Intelligent energy management of electrical power systems with distributed feeding on the basis of forecasts of demand and generation

Abstract

The intelligent energy management system described in this paper allows optimum short-term scheduling, especially of power systems with distributed feeding, considering energetic, ecological and economical criteria. With it, the demand of the loads is covered with the required reliability of supply regarding all technical constraints. The energy management work with evolutionary methods based on forecasts of the probability distribution of the power generated by the regenerative sources of energy as well as the demand of the load. For the prediction task the use of artificial neural networks is recommended. Precise rules for the design of the prediction system, which have been determined by systematic investigations, are presented. Even though forecasts of renewable energy generation show higher prediction errors than forecasting the load, it is possible to include renewable energy sources into power scheduling.