 |
| Titel |
Incorporating routing into reservoir planning optimization models |
| VerfasserIn |
Nicholas Zmijewski, Anders Wörman, Andrea Bottacin-Busolin |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250113759
|
| Publikation (Nr.) |
 EGU/EGU2015-13978.pdf |
|
|
|
|
|
| Zusammenfassung |
| To achieve the best overall operation result in a reservoir network, optimization models are used. For larger reservoir networks the computational cost increases, making simplification of the hydrodynamic description necessary. In-accuracy in flow prediction can be related to an incurred sub-optimality in production planning. Flow behavior in a management optimization model is often described using a constant time-lag model. A simplified hydraulic model was used, describing the stream flow in a reservoir network for short term production planning of a case-study reservoir network (Dalälven River). In this study, the importance of incorporating hydrodynamic wave diffusion for optimized hydropower production planning in a regulated water system was examined, comparing the kinematic-wave model to the constant time-lag. The receding horizon optimization procedure was applied, emulating the data-assimilation procedure present in modern operations. Power production was shown to deviate from the planned production while considering a single time-lag, as a function of the stream description. The simplification of using a constant time-lag could be considered acceptable for streams characterized by high Peclet number. Examining the effect of the effect of the length of the decision time-step demonstrated the importance of high frequency data assimilation for streams characterized by low Peclet numbers. Further, it was shown that the variability in flow becomes more ordered as a result of management and that the Peclet number contributes to that goal. |
|
|
|
|
|
|