Optimising electricity production and consumption requires fuel suppliers, electrical grid operators, trading operators and demand response aggregators to monitor and predict the wholesale electricity price. It is also essential for integrating renewable energies and dispatching electricity generators. Using predictive models to forecast wholesale electricity prices is highly beneficial to these businesses to prepare for cases where the price deviates from normal patterns. This report focuses on predicting the wholesale electricity price in Ireland using cleaned data from 2015 to 2021, containing specific features. With a literature review of the related work in this field, two models were selected to analyse the dataset an ARIMA and ETS. These concepts are explained, and the results of the models are discussed for forecasting 14 days and 30 days. The predictive accuracy of these models was compared by using the evaluation metrics MAPE, MAE and RMSE. The results show how the ETS model best predicted wholesale electricity prices in both time horizons. Although there are limitations associated with the data and predictive models, this paper highlights the potentiality of these predictive models in wholesale electricity price forecasting and envisions future steps to improve their performance.

COMPARATIVE ANALYSES OF FORECASTING TECHNIQUES FOR ELECTRICITY WHOLESALE PRICE UNDER HIGH PENETRATION OF RENEWABLE ENERGY SYSTEMS

De Rosa M.;
2022-01-01

Abstract

Optimising electricity production and consumption requires fuel suppliers, electrical grid operators, trading operators and demand response aggregators to monitor and predict the wholesale electricity price. It is also essential for integrating renewable energies and dispatching electricity generators. Using predictive models to forecast wholesale electricity prices is highly beneficial to these businesses to prepare for cases where the price deviates from normal patterns. This report focuses on predicting the wholesale electricity price in Ireland using cleaned data from 2015 to 2021, containing specific features. With a literature review of the related work in this field, two models were selected to analyse the dataset an ARIMA and ETS. These concepts are explained, and the results of the models are discussed for forecasting 14 days and 30 days. The predictive accuracy of these models was compared by using the evaluation metrics MAPE, MAE and RMSE. The results show how the ETS model best predicted wholesale electricity prices in both time horizons. Although there are limitations associated with the data and predictive models, this paper highlights the potentiality of these predictive models in wholesale electricity price forecasting and envisions future steps to improve their performance.
File in questo prodotto:
File Dimensione Formato  
Comparative_Analysis_of_Forecasting_Techniques_for_Electricity_Wholesale_Price.pdf

accesso chiuso

Tipologia: Documento in Post-print
Dimensione 732.71 kB
Formato Adobe PDF
732.71 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1156697
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact