Abstract

The aim of this study is to improve the accuracy of localization of defects in insulators and determine their type. This goal is achieved by solving the problem of combining contact and remote methods for polymer and porcelain insulators using model partial discharges. The most significant results are the regularities of the dynamics of the characteristics of partial discharges up to the prebreakdown situation for porcelain insulators, the study of the statistical distributions of partial discharges depending on their intensity, and the identification of the features of the statistical distribution of surface discharges. Part of the work is devoted to the study of the characteristics of partial discharges and their sources by spectra, polarity, statistical distributions, oscillograms, which is important from the point of view of automating the recognition of corona and internal partial discharges, as well as for the recognition of porcelain insulators destroyed by partial discharges. Regularities of changes in the statistical distribution of partial discharges up to the pre-breakdown situation were established. At the same time, the breakdown signs of the model discharge gap, the breakdown voltage values for defective and operable porcelain insulators are determined, which can be used to train models of artificial neural networks and recognize the pre-breakdown situation based on them. The most significant results were: assessment of the ohmic resistance of porcelain insulators by the characteristics of partial discharges, recognition of corona, internal and surface partial discharges of polymer insulators, localization of defects, using electromagnetic radiation sensors.

Highlights

  • Metodă complexă de evaluare a caracterului deteriorării, localizare și predicție a defectării izolatorului în echipamentele electrice de înaltă tensiune pe baza caracteristicilor descărcărilor parțiale Gataullin А.М., Gubaev D.F

  • Область интересов: релейная защита, диагностика изоляторов высоковольтного электрооборудования методом частичных разрядов

Read more

Summary

Обзор литературы и описание нерешенных проблем

В ряде работ [1-2] искусственная нейронная сеть (ИНР) обучалась распознаванию внутренних. ЧР, поверхностных ЧР (ПЧР) и коронных разрядов, источником которых была система поверхность-игла. При этом ИНР обучали на основе осциллограмм ЧР и распределений ЧР по фазе приложенного напряжения, так называемых амплитудно-фазовых диаграммам (АФД). Недостатком этих работ является то, что в них не рассмотрены способы обучения ИНР для локализации источника ЧР

Ряд работ посвящены классификации типов
Цель исследования
Переменное напряжение
Transactions on Electrical and Electronic
Temperature in Moving Transformer Oil
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call