CN-116977879-B - Overhead transmission line defect detection method and device, electronic equipment and storage medium

CN116977879BCN 116977879 BCN116977879 BCN 116977879BCN-116977879-B

Abstract

The invention discloses a method and a device for detecting defects of an overhead transmission line, electronic equipment and a storage medium. The method is characterized by comprising the steps of obtaining a simulation sample image of a simulation target power transmission line and a real sample image of the target power transmission line, carrying out model training on a pre-detection model established in advance according to the simulation sample image and the real sample image to obtain a defect detection model, and determining the power transmission line defect of the target power transmission line according to the defect detection model and the real sample image. The defect detection method and device for the overhead transmission line are capable of detecting the defect in the overhead transmission line through the unmanned aerial vehicle, and the accuracy rate of identifying the defect of the overhead transmission line is improved.

Inventors

WANG KAI
XIE HAOBIN
CHEN SHAOJUN
WU XUDONG
LIU JIAN
YANG JIALIN
LIU JIAQI
WANG ZHAORONG

Assignees

广东电网有限责任公司
广东电网有限责任公司惠州供电局

Dates

Publication Date: 20260505
Application Date: 20230807

Claims (7)

1. A method for detecting defects of an overhead transmission line, comprising: Obtaining a simulation sample image of a simulation target power line and a real sample image of the target power line; performing model training on a pre-established pre-detection model according to the simulated sample image and the real sample image to obtain a defect detection model; determining a power transmission line defect of the target power transmission line according to the defect detection model and the real sample image; performing model training on a pre-built pre-detection model according to the simulated sample image and the real sample image to obtain a defect detection model, wherein the method comprises the following steps of: performing image sample mixing according to the simulated sample image and the real sample image, and determining a mixed sample image set; Inputting the mixed sample image set into a pre-established pre-detection model for model training to obtain a first preliminary detection model; Inputting the real sample image into a pre-established pre-detection model for model training, and determining a second preliminary detection model; Determining a defect detection model according to the first preliminary detection model and the second preliminary detection model; The determining a defect detection model according to the first preliminary detection model and the second preliminary detection model comprises: inputting the simulated sample image into the first preliminary detection model for model test, and determining a first test result; Acquiring a target sample image from the real sample image; Inputting the target sample image into the second preliminary detection model for model test, and determining a second test result; determining a defect detection model according to the first test result and the second test result; The first test result comprises a first test curve and a first test defect result, the second test result comprises a second test curve and a second test defect result, and the determining the defect detection model according to the first test result and the second test result comprises the following steps: The curve change of the first test curve and the curve change of the second test curve are subjected to curve comparison to determine a curve comparison result, wherein the first test curve is an average accuracy rate change curve corresponding to the first test defect result, the first test defect result is a test defect result output by a first preliminary detection model, the second test curve is an average accuracy rate change curve corresponding to the second test defect result, the second test defect result is a test defect result output by a second preliminary detection model, the curve change is a curve trend, and the curve comparison result is a comparison result of the curve trend of the first test curve and the curve trend of the second test curve; determining a redundant image in a simulation sample image according to the first test defect result, the second test defect result and the curve comparison result, wherein the redundant image is redundant data in the simulation sample image; sample adjustment is carried out on the simulated sample image according to the redundant image, the adjusted simulated sample image is determined, and the adjustment times of sample adjustment are recorded; Returning to execute the operation of model training on a pre-detection model established in advance according to the simulation sample image and the real sample image under the condition that the adjustment times are smaller than a preset adjustment threshold value; and under the condition that the adjustment times reach a preset times threshold value, determining the first preliminary detection model as the defect detection model.
2. The method of claim 1, wherein the acquiring a simulated sample image of the simulated target power line comprises: Acquiring a connection structure and a connection mode of a target power transmission line; Creating a simulated power transmission line simulating the target power transmission line according to the connection structure and the connection mode, and simulating the power transmission line defect of the target power transmission line in the simulated power transmission line; and shooting the defect of the transmission line in the simulated transmission line through the unmanned aerial vehicle to obtain the simulated sample image.
3. The method of claim 1, wherein the acquiring a target sample image from the real sample image comprises: and performing image screening in the real sample image according to the preset target power line defect, and determining a target sample image corresponding to the target power line defect.
4. The method of claim 1, wherein the power line defect comprises at least one of a power line defect type and a power line defect location.
5. An overhead transmission line defect detection device, characterized by comprising: The data acquisition module is used for acquiring a simulation sample image of the simulated target power transmission line and a real sample image of the target power transmission line; The model training module is used for carrying out model training on a pre-built pre-detection model according to the simulated sample image and the real sample image to obtain a defect detection model; The defect detection module is used for determining the transmission line defect of the target transmission line according to the defect detection model and the real sample image; the model training module is specifically used for: performing image sample mixing according to the simulated sample image and the real sample image, and determining a mixed sample image set; Inputting the mixed sample image set into a pre-established pre-detection model for model training to obtain a first preliminary detection model; Inputting the real sample image into a pre-established pre-detection model for model training, and determining a second preliminary detection model; Determining a defect detection model according to the first preliminary detection model and the second preliminary detection model; the model training module is specifically further configured to: inputting the simulated sample image into the first preliminary detection model for model test, and determining a first test result; Acquiring a target sample image from the real sample image; Inputting the target sample image into the second preliminary detection model for model test, and determining a second test result; determining a defect detection model according to the first test result and the second test result; the model training module is specifically further configured to: The curve change of the first test curve and the curve change of the second test curve are subjected to curve comparison to determine a curve comparison result, wherein the first test curve is an average accuracy rate change curve corresponding to the first test defect result, the first test defect result is a test defect result output by a first preliminary detection model, the second test curve is an average accuracy rate change curve corresponding to the second test defect result, the second test defect result is a test defect result output by a second preliminary detection model, the curve change is a curve trend, and the curve comparison result is a comparison result of the curve trend of the first test curve and the curve trend of the second test curve; determining a redundant image in a simulation sample image according to the first test defect result, the second test defect result and the curve comparison result, wherein the redundant image is redundant data in the simulation sample image; sample adjustment is carried out on the simulated sample image according to the redundant image, the adjusted simulated sample image is determined, and the adjustment times of sample adjustment are recorded; Returning to execute the operation of model training on a pre-detection model established in advance according to the simulation sample image and the real sample image under the condition that the adjustment times are smaller than a preset adjustment threshold value; and under the condition that the adjustment times reach a preset times threshold value, determining the first preliminary detection model as the defect detection model.
6. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the overhead transmission line defect detection method of any one of claims 1-4.
7. A computer readable storage medium storing computer instructions for causing a processor to perform the overhead transmission line defect detection method of any one of claims 1-4.

