Otimização estocástica implícita e algoritmos genéticos como suporte na operação mensal do reservatório São Gonçalo - PB.

Abstract

This paper aims at developing a model based on Implicit Stochastic Optimization (ISO) and Genetic Algorithms (GA) for deriving monthly reservoir hedging rules. The ISO technique consists of optimizing the reservoir system operation using a set of possible scenarios of inflows and using the optimal data for construction operating rules. In this research, the optimal data obtained with ISO was used by GA in order to calibrate operating rules based on the volume of available water for the current month. Synthetic scenarios inflows were obtained by the Method of Fragments. The proposed methodology was applied to the operation of São Gonçalo reservoir, which is located in the semiarid region of Paraíba State, Brazil. A mathematical function comprising the cost of water scarcity was evaluation and the results indicate that the ISO–GA rules are superior to a standard rule of operation. As a result, the generated rules may support the decision making process for the monthly operation of São Gonçalo water system.