Using Predictive Model for Strategic Control of Multi-reservoir System Storage Capacity

Abstract

The paper will describe the algorithm based on adaptive optimization of multi reservoir control, which the medium-term water flow predictions into the reservoirs for several months ahead repeatedly use. Hydrological prediction model was created using ANN method and values of control outflows are searching by optimization based on evolutionary algorithms optimization technique. The objective function was descripted as the sum of squares deviations between required and actual controlled water outflow from reservoirs where objective function is minimized. The algorithm of adaptive control is applied to the operation storage control of selected reservoir system, which open water reservoirs Vir and Brno are created.

The paper will describe the algorithm based on adaptive optimization of multi reservoir control, which the medium-term water flow predictions into the reservoirs for several months ahead repeatedly use. Hydrological prediction model was created using ANN method and values of control outflows are searching by optimization based on evolutionary algorithms optimization technique. The objective function was descripted as the sum of squares deviations between required and actual controlled water outflow from reservoirs where objective function is minimized. The algorithm of adaptive control is applied to the operation storage control of selected reservoir system, which open water reservoirs Vir and Brno are created.