Insulator abnormal automatic detection method based on ensemble classifier online learning

Abstract

The invention discloses an insulator abnormal automatic detection method based on ensemble classifier online learning. The method comprises the following concrete steps of carrying a high-definition PTZ camera by using a unmanned aerial vehicle / pole tower as a carrier; shooting a power transmission line visible light insulator image; performing preprocessing, image division, image multi-feature extraction and the like on the obtained image to obtain various typical features of abnormal isolator images and to perform normalization processing; next, selecting decision trees to be used as a weakclassifier; using the normalization features as the classification attribute to train the weak classifier; performing repeated training to obtain a plurality of decision trees and the classification weight; finally, generating a strong classifier with high classification accuracy through weighted voting; meanwhile, timely updating the strong classifier by combining the online learning technology;and separating each type of abnormal insulator images for subsequent processing. The method has the advantages that the principle is simple and visual; the implementation is easy; the image processingtechnology and the machine learning algorithm are combined for intelligently recognizing the insulator abnormality; and a novel idea and method is provided for the insulator abnormal operation monitoring.

Description

technical field [0001] The invention belongs to the technical fields of transmission line insulator anomaly detection, machine learning and image processing, and in particular relates to an insulator anomaly automatic detection method based on integrated classifier online learning. Background technique [0002] Insulators are widely used in power grids and are of various types. They are one of the key components of high-voltage overhead transmission lines and occupy a very important position. On the one hand, it is responsible for the mechanical support of the wire; on the other hand, it plays the role of insulation to prevent the current from forming a channel to the ground. However, due to the long-term exposure of insulators to the wild environment, various electrical and mechanical failures are likely to occur, which may lead to interruption of the power supply system, and even paralysis of the power grid in severe cases. Therefore, insulators are directly related to the...

AI Technical Summary

Problems solved by technology

However, there are still few studies on online classification and detection of insulator anomalies using machine learning combined with digital image processing technology.

Images