Monthly Reservoir Operating Rules Generated by Implicit Stochastic Optimization

The main purpose of this paper is to develop the monthly operating rules for the multi reservoirs such as Hwacheon, Soyanggang, and Chungju located at the upstream of the Han River Basin in Korea. The monthly operating rules by an implicit stochastic optimization and a regression analysis are developed. Since it is necessary to generate the inflows having the same statistical properties of historical data for applying an implicit stochastic optimization, the synthetic inflow data of three reservoirs are generated by periodic ARMA (1,1) model. And then an implicit stochastic optimization is applied to get optimal multi reservoir operating results. These results are used to derive operating rules by a regression analysis. The regression equations among the beginning storage, the synthetic inflow and the release are obtained. Using the graphical analysis of this operating rules, the applicability to the flood season and non-flood season is accomplished.