Learning technique for real-time congestion management in an active distribution networks

M.S. Babar, A.N.M.M. Haque, H.P. Nguyen, V. Cuk, I.G. Kamphuis, J.G. Slootweg, M. Bongaerts

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Downloads (Pure)

Abstract

During the last few decades, the concept of demand response (DR) in the energy sector has gained substantial momentum. Research has led to a range of DR solutions. These solutions mostly differ in their applications, the hosting power system, the energy market etc. Moreover, as per the EU directive, DR aggregators should be allowed to trade DR alongside supply in both day-ahead and real time electricity markets. Meanwhile, independent aggregators do not consider physical limitations of a network, thus setting up new a challenges for network operation. In this paper, an active learning technique for real-time congestion management is proposed to tackle this challenge. This enables distributed system operator (DSO) to incenticize independent aggregators efficiently in order to use DR for overloading mitigation. Lastly, a case study is simulated which verifies the performance of a new approach for congestion management.

Original languageEnglish
Title of host publication2017 IEEE Manchester PowerTech, 18-22 June 2017, Manchester, United Kingdom
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages1-6
ISBN (Electronic)978-1-5090-4237-1
ISBN (Print)978-1-5090-4238-8
DOIs
Publication statusPublished - 18 Jun 2017
Event12th IEEE PES PowerTech Conference - University of Manchester, Manchester, United Kingdom
Duration: 18 Jun 201722 Jun 2017
Conference number: 12
http://ieee-powertech.org/

Conference

Conference12th IEEE PES PowerTech Conference
Abbreviated titlePowerTech 2017
CountryUnited Kingdom
CityManchester
Period18/06/1722/06/17
OtherTowards and Beyond Sustainable Energy Systems
Internet address

Keywords

  • active distribution network
  • congestion management
  • demand response
  • market-based control mechanism
  • reinforcement learning

Fingerprint Dive into the research topics of 'Learning technique for real-time congestion management in an active distribution networks'. Together they form a unique fingerprint.

Cite this