance sequence set and a component total movement distance based on the component safety distance, the component inspection image overlapping rate, and the component inspection photographing times;

The component inspection image acquisition module 1960 is configured to acquire an image based on the component reference flight direction, the component single-step movement distance sequence set and the component total movement distance, and obtain a component inspection image set.

For a specific description of the unmanned aerial vehicle-based wind turbine inspection device, reference may be made to the description of the unmanned aerial vehicle-based wind turbine inspection method hereinabove, and no further description is given here.

In some embodiments, a computer device is provided, which may be a terminal, and an internal structure diagram thereof may be as shown in fig. 20. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program when executed by the processor is used for realizing a wind driven generator inspection method based on the unmanned aerial vehicle. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in fig. 20 is merely a block diagram of a portion of the structure associated with the aspects disclosed herein and is not limiting of the computer device to which the aspects disclosed herein apply, and in particular, the computer device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In some embodiments, a computer device is provided, comprising a memory in which a computer program is stored, and a processor which, when executing the computer program, carries out the method steps of the above embodiments.

The present description embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method of any of the above embodiments.

An embodiment of the present specification provides a computer program product comprising instructions which, when executed by a processor of a computer device, enable the computer device to perform the steps of the method of any one of the embodiments described above.

It should be noted that the logic and/or steps represented in the flowcharts or otherwise described herein, for example, may be considered as a ordered listing of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include an electrical connection (an electronic device) having one or more wires, a portable computer diskette (a magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium may even be paper or other suitable medium upon which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

Claims

1. An unmanned aerial vehicle-based wind driven generator inspection method is characterized by comprising the following steps:

2. The method of claim 1, wherein the fan top image corresponds to a top image center point, the controlling the drone to fly to a fan blade center point of the wind turbine and collecting a fan swept surface image based on the fan top image, comprising:

identifying based on the fan top image, and determining a first target area outline;

calculating based on the first target area outline, and determining a blade foot drop coordinate;

determining a top actual error distance based on the first target area profile, the blade foot drop coordinates, and the top image center point;

Under the condition that the actual error distance of the top meets the preset top condition, determining the axial flight direction of the wind wheel based on the vertical foot coordinates of the blades;

Based on the axial flight direction of the wind wheel, the horizontal safety distance and the safety height of the blades, the unmanned aerial vehicle is controlled to fly to the center point of the fan blades and the image of the sweeping surface of the fan is acquired.

3. The method of claim 1, wherein the fan swept surface image corresponds to a swept surface image center point, the determining a swept surface actual error distance based on the fan swept surface image comprising:

identifying based on the fan sweep surface image, and determining a second target area contour;

Determining the center structure centroid coordinates of the wind wheel based on the second target area outline;

and determining the actual error distance of the swept surface based on the second target area outline, the center of mass coordinates of the wind wheel center structure and the center point of the image of the swept surface.

4. The method of claim 2, wherein the first target area profile comprises a first blade-split profile, a second blade-split profile, and a nacelle-split profile, the calculating based on the first target area profile, determining blade-drop coordinates, comprising:

calculating based on the first blade segmentation contour, and determining a first blade centroid coordinate;

calculating based on the second blade segmentation contour, and determining a second blade centroid coordinate;

calculating based on the cabin segmentation profile, and determining cabin centroid coordinates;

And calculating based on the first blade centroid coordinates, the second blade centroid coordinates and the cabin centroid coordinates, and determining the blade drop foot coordinates.

5. The method of claim 4, wherein the determining a top actual error distance based on the first target region profile, the blade foot drop coordinates, and the top image center point comprises:

Calculating based on the blade foot drop coordinates and the top image center point, and determining a top image error distance;

Determining a top scaling factor based on the area of the nacelle dividing contour and the actual area of the nacelle top;

The top actual error distance is determined based on the top image error distance and the top scaling factor.

6. The method according to claim 2, wherein the method further comprises:

Under the condition that the actual error distance of the top does not meet the preset condition of the top, calculating based on the vertical foot coordinates of the blade and the center point of the top image, and determining to correct the flight direction;

After controlling the unmanned aerial vehicle to fly the top actual error distance based on the corrected flight direction, collecting corrected fan top images, wherein the corrected fan top images correspond to corrected image center points;

Identifying based on the corrected fan top image, and determining a corrected target area outline, wherein the corrected target area outline corresponds to corrected cabin centroid coordinates;

Calculating based on the correction target area outline, and determining correction blade drop foot coordinates;

determining a corrected top actual error distance based on the corrected target area outline, the correction blade foot drop coordinates and the corrected image center point;

and comparing the corrected top actual error distance with the top preset condition, and performing iteration execution on the condition that the corrected top actual error distance does not meet the top preset condition and the iteration number does not meet the iteration preset number until the corrected top actual error distance meets the top preset condition and the corrected cabin centroid coordinate is taken as the cabin centroid coordinate, and taking the corrected blade vertical foot coordinate as the blade vertical foot coordinate.

