CN-122016832-A - Unmanned aerial vehicle inspection-based power equipment defect detection method

CN122016832ACN 122016832 ACN122016832 ACN 122016832ACN-122016832-A

Abstract

The invention relates to the technical field of inspection and detection of power equipment, in particular to a power equipment defect detection method based on unmanned aerial vehicle inspection. According to the invention, the radial detection distance of the unmanned aerial vehicle is determined through the fan blade size data of the wind driven generator, the inspection flight speed of the unmanned aerial vehicle is combined, the inspection flight path is planned, whether an abnormal area exists or not is judged according to the collected fan blade inspection images, the boundary outline of the abnormal area is determined, the annular flight speed of the unmanned aerial vehicle at the directional inspection position is analyzed according to the current fan blade real-time rotating speed, the directional refined inspection of the fan blade defect area is realized, the image acquisition quality of the fan blade surface is ensured, meanwhile, the surface image with qualified definition is screened, the defect area characteristic data in the surface image is identified, the health state of the fan blade is judged, the overall balance stability of the wind driven generator set is analyzed when all the fan blade health states are in a normal state, the overall control of the overall balance stability analysis of the set is realized, and the overall operation life of the wind driven generator set is prolonged.

Inventors

Yao Tanchun
LI RUI

Assignees

长沙修能信息科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260312

Claims (10)

1. The utility model provides a power equipment defect detection method based on unmanned aerial vehicle inspection which is characterized in that includes: Based on the fan blade size data of the wind driven generator, determining the radial detection distance between the unmanned aerial vehicle and the fan blade, and combining the current fan blade real-time rotating speed analysis to obtain the inspection flying speed of the unmanned aerial vehicle, and planning the inspection route of the corresponding unmanned aerial vehicle; collecting fan blade inspection images of the unmanned aerial vehicle flying along an inspection route, judging whether the fan blade inspection images have abnormal areas or not, and determining boundary contours of the abnormal areas; determining the directional inspection position of the unmanned aerial vehicle based on the boundary contour of the abnormal region, analyzing the annular flight speed of the unmanned aerial vehicle at the directional inspection position by combining the current fan blade real-time rotating speed, and collecting surface image sequences covering different directional inspection positions; performing definition analysis on the surface image sequence, screening surface images with qualified definition, identifying defect area characteristic data in the surface images, and judging the health state of the fan blades; And when all the fan blade health states are normal states, analyzing the overall balance stability of the wind generating set based on the fan blade assembly parameters of the wind generating set.
2. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 1, wherein the step of planning the inspection route of the unmanned aerial vehicle is as follows: extracting a fan blade boundary outline from fan blade size data of a wind driven generator, and determining a radial detection distance between the unmanned aerial vehicle and the fan blade based on the diameters of reference shooting ranges corresponding to different shooting distances of the unmanned aerial vehicle; dividing the boundary contour of the fan blade into different boundary contour areas according to the reference shooting range of the unmanned aerial vehicle at the corresponding radial detection distance; establishing a three-dimensional space coordinate system by taking the hub center of the wind driven generator as a coordinate origin, and determining the coordinate time sequences of the central positions of the different boundary contour areas based on the real-time rotating speed of the current fan blades; Obtaining displacement of a central position of the same boundary contour area corresponding to unit time and displacement of a central position of an adjacent boundary contour area corresponding to unit time interval, and calculating a deflection angle of the unmanned aerial vehicle in a routing inspection direction and a routing inspection flying speed of the unmanned aerial vehicle; And combining radial detection distances between the unmanned aerial vehicle and the fan blades, establishing a routing inspection path node sequence according to the direction of the rotation plane of the fan blades, and sequentially connecting to form a routing inspection route of the unmanned aerial vehicle.
3. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 2, wherein the method for determining the radial detection distance between the unmanned aerial vehicle and the fan blade is as follows: screening the maximum outline width of the fan blade boundary in the outline of the fan blade boundary, and matching the corresponding shooting distance of the maximum outline width from the diameters of the reference shooting ranges corresponding to different shooting distances of the unmanned aerial vehicle; If the shooting distance corresponding to the maximum profile width is larger than the safe shooting distance of the fan blade of the wind driven generator, taking the shooting distance as the radial detection distance between the unmanned aerial vehicle and the fan blade; otherwise, the safety shooting distance of the fan blade of the wind driven generator is used as the radial detection distance between the unmanned aerial vehicle and the fan blade.
4. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 1, wherein the boundary contour of the abnormal area is determined by the following steps: splicing and integrating fan blade inspection image sets acquired by the unmanned aerial vehicle along the inspection route to obtain integrated fan blade integral images; Comparing the gray level of the pixel points of the fan blade integral image with the gray level of the fan blade standard defect-free image, calculating the gray level deviation value of the pixel points at the positions corresponding to the fan blade integral image and the standard defect-free image, and marking the pixel points as gray level abnormal points if the gray level deviation value of a certain pixel point exceeds a preset gray level deviation allowable range; And (3) carrying out adjacent region communication processing on all the gray abnormal points, if the gray abnormal communication region exists, judging that the fan blade inspection image has an abnormal region, carrying out edge detection on the abnormal region, and fitting to obtain the boundary contour of the abnormal region.
5. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 1, wherein the method for determining the directional inspection position of the unmanned aerial vehicle is as follows: based on the boundary contour of the abnormal region, determining the corresponding directional inspection radial distance of the unmanned aerial vehicle by combining the diameters of the reference shooting ranges of the unmanned aerial vehicle corresponding to different shooting distances; Extracting geometric center coordinates of boundary outlines of the abnormal areas, and determining a reference position of the unmanned aerial vehicle directional inspection by taking the geometric center coordinates as a reference and combining the corresponding directional inspection radial distance of the unmanned aerial vehicle; and extending a preset distance from the reference position to the boundary contour edge of the abnormal area, determining a plurality of auxiliary inspection positions, and forming the reference position and the auxiliary inspection positions into an unmanned aerial vehicle directional inspection position.
6. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 5, wherein the analysis mode of the annular flying speed of the unmanned aerial vehicle at the directional inspection position is as follows: Dividing the fan blade into a plurality of sections along the radial direction according to different rotating radiuses, obtaining the surface tangential linear velocity of each section according to the current real-time rotating speed of the fan blade and the rotating radius of each section, and extracting the surface tangential linear velocity at the geometric center position of the boundary contour; calculating a reference annular flight speed required by the unmanned aerial vehicle for maintaining relative static observation based on the annular inspection track radius of the unmanned aerial vehicle at the directional inspection position; And carrying out vector synthesis analysis according to the surface tangential linear velocity at the geometric center position of the boundary contour and the reference annular flying velocity to obtain the annular flying velocity of the unmanned aerial vehicle at the directional inspection position.
7. The method for detecting the defects of the power equipment based on the inspection of the unmanned aerial vehicle according to claim 6, wherein the calculation mode of the reference annular flying speed required by the unmanned aerial vehicle for maintaining the relative static observation is as follows: Acquiring a horizontal projection distance of the unmanned aerial vehicle directional inspection position relative to the hub center of the wind driven generator, and taking the horizontal projection distance as the annular inspection track radius of the unmanned aerial vehicle; carrying out ratio analysis on the tangential linear velocity of the surface at the geometric center position of the boundary contour and the annular inspection track radius of the unmanned aerial vehicle to obtain the angular velocity required by maintaining the relative static observation of the unmanned aerial vehicle; Comparing the required angular velocity with the allowable range of the safe flight angular velocity of the unmanned aerial vehicle, determining the reference annular flight angular velocity of the unmanned aerial vehicle, and multiplying the reference annular flight angular velocity by the annular inspection track radius of the unmanned aerial vehicle to obtain the vector module length of the reference annular flight velocity; And the tangential direction of the annular inspection track is taken as the vector direction of the reference annular flying speed, and the reference annular flying speed is formed by combining the vector module length.
8. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 1, wherein the method for screening the surface images with qualified definition is as follows: carrying out gray processing on each frame of image in the surface image sequence to obtain gradient amplitude values of each pixel point in the image, and establishing a gradient amplitude distribution matrix; Carrying out statistical feature extraction on the gradient amplitude distribution matrix to obtain a gradient amplitude distribution statistical feature index, wherein the gradient amplitude distribution statistical feature index comprises a mean value and a variance of gradient amplitude and a pixel point duty ratio of the gradient amplitude exceeding the mean value; and screening surface images with gradient amplitude distribution statistical characteristic indexes larger than the corresponding preset clear index threshold value based on the preset clear index threshold value corresponding to the clear surface image set of the unmanned aerial vehicle inspection fan blade, and taking the surface images as surface images with qualified definition.
9. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 8, wherein the fan blade health state judging mode is as follows: Carrying out image recognition on the surface image with qualified definition, dividing a defect area in the surface image, extracting outline size, texture characteristics and gray level characteristics of the defect area, and recognizing the type of the defect area and a corresponding characteristic data set; based on the material type and the operation life of the fan blade of the wind driven generator, a characteristic threshold value of performance degradation of a corresponding fan blade sample under the action of various defects is called from a fan blade structural performance degradation database; if any feature data in the feature data set corresponding to the defect area exceeds the corresponding feature threshold value, judging that the fan blade health state is a defect state, otherwise, judging that the fan blade health state is a normal state.
10. The method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection according to claim 9, wherein the method for analyzing the overall balance stability of the wind generating set is characterized by comprising the following specific contents: Extracting outline dimensions in feature data of defect areas corresponding to all blades of the wind driven generator, and calculating defect mass loss quantity corresponding to each blade by combining the corresponding density of the blade materials; Determining the vertical distance from the center of a defective area of the fan blade to the fan blade rotating axis based on the fan blade rotating axis in the fan blade assembling parameters of the wind generating set, and performing product operation on the vertical distance and the defective mass loss to obtain the mass eccentric moment of the defective area; Based on the installation angle of each fan blade in the fan blade assembly parameters, carrying out coordinate rotation transformation on the radial component and the tangential component of the mass eccentric moment of each fan blade defect area, and carrying out summation calculation on the radial component and the tangential component after all fan blade conversion to obtain the radial offset and the tangential offset of the mass center of the unit; Vector synthesis operation is carried out on the radial offset of the mass center of the unit and the tangential offset of the mass center of the unit, so that the overall mass center offset of the unit is obtained; and comparing the integral mass center offset of the wind generating set with a preset stability threshold value, and judging the integral balance stability of the wind generating set.

Description

Unmanned aerial vehicle inspection-based power equipment defect detection method Technical Field The invention relates to the technical field of inspection and detection of power equipment, in particular to a power equipment defect detection method based on unmanned aerial vehicle inspection. Background The fan blade of the wind driven generator is used as a core component for capturing wind energy, bears the effects of alternating load, environmental erosion and fatigue stress for a long time, is easy to generate defects such as cracks, corrosion and layering, and can cause serious equipment faults and even safety accidents if the defects cannot be found and processed in time. Therefore, the high-efficiency and accurate detection of the defects of the fan blades is of great significance in guaranteeing the safe operation of the wind generating set. With the rapid development of unmanned aerial vehicle technology, fan blade inspection technology based on unmanned aerial vehicle gradually becomes industry research hotspot. In the prior art, chinese patent publication No. CN120747780A discloses a fan blade defect detection method and device based on unmanned aerial vehicle dynamic inspection, the method is based on blade motion parameters, unmanned aerial vehicle flight state parameters and shooting time parameters, deblurring and splicing processing are carried out on an initial image of a blade acquired in unmanned aerial vehicle dynamic inspection to obtain a second image, feature extraction is carried out on the second image by adopting a SIFT feature matching algorithm to obtain a third image, defect detection is carried out on the third image by adopting a double-channel convolution neural network, and image definition processing and defect detection of a rotating blade are realized. The prior art has the following problems: 1, prior art only predicts the motion vector in order to handle the fuzzy problem of image through blade motion parameter and unmanned aerial vehicle flight parameter, does not combine flabellum size data to confirm unmanned aerial vehicle radial detection distance, also does not adopt general route collection image of patrolling and examining according to the route of patrolling and examining of flabellum rotational speed planning adaptation yet, has the problem of patrolling and examining route suitability inadequately, and image acquisition is incomplete when leading to unmanned aerial vehicle to patrol and examine, partial flabellum regional hourglass examine, easily causes the collision risk because of too closely with the flabellum distance simultaneously, influences and patrols and examines security and integrality. 2. In the prior art, defects are directly detected after pretreatment such as deblurring, alignment and the like are carried out on collected images, abnormal areas in the images are not recognized first, directional refined inspection is carried out, defect judgment is carried out only by means of single image collection, the problem of insufficient defect detection accuracy exists, the tiny and concealed defects on the surfaces of the fan blades cannot be effectively recognized, the conditions of missed judgment and misjudgment are easy to occur, and early defects of the fan blades cannot be found in time. 3. The prior art focuses on fan blade defect detection, the overall balance stability of the wind generating set is not analyzed, the running condition of the equipment is judged only through the single fan blade defect state, the problem of single detection and analysis dimension exists, the potential risks of unit centroid deviation, balance unbalance and the like caused by the fact that local defects of fan blades cannot be identified are caused, the running vibration of the unit is easy to aggravate, the abrasion of parts is accelerated, and the overall running life of the wind generating set is reduced. Disclosure of Invention The invention aims to overcome the defects in the prior art, and provides a power equipment defect detection method based on unmanned aerial vehicle inspection, which realizes intelligent and accurate detection of the fan blade defects of a wind driven generator and ensures safe and stable operation of the wind driven generator. The technical scheme includes that the method for detecting the defects of the power equipment based on the unmanned aerial vehicle inspection comprises the steps of determining radial detection distances between the unmanned aerial vehicle and blades based on fan blade size data of a wind driven generator, obtaining inspection flight speed of the unmanned aerial vehicle by combining current real-time rotating speed analysis of the blades, and planning inspection routes of corresponding unmanned aerial vehicles. And collecting fan blade inspection images of the unmanned aerial vehicle flying along the inspection route, judging whether the fan blade inspection images have abnormal areas, and determining the b