SolverSIN: A Practical Model for Large Hydrothermal System Analysis

Almost 90% of the electrical energy in Brazil is produced by hydropower plants, but to cover national demand and to ensure production during dry periods the system operates in conjunction with a set of thermal power plants. The Brazilian hydrothermal system is completely linked together and the main purpose of operation is to minimize the cost of energy production considering a long-term horizon, generally five years. Therefore the complexity of the planning operation involves a large set of variables and many decisions have to be made to ensure maximum efficiency. Today the Brazilian National System Operator (ONS) runs a set of models to define the dispatch and the marginal cost of the energy in different parts of the country in a weekly time step. These models are very complex and are not practical to support basic information for the electrical agents in a short time. For this reason a very practical decision support system called SolverSIN was developed. This paper describes the SolverSIN model and shows its performance in dealing with a very complex hydrothermal systems analysis. The model was prepared to run with the Brazilian system considering its equivalent energy reservoirs scheme but its methodology can be adapted for any system. The main variables of the planning operation are estimated in a very short time, and different situations can be explored to conduct sensitivity analysis of the main ones. Another important SolverSIN characteristic is its ability to run different hydrological scenarios to evaluate the impacts on dispatches and energy costs. This is very important considering that the greater the safe thermal energy production of the system, the lower the electrical costs and the greater the protection to the environment due to minimization of oil consumption.