7. The method according to claim 2 or 6, wherein the wind wheel axial flight direction comprises a windward flight direction and a leeward flight direction, and wherein determining the wind wheel axial flight direction based on the blade foot drop coordinates in case the top actual error distance meets a top preset condition comprises:

determining the windward flight direction based on cabin centroid coordinates and the blade foot drop coordinates;

and determining the leeward flight direction based on the blade foot drop coordinates and the cabin centroid coordinates.

8. The method of claim 7, wherein the fan swept image comprises a fan windward image, wherein controlling the drone to fly against a fan blade center point and acquiring a fan swept image based on the rotor axial flight direction, a horizontal safety distance, and a blade safety height comprises:

Controlling the unmanned aerial vehicle to horizontally fly by the horizontal safety distance based on the windward flight direction, and then downwardly flying the blade safety height to enable the unmanned aerial vehicle to fly to the center point of the fan blade;

And adjusting the head orientation of the unmanned aerial vehicle based on the windward flight direction and collecting the windward image of the fan, wherein the windward image of the fan corresponds to a center point of the windward image.

9. The method of claim 8, wherein the second target area profile comprises a hub split profile, and wherein determining rotor center structure centroid coordinates based on the second target area profile comprises:

calculating based on the hub segmentation profile, and determining a hub centroid coordinate;

Determining a swept surface actual error distance based on the second target area profile, the rotor wheel center structure centroid coordinates and a swept surface image center point, including:

Calculating based on the hub centroid coordinates and the windward image center point, and determining a windward image error distance;

determining a windward side proportionality coefficient based on the area of the hub segmentation contour and the actual area of the hub;

And determining the actual error distance of the windward side based on the error distance of the windward side image and the windward side proportionality coefficient.

10. The method according to claim 9, wherein the method further comprises:

Under the condition that the actual error distance of the windward side does not meet the preset condition of the swept side, calculating based on the center of mass coordinate of the hub and the center point of the windward side image, and determining the windward correction flight direction;

After controlling the unmanned aerial vehicle to fly the windward actual error distance based on the windward correction flying direction, collecting corrected windward images, wherein the corrected windward images correspond to windward correction image center points;

identifying based on the corrected windward image, and determining a windward correction target area outline, wherein the windward correction target area outline corresponds to a correction hub centroid coordinate;

Determining a corrected windward actual error distance based on the windward correction target area outline, the correction hub centroid coordinates and the windward correction image center point;

and comparing the corrected windward actual error distance with the preset condition of the swept surface, and iteratively executing the process until the corrected windward actual error distance meets the preset condition of the swept surface and the corrected windward image is used as a fan windward image under the condition that the corrected windward actual error distance does not meet the preset condition of the swept surface and the iteration times meet the preset iteration times of the windward.

11. A method according to claim 3, wherein the second target area further comprises a third blade-split profile, a fourth blade-split profile, a fifth blade-split profile and a tower-split profile, and wherein determining the component reference flight direction of the wind turbine in case the swept surface actual error distance meets a swept surface preset condition comprises:

Calculating based on the third blade segmentation contour, and determining a third blade centroid coordinate;

calculating based on the fourth blade segmentation contour, and determining a fourth blade centroid coordinate;

Calculating based on the fifth blade segmentation contour, and determining a fifth blade centroid coordinate;

calculating based on the tower barrel segmentation contour, and determining a tower barrel centroid coordinate;

determining a third blade flight direction based on the third blade centroid coordinates and the wind wheel center structure centroid coordinates;

determining a fourth blade flight direction based on the fourth blade centroid coordinates and the wind wheel center structure centroid coordinates;

Determining a fifth blade flight direction based on the fifth blade centroid coordinates and the wind wheel center structure centroid coordinates;

and determining the flight direction of the tower barrel based on the barycenter coordinates of the tower barrel and the barycenter coordinates of the central structure of the wind wheel.

12. The method of claim 1, wherein the part safety distance comprises a blade length, the part inspection image overlap rate comprises a blade inspection image overlap rate, the part inspection shots comprise a blade inspection shots, and the determining the part single step distance sequence set and the part total movement distance based on the part safety distance, the part inspection image overlap rate, and the part inspection shots comprises:

determining a single shooting coverage length of the blade based on the blade length, the blade inspection image overlapping rate and the blade inspection shooting times;

calculating based on the single shooting coverage length and the overlapping rate of the blade inspection images, and determining the first step distance of the blade and the subsequent step length of the blade;

determining a single-step moving distance sequence set of the blade based on the first step distance of the blade, the follow-up step length of the blade and the inspection shooting times of the blade;

and determining the total moving distance of the blade based on the blade length and the single shooting coverage length of the blade.

