Reability of the Forecasting of the Monthly Demand of Electric Energy with Neural Networks.

Abstract

Electric energy demand forecasting represents a fundamental tool to plan the activities of the companies that generate and distribute it. So a good prediction of its demand will provide an invaluable information to plan the production and purchase policies of these companies. This demand may be seen as a temporal series when these data are conveniently arranged. In this way the prediction of a future value may be performed studying the past ones. Neural networks have proved to be a very powerful tool to do this. They are mathematical structures that mimic that of the nervous system of living beings and are used extensively for system identification and prediction of their future evolution. In this work a neural network is presented to predict the evolution of the monthly demand of electric consumption. A Feedforward Multilayer Perceptron (MLP) with three hidden layers has been used as neural model with Backpropagation as learning strategy. The consumption data have been normalized to avoid their rising trend. Several procedures have been tested in order to find out those performing the best. Errors smaller than 5% have been obtained in most of the predictions.