Power Transformers: Predictive Maintenance

Thumbnail Image

Files

CIRED 2019 - 995.pdf (647.26 KB)

Paper number

995

Working Group Number

Conference name

CIRED 2019

Conference date

3-6 June 2019

Conference location

Madrid, Spain

Peer-reviewed

Yes

Short title

Convener

Authors

Rodrigues, Sílvio, Jungle, Portugal
Verdelho, Maria Inês, EDP Distribuição, Portugal
Ribeiro, Ana Filipa, EDP Distribuição, Portugal
Cordeiro, Luís, EDP Inovação, Portugal

Abstract

This paper presents a holistic machine learning solution to forecast current and future Power Transformers (PT) health condition. Current PT monitoring and maintenance standards recommendations followed by utility companies depend mostly on the type of the PT and past results. We propose a health framework that is capable of accurately deriving key actionable insights. It leverages data from different sources and distinct business units. Current and future oil health conditions are forecasted and summarized in a Key Performance Indicator (KPI). The results show that key oil properties are forecasted up to five years in advance with high accuracy and that the predictions are robust to data outliers. Future oil interventions are then recommended based on the KPI value of the forecasted measurements. The proposed framework allows utility companies to save considerable resources by reducing and improving the scheduling of oil analyses, as well as increasing the quality of service (QoS) standards of energy grids.

Table of content

Keywords

Publisher

AIM

Date

2019-06-03

Published in

CIRED 2019 Conference

Permanent link to this record

https://cired-repository.org/handle/20.500.12455/239
http://dx.doi.org/10.34890/465

ISSN

2032-9644

ISBN

978-2-9602415-0-1