Description

Overhead transmission line defect detection method and device, electronic equipment and storage medium Technical Field The present invention relates to the field of power transmission line detection, and in particular, to a method and apparatus for detecting defects of an overhead power transmission line, an electronic device, and a storage medium. Background The transmission network is used as an important task for conveying electric energy in the power network, the most important transmission line in the transmission network of the overhead transmission line is used, the safe conveying of the overhead transmission network is very important for transmission network staff, the transmission network staff needs to patrol the overhead transmission network, at present, the transmission network staff generally uses an unmanned plane to intelligently patrol the overhead transmission network, a large number of visible light images are shot for the transmission line through camera equipment in the unmanned plane, and defects in the transmission line are determined through the images. In the prior art, because a large number of images shot by the unmanned aerial vehicle are required to be checked one by power transmission network staff, and part of defects in the power transmission network are difficult to identify manually, the identification capability of the defects in the power transmission network in the prior art is too low. Disclosure of Invention The invention provides a method, a device, electronic equipment and a storage medium for detecting defects of an overhead transmission line, so as to realize accurate identification of the defects in the overhead transmission line. According to an aspect of the present invention, there is provided a method for detecting a defect of an overhead transmission line, including: Obtaining a simulation sample image of a simulation target power line and a real sample image of the target power line; performing model training on a pre-established pre-detection model according to the simulated sample image and the real sample image to obtain a defect detection model; And determining the transmission line defect of the target transmission line according to the defect detection model and the real sample image. According to another aspect of the present invention, there is provided an overhead transmission line defect detecting apparatus including: The data acquisition module is used for acquiring a simulation sample image of the simulated target power transmission line and a real sample image of the target power transmission line; The model training module is used for carrying out model training on a pre-built pre-detection model according to the simulated sample image and the real sample image to obtain a defect detection model; and the defect detection module is used for determining the transmission line defect of the target transmission line according to the defect detection model and the real sample image. According to another aspect of the present invention, there is provided an electronic apparatus including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the overhead transmission line defect detection method according to any one of the embodiments of the present invention. According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the overhead transmission line defect detection method according to any one of the embodiments of the present invention when executed. According to the technical scheme, the simulation sample image of the simulated target power transmission line and the real sample image of the target power transmission line are obtained, the identified sample data are increased through the simulation sample image of the simulated target power transmission line, the accuracy of identification is improved through the real sample image, the pre-detection model which is built in advance is subjected to model training according to the simulation sample image and the real sample image to obtain a defect detection model, the prediction accuracy of the model can be improved according to the real sample image and the simulation sample image through the defect detection model, the power transmission line defect of the target power transmission line is determined according to the defect detection model and the real sample image, the real sample image is detected through the defect detection model, and the power transmission line defect is identified. The defect detection method and the device for the overhead transmission line are used for detecting the defect of the overhead transmission line through the unmanned aerial vehicle, improving the accuracy of identifying the defect of t