Abstracts

The study explores the optimization of reservoir operations based on weather forecasts to maximize hydropower production, without compromising other competing objectives. Two dam sites with small to medium storage receiving unregulated inflow were selected. Short-term weather forecasts from the Global Forecast System (GFS) and dynamically downscaled by the Weather Research Forecasting (WRF) Model were used to forecast reservoir inflow. The resulting forecast inflow information was used to optimize reservoir operations aimed at maximizing hydropower with flood control, environmental flow and dam safety as key constraints. Three strategies of optimization were tested to study the effect of lead time and forecasting skill on derived benefits. Results suggest that significant hydropower benefits can be obtained by forecasting the inflow peak early and maintaining the reservoir levels accordingly. Despite reduced forecasting skill at longer lead times, hydropower maximization is found to be greater when the dam operator optimizes storage and releases earlier based on forecasted inflow. The reservoir state at the end of a flood event is found to be closer to the historical rule-curve of the dams, thereby leaving sufficient pool to handle future and unexpected flood events. The study clearly highlights the added value of weather forecasts for hydropower maximization when compared with conventional operations using the rigid rule curves for medium and small storage dams that represent 98% of US dams. Because of the significant amount of additional hydropower that is generated, the use of weather forecasts is a clear source of additional economic benefits for society that should be scaled up across the nation to further reduce dependence on fossil-fuel based energy production in the long run.Advisors/Committee Members: Hossain, Faisal (advisor).