Details
Originalsprache | Englisch |
---|---|
Titel des Sammelwerks | Proceedings of Building Simulation 2021 |
Untertitel | 17th Conference of IBPSA |
Herausgeber/-innen | Dirk Saelens, Jelle Laverge, Wim Boydens, Lieve Helsen |
Seiten | 1311-1316 |
Seitenumfang | 6 |
ISBN (elektronisch) | 978-1-7750520-2-9 |
Publikationsstatus | Veröffentlicht - Sept. 2021 |
Extern publiziert | Ja |
Veranstaltung | 17th IBPSA Conference - Bruges, Belgien Dauer: 1 Sept. 2021 → 3 Sept. 2021 |
Publikationsreihe
Name | Building Simulation Conference Proceedings |
---|---|
ISSN (Print) | 2522-2708 |
Abstract
ASJC Scopus Sachgebiete
- Ingenieurwesen (insg.)
- Bauwesen
- Informatik (insg.)
- Angewandte Informatik
- Ingenieurwesen (insg.)
- Architektur
- Mathematik (insg.)
- Modellierung und Simulation
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
Proceedings of Building Simulation 2021: 17th Conference of IBPSA. Hrsg. / Dirk Saelens; Jelle Laverge; Wim Boydens; Lieve Helsen. 2021. S. 1311-1316 (Building Simulation Conference Proceedings).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung
}
TY - GEN
T1 - CNN-based Quick Energy Prediction Model using Image Analysis for Shape Information
AU - Geyer, Philipp Florian
AU - Singh, Manav Mahan
N1 - Publisher Copyright: © International Building Performance Simulation Association, 2022
PY - 2021/9
Y1 - 2021/9
N2 - Data-driven approaches are useful to substitute computationally expensive tools used for energy prediction. These approaches allow developing quick energy prediction models, essential to promote energy analysis at the early stages. However, developing a wellgeneralising model is challenging due to varying building shape. This article develops such a model using a deep learning approach. A convolutional neural network (CNN) with modified architecture is used to capture the shape and technical specifications. The model predicts energy use intensity with a mean-absolute-percentageerror of 1.51% and root-mean-square-error of 1.06 kWh/m2a. Integrated with building information modelling (BIM), the model predicts probabilistic energy performance for 5000 samples in 65 seconds to support informed decision-making under uncertain scenario.
AB - Data-driven approaches are useful to substitute computationally expensive tools used for energy prediction. These approaches allow developing quick energy prediction models, essential to promote energy analysis at the early stages. However, developing a wellgeneralising model is challenging due to varying building shape. This article develops such a model using a deep learning approach. A convolutional neural network (CNN) with modified architecture is used to capture the shape and technical specifications. The model predicts energy use intensity with a mean-absolute-percentageerror of 1.51% and root-mean-square-error of 1.06 kWh/m2a. Integrated with building information modelling (BIM), the model predicts probabilistic energy performance for 5000 samples in 65 seconds to support informed decision-making under uncertain scenario.
UR - http://www.scopus.com/inward/record.url?scp=85151536066&partnerID=8YFLogxK
U2 - 10.26868/25222708.2021.30250
DO - 10.26868/25222708.2021.30250
M3 - Conference contribution
T3 - Building Simulation Conference Proceedings
SP - 1311
EP - 1316
BT - Proceedings of Building Simulation 2021
A2 - Saelens, Dirk
A2 - Laverge, Jelle
A2 - Boydens, Wim
A2 - Helsen, Lieve
T2 - 17th IBPSA Conference
Y2 - 1 September 2021 through 3 September 2021
ER -