Lightweight hybrid local PV short term forecasting using both physics and data - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Lightweight hybrid local PV short term forecasting using both physics and data

Résumé

There is an increase of PV generation on many distributed sites, especially rooftop PV on buildings. The emergence of aware end-users, that are doing their best to maximize their auto-consumption in an energy community using energy management systems, leads to a need for new adapted forecasting models. We ask the question in this article of the ability to produce with few data, a robust day ahead PV generation forecasting. To do that, we have applied a hybrid methodology taking advantage of both physical model and data-based model in order to reduce the amount and kind of data. We reach improvements from 5% to 12% depending on the model used, compared to pure data-based approaches.
Fichier principal
Vignette du fichier
APPEEC - DELINCHANT (before submission).pdf (1014.39 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04339309 , version 1 (13-12-2023)

Identifiants

  • HAL Id : hal-04339309 , version 1

Citer

Benoit Delinchant, Vincent Imard, Yaiza Bermudez, Frédéric Wurtz. Lightweight hybrid local PV short term forecasting using both physics and data. APPEEC, 15th Asia-Pacific Power and Energy Engineering Conference, IEEE PES, Dec 2023, Chiang Mai, Thailand. ⟨hal-04339309⟩
141 Consultations
58 Téléchargements

Partager

Gmail Facebook X LinkedIn More