Segmentation and Defect Classification of the Power Line Insulators: A Deep Learning-based Approach

TL;DR

This paper presents a two-stage deep learning model for automatic segmentation and classification of power line insulators, effectively identifying various fault types to improve inspection efficiency and reliability.

Contribution

It introduces a novel two-stage deep learning approach that segments insulators and classifies their faults into four categories, addressing limitations of previous studies.

Findings

High accuracy in insulator segmentation

Effective detection of multiple fault types

Reduces inspection time and costs

Abstract

Power transmission networks physically connect the power generators to the electric consumers. Such systems extend over hundreds of kilometers. There are many components in the transmission infrastructure that require a proper inspection to guarantee flawless performance and reliable delivery, which, if done manually, can be very costly and time consuming. One essential component is the insulator. Its failure can cause an interruption of the entire transmission line or a widespread power failure. Automated fault detection could significantly decrease inspection time and related costs. Recently, several works have been proposed based on convolutional neural networks, which address the issue mentioned above. However, existing studies focus on a specific type of insulator faults. Thus, in this study, we introduce a two-stage model that segments insulators from their background to then classify their states based on four different categories, namely: healthy, broken, burned/corroded and missing cap. The test results show that the proposed approach can realize the effective segmentation of insulators and achieve high accuracy in detecting several types of faults.