Annual Electricity Consumption Forecasting with Neural Network in High Energy Consuming Industrial Sectors of Iran

Abstract

Due to various changes in electricity consumption in Iran, it is hard to model with conventional methods and makes it suitable to estimate with Artificial Neural Network. Altough this method typically has been used to forecast short term consumptions, we use Neural Network to forecast annual consumption This paper illustrates an Artificial Neural Network (ANN) approach based on supervised multi layer perceptron (MLP) network for the electrical consumption forecasting and shows that it can estimate the annual consumption with lesser error. This study shows the advantage of Neural Network methodology through analysis of variance (ANOVA). Furthermore, actual data is compared with ANN and conventional regression model.