Day-ahead forecasting of wholesale electricity pricing using extreme learning machine

J. E.Christine Tee, T. T. Teo, T. Logenthiran, W. L. Woo, K. Abidi

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

8 Citations (Scopus)

Abstract

In a deregulated electricity market where consumers can prepare bidding plans and purchase electricity directly from supplies, consumers can expect the price to fluctuate based on the demand. The consumers can also make economic beneficial decision to use electricity when the price is low. In this context, accurate forecast of the electricity price enable the consumers to plan and make such decisions. This paper proposes a methodology to forecast day-ahead electricity pricing using extreme learning machine. An artificial neural network forecasting model enables inputs variables that affect the output variable. The forecasting model is implemented in MATLAB/Simulink software. The proposed methodology is compared with a simple moving average model, and empirical evidence shows that the proposed methodology has a higher accuracy.
Original languageEnglish
Title of host publicationIEEE Region 10 Annual International Conference, Proceedings/TENCON
PublisherIEEE
Pages2973-2977
Number of pages5
ISBN (Print)9781509011339
DOIs
Publication statusPublished - 21 Dec 2017

Publication series

NameIEEE Region 10 Annual International Conference, Proceedings/TENCON
Volume2017-December

Keywords

  • Artificial Neural Network
  • Electricity Price Forecasting
  • Extreme Learning Machine
  • Full Retail Competition
  • Wholesale Electricity Pricing

Fingerprint

Dive into the research topics of 'Day-ahead forecasting of wholesale electricity pricing using extreme learning machine'. Together they form a unique fingerprint.

Cite this