Applying Genetic Algorithms for Predicting Distribution Companies Power Contracting

The rules defined by the Decree 5.163/2004 bring incentives and penalties for Distribution Companies (Discos) when presenting their power purchase necessity declaration to the Ministry of Mines and Energy. Therefore, Independent Power Producers (IPPs) must try to predict the behavior of those agents in order to prepare themselves for the auctions in which they intend to sell. This paper analyses those rules, presenting an optimization model using Genetic Algorithms, which simulate the behavior of Discos, getting an important tool to support an IPP selling strategy.