Abstract
In this contribution, we used machine learning to forecast both the HVAC energy consumption of an office building and the thermal comfort in its office spaces. To a limited extent, we used the forecasts to optimize control values of the building’s automation system for minimal HVAC energy consumption. Our investigation was based on data collected in an office building in Vienna, Austria. To structure the available data, we leveraged an existing semantic model of the building. We found that the forecasts for the HVAC energy consumption were more reliable than those for thermal comfort, and that the control value optimization has potential for real-life application. The results of this work are intended to support the facility management of the building in the operation of the building.
Originalsprache | Englisch |
---|---|
Titel | Intelligente Energie- und Klimastrategien |
Untertitel | Energie – Gebäude – Umwelt |
Redakteure/-innen | Hildegard Gremmel-Simon |
Erscheinungsort | Wien |
Seiten | 41-48 |
Seitenumfang | 8 |
ISBN (elektronisch) | 978-3-903207-89-9 |
DOIs | |
Publikationsstatus | Veröffentlicht - Juni 2024 |
Veranstaltung | e·nova: Intelligente Energie- und Klimastrategien - FH Burgenland, Pinkafeld, Österreich Dauer: 12 Juni 2024 → 13 Juni 2024 https://www.fh-burgenland.at/bachelor-energie-und-umweltmanagement/enova/ |
Konferenz
Konferenz | e·nova |
---|---|
Kurztitel | e-nova 2024 |
Land/Gebiet | Österreich |
Stadt | Pinkafeld |
Zeitraum | 12/06/24 → 13/06/24 |
Internetadresse |
Research Field
- Efficient Buildings and HVAC Technologies