Short-term solar irradiance and irradiation forecasts via different time series techniques: A preliminary study

Cédric Join (1, 2, 3) , Cyril Voyant (4, 5) , Michel Fliess (3, 6) , Marc Muselli (4) , Marie-Laure Nivet (4) , Christophe Paoli (7) , Frédéric Chaxel (2)

1 NON-A - Non-Asymptotic estimation for online systems
2 CRAN - Centre de Recherche en Automatique de Nancy
3 AL.I.E.N. - ALgèbre pour Identification & Estimation Numériques
4 SPE - Sciences pour l'environnement
5 Unité de Radiothérapie
6 LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
7 Département de Génie Informatique

This communication is devoted to solar irradiance and irradiation short-term forecasts, which are useful for electricity production. Several different time series approaches are employed. Our results and the corresponding numerical simulations show that techniques which do not need a large amount of historical data behave better than those which need them, especially when those data are quite noisy.

Keywords: Meteorological forecasts; machine learning; quick fluctuations; solar irradiance and irradiation; electricity production; time series; big data.; multilayer perceptron; big data; trends

Domains: Environmental Sciences; Environmental Engineering; Computer Science [cs]; Machine Learning [cs.LG]; Computer Science [cs]; Databases [cs.DB]; Computer Science [cs]; Neural and Evolutionary Computing [cs.NE]; Statistics [stat]; Machine Learning [stat.ML]; Mathematics [math]; Statistics [math.ST]; Statistics [stat]; Statistics Theory [stat.TH]; Mathematics [math]; Logic [math.LO]

Cédric Join, Cyril Voyant, Michel Fliess, Marc Muselli, Marie-Laure Nivet, et al.. Short-term solar irradiance and irradiation forecasts via different time series techniques: A preliminary study. 3rd International Symposium on Environment-Friendly Energies and Applications, EFEA 2014, Nov 2014, Paris, France. ⟨hal-01068569⟩