Details

Autor(en) / Beteiligte

• Uysal, Gokcen
• Akkol, Bulut
• Topcu, M. Irem
• Sensoy, Aynur
• Schwanenberg, Dirk

Titel

Comparison of Different Reservoir Models for Short Term Operation of Flood Management

Ist Teil von

• Procedia engineering, 2016, Vol.154, p.1385-1392

Ort / Verlag

Elsevier Ltd

Erscheinungsjahr

2016

Link zum Volltext

Quelle

EZB Electronic Journals Library

Beschreibungen/Notizen

• Short-term operation of reservoir systems is challenging due to conflicting objectives and constraints as well as the need for taking robust decisions in real-time. This study compares simulation and optimization based decision support techniques in application to the mitigation of flood events. Two models are employed in the study to support the operators’ decisions: (1) HEC-ResSim of USACE as a representative of a simulation-based approach, and (2) the RTC-Tools package of Deltares with an optimization approach. The methods are applied to a complex flood management problem at Yuvacik Dam, Turkey. A worst case scenario of an extreme flood event is used to evaluate the pros and cons of the models including a high initial water level exceeding the flood control pool. Objectives of the control are the maximization of water supply benefits, i.e. a full reservoir, at the end of the event as well as flood mitigation in the downstream river reach. In the first method, a script-based rule is defined in the GUI with user access to its parameters. The refinement of the reservoir operation is conducted manually by trial and error. Secondly, an optimization approach using Model Predictive Control (MPC) is used in combination with the IPOPT optimizer. The advantage of HEC-ResSim is the detailed representation of the gate management on the level of the individual gates. However, the implementation of the total release is partially up to user interaction and not necessarily optimal. RTC-Tools provides optimum releases on the project level, but not on the level of individual gates. Both approaches consider system constraints. Furthermore, an advantage of the optimization approach is its extension to probabilistic ensemble forecasts to consider forecast uncertainty in the decision by use stochastic optimization.