Nouri, Bijan und Lezaca Galeano, Jorge Enrique und Fabel, Yann und Hammer, Annette und Blum, Niklas und Wilbert, Stefan (2026) 2026 EnMet Fachtagung: ENHANCING INTRA-HOUR SOLAR IRRADIANCE FORECASTING USING A HYBRID MODEL OF SATELLITE, SKY IMAGER AND PERSISTENCE. Fachtagung Energiemeteorologie 2026, 2026-05-05 - 2026-05-07, Bad Staffelstein, Germany.
|
PDF
3MB |
Kurzfassung
This study develops a hybrid forecasting approach to improve intra-hour solar irradiance predictions up to 1 hour ahead by combining all-sky imager (ASI)-based deep learning forecasts, satellite-based forecasts (Heliosat-3), and persistence models. Different blending techniques, including linear regression and advanced machine learning algorithms, are evaluated under varying sky conditions and spatial and temporal configurations.
The results show that the multisource approach significantly outperforms the standalone satellite forecast, reducing RMSE by 13.6% and MAE by 17.0%. The improvements result from the complementary strengths of the individual forecasting sources: ASI models capture rapidly changing cloud conditions, satellite forecasts provide broader spatial coverage, and persistence offers a robust short-term baseline. Furthermore, the ASI model demonstrates strong transferability, maintaining good performance when trained in southern Spain and validated at a climatically different site in northern Germany.
| elib-URL des Eintrags: | https://elib.dlr.de/228995/ | ||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dokumentart: | Konferenzbeitrag (Vortrag) | ||||||||||||||||||||||||||||
| Titel: | 2026 EnMet Fachtagung: ENHANCING INTRA-HOUR SOLAR IRRADIANCE FORECASTING USING A HYBRID MODEL OF SATELLITE, SKY IMAGER AND PERSISTENCE | ||||||||||||||||||||||||||||
| Autoren: |
| ||||||||||||||||||||||||||||
| Datum: | 2026 | ||||||||||||||||||||||||||||
| Referierte Publikation: | Ja | ||||||||||||||||||||||||||||
| Open Access: | Ja | ||||||||||||||||||||||||||||
| Gold Open Access: | Nein | ||||||||||||||||||||||||||||
| In SCOPUS: | Nein | ||||||||||||||||||||||||||||
| In ISI Web of Science: | Nein | ||||||||||||||||||||||||||||
| Status: | veröffentlicht | ||||||||||||||||||||||||||||
| Stichwörter: | all-sky imagers, hybrid models, irradiance, machine learning, persistence, satellite data, solar forecasting | ||||||||||||||||||||||||||||
| Veranstaltungstitel: | Fachtagung Energiemeteorologie 2026 | ||||||||||||||||||||||||||||
| Veranstaltungsort: | Bad Staffelstein, Germany | ||||||||||||||||||||||||||||
| Veranstaltungsart: | nationale Konferenz | ||||||||||||||||||||||||||||
| Veranstaltungsbeginn: | 5 Mai 2026 | ||||||||||||||||||||||||||||
| Veranstaltungsende: | 7 Mai 2026 | ||||||||||||||||||||||||||||
| HGF - Forschungsbereich: | Energie | ||||||||||||||||||||||||||||
| HGF - Programm: | Materialien und Technologien für die Energiewende | ||||||||||||||||||||||||||||
| HGF - Programmthema: | Thermische Hochtemperaturtechnologien | ||||||||||||||||||||||||||||
| DLR - Schwerpunkt: | Energie | ||||||||||||||||||||||||||||
| DLR - Forschungsgebiet: | E SW - Solar- und Windenergie | ||||||||||||||||||||||||||||
| DLR - Teilgebiet (Projekt, Vorhaben): | E - Condition Monitoring | ||||||||||||||||||||||||||||
| Standort: | Köln-Porz | ||||||||||||||||||||||||||||
| Institute & Einrichtungen: | Institut für Solarforschung > Qualifizierung Institut für Vernetzte Energiesysteme > Energiesystemanalyse, OL | ||||||||||||||||||||||||||||
| Hinterlegt von: | Nouri, Bijan | ||||||||||||||||||||||||||||
| Hinterlegt am: | 09 Nov 2026 08:20 | ||||||||||||||||||||||||||||
| Letzte Änderung: | 09 Nov 2026 08:20 |
Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags