The estimation of monthly electrical energy consumption with feed forward neural networks

Abstract

In this paper we present a method for the estimation of the monthly electric energy consumption, in a house, using a feed forward neural network. The network is trained with the backpropagation algorithm and the patterns used for training are the previous months' consumption. The network was trained using real consumptions from a house, from the previous three years. We used for the network three input neurons, and on these three inputs will be placed the consumptions for the previous three months, and the network will estimate on its output the consumption for the current month.