Deriving operating rules for inter-basin water transfer projects incorporating a scenario reduction strategy

Abstract

Streamflow uncertainty is one of the main challenges in managing reservoir systems. To accurately characterize the streamflow uncertainty in a reservoir operation model, numerous streamflow scenarios should be considered, but this results in a huge computational burden for multi-objective optimization. Here, we propose a methodology for deriving the operating rules for an inter-basin water transfer project (IBWT). The methodology adopts a scenario-reduction strategy to balance accuracy against the computational burden. The proposed method first reduces a large number of streamflow scenarios to several typical scenarios and the corresponding occurrence probabilities using a simultaneous backward reduction method. A multi-objective optimal operation model is then formulated to optimize the expected values of performance indices of the IBWT system under the typical scenarios. Finally, a simulation-based optimization framework is adopted to derive the multi-objective operating rules. To test the method, the Han to Wei IBWT project in northwest China is considered as a case study. The results show that the proposed methodology produces effective reservoir operating rules with a relatively low computational burden. After the initial 1,000 streamflow scenarios have been reduced to 20, the model solution time decreases significantly, from 49.17 h to 1.05 h, while the changes in mean value (0.47%) and standard deviation (0.93%) of these scenarios are less than 1%. Compared with an operation scheme that uses only historical streamflow data, the operation performance is further improved in terms of energy production, energy consumption, and water shortage index when the streamflow uncertainty is considered.