Research on Transmission Line Defect Detection Based on Deep Learning

Main Article Content

Liu Wuneng, Liu Lilong, Luo Changbing, Li Bin

Abstract

At present, the inspection method of transmission lines in China is manual inspection assisted by unmanned aerial
vehicles, which has a low degree of intelligence. In order to improve the intelligent level of patrol inspection, the
main defects of transmission lines, such as insulator self-explosion and bird's nest, are taken as detection objects,
aiming to explore a detection method of transmission line defects with high detection accuracy and high speed.
Aiming at the problems of heavy workload and low efficiency of manual defect image recognition, deep learning
technology is introduced into the defect recognition module of transmission line engineering acceptance. By
integrating the optimized Faster-R CNN image recognition algorithm to learn and recognize the collected images,
a lightweight transmission line defect detection method is constructed by combining depth separable convolution
and SVD decomposition. Experimental results show that the effectiveness and reliability of the deep learning
method in the identification and defect detection of high-voltage transmission line components are very high, and
Faster-R CNN can achieve the identification speed of nearly 0.147 s per piece.

Article Details

How to Cite
Liu Wuneng, Liu Lilong, Luo Changbing, Li Bin. (2021). Research on Transmission Line Defect Detection Based on Deep Learning. CONVERTER, 2021(6), 854-861. Retrieved from https://converter-magazine.info/index.php/converter/article/view/460
Section
Articles