13. The method of claim 1, wherein the part safety distance comprises a tower safety height, the part inspection image overlap ratio comprises a tower inspection image overlap ratio, the part inspection shots comprise tower inspection shots, and the determining the part single step distance sequence set and the part total movement distance based on the part safety distance, the part inspection image overlap ratio, and the part inspection shots comprises:

Determining single shooting coverage length of the tower barrel based on the tower barrel safety height, the tower barrel inspection image overlapping rate and the tower barrel inspection shooting times;

calculating based on the single shooting coverage length of the tower and the overlapping rate of the tower inspection images, and determining the first step distance of the tower and the subsequent step length of the tower;

determining a single-step moving distance sequence set of the tower barrel based on the first step distance of the tower barrel, the follow-up step length of the tower barrel and the inspection shooting times of the tower barrel;

and determining the total moving distance of the tower barrel based on the tower barrel safety height and the single shooting coverage length of the tower barrel.

14. The method of claim 12, wherein the performing image acquisition based on the component reference flight direction, the component single step distance sequence set, and the component total distance of movement to obtain a component inspection image set comprises:

Controlling the unmanned aerial vehicle to fly from the center point of an actual blade based on the third blade flying direction and the single-step moving distance sequence set of the blade, and acquiring images to obtain a third blade inspection image set;

controlling the unmanned aerial vehicle to fly to the actual blade center point based on the total moving distance of the blade;

controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the fourth blade flying direction and the single-step moving distance sequence set of the blade, and acquiring images to obtain a fourth blade inspection image set;

Controlling the unmanned aerial vehicle to fly from the center point of the actual blade based on the flight direction of the fifth blade and the single-step moving distance sequence set of the blade, and acquiring images to obtain a fifth blade inspection image set;

And controlling the unmanned aerial vehicle to fly to the actual blade center point based on the total moving distance of the blade.

15. The method of claim 13, wherein the performing image acquisition based on the component reference flight direction, the component single step distance sequence set, and the component total distance of movement to obtain a component inspection image set comprises:

controlling the unmanned aerial vehicle to fly from the center point of an actual blade based on the tower flight direction and the single-step movement distance sequence set of the tower, and acquiring images to obtain a tower inspection image set;

And controlling the unmanned aerial vehicle to fly to the center point of the actual blade based on the total moving distance of the tower.

16. The method of claim 7, wherein the fan swept surface image comprises a fan lee surface image, wherein the controlling the drone to fly to a fan blade center point and collecting the fan swept surface image based on the wind wheel axial flight direction, a horizontal safety distance, and a blade safety height comprises:

controlling the unmanned aerial vehicle to horizontally fly the horizontal safety distance based on the leeward flight direction, and then downwardly flying the blade safety height to enable the unmanned aerial vehicle to fly to the center point of the fan blade;

And adjusting the head direction of the unmanned aerial vehicle based on the leeward flight direction and collecting the leeward image of the fan, wherein the leeward image of the fan corresponds to a leeward image center point.

17. The method of claim 16, wherein the second target area profile comprises a further nacelle-split profile, determining rotor-center-structure centroid coordinates based on the second target area profile, comprising:

Calculating based on the cabin centroid coordinates and the leeward image center point, and determining an leeward image error distance;

determining a leeward scaling factor based on the area of the cabin segmentation contour and the actual cabin area;

and determining the actual error distance of the leeward surface based on the leeward surface image error distance and the leeward surface proportionality coefficient.

18. The method of claim 17, wherein the method further comprises:

Under the condition that the actual error distance of the leeward does not meet the preset condition of the swept surface, calculating based on the cabin centroid coordinates and the leeward image center point, and determining the leeward correction flight direction;

after controlling the unmanned aerial vehicle to fly the actual error distance of the leeward based on the leeward correction flying direction, collecting corrected leeward images, wherein the corrected leeward images correspond to leeward correction image center points;

Identifying based on the corrected leeward image, and determining a leeward correction target area outline, wherein the leeward correction target area outline corresponds to corrected cabin centroid coordinates;

Determining a corrected leeward actual error distance based on the leeward corrected target area contour, the corrected cabin centroid coordinates and the leeward corrected image center point;

And comparing the corrected leeward actual error distance with the preset condition of the swept surface, and iteratively executing the process until the corrected leeward actual error distance meets the preset condition of the swept surface and the corrected leeward image is used as a fan leeward image under the condition that the corrected leeward actual error distance does not meet the preset condition of the swept surface and the iteration number meets the preset iteration number of the leeward surface.

19. Unmanned aerial vehicle-based wind driven generator inspection device, characterized in that the device comprises:

20. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 18 when the computer program is executed.

21. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 18.