Resumo:
The introduction of competition on the electricity sector has turned the electricity price the most important variable in the energy market. On this way, efficient estimating methods of spot prices have became crucial to maximize the agent benefits. In Brazil the electricity price is formed by the marginal cost provided by an optimization software (NEWAVE). Forecasting the Marginal Cost (CMO) and its volatility has been the major problem in the Brazilian market because each simulation on NEWAVE takes about four hours of computational time (in a Pentium IV 2GHz with 512Mbytes of RAM memory). This work presents a fast and efficient model to simulate the spot price based on the Brazilian electricity market, using DOE (Design of Experiments) and ANN (Artificial Neural Networks) techniques. The work proved that the combined techniques provided promising results and should be applied to risk management and investment analysis.
