Monthly Joint Operations for the Nakdong Multi-reservoir System in Korea

This study applies a state-of-art optimization technique, SSDP/ESP (Sampling Stochastic Dynamic Programming with Ensemble Streamflow Prediction), to derive a monthly joint operating policy for the Nakdong multi-reservoir system in Korea. A rainfall-runoff model, SSARR (Streamflow Synthesis And Reser...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Water international Jg. 32; H. 3; S. 416 - 427
Hauptverfasser: Kim, Young-Oh, Eum, Hyung-Il, Kim, Sang Uk, Lee, Kil Seong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Taylor & Francis Group 01.09.2007
Schlagworte:
ISSN:0250-8060, 1941-1707
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This study applies a state-of-art optimization technique, SSDP/ESP (Sampling Stochastic Dynamic Programming with Ensemble Streamflow Prediction), to derive a monthly joint operating policy for the Nakdong multi-reservoir system in Korea. A rainfall-runoff model, SSARR (Streamflow Synthesis And Reservoir Regulation), is linked to the SSDP/ESP model to provide ESP scenarios for runoff during the next month in the Nakdong River basin. The primary advantage of the SSDP/ESP is that it updates the derived operating policy as new ESP forecasts become available. Another SSDP model that employs historical runoff scenarios (SSDP/Hist) is also developed. The main difference between the two SSDP models is that SSDP/Hist is an off-line model whereas the SSDP/ESP is on-line. The developed operating policies are tested with a simulation model using an object-oriented simulation software, STELLA. The simulation results show that SSDP/ESP is superior to SSDP/Hist with respect to the water supply criterion, although both models perform similarly with respect to the hydroelectric energy production criterion.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ObjectType-Article-2
ObjectType-Feature-1
ISSN:0250-8060
1941-1707
DOI:10.1080/02508060708692221