Hybrid probabilistic wind power forecasting using temporally local Gaussian process

Juan Yan, Kang Li, Er-Wei Bai, Jing Deng, Aoife Foley

Research output: Contribution to journalArticle

61 Citations (Scopus)
647 Downloads (Pure)

Abstract

The demand for sustainable development has resulted in a rapid growth in wind power worldwide. Despite various approaches have been proposed to improve the accuracy and to overcome the uncertainties associated with traditional methods, the stochastic and variable nature of wind still remains the most challenging issue in accurately forecasting wind power. This paper presents a hybrid deterministic-probabilistic method where a temporally local ‘moving window’ technique is used in Gaussian Process to examine estimated forecasting errors. This temporally local Gaussian Process employs less measurement data while faster and better predicts wind power at two wind farms, one in the USA and the other in Ireland. Statistical analysis on the results shows that the method can substantially reduce the forecasting error while more likely generate Gaussian-distributed residuals, particularly for short-term forecast horizons due to its capability to handle the time-varying characteristics of wind power.
Original languageEnglish
Pages (from-to)87 - 95
Number of pages9
JournalIEEE Transactions on Sustainable Energy
Volume7
Issue number1
Early online date23 Sep 2015
DOIs
Publication statusPublished - Jan 2016

Bibliographical note

The demand for sustainable development has resulted in a rapid growth in wind power worldwide. The stochastic and variable nature of wind remains the most challenging issue in accurately forecasting wind power. As a joint effort of international collaborations between UK, US and China, under the auspice of the EPSRC-NSFC jointly funded large scale project on electric vehicles and smart grid (EP/L001063/1), this paper presents the development of a hybrid deterministic-probabilistic method, leading to faster and better predictions of wind power, with the method being verified at two wind farms from the USA and Ireland.

Keywords

  • Error analysis
  • forecasting
  • Gaussian process (GP)
  • wind power

Fingerprint Dive into the research topics of 'Hybrid probabilistic wind power forecasting using temporally local Gaussian process'. Together they form a unique fingerprint.

  • Cite this