Artificial neural network modeling for forecasting gas consumption

Gorucu F., Geris P., Gumrah F.

ENERGY SOURCES, vol.26, no.3, pp.299-307, 2004 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 26 Issue: 3
  • Publication Date: 2004
  • Doi Number: 10.1080/00908310490256626
  • Journal Name: ENERGY SOURCES
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.299-307
  • Keywords: artificial neural network, gas consumption, demand forecasting
  • Middle East Technical University Affiliated: No


This study includes an approach to evaluate and forecast gas consumption by Artificial Neural Network (ANN) modeling for the capital city of Ankara, Turkey. ANN models have been trained to perform complex functions in various fields of application including the forecasting process. The process of the study is examining the factors affecting the output and training the ANNs to decide the optimum parameters to be used in forecasting the gas consumption for the remaining days of 2002 and the year 2005. During the project, some optimistic (assuming the stable economical conditions) and the pessimistic (considering an economical crisis) scenarios are handled to get the idea for the following years' gas consumption amount in a range that will remain between these two scenarios.