“Outage Forecast” – A Real Application of Machine Learning on Grid Operation Management Strategies

Thumbnail Image

Paper number

1209

Working Group Number

Conference name

CIRED 2019

Conference date

3-6 June 2019

Conference location

Madrid, Spain

Peer-reviewed

Yes

Short title

Convener

Authors

Almeida, Bernardo, EDP Distribuição, Portugal
Faria, Gonçalo, EDP Distribuição, Portugal
Soares, Tiago, EDP Distribuição, Portugal
Santos, Ricardo Jorge, EDP Distribuição, Portugal
Ferreira Pinto, José, EDP Distribuição, Portugal
Santos, Tiago, Smartwatt, Portugal
Preto, Isabel, Smartwatt, Portugal
Monteiro, Cláudio, FEUP, Portugal

Abstract

The energy sector is under one of the biggest transformations ever. Driven by the current digitalization process being held in all sectors, energy utilities are embracing data analytics and artificial intelligence to face current challenges and create new opportunities. Utilities need to act fast to unlock the true potential of the digital grid. The use of real-time data to monitor and operate grid infrastructure under efforts to improve the efficiency and reliability of energy networks is increasing at a global scale.A lot of utilities experience challenges associated with extreme weather conditions and its implications in quality of service, so a special focus on storm analytics, reliability, resiliency and outage management is required. With that in mind and being work force management (WFM) an extremely value-driving activity, EDP Distribuição (EDPD), the Portuguese DSO, started investing on new methodologies based on data analytics and machine learning to aid in the decision-making process of operational planning activities like the ones presented in this paper.This paper presents the newly developed tool that uses machine learning algorithms to predict the number and location of outages on EDPD’s high and medium voltage grid based on weather forecast. The paper shows in detail the used methodology and presents the project’s first results.

Table of content

Keywords

Publisher

AIM

Date

2019-06-03

Published in

CIRED 2019 Conference
Load more

Permanent link to this record

https://cired-repository.org/handle/20.500.12455/344
http://dx.doi.org/10.34890/570

ISSN

2032-9644

ISBN

978-2-9602415-0-1