Evaluation of the level of prediction errors and sub-hourly variability of PV and wind generation in a future with a large amount of renewables

Robin Girard (1) , Arthur Bossavy (1) , Loïc Le Gars (1) , Georges Kariniotakis (1)

1 PERSEE - Centre Procédés, Énergies Renouvelables, Systèmes Énergétiques

In this paper we propose a method for the simulation of errors in renewable energy sources generation forecasting (photovoltaic and wind) for use in power system planning studies. The proposed methodology relies on 5 elementary simulation steps. The first step is the simulation of photovoltaic plant and wind farm power production, with a sufficient spatial and temporal resolution (few km and hourly time step), the second is the simulation of the localisation of production sites, the third step is the generation of forecast errors using historic data of numerical weather predictions, and the last step is the simulation of intra-hourly variations of photovoltaic production. Finally, it is discussed how these simulation tools can assist the evaluation of the required tertiary reserves in a power system with a large share of renewable energies into the mix.

Mots clés: renewable energy; simulation

Domaines: Energétique

Robin Girard, Arthur Bossavy, Loïc Le Gars, Georges Kariniotakis. Evaluation of the level of prediction errors and sub-hourly variability of PV and wind generation in a future with a large amount of renewables. 23rd International Conference on Electricity Distribution - CIRED 2015, Jun 2015, Lyon, France. ⟨hal-01255039⟩