Details
| Original language | English |
|---|---|
| Article number | 3 |
| Number of pages | 26 |
| Journal | Energy Informatics |
| Volume | 7 |
| Publication status | Published - 22 Jan 2024 |
Abstract
This paper introduces an univariate application-independent set of load profiles or time series derived from real-world energy system data. The generation involved a two-step process: manifolding the initial dataset through signal processors to increase diversity and heterogeneity, followed by a declustering process that removes data redundancy. The study employed common feature engineering and machine learning techniques: the time series are transformed into a normalized feature space, followed by a dimensionality reduction via hierarchical clustering, and optimization. The resulting dataset is uniformly distributed across multiple feature space dimensions while retaining typical time and frequency domain characteristics inherent in energy system time series. This data serves various purposes, including algorithm testing, uncovering functional relationships between time series features and system performance, and training machine learning models. Two case studies demonstrate the claims: one focused on the suitability of hybrid energy storage systems and the other on quantifying the onsite hydrogen supply cost in green hydrogen production sites. The declustering algorithm, although a bys study, shows promise for further scientific exploration. The data and source code are openly accessible, providing a robust platform for future comparative studies. This work also offers smaller subsets for computationally intensive research. Data and source code can be found at https://github.com/s-guenther/estss and https://zenodo.org/records/10213145 .
Keywords
- Energy systems, Feature engineering, Load profiles, Machine learning, Statistical analysis, Systems modeling, Time series, Time series analysis, Time series features
ASJC Scopus subject areas
- Computer Science(all)
- Information Systems
- Energy(all)
- Energy Engineering and Power Technology
- Computer Science(all)
- Computer Networks and Communications
Sustainable Development Goals
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Energy Informatics, Vol. 7, 3, 22.01.2024.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - ESTSS—energy system time series suite
T2 - a declustered, application-independent, semi-artificial load profile benchmark set
AU - Günther, Sebastian
AU - Brandt, Jonathan
AU - Bensmann, Astrid
AU - Hanke-Rauschenbach, Richard
N1 - Funding Information: SG developed, realized, and implemented the presented concept and method and designed and provided the first application example. JB designed and provided the second application example. The manuscript writing and visualization is in line with the specified contributions. The presented work was supported and supervised by AB and RHR. All authors read and approved the final manuscript.
PY - 2024/1/22
Y1 - 2024/1/22
N2 - This paper introduces an univariate application-independent set of load profiles or time series derived from real-world energy system data. The generation involved a two-step process: manifolding the initial dataset through signal processors to increase diversity and heterogeneity, followed by a declustering process that removes data redundancy. The study employed common feature engineering and machine learning techniques: the time series are transformed into a normalized feature space, followed by a dimensionality reduction via hierarchical clustering, and optimization. The resulting dataset is uniformly distributed across multiple feature space dimensions while retaining typical time and frequency domain characteristics inherent in energy system time series. This data serves various purposes, including algorithm testing, uncovering functional relationships between time series features and system performance, and training machine learning models. Two case studies demonstrate the claims: one focused on the suitability of hybrid energy storage systems and the other on quantifying the onsite hydrogen supply cost in green hydrogen production sites. The declustering algorithm, although a bys study, shows promise for further scientific exploration. The data and source code are openly accessible, providing a robust platform for future comparative studies. This work also offers smaller subsets for computationally intensive research. Data and source code can be found at https://github.com/s-guenther/estss and https://zenodo.org/records/10213145 .
AB - This paper introduces an univariate application-independent set of load profiles or time series derived from real-world energy system data. The generation involved a two-step process: manifolding the initial dataset through signal processors to increase diversity and heterogeneity, followed by a declustering process that removes data redundancy. The study employed common feature engineering and machine learning techniques: the time series are transformed into a normalized feature space, followed by a dimensionality reduction via hierarchical clustering, and optimization. The resulting dataset is uniformly distributed across multiple feature space dimensions while retaining typical time and frequency domain characteristics inherent in energy system time series. This data serves various purposes, including algorithm testing, uncovering functional relationships between time series features and system performance, and training machine learning models. Two case studies demonstrate the claims: one focused on the suitability of hybrid energy storage systems and the other on quantifying the onsite hydrogen supply cost in green hydrogen production sites. The declustering algorithm, although a bys study, shows promise for further scientific exploration. The data and source code are openly accessible, providing a robust platform for future comparative studies. This work also offers smaller subsets for computationally intensive research. Data and source code can be found at https://github.com/s-guenther/estss and https://zenodo.org/records/10213145 .
KW - Energy systems
KW - Feature engineering
KW - Load profiles
KW - Machine learning
KW - Statistical analysis
KW - Systems modeling
KW - Time series
KW - Time series analysis
KW - Time series features
UR - http://www.scopus.com/inward/record.url?scp=85182858574&partnerID=8YFLogxK
U2 - 10.1186/s42162-024-00304-8
DO - 10.1186/s42162-024-00304-8
M3 - Article
AN - SCOPUS:85182858574
VL - 7
JO - Energy Informatics
JF - Energy Informatics
M1 - 3
ER -