A Stochastic-Conceptual Reservoir Inflow Forecasting Model

by Walter Schoel, III, Walter Schoel Engineering Co, Birmingham, AL, USA,
Harold R. Henry, Walter Schoel Engineering Co, Birmingham, AL, USA,
David E. Miller, Walter Schoel Engineering Co, Birmingham, AL, USA,
David K. Hains, Walter Schoel Engineering Co, Birmingham, AL, USA,



Document Type: Proceeding Paper

Part of: Waterpower '85

Abstract:

The development of a method to improve real-time forecasting of reservoir inflows for the purpose of minimizing water loss through spillage is presented. Streamflow forecasts are computed from rainfall data by means of a stochastic-conceptual peak flow model. Computed values are updated using feedback from streamflow measurements weighted by the Kalman filter algorithm. Forecasts are then made by integrating the state-space equations over the desired lead time. The model comes on-line only when it is triggered by a threshold rainfall and the time step can be as short as 30 minutes. The model is initially evaluated on a 39 square kilometer drainage basin. It is then applied to a basin of 3,763 square kilometers.

Subject Headings: Forecasting | Stochastic processes | Reservoirs | Inflow | Streamflow | Rainfall | Mathematical models

Services: Buy this book/Buy this article

 

Return to search