View source for Publications/tschora.22.apen

{{Publication
| published = true
| date = 2022-03-10
| authors = Léonard Tschora, Erwan Pierre, Marc Plantevit, Céline Robardet
| title = Electricity Price Forecasting on the Day-Ahead Market using Machine Learning
| journal = Applied Energy
| volume = 313
| number = 118752
| None = Electricity price forecasting, Machine learning, Forecast evaluation, Open-access benchmark, Explainable AI (XAI)
| abstract = The price of electricity on the European market is very volatile. This is due both to its mode of production by different sources, each with its own constraints (volume of production, dependence on the weather, or production inertia), and by the difficulty of its storage. Being able to predict the prices of the next day is an important issue, to allow the development of intelligent uses of electricity. In this article, we investigate the capabilities of different machine learning techniques to accurately predict electricity prices. Specifically, we extend current state-of-the-art approaches by considering previously unused predictive features such as price histories of neighboring countries. We show that these features significantly improve the quality of forecastseven in the current period when sudden changes are occurring. We also develop an analysis of the contribution of the different features in model prediction using Shap values, in order to shed light on how models make their prediction and to build user confidence in models.
| lrdekeywords = IA
| lrdepaper = http://www.lrde.epita.fr/dload/papers/tschora.22.apen.pdf
| lrdenewsdate = 2022-03-10
| type = article
| id = tschora.22.apen
| identifier = doi:10.1016/j.apenergy.2022.118752
| bibtex = 
@Article<nowiki>{</nowiki>	  tschora.22.apen,
  author	= <nowiki>{</nowiki>L\'eonard Tschora and Erwan Pierre and Marc Plantevit and
		  C\'eline Robardet<nowiki>}</nowiki>,
  title		= <nowiki>{</nowiki>Electricity Price Forecasting on the Day-Ahead Market
		  using Machine Learning<nowiki>}</nowiki>,
  journal	= <nowiki>{</nowiki>Applied Energy<nowiki>}</nowiki>,
  volume	= <nowiki>{</nowiki>313<nowiki>}</nowiki>,
  number	= <nowiki>{</nowiki>118752<nowiki>}</nowiki>,
  year		= <nowiki>{</nowiki>2022<nowiki>}</nowiki>,
  month		= may,
  doi		= <nowiki>{</nowiki>10.1016/j.apenergy.2022.118752<nowiki>}</nowiki>,
  keywords	= <nowiki>{</nowiki>Electricity price forecasting, Machine learning, Forecast
		  evaluation, Open-access benchmark, Explainable AI (XAI)<nowiki>}</nowiki>,
  abstract	= <nowiki>{</nowiki>The price of electricity on the European market is very
		  volatile. This is due both to its mode of production by
		  different sources, each with its own constraints (volume of
		  production, dependence on the weather, or production
		  inertia), and by the difficulty of its storage. Being able
		  to predict the prices of the next day is an important
		  issue, to allow the development of intelligent uses of
		  electricity. In this article, we investigate the
		  capabilities of different machine learning techniques to
		  accurately predict electricity prices. Specifically, we
		  extend current state-of-the-art approaches by considering
		  previously unused predictive features such as price
		  histories of neighboring countries. We show that these
		  features significantly improve the quality of forecasts,
		  even in the current period when sudden changes are
		  occurring. We also develop an analysis of the contribution
		  of the different features in model prediction using Shap
		  values, in order to shed light on how models make their
		  prediction and to build user confidence in models.<nowiki>}</nowiki>
<nowiki>}</nowiki>

}}