Optimizing Multidam Releases in Large River Basins by Combining Distributed Hydrological Inflow Predictions with Rolling-Horizon Decision Making

For the purpose of real-time dam operation in large river basins, an integrated simulation and optimization system (ISOS) has been constructed by combining distributed hydrological inflow predictions with rolling horizon decision making. First, with operational quantitative precipitation forecasts (QPFs) over a forecast horizon (FH), the ISOS is applied to obtain an optimal rule for dam releases; second, with the corrected satellite precipitation, the prescribed rule is employed for the reservoir operation over a decision horizon (DH). The ISOS is applied to the Red River Basin (169,000 km(2)), to test its performance for optimizing multidam releases in the large river basin. Results show that by using the ISOS, two things were simultaneously achieved, reducing the water level at Hanoi to avoid flooding while raising water storage in Hoa Binh reservoir at the end of flood season for better hydropower generation. Through comparing reservoir performances with different FHs, the effective forecast horizon (EFH) was derived for the 3-reservoir system in the Red River. (C) 2014 American Society of Civil Engineers.