Short-term Photovoltaic Power Forecasting Enhanced by Heterogeneous Sources of Spatio-temporal Information - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail (Preprint/Prepublication) Année : 2022

Short-term Photovoltaic Power Forecasting Enhanced by Heterogeneous Sources of Spatio-temporal Information

Résumé

The power output of Photovoltaic (PV) plants is weather-dependent, which leads to inherent uncertainties regarding the future production. With the growing share of renewables in the energy mix, such a characteristic raises technical challenges regarding the safe operation of the grid, and impacts the profitability of producers involved in electricity markets. This motivates the development of accurate forecasting tools for horizons ranging from a few minutes up to several days ahead. Two levers are usually available to improve the accuracy of predictions: forecasting tools, and input data. With the development of PV plants, and the advances in smart monitoring and measurements, we observe a paradigm shift from temporal- to spatio-temporal-based forecasting models. This family of models considers features that exploit spatio-temporal correlations in the data, such as observations from spatially distributed portfolios of PV plants or satellite-derived information. In this paper we provide a full assessment of the value of spatio-temporal data. First, the limits of a PV portfolio are highlighted through an analysis of the local topography and wind distribution at several altitudes. This motivates the use of 2D satellite-based maps. A features selection approach that we originally applied to the PV forecasting field is implemented to deal with the induced dimensionality burden. This approach enables the derivation of low-redundant features fairly distributed around the power plant. Lastly, we consider cloud opacity maps obtained from infrared channels. Despite being under-represented in the literature (only two studies have been found), infrared channel-based data present the advantage of offering nighttime observations of cloud cover, which contributes to improving early morning forecasts. This paper demonstrates the scientific interest of opacity maps compared with satellite-derived irradiance and irradiance forecasts for the field of short-term PV power forecasting. Evaluations are performed on real-world datasets composed of nine PV plants.
Fichier non déposé

Dates et versions

hal-03839690 , version 1 (04-11-2022)

Identifiants

  • HAL Id : hal-03839690 , version 1

Citer

Kevin Bellinguer, Robin Girard, Guillaume Bontron, Georges Kariniotakis. Short-term Photovoltaic Power Forecasting Enhanced by Heterogeneous Sources of Spatio-temporal Information. 2022. ⟨hal-03839690⟩
55 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More