A method for detecting pollution flashover status of insulators based on aerial images of distribution lines

Abstract

The invention relates to a method for detecting the state of insulator pollution flashover based on aerial images of power distribution lines. The method for training the detection and classification model of insulator image pollution flashover is as follows: convert the RGB space of the aerial image during line patrol to YCbCr space; set the segmentation threshold, Binarize the image according to the threshold; separate the insulator image, specify the image of the background and normal insulator as a positive sample, and the image containing the insulator pollution flashover as a negative sample; data enhancement; 6) extract all insulator images in the V channel color component Five features including mean, variance, slope, kurtosis, and entropy; design and train an SVM classifier to obtain a classifier model, thereby obtaining a classification model for insulator image pollution flashover detection.

Description

technical field [0001] The invention belongs to the fields of electric power technology and computer vision, and relates to a method for detecting the pollution flashover state of an insulator in an aerial image of an unmanned aerial vehicle inspection based on image processing technology and machine learning technology. Background technique [0002] The power distribution line refers to the line that sends the power from the step-down substation to the distribution transformer or the power from the distribution substation to the power unit. As the end of the power network, it is directly connected to the user. Safety, quality and other requirements, its operation safety, reliability and economy are directly related to social production and people's life. [0003] In the distribution line, the insulator plays the role of supporting the wire and preventing the current from returning to the ground, and is one of the essential equipment in the high-voltage transmission line. T...

AI Technical Summary

Problems solved by technology

In addition, the equipment cleaning work is in a passive state, which cannot guarantee that the equipment will be cleaned in a timely and effective manner.

[0008] ③The use of anti-pollution insulators and composite insulators, the selection of appropriate creepage distances, the step-down (limited voltage) operation of the power grid in bad weather, and the application of hydrophobic coatings on the surface of insulators have all played a certain role, but from Fundamentally speaking, it is all passive prevention, without taking the initiative. One of the reasons is the lack of comprehensive and true information on the degree of pollution flashover of insulators, and the effect on reducing pollution flashover accidents is not very satisfactory.

However, after the inspection, these systems will obtain a large amount of image information, and a large number of images are still inspected by humans, and the degree of intelligence is not high, and if the subjective judgment of the staff is used for these data without the automatic image analysis function, Misjudgments or missed judgments are very easy to occur, and it is difficult to accurately detect potential safety hazards of power distribution equipment, which cannot meet the needs of smart grid construction

Images