Cluster-based regime-switching AR for the EEM 2017 Wind Power Forecasting Competition

Jethro Browell, Ciaran P Gilbert

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

217 Downloads (Pure)

Abstract

This paper describes the regime-switching autoregressive models used to win the EEM 2017 Wind Power Forecasting Competition. The competition required participants to produce daily forecast wind power production for a portfolio of wind farms from 2 to 38 hours-ahead based on historic generation and numerical weather prediction analysis data only. The regimes used in the methodology presented are defined on the previous day’s weather conditions using the k-medians clustering algorithm. Cross-validation is used to identify models with the best predictive power from a pool of candidate models. The final methodology produced a final weighted mean absolute error 4.5% lower than the second place team during the two-week competition period.
Original languageEnglish
Title of host publication2017 14th International Conference on the European Electricity Market Conference (EEM)
Place of PublicationPiscataway
PublisherIEEE
ISBN (Print)9781509054992
Publication statusAccepted/In press - 19 May 2017
Event14th International Conference on the European Energy Market - Technische Universität Dresden, Dresden, Germany
Duration: 6 Jun 20179 Jun 2017

Conference

Conference14th International Conference on the European Energy Market
Abbreviated titleEEM 17
Country/TerritoryGermany
CityDresden
Period6/06/179/06/17

Keywords

  • wind power
  • forecasting
  • time series
  • clustering
  • autoregression
  • regime switching

Fingerprint

Dive into the research topics of 'Cluster-based regime-switching AR for the EEM 2017 Wind Power Forecasting Competition'. Together they form a unique fingerprint.

Cite this