Sensitivity Based Feature Selection for Recurrent Neural Network Applied to Forecasting of Heating Gas Consumption

  • Martin Macas
  • Fiorella Lauro
  • Fabio Moretti
  • Stefano Pizzuti
  • Mauro Annunziato
  • Alessandro Fonti
  • Gabriele Comodi
  • Andrea Giantomassi
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 299)

Abstract

The paper demonstrates the importance of feature selection for recurrent neural network applied to problem of one hour ahead forecasting of gas consumption for office building heating. Although the accuracy of the forecasting is similar for both the feed-forward and the recurrent network, the removal of features leads to accuracy reduction much earlier for the feed-forward network. The recurrent network can perform well even with 50% of features. This brings significant benefits in scenarios, where the neural network is used as a blackbox model of building consumption, which is called by an optimizer that minimizes the consumption. The reduction of input dimensionality leads to reduction of costs related to measurement equipment, but also costs related to data transfer.

Keywords

forecasting consumption gas heating neural networks feature selection 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Villar, J.R., González, S., Sedano, J., Corchado, E., Puigpinós, L., de Ciurana, J.: Meta-heuristic improvements applied for steel sheet incremental cold shaping. Memetic Computing 4(4), 249–261 (2012)CrossRefGoogle Scholar
  2. 2.
    Calvo-Rolle, J.L., Corchado, E.: A bio-inspired knowledge system for improving combined cycle plant control tuning. Neurocomputing 126(0), 95–105 (2014)CrossRefGoogle Scholar
  3. 3.
    Macaš, M., Lhotská, L.: Wrapper feature selection significantly improves nonlinear prediction of electricity spot prices. In: 2013 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 1171–1174 (2013)Google Scholar
  4. 4.
    Sarak, H., Satman, A.: The degree-day method to estimate the residential heating natural gas consumption in turkey: a case study. Energy 28(9), 929–939 (2003)CrossRefGoogle Scholar
  5. 5.
    Kalogirou, S.A.: Applications of artificial neural-networks for energy systems. Applied Energy 67(1-2), 17–35 (2000)CrossRefGoogle Scholar
  6. 6.
    Khotanzad, A., Elragal, H., Lu, T.L.: Combination of artificial neural-network forecasters for prediction of natural gas consumption. IEEE Transactions on Neural Networks 11(2), 464–473 (2000)CrossRefGoogle Scholar
  7. 7.
    Kalogirou, S.A., Bojic, M.: Artificial neural networks for the prediction of the energy consumption of a passive solar building. Energy 25(5), 479–491 (2000)CrossRefGoogle Scholar
  8. 8.
    Schijndel, A.W.M.V.: HAMLab: Integrated heat air and moisture modeling and simulation. PhD thesis, Eindhoven: Technische Universiteit (2007)Google Scholar
  9. 9.
    de Wit, M.: HAMBASE: Heat, Air and Moisture Model for Building And Systems Evaluation. Technische Universiteit Eindhoven, Faculteit Bouwkunde (2006)Google Scholar
  10. 10.
    Elman, J.L.: Finding structure in time. Cognitive Science 14(2), 179–211 (1990)CrossRefGoogle Scholar
  11. 11.
    Mathworks: Neural Network Toolbox for Matlab ver. 2012b (2012)Google Scholar
  12. 12.
    Moody, J.E.: The effective number of parameters: An analysis of generalization and regularization in nonlinear learning systems. In: NIPS, pp. 847–854. Morgan Kaufmann (1991)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Martin Macas
    • 1
  • Fiorella Lauro
    • 2
  • Fabio Moretti
    • 2
  • Stefano Pizzuti
    • 2
  • Mauro Annunziato
    • 2
  • Alessandro Fonti
    • 3
  • Gabriele Comodi
    • 3
  • Andrea Giantomassi
    • 4
  1. 1.Department of CyberneticsCzech Technical University in PraguePragueCzech Republic
  2. 2.Unità Tecnica Tecnologie Avanzate per l’Energia e l’Industria, ENEA (Italian National Agency for New Technologies, Energy and Sustainable Economic Development)Cassacia Research CenterRomaItaly
  3. 3.Dipartimento di Ingegneria Industriale e Scienze MatematicheUniversità Politecnica delle MarcheAnconaItaly
  4. 4.Dipartimento di Ingegneria dell’InformazioneUniversità Politecnica delle MarcheAnconaItaly

Personalised recommendations