CN-121982089-A - Method and device for detecting vibration damper distance of power transmission tangent tower based on machine inspection image

CN121982089ACN 121982089 ACN121982089 ACN 121982089ACN-121982089-A

Abstract

The invention provides a method and a device for detecting the distance of a vibration damper of a power transmission tangent tower based on an inspection image, and belongs to the technical field of operation and maintenance of power transmission lines. According to the invention, the unmanned aerial vehicle is utilized to acquire the inspection image of the power transmission tower, and the design parameters of the inspection image are collected. And then, a pre-trained part detection network is used for identification, and detection frames of the damper and the suspension clamp are respectively extracted. Next, center point correction and tilt correction are performed on these detection frames. The actual distance represented by the unit pixel is determined using the standard design parameters of the component and the length of the pixel in the detection frame. And finally, calculating the actual distance between the damper and the suspension clamp according to the corrected connecting line length between the detection frames and the actual distance of the unit pixels. According to the invention, by combining the space information in the unmanned aerial vehicle inspection image and the power transmission line account data, the automatic detection of the vibration damper distance of the tangent tower is realized, and the defects of low efficiency, unstable quality, high risk and easiness in causing fatigue of an operator in the traditional measuring method are effectively overcome.

Inventors

SHENG SHIPING
HAN CHUNYU
HUANG YINGSHENG
ZHANG HU
ZENG YIHUI
LIU GAO
MAI JUNJIA
LUO XIANGYUAN
LIN ZIJIAN
XU ZHENLEI
HUANG FENG

Assignees

广东电网有限责任公司佛山供电局

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. The method for detecting the vibration damper distance of the power transmission tangent tower based on the machine inspection image is characterized by comprising the following steps: acquiring a machine inspection image of a key component of a transmission tangent tower to be detected; obtaining standard design parameters of the transmission tangent tower to be detected; Performing image recognition on the inspection images of the key components by utilizing a pre-trained component detection network to respectively obtain a damper detection frame and a suspension clamp detection frame; The center point correction is used for correcting the non-coincident error of a connecting line between the damper and the suspension clamp and the transmission tangent tower ground lead, and the inclination correction is used for correcting the inclination error during image inspection of a corrector; And calculating the actual distance between the damper and the suspension clamp based on the corrected connecting line length between the damper detection frame and the suspension clamp detection frame and the actual distance of the unit pixel, wherein the actual distance of the unit pixel is determined based on the standard design parameters.
2. The method for detecting the vibration damper distance of the power transmission tangent tower based on the machine inspection image according to claim 1, wherein in the step of obtaining the machine inspection image of the key component of the power transmission tangent tower to be detected, the method further comprises the steps of naming the machine inspection image of the key component by combining the power transmission tangent tower to be detected; the obtaining of the standard design parameters of the transmission tangent tower to be detected comprises the following steps: Extracting a line name and a tower number from the name of the key component machine inspection image; And inquiring the size information and the design installation distance of the corresponding pole tower installation damper from a database of the equipment by utilizing the line name and the pole tower number, and acquiring the longitude and latitude coordinates of the current tangent tower, the longitude and latitude coordinates of the small-size side pole tower and the longitude and latitude coordinates of the large-size side pole tower.
3. The machine inspection image-based transmission tangent tower damper distance detection method as defined in claim 1, wherein prior to the step of image recognition of the critical component machine inspection image by means of a pre-trained component detection network, an algorithm suitability check is further performed, comprising the steps of: extracting the yaw angle of the unmanned aerial vehicle when the critical component machine inspection image is imaged; converting the unmanned aerial vehicle yaw angle into a first direction vector; Calculating the direction vector of the connecting line of the big side pole and the small side pole according to the pre-acquired current longitude and latitude coordinates of the tangent tower, the longitude and latitude coordinates of the small side pole and the longitude and latitude coordinates of the big side pole; calculating a second direction vector of an angle bisector of the line included angle of the size side lever tower and the tangent tower based on the direction vector of the connecting line of the size side lever tower and the tangent tower; Calculating an included angle cosine between the first direction vector and the second direction vector; judging whether the cosine of the included angle is larger than a set value or not; If yes, the algorithm adaptability check passes, the subsequent steps are continued, and if not, the algorithm adaptability check does not pass.
4. The method for detecting the vibration damper distance of the power transmission tangent tower based on the machine inspection image according to claim 1, wherein in the step of performing image recognition on the critical component machine inspection image by using a pre-trained component detection network, the method comprises performing background filtering processing on the critical component machine inspection image by using a background filtering algorithm, wherein the background filtering processing comprises the following steps: grouping all the detection frames according to the component types; setting a threshold value for each group by taking the maximum widths of a plurality of detection frames under each group as references; and screening the detection frames with the detection frame width larger than the corresponding set threshold value from each group.
5. The method for detecting the distance of the damper to the power transmission tangent tower based on the machine inspection image according to claim 1, wherein the step of correcting the center point of the damper detection frame and the suspension clamp detection frame comprises: and according to the characteristics of the shockproof hammer and the suspension clamp, respectively adopting different correction factors to correct the ordinate of the central point of each detection frame to the direction of the ground wire.
6. The method for detecting the distance between the damper and the suspension clamp of the power transmission tangent tower based on the machine inspection image according to claim 1, wherein the step of performing inclination correction on the damper detection frame and the suspension clamp detection frame for a group of matched dampers and suspension clamps comprises: Calculating a first vector of the damper and the suspension clamp according to the coordinates of the center point of the detection frame; Calculating a second vector in the horizontal direction of the damper and a third vector in the horizontal direction of the suspension clamp by utilizing the widths of the detection frames; calculating a first angle cosine value between the first vector and the second vector and a second angle cosine value between the first vector and the third vector; Dividing the module of the second vector by the cosine value of the first included angle to obtain the corrected vibration damper length taking the pixel as a unit; dividing the modulus of the third vector by the cosine value of the second included angle to obtain the length of the corrected suspension clamp taking the pixel as a unit.
7. The method for detecting the distance of the damper of the power transmission tangent tower based on the machine inspection image according to claim 1, wherein if the form of the wire of the power transmission tangent tower to be detected is a double split wire, before the step of performing the inclination correction on the damper detection frame and the suspension clamp detection frame, the step of performing the component matching when the double split wire is performed further comprises the steps of: Calculating the geometric center coordinates of the suspension clamp by using the suspension clamp detection frame; Carrying out two classifications on the damper in the horizontal direction based on the geometric center of the suspension clamp to obtain dampers on the left side and the right side of the suspension clamp; Respectively calculating vectors between geometric center points of the vibration-proof hammers at two sides and the geometric center points of the suspension clamp to obtain connecting line vectors of the center points at two sides; Respectively calculating the vectors of the connecting lines of the shock-proof hammers on the left side and the right side and each suspension clamp to obtain a plurality of connecting line vectors of the shock-proof hammers on the two sides; For any one damper, calculating the cosine values of included angles between the connecting line vectors and the connecting line vectors of the central points on the same side respectively, and judging the suspension clamp corresponding to the cosine values of the included angles larger than a set threshold value as belonging to the same sub-wire as the damper; Calculating the pixel distance between the shockproof hammer and each suspension clamp belonging to the same sub-wire; selecting a suspension clamp with the minimum pixel distance as a clamp matched with the damper; And (3) calculating corresponding cosine values of included angles for each of the anti-vibration hammers at two sides and then matching, so that the matching relation between all the anti-vibration hammers and the suspension clamp is obtained.
8. The method for detecting the distance between the damper and the suspension clamp of the power transmission tangent tower based on the machine inspection image according to claim 1, wherein for a group of matched dampers and suspension clamps, the actual distance between the damper and the suspension clamp is calculated based on the corrected connection length between the damper detection frame and the suspension clamp detection frame and the actual distance of unit pixels, comprising: Calculating a connection line vector between the damper and the suspension clamp according to the coordinates of the central points of the detection frames; Calculating the actual distance of the unit pixel according to the actual length of the damper and the corrected damper length, wherein the actual length of the damper is determined according to the standard design parameter, and the corrected damper length is determined according to the tilt-corrected damper detection frame; Calculating the connection length between the center point of the damper and the edge of the suspension clamp by using the connection vector; Dividing the length of the connecting line by the actual distance of the unit pixel to obtain the actual distance between the damper and the suspension clamp.
9. The machine inspection image-based transmission tangent tower damper distance detection method as defined in claim 1, further comprising: And comparing the actual distance between the damper and the suspension clamp with the standard design parameters, and calculating an absolute displacement error and an absolute displacement error percentage to judge whether defects exist.
10. The utility model provides a transmission of electricity tangent tower damper distance detection device based on machine inspection image which characterized in that includes: the image acquisition module is used for acquiring an inspection image of a key component of the transmission tangent tower to be detected; the parameter acquisition module is used for acquiring standard design parameters of the transmission tangent tower to be detected; the image recognition module is used for carrying out image recognition on the inspection images of the key components by utilizing a pre-trained component detection network to respectively obtain a damper detection frame and a suspension clamp detection frame; The correction module is used for carrying out center point correction and inclination correction on the damper detection frame and the suspension clamp detection frame, wherein the center point correction is used for correcting the non-superposition error of a connecting line between the damper and the suspension clamp and the transmission tangent tower ground lead, and the inclination correction is used for correcting the inclination error during image imaging of the inspection images of the corrector; The distance calculation module is used for calculating the actual distance between the damper and the suspension clamp based on the corrected connecting line length between the damper detection frame and the suspension clamp detection frame and the actual distance of the unit pixel, and the actual distance of the unit pixel is determined based on the standard design parameters.

Description

Method and device for detecting vibration damper distance of power transmission tangent tower based on machine inspection image Technical Field The invention belongs to the technical field of operation and maintenance of power transmission lines, and particularly relates to a power transmission tangent tower damper distance detection method and device based on machine inspection images. Background In a modern power system, a power transmission line is used as a key channel for power transmission, and safe and stable operation of the power transmission line is important for guaranteeing normal power utilization of social production and life. With the rapid development of economy and the continuous increase of the power demand of society, the power transmission line is continuously expanded in scale, and the coverage range is wider and wider, so that the power transmission line can traverse various complicated geographic environments and climate conditions. In order to ensure that the transmission line can reliably run under various working conditions, a series of auxiliary equipment is required to ensure the performance of the transmission line, and a damper is an indispensable important device. The vibration damper is arranged on the wire of the power transmission line, and mainly aims to absorb and consume vibration energy generated on the wire due to wind force, so that vibration of the wire is effectively restrained, fatigue strand breakage and even wire breakage of the wire due to long-term vibration are prevented, and the safety and stability of the power transmission line are greatly improved. However, during the long-term operation of the power transmission line, the damper is easily displaced due to the comprehensive influence of various factors such as natural environment, mechanical stress, and the waving of the line itself. Once the damper is shifted, the originally designed energy absorption and shock absorption functions of the damper are greatly reduced, the effect of protecting the lead cannot be effectively exerted, and potential threat is formed to the safe operation of the transmission line. Therefore, the accurate detection of the installation distance of the damper and the timely discovery of the displacement problem of the damper become a key task in the operation and maintenance work of the power transmission line. At present, the detection method for the displacement defect of the damper mainly depends on manual operation, and has a plurality of defects. On the one hand, the manual inspection mode faces a plurality of practical difficulties. The inspection personnel need walk along the long transmission line, and the distance is far away and working strength is big, very easily produces tired sense, leads to the attention to descend, increases the risk of missing to examine. Moreover, complicated topography such as mountains, jungles, marshes, etc. can severely limit the actions and observation angles of inspection personnel, and it is difficult to comprehensively and clearly observe the actual state of the damper. In addition, when looking up, the sunlight directly directs and causes the phenomenon in a poor light easily for the details of damper are difficult to discern, further influence accuracy and the reliability that detects. On the other hand, the manual comparison analysis mode is adopted for the machine inspection images one by one, and because the unmanned aerial vehicle automatically inspects the collected image data volume is huge, a large amount of labor and time cost are required to be consumed, the analysis efficiency is extremely low, and the requirements of quick and efficient operation and maintenance of a modern large-scale power transmission line cannot be met. Disclosure of Invention In view of the above, the invention aims to provide a method and a device for detecting the vibration damper distance of a power transmission tangent tower based on an inspection image of an aircraft, which solve the problems of low efficiency, poor quality, high risk and easy fatigue existing in the traditional measurement method by utilizing the space information data in the inspection image data of the unmanned aircraft and the standing account data of a power transmission line to automatically detect the vibration damper distance of the tangent tower. In order to achieve the above purpose, the technical scheme provided by the invention is as follows: In a first aspect, the invention provides a transmission tangent tower damper distance detection method based on machine inspection images, which comprises the following steps: acquiring a machine inspection image of a key component of a transmission tangent tower to be detected; obtaining standard design parameters of a transmission tangent tower to be detected; Carrying out image recognition on the inspection images of the key components by utilizing a pre-trained component detection network to respectively obtain a damper detectio