Statistical parameters as a means to a priori assess the accuracy of solar forecasting models

Cyril Voyant (1) , Ted Soubdhan (2) , Philippe Lauret (3) , Mathieu David (3) , Marc Muselli (1)

1 SPE - Sciences pour l'environnement
2 LARGE - Laboratoire de Recherche en Géosciences et Energies [UR2_1]
3 PIMENT - Physique et Ingénierie Mathématique pour l'Énergie, l'environnemeNt et le bâtimenT

In this paper we propose to determinate and to test a set ofstatistical parameters (20)to estimate the predictability of the global horizontal irradiation time series and thereby propose a new prospective tool indicating the expected error regardlessthe forecasting methodsa modeller can possibly implement. The mean absolute log return, which is a tool usually used in econometry, proves to be a very good estimator. Some examples of the use of this tool are exposed, showing the interest of this statistical parameter in concrete cases of predictions or optimizations.

Mots clés: Solar forecasting; time series; clear sky models; fractal dimension; mutual information; log-return

Domaines: Recherche d'information [cs.IR]; Sciences de l'environnement

Cyril Voyant, Ted Soubdhan, Philippe Lauret, Mathieu David, Marc Muselli. Statistical parameters as a means to a priori assess the accuracy of solar forecasting models. Energy Journal, 2015, pp.1. ⟨hal-01180041⟩