Datum
2022-03-29Schlagwort
600 Technik LastprofilElektrizitätHitzeIndustrieMaschinelles LernenPrognoseAnomalieerkennungSupplement
doi:10.48662/daks-9Metadata
Zur Langanzeige
Aufsatz
Heat Load Profiles in Industry and the Tertiary Sector: Correlation with Electricity Consumption and Ex Post Modeling
Zusammenfassung
The accurate prediction of heat load profiles with a daily resolution is required for a broad range of applications, such as potential studies, design, or predictive operation of heating systems. If the heat demand of a consumer mainly originates from (production) processes independent of the ambient temperature, existing load profile prediction methods fail. To close this gap, this study develops two ex post machine learning models for the prediction of the heat demand with a daily resolution. The selected input features are commonly available to each consumer connected to public natural gas and electricity grids or operating an energy monitoring system: Ambient temperature, weekday, electricity consumption, and heat consumption of the last seven days directly before the predicted day. The study’s database covers electricity and natural gas consumption load profiles from 82 German consumers over a period of two years. Electricity and heat consumption correlate strongly with individual patterns for many consumers. Both shallow and deep learning algorithms from the Python libraries Scikit-Learn and Keras are evaluated. A Long Short-Term Memory (LSTM) model achieves the best results (the median of R2 is 0.94). The ex post model architecture makes the model suitable for anomaly detection in energy monitoring systems.
Zitierform
In: Sustainability Volume 14 / Issue 7 (2022-03-29) eissn:2071-1050Förderhinweis
Gefördert durch den Publikationsfonds der Universität KasselZitieren
@article{doi:10.17170/kobra-202205176194,
author={Jesper, Mateo and Pag, Felix and Vajen, Klaus and Jordan, Ulrike},
title={Heat Load Profiles in Industry and the Tertiary Sector: Correlation with Electricity Consumption and Ex Post Modeling},
journal={Sustainability},
year={2022}
}
0500 Oax 0501 Text $btxt$2rdacontent 0502 Computermedien $bc$2rdacarrier 1100 2022$n2022 1500 1/eng 2050 ##0##http://hdl.handle.net/123456789/13876 3000 Jesper, Mateo 3010 Pag, Felix 3010 Vajen, Klaus 3010 Jordan, Ulrike 4000 Heat Load Profiles in Industry and the Tertiary Sector: Correlation with Electricity Consumption and Ex Post Modeling / Jesper, Mateo 4030 4060 Online-Ressource 4085 ##0##=u http://nbn-resolving.de/http://hdl.handle.net/123456789/13876=x R 4204 \$dAufsatz 4170 5550 {{Lastprofil}} 5550 {{Elektrizität}} 5550 {{Hitze}} 5550 {{Industrie}} 5550 {{Maschinelles Lernen}} 5550 {{Prognose}} 5550 {{Anomalieerkennung}} 7136 ##0##http://hdl.handle.net/123456789/13876
2022-05-30T13:42:26Z 2022-05-30T13:42:26Z 2022-03-29 doi:10.17170/kobra-202205176194 http://hdl.handle.net/123456789/13876 Gefördert durch den Publikationsfonds der Universität Kassel eng Namensnennung 4.0 International http://creativecommons.org/licenses/by/4.0/ load profiles electricity heat industry correlation machine learning LSTM prediction anomaly detection 600 Heat Load Profiles in Industry and the Tertiary Sector: Correlation with Electricity Consumption and Ex Post Modeling Aufsatz The accurate prediction of heat load profiles with a daily resolution is required for a broad range of applications, such as potential studies, design, or predictive operation of heating systems. If the heat demand of a consumer mainly originates from (production) processes independent of the ambient temperature, existing load profile prediction methods fail. To close this gap, this study develops two ex post machine learning models for the prediction of the heat demand with a daily resolution. The selected input features are commonly available to each consumer connected to public natural gas and electricity grids or operating an energy monitoring system: Ambient temperature, weekday, electricity consumption, and heat consumption of the last seven days directly before the predicted day. The study’s database covers electricity and natural gas consumption load profiles from 82 German consumers over a period of two years. Electricity and heat consumption correlate strongly with individual patterns for many consumers. Both shallow and deep learning algorithms from the Python libraries Scikit-Learn and Keras are evaluated. A Long Short-Term Memory (LSTM) model achieves the best results (the median of R2 is 0.94). The ex post model architecture makes the model suitable for anomaly detection in energy monitoring systems. open access Jesper, Mateo Pag, Felix Vajen, Klaus Jordan, Ulrike doi:10.3390/su14074033 doi:10.48662/daks-9 Lastprofil Elektrizität Hitze Industrie Maschinelles Lernen Prognose Anomalieerkennung publishedVersion eissn:2071-1050 Issue 7 Sustainability Volume 14 false 4033
Die folgenden Lizenzbestimmungen sind mit dieser Ressource verbunden: