📆 Embargo bis 2024-08-31
Öffnen
Datum
2023Schlagwort
510 Mathematik ZeitreihenanalyseKapazitätsplanungOrdnungsreduktionHauptkomponentenanalyseSpeicherungMetadata
Zur Langanzeige
Konferenzveröffentlichung
Identifying critical demand periods in capacity planning for networks including storage
Zusammenfassung
We consider a capacity planning problem for networks including storage. Given a graph and a time series of demands and supplies, we seek for integer link and storage capacities that permit a single commodity flow with valid storage in- and outtakes over all time steps. This problem arises, for example, in power systems planning, where storage can be used to buffer peaks of varying supplies and demands. For typical time series spanning a full year at hourly resolution, this leads to huge optimization models. To reduce the model size, time series aggregation is commonly used. The time horizon is sliced into fixed size periods, e.g. days or weeks, a small set of representative periods is chosen via clustering methods, and a much smaller model involving only the chosen periods is solved. Representative periods, however, typically do not contain the situations with the most extreme demands and supplies and the strongest effects on storage. In this paper, we show how to identify such critical periods using principal component analysis (PCA) and convex hull computations and we compare the quality and solution time of the reduced models to the original ones for benchmark instances derived from power systems planning.
Zusätzliche Informationen
Erschienen in: Grothe, O., Nickel, S., Rebennack, S., Stein, O. (eds) Operations Research Proceedings 2022. OR 2022. Lecture Notes in Operations Research. Springer, Cham. https://doi.org/10.1007/978-3-031-24907-5_27Zitieren
@inproceedings{doi:10.17170/kobra-202311108996,
author={Bley, Andreas and Hahn, Philipp},
title={Identifying critical demand periods in capacity planning for networks including storage},
year={2023}
}
0500 Oax 0501 Text $btxt$2rdacontent 0502 Computermedien $bc$2rdacarrier 1100 2023$n2023 1500 1/eng 2050 ##0##http://hdl.handle.net/123456789/15179 3000 Bley, Andreas 3010 Hahn, Philipp 4000 Identifying critical demand periods in capacity planning for networks including storage / Bley, Andreas 4030 4060 Online-Ressource 4085 ##0##=u http://nbn-resolving.de/http://hdl.handle.net/123456789/15179=x R 4204 \$dKonferenzveröffentlichung 4170 5550 {{Zeitreihenanalyse}} 5550 {{Kapazitätsplanung}} 5550 {{Ordnungsreduktion}} 5550 {{Hauptkomponentenanalyse}} 5550 {{Speicherung}} 7136 ##0##http://hdl.handle.net/123456789/15179
2023-11-15T08:27:37Z 2023 doi:10.17170/kobra-202311108996 http://hdl.handle.net/123456789/15179 Erschienen in: Grothe, O., Nickel, S., Rebennack, S., Stein, O. (eds) Operations Research Proceedings 2022. OR 2022. Lecture Notes in Operations Research. Springer, Cham. https://doi.org/10.1007/978-3-031-24907-5_27 eng Urheberrechtlich geschützt https://rightsstatements.org/page/InC/1.0/ time series analysis capacity planning model reduction 510 Identifying critical demand periods in capacity planning for networks including storage Konferenzveröffentlichung We consider a capacity planning problem for networks including storage. Given a graph and a time series of demands and supplies, we seek for integer link and storage capacities that permit a single commodity flow with valid storage in- and outtakes over all time steps. This problem arises, for example, in power systems planning, where storage can be used to buffer peaks of varying supplies and demands. For typical time series spanning a full year at hourly resolution, this leads to huge optimization models. To reduce the model size, time series aggregation is commonly used. The time horizon is sliced into fixed size periods, e.g. days or weeks, a small set of representative periods is chosen via clustering methods, and a much smaller model involving only the chosen periods is solved. Representative periods, however, typically do not contain the situations with the most extreme demands and supplies and the strongest effects on storage. In this paper, we show how to identify such critical periods using principal component analysis (PCA) and convex hull computations and we compare the quality and solution time of the reduced models to the original ones for benchmark instances derived from power systems planning. restricted access Bley, Andreas Hahn, Philipp 7 Seiten Zeitreihenanalyse Kapazitätsplanung Ordnungsreduktion Hauptkomponentenanalyse Speicherung submittedVersion 2022-09-06 - 2022-09-09 Karlsruhe 2024-08-31 2024-08-31 false Operations Research Society of Germany
Die folgenden Lizenzbestimmungen sind mit dieser Ressource verbunden:
:Urheberrechtlich geschützt