Reservoir Operation by the Use of an Expert System

See related content

by Tiao J. Chang, Ohio Univ, Athens, United States,
David Moore, Ohio Univ, Athens, United States,

Document Type: Proceeding Paper

Part of: Hydraulic Engineering

Abstract: Most reservoirs, designed for the purpose of flood control, face operational problems during periods of drought. This study uses five indicators, streamflow, precipitation, temperature, groundwater and lake levels to define drought. An optimization model for water releases from flood-control reservoirs for drought management (Chang and Kleopa, 1991) was applied to a four-reservoir system in the Scioto River Basin in Ohio for the period from April 1 to September 30, 1988. Daily optimal water releases for 70%, 80%, 90%, and 95% levels of drought severity were obtained by the optimization model. The expert system of this study was developed using rules and heuristic knowledge from expert opinions, reservoir control manuals, and the optimization model. Based on the drought evaluation, the developed Daily Drought Management Expert System was used to recommend water releases for the operation of four reservoirs involved. These results were compared with their counterparts of the optimization model and the actual operation of reservoirs.

Subject Headings: Reservoirs | Hydrologic models | Optimization models | Expert systems | Droughts | Water supply systems | North America | United States | Ohio

Services: Buy this book/Buy this article


Return to search