A novel neural model to electrical load forecasting in transformers

Abstract

The paper describes a novel neural model to electrical load forecasting in transformers. The network acts as identifier of structural features to forecast process. So that output parameters can be estimated and generalized from an input parameter set. The model was trained and assessed through load data extracted from a Brazilian Electric Utility taking into account time, current, tension, active power in the three phases of the system. The results obtained in the simulations show that the developed technique can be used as an alternative tool to become more appropriate for planning of electric power